Guide to Research Techniques in Neuroscience Second Edition Matt Carter Jennifer Shieh AMSTERDAM • BOSTON • HEIDELBERG • LONDON NEW YORK • OXFORD • PARIS • SAN DIEGO SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO Academic Press is an imprint of Elsevier Publisher: Mica Haley Acquisition Editor: Natalie Farra Editorial Project Manager: Kathy Padilla Project Managers: Karen East and Kirsty Halterman Designer: Greg Harris Typeset by TNQ Books and Journals www.tnq.co.in Printed and bound in the United States of America Academic Press is an imprint of Elsevier 125 London Wall, London EC2Y 5AS, UK 525 B Street, Suite 1800, San Diego, CA 92101-4495, USA 225 Wyman Street, Waltham, MA 02451, USA The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK Second Edition Copyright © 2015, 2010 Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. ISBN: 978-0-12-800511-8 British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress For information on all Academic Press publications visit our website at http://store.elsevier.com/ xiii Foreword to the Second Edition When Guide to Research Techniques in Neuroscience was published in 2010, the field was already vibrant with an abundance of experimental approaches and research methods for studying the brain. Amazingly, the ingenuity of neuroscientists to develop more diverse and more powerful techniques in the past 5 years has already rendered the first edition somewhat out of date. New technologies allow neuroscientists to measure and manipulate neural activity with unprecedented spatial and temporal precision, to create brain–machine interfaces in a variety of animals, including humans, and even to turn the brains of animals completely transparent for remarkably detailed anatomical characterization. The enthusiasm for advancing techniques in neuroscience has never been stronger. In 2013, US President Barack Obama launched the BRAIN (Brain Research through Advancing Innovative Neurotechnologies) Initiative, a bold enterprise spanning multiple federal agencies and private sector partners to accelerate the development and application of new, sophisticated tools to understand the brain. We have enormous potential to continue to build on technology in imaging and electrophysiology to better examine the collective actions of thousands or even tens of thousands of neurons; what we have seen to date is only the tip of the iceberg. Additionally, the ability to manipulate neural activity in highly specific populations of neurons is allowing for remarkable advances in basic neuroscience research in animal models as well as potentially therapeutic applications in humans. Like the US, a number of other countries are also pursuing exciting neuroscience initiatives. We are at a unique point in the history of neuroscience—in the next several years, we have an opportunity to amplify our previous achievements by combining exciting new technologies with important research questions. In the face of these exciting new techniques, classical approaches to brain research still have an important role. New techniques are not replacing traditional methods so much as they are standing on their solid foundation. Excellent studies continue to rely on behavioral, histological, and electrophysiological methods that have been the hallmark of neuroscience research for dozens of years. Therefore, this book is even more important now than when it was first published. Future neuroscience studies will depend on a common framework of techniques, and this guide provides that solid foundation. As before, this book serves as an excellent resource for anyone from the beginning graduate student to the seasoned faculty member who could use an efficient, cogent guidebook xiv  Foreword to the Second Edition by their side. Here’s to the amazing advances we have made in the past and the exciting possibilities we have for the future! William T. Newsome, PhD Professor, Department of Neurobiology Stanford University Investigator, Howard Hughes Medical Institute xv Foreword to the First Edition Understanding how the brain works is perhaps the greatest challenge facing contemporary science. How mental life is rooted in the biology of the brain is intrinsically fascinating, and researchers from remarkably diverse scientific backgrounds are being drawn increasingly to the field of neuroscience. Psychologists, molecular biologists, physiologists, physicists, engineers, computer scientists (and more) are all contributing importantly to the richness of modern neuroscience research. The brain remains the most complex and enigmatic entity in the known universe, and all relevant scientific techniques and perspectives must ultimately be brought to bear to solve its mysteries. The quickening pace and diversity of neuroscience research pose substantial challenges, however, for new students and established researchers seeking to enter the field for the first time. The techniques used to measure and manipulate the nervous system are dizzying in their scope and complexity, and some of the important new approaches are accompanied by clouds of jargon that are all but impenetrable to outsiders. Fortunately for those of us who are perplexed, a concise, no-nonsense guidebook has now arrived. The Guide to Research Techniques in Neuroscience presents the central experimental techniques in contemporary neuroscience in a highly readable form. It is all there—from brain imaging to electrophysiology to microscopy to transgenic technologies. Matt Carter and Jennifer Shieh take us on a very practical cook’s tour of research techniques, all the while providing concise overviews of exactly where and how particular techniques fit into the grand scheme of the basic questions being asked of the nervous system. As a neurophysiologist, I read quickly through the electrophysiology and brain imaging chapters, noting with appreciation the efficient presentation of basic information that should be in the intellectual repertoire of any aspiring neuroscientist. But I read avidly the chapters on cloning, gene delivery, and use of transgenic animals—the exotica (to me!) of molecular biological approaches made clean, simple, and pleasingly devoid of specialist vocabulary. I rather suspect that most neuroscience professionals will have the same experience reading this book, encountering high-yield veins of useful information interspersed with much they already know. For the student or postdoc just entering the field, however, the majority of the book will be very high yield. This book is an essential resource for anyone—from the beginning graduate student to the seasoned faculty member—who could use an efficient ­guidebook at his or her side while reading papers written by any of their xvi  Foreword to the First Edition colleagues, irrespective of the level of analysis, from small molecules to spike trains. It may not render the mysterious raster plots of systems lectures immediately clear, but it will orient you to the key types of information you want to take home from that talk. The book is a little gem. Read it, rely on it, and pass it on to others. William T. Newsome, PhD Professor, Department of Neurobiology Stanford University Investigator, Howard Hughes Medical Institute xvii Preface to the Second Edition It has been only 5 years since the publication of the first edition of Guide to Research Techniques in Neuroscience, yet the rapid pace of biological discovery, enabled by finer tools and more powerful techniques, has already made a second edition necessary. In this new edition, we have added recently developed methods as well as older methods that have been more widely adopted among neuroscience labs. Many of these methods are used to stimulate or inhibit neural activity, and therefore we decided to add an entirely new chapter, Manipulating Neural Activity. In addition to including new information, we have made this latest edition a full-color book and redesigned most of the nearly 200 figures to increase detail and clarity. In the past 5 years, we have also come a long way in our professional lives, leaving graduate school, and attaining full-time scientific positions. Matt completed a postdoc in the lab of Richard Palmiter at the University of Washington and is now an assistant professor of biology at Williams College in Williamstown, MA. Jennifer advanced to become an AAAS Science and Technology Policy Fellow in Washington, DC and is now a program officer at the National Institutes of Health. While we look ahead to our futures, we are excited that “Understanding Techniques in Neuroscience,” the course we started at Stanford almost a decade ago, lives on among the current graduate students, with versions being taught at many other colleges and universities throughout the country. Thank you to the Elsevier team for their support and patience: Natalie Farra, for seeing that it was time to write the second edition, and Kathy Padilla, for making sure that our project stayed on track. Obviously, our amazing contributors to the first edition still earn our gratitude. For updates to this second edition, we especially thank Moriah Thomason, Matt Kaufman, Astra Bryant, and Matthew Goodman for their valuable input. Jennifer would also like to thank the AAAS Science and Technology Policy Fellow community and its vibrant Neuropolicy group for expansive discussions, helpful advice, and allowing her to stay connected to her first love—the brain. Finally, nothing would be possible without the love, encouragement, and support from our significant others. Jennifer offers a big thank you to Steve, while Matt owes his wife Ali big time. And 4-year-old Liam and baby Hadley as well! Thank you! The next 5 years are bound to be even more exciting, especially because of new funding efforts such as the BRAIN (Brain Research through Advancing xviii  Preface to the Second Edition Innovative Neurotechnologies) Initiative to spur the advancement of new technologies to better understand the nervous system at multiple levels. If the pace of the next 5 years matches that of the preceding 5 years, we are in for a fun ride! Matt Carter, Williamstown, MA Jennifer C. Shieh, Washington, DC1 September 2014 1. Contributions provided by Jennifer C. Shieh to this work were written in a personal capacity and do not necessarily represent the opinions or endorsement of the National Institutes of Health, Department of Human Health Services, or the Federal Government. xix Preface to the First Edition The story of why we decided to write this book can essentially be summarized as follows: We tried to find a book just like it, could not find one anywhere, and ultimately decided the book would be so useful that we would write it ourselves. When we were advanced undergraduates and beginning graduate students, we found the number of techniques used in neuroscience research daunting. In no other biological field are students required to learn biochemistry, molecular biology, genetics, electrophysiology, microscopy, histology, behavioral assays, and human imaging technologies, not to mention the basic biological and physical properties that allow these methodologies to work. It is true that each neuroscientist only practices a handful of these general techniques. However, all neuroscientists must learn to understand and appreciate each other’s research, even if they do not perform the specific experiments themselves. For example, consider the study of the auditory system. Auditory research can be performed in humans using functional magnetic resonance imaging (fMRI) and other methods of whole-brain imaging. Auditory research can also be performed in animal models, from standard laboratory animals (mice, rats, flies) to animals that have unique auditory properties that make them particularly well-suited for auditory research (bats, barn owls). These animal models can be studied using in vivo electrophysiology while they are anesthetized or while performing behavioral tasks. Additionally, the cells responsible for hearing and interpreting auditory stimuli can be examined in vitro, for example, by recording electrical signals from isolated hair cells. Histological and biochemical techniques can highlight the specific ion channels and other proteins that provide auditory cells their physiological properties. Genetic methods can be used to investigate the role of molecules that make hearing possible or cause auditory deficits. Thus, anyone interested in the auditory system, or any other specific subfield of neuroscience, must be prepared to read about and analyze the contributions of studies using a wide spectrum of techniques. As students new to neuroscience research, we were overwhelmed with the expectation that we understand all of these methods. Moreover, we were surprised by the number of advanced students who seemed to pretend that they understood all these techniques, when they really only understood the specific techniques they used in their own studies. Sometimes, these students and even some postdocs and faculty would confess to us, “I should have learned that long ago but I just don’t really understand it.” We would hear scientists who use electrophysiology and fMRI techniques sometimes say, “Who cares about a blob on xx  Preface to the First Edition a gel?” Likewise, we would hear geneticists and molecular biologists confess “I don’t know what the squiggly lines in electrophysiology studies mean,” or “I don’t know why fMRI experiments are so hard—can’t you just put someone in a scanner and turn on the machine?” In short, it became obvious that there should be a guide to neuroscience techniques. So we tried to find one. Finding a guide to neuroscience techniques turned out to be more difficult than we imagined. During our first trips to the annual Society for Neuroscience conference, we tried to find a simple techniques book among the hundreds of books being promoted by different publishers. Amazingly, there was no single book that aimed to explain the breadth of neuroscience techniques at a basic level. There were detailed books written about single techniques (“PCR Methods” or “Making Transgenic Animals”) and recipe books filled with protocols, but no book addressed techniques as a subject themselves. Even the almighty Internet did not provide thorough insight into standard techniques. On the Internet, you can find protocols for doing a western blot and learn that it is a technique that measures the amount of protein in a sample. But you cannot find information about other ways a scientist can measure protein expression, or why you might choose a western blot over these other methods. There are no example figures from the literature, nor are there suggestions of what to look for in these figures. Our dream book would contain a description of western blots, a comparison of western blots with similar methods, and an example of what the data looked like in the literature. No such book existed, and we wanted one. With the encouragement of faculty members and students at Stanford, we decided to write this book. We also decided that the information would make an excellent seminar course. Along with a talented student colleague of ours, Saul Villeda, we created a 9-week techniques course called “Understanding Techniques in Neuroscience.” This class surveys neuroscience techniques and provides examples from the literature. We wrote a 110+ page course reader to accompany the class and quickly decided that this text should also form the basis of a future book that could exist independently of the class. The class was amazingly well received. The first year, our class had 15 students. Through word of mouth, the class doubled in size the next year. The third year, over 100 students attended our lectures, including undergraduates, graduate students, postdocs, and occasionally a faculty member. We believe the number of people taking the course demonstrates that neuroscientists are interested in learning about each other’s methods, as well as the need for formalized education about neuroscience techniques. Therefore, we decided to adapt our course reader into this book. It is the book that we so desperately wanted when we were beginning neuroscience students, and we are happy that it is now possible to find at Society for Neuroscience meetings. We learned a tremendous amount of information while researching and writing this book, and we hope that you find it helpful for your own education. Preface to the First Edition  xxi We would like to thank the editorial staff at Elsevier/Academic Press for making this book a reality. From Susan Lee, who initially saw the potential of our course reader, to Melissa Turner and Mica Haley, who managed to get the final product with only minor hair pulling. This book would not have been possible without the guidance and help of the incredible faculty at Stanford. In particular, Bill Newsome provided tremendous encouragement and support throughout the writing process. As promised, he will receive a fine bottle of Woodford Reserve upon the successful publication of this book. We must also acknowledge the understanding of our research advisors Luis de Lecea and Sue McConnell. As promised, they will receive our theses without further delay. Finally, we thank Mehrdad Shamloo and Mehrdad Faizi at Stanford’s Behavioral and Functional Neuroscience Laboratory and Jon Mulholland of the Cell Sciences Imaging Facility for their expert advice. Our teaching partner Saul Villeda researched many of the methods that are detailed in this book and helped prepare the original course reader. Saul is an amazing teacher and it has been our pleasure to work with him over the years, both as colleagues and friends. We also received substantial guidance and support from the Stanford neuroscience community. So many students and postdocs spent their valuable time and energy to help us understand techniques, edit our chapters, and provide encouragement as we finished the book. Special thanks to Raag Airan, Bjorn Brembs, Brittany Burrows, Laurie Burns, Kelsey Clark, Emily Drabant, Lydia-Marie Joubert, Rachel Kalmar, Jocelyn Krey, Dino Leone, Jon Mulholland, Scott Owen, Georgia Panagiotakos, Chris Potter, Elizabeth Race, Victoria Rafalski, Andreas Rauschecker, Magali Rowan, Rory Sayres, Bob Schafer, Jana Schaich Borg, John Wilson, and Sandra Wilson, who provided substantial suggestions to improve the content and text. We completed this book with our neuroscience colleagues in mind, and we hope that all future neuroscience students find this book helpful to their education and throughout their careers. Finally, we thank our significant others, Vishal Srivastava and Alison Cross. Thank you SO much for giving us time, love, and encouragement. Matt Carter Jennifer C. Shieh Stanford, CA, 2009 xxiii Introduction The human mind has been studied for thousands of years, but the human brain, as well as the brains of other species, has only been studied for about a century. Only 150 years ago, the ability to study the nervous systems of humans and other animals was limited to direct observation and by examining the effects of brain damage in people and other organisms. With the advent of histology came the ability to visualize and differentiate between neurons based on morphology. The great neuroscientist Santiago Ramón y Cajal used a method called Golgi staining to visualize the morphology and architecture of neurons and their circuits throughout the brain. Cajal used the Golgi stain to propel the field of neuroscience into its modern state. Indeed, in the history of neuroscience, each leap forward in knowledge has been based on a leap forward in techniques and technology. Just as Ramón y Cajal used Golgi staining to greatly advance our understanding of the structure of the nervous system, scientists throughout the twentieth century used more and more advanced techniques to contribute to our understanding of the function of the nervous system: Eccles, Hodgkin, and Huxley used intracellular recording technology to investigate the ionic basis of membrane potentials; Hubel and Wiesel used extracellular recording technology to investigate how information is processed and recorded in the visual system; Neher and Sakmann used patch-clamp technology to investigate the physiology of single-ion channels. In the latter half of the twentieth century, the explosion of molecular biology techniques and methods of genetically manipulating model organisms allowed neuroscientists to study individual genes, proteins, and cell types. Technology has progressed so far in the past 100 years that the Golgi stain itself seems to have been reinvented through powerful technologies (Chapter 12) that allow investigators to turn specific neurons different colors to further investigate the structure and connectivity of the nervous system. In fact, technologies are developing at such a rapid pace that in 2013, U.S. President Barack Obama announced the BRAIN (Brain Research through Advancing Innovative Neurotechnologies) Initiative to double federal funding for new techniques and technologies to accelerate our understanding of the nervous system. The modern neuroscientist now has hundreds of techniques that can be used to answer specific scientific questions. This book contains 15 chapters that provide an overview of the most commonly used techniques. Although there are dozens of techniques that seem very different at first glance, many of them xxiv Introduction attempt to study the nervous system in the same way. For example, transcranial magnetic stimulation (Chapter 1), physical lesions (Chapter 8), optogenetic inhibition (Chapter 8), and genetic knockdown or knockouts (Chapter 13) are all attempts to test the effect of a loss-of-function of some aspect of the nervous system on another aspect of the nervous system. For each level of investigation (whole brains to individual genes), research strategies can be similar in concept even if the techniques used are very different in practice. LEVELS OF INVESTIGATION Something immediately obvious to all students of neuroscience is that the nervous system is exceptionally complicated and can be examined at multiple levels of investigation. The basic functional unit of the nervous system is the neuron. The human brain is composed of approximately 100 billion neurons that are connected into circuits via approximately 100 trillion synapses. Neural circuits are organized into anatomical structures and larger networks of neurons that can integrate information across modalities from many different parts of the brain. These networks process sensory information from the external and internal environment and provide the neural basis of cognition—learning, memory, perception, decision-making, emotion, and other higher-order processes. The final output of the nervous system is a behavior composed of a coordinated motor action. This behavior can be either extremely simple, such as a motor reflex, or incredibly complicated, such as dancing, typing, or playing a musical instrument. Behavior is usually defined not just by what an organism does, but what it chooses to do. Therefore, except in rare circumstances of lesion or disease, cognition and behavior are inseparably linked, and in animals other than humans, behavior is used as a readout of animal cognition. Just as one can start with a neuron and scale up toward circuits, cognition, and behavior, a scientist can also scale down and examine the components that make up a neuron. A neuron is itself defined as having a cell body (soma), axon, and dendrites. These neuronal components contain subcellular specializations that make the neuron unique among other cell types. Specialized organelles in a neuron, such as vesicles containing neurotransmitters, provide the cell with the ability to signal to other neurons. Specialized cytoskeletal elements allow a neural process to extend great distances throughout the brain and body. Several proteins provide neurons with their intercellular signaling abilities and physiological characteristics. For example, biosynthetic enzymes produce neurotransmitters, while other proteins serve as receptors for these signaling molecules. One of the most important types of proteins in the nervous system form ion channels, the transmembrane structures that allow neurons to become electrically active under certain conditions. All of these proteins are the products of genes, the functional units of an organism’s genome. The human genome contains approximately 30,000 genes, with each neural subtype expressing its own subset of these genes. Introduction  xxv The complexity of the nervous system is awesome in scope. It is amazing that a mutation in a single gene, such as a gene that codes for a transmembrane ion channel, can produce effects that alter the electrical properties of a neuron, in turn altering the normal firing patterns of a neural circuit and thus causing an abnormal behavior. A neuroscientist can approach the study of the nervous system through any of these levels of organization. The 15 chapters of this book provide a guide to the types of experiments that can be performed at each level. However, irrespective of technique, the basic scientific approach one can use to study the nervous system is consistent from level to level, whether the subject is human cognition or axon guidance in cell culture. METHODS OF STUDYING THE NERVOUS SYSTEM There are four general methods of studying the nervous system: 1. Examining case studies—identifying interesting events that have occurred naturally and using these events to develop hypotheses that can be tested in future experiments 2. Screens—performing unbiased searches for anatomical structures, neurons, proteins, or genes that could play a role in a subject of interest 3. Description—using techniques that allow a scientist to observe the nervous system without manipulating any variables 4. Manipulation—testing hypotheses by determining the effect of an independent variable on a dependent variable Each of these four methods is described in more detail below. Examining Case Studies A case study is an example of an event that happened to a subject (most often a human or group of humans) that demonstrates an important role for an aspect of the nervous system. The circumstances surrounding the event are usually nonrepeatable and cannot be precisely recreated in a laboratory setting. Such demonstrations are, therefore, not true experiments in that no variables are deliberately controlled by a scientist. However, these events can often reveal substantial information about an aspect of neural function that was previously unknown. For example, consider the case of Phineas Gage, a railroad worker who was involved in an accident in 1848 that caused an iron rod to pass through his skull. The rod entered the left side of his face, passed just behind his left eye, and exited through the top of his head, completely lesioning his frontal lobes. This is an amazing event, not only because Gage survived (and lived for another 12 years), but also because it informed scientists about the function of the frontal lobe of the brain. The event allowed investigators to retrospectively ask the question: “What is the effect of removing the frontal lobe on consciousness and behavior?” According to Gage’s friends, family, and coworkers, he was “no xxvi Introduction longer Gage.” He retained the ability to learn, remember, sense, and perceive his environment, to execute motor functions, and to live a fairly normal life, but it seemed to people who knew him that his personality had changed completely. After the accident, Gage was described as impolite, erratic, unreliable, and offensive to others. He wound up losing his job at the railroad, not because of any physical or mental incapacity, but because he was simply so disrespectful and offensive that people could not stand to work with him. This case study is not a true experiment; no scientist decided to test the removal of the frontal lobe on personality. But the incident, and others like it, allows neuroscientists to form hypotheses based on naturally occurring events. Because of Gage’s story, neuroscientists could hypothesize about the contribution of the frontal lobe to human personality. Future experiments could test these hypotheses on animal models (that share certain human personality traits) and even attempt to identify neural circuits that contribute to human behaviors. Screens A screen is a method that allows an investigator to determine which nuclei, neurons, or genes/proteins may be involved in a particular biological process. Such experiments are not necessarily driven by a hypothesis, but the experiment identifies candidates that can form the basis for future hypothesis-driven research. For example, a neuroscientist who wants to identify genes involved in body weight regulation may compare gene expression profiles in central feeding centers of the brain in both fed and starved animals: genes expressed in starved animals relative to fed animals may be important for generating the motivation to eat. Screens can be performed at multiple levels of investigation. When a cognitive neuroscientist places a human subject into an fMRI scanner and examines which brain areas show increased activation in response to a specific stimulus or task, the scientist is essentially performing a screen of different brain regions. When a fly geneticist examines thousands of mutagenized flies for deficiency in a behavioral task, the scientist is attempting to identify genes necessary for that behavior to occur. Such genes can then be tested in future experiments. Thus, screens can be performed to identify interesting molecules or entire brain regions. Description Descriptive science is the act of observing properties of the nervous system without manipulation. This type of research is usually the first step in acquiring knowledge about a newly discovered gene, protein, or neuronal subtype. For example, an investigator could describe the sequence of a gene and where in the brain the gene is expressed. Likewise, in the case of proteins, an investigator could describe the amino acid sequence of the protein and where in the brain the protein is expressed. Neurons can be described in terms of what genes/proteins they express, their morphology, how many neurons make up a population of neurons, and their electrophysiological properties. Introduction  xxvii It is important to note that just because a study may be descriptive does not mean that it is necessarily easier than other types of experiments. Observation and description form the foundation for understanding the relationship between structure and function, as well as providing insight about what elements to manipulate in future experiments. Thus, descriptive neuroscience plays just as important a role in modern research as when Ramón y Cajal observed the structure of neurons 100 years ago. Manipulation Manipulating an aspect of the nervous system or environment and examining the effect this perturbation has on a separate aspect of the nervous system is the only way to test a hypothesis in neuroscience. A manipulation experiment tests the effect of X onY. The variable that is manipulated, X, is referred to as the independent variable. The part of the system that is measured, Y, is referred to as the dependent variable. Two of the most common types of manipulation experiments are loss-offunction and gain-of-function experiments. In a loss-of-function (also called “necessity”) experiment, a part of the nervous system is diminished or removed in an attempt to determine if it is necessary for a certain process to occur. The following questions are all loss-of-function questions: l Is a normal copy of the gene Fezf2 necessary for the proper development of the cerebral cortex? l Is the receptor for the hypocretin neuropeptide necessary for normal sleep/ wake transitions in mammals? l Is electrical activity in the medial geniculate nucleus required for auditorydriven spikes in the auditory cortex? l Can human patients with damage to the cerebellar vermis perform as well as healthy controls on a verbal-memory task? In all of these experiments, an aspect of the nervous system is partially or totally disrupted, whether it is a gene, a protein, electrical activity, or an entire brain structure. The independent variable is the loss of the structure, and the dependent variable is the effect on another aspect of the nervous system. Sometimes, a good follow-up for a loss-of-function experiment is a rescue experiment in which the aspect of the nervous system that is lost is deliberately returned. For example, if it is found that a certain line of fruit flies lacks a gene that is necessary for proper development of an eye, the specific gene can be reintroduced to the flies in transgenic experiments to see if the functional gene can rescue the aberrant phenotype. In a gain-of-function (also called “sufficiency”) experiment, an aspect of the nervous system is increased relative to normal. This may include an increased expression of a gene or protein, an increase in electrical activity in a brain region, or an increase of a particular neurotransmitter in the extracellular medium. The aspect of the nervous system that is increased is the independent xxviii Introduction variable, and the effect on another part of the nervous system is the dependent variable. The following questions are all gain-of-function questions: l Can an increase in the gene TrpA1 cause mice to be hypersensitive to cold temperatures? l Can an intracerebroventricular injection of neuropeptide Y cause an increase in feeding behavior in rats? l Can electrical microstimulation of the lateral geniculate nucleus increase spike frequencies over time in area V4 of the visual cortex? l Can stimulation of the motor cortex using transcranial magnetic stimulation induce motor behaviors? In both loss-of-function and gain-of-function experiments, it is important not to overstate the conclusions of the experiments. For example, consider a loss-offunction experiment in which a mouse that lacks a gene is unresponsive to painful stimuli. The investigator could conclude that this gene is necessary for proper performance on an assay for pain detection. However, an inappropriate conclusion would be that this gene regulates pain detection. Perhaps this gene is responsible for normal development of the spinal cord and the mouse lacks all peripheral sensation. Alternatively, this gene may code for a protein that is necessary for normal development of the thalamus; if improper development of the part of the thalamus that receives information about painful stimuli causes the stimuli not to reach the somatosensory cortex, the animal will not perform normally on the pain-detection task. Careful controls are necessary to reach appropriate conclusions. UNDERSTANDING TECHNIQUES IN NEUROSCIENCE We hope that these 15 chapters serve as a useful guide to studying the nervous system. There are two important points to keep in mind while reading these chapters: 1. All of the techniques described throughout this book depend on the principles just mentioned. For each level of investigation that a scientist may choose to study, the same four general categories of methods exist: examining case studies, screens, description, and manipulation. The same principles that are used to study brain activity in awake human subjects are used to study genes and proteins in tissue samples. The methods may vary, but the principles remain the same. 2. It is also important to remember that techniques and methods should never be the guiding force behind doing experiments in research. Ideally, experiments should be performed to answer an interesting question, not the other way around. A technique should not be used for its own sake but because it is the best technique available to answer a particular research question. We hope this book answers your questions about what techniques are available in modern neuroscience research, but more importantly, we hope these techniques answer your specific research questions as well! 1 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00001-0 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 1 Whole-Brain Imaging After Reading This Chapter, You Should be Able to: l Compare the relative strengths and limitations of different structural and functional brain imaging techniques l Explain the physical and physiological basis of (MRI/fMRI) technology l Describe the design of a functional brain imaging experiment: formulating a hypothesis, choosing task paradigms, performing the experiment, acquiring and analyzing data, and constructing figures Techniques covered: l Structural techniques: cerebral angiography, computerized tomography (CT), MRI, and diffusion MR imaging l Functional techniques: fMRI, positron emission tomography (PET), single-proton emission computerized tomography (SPECT), electroencephalography (EEG), magnetoencephalography (MEG), and optical imaging l Techniques used to investigate the necessity and sufficiency of a specific brain region for a cognitive function: transcranial magnetic stimulation (TMS), ultrasonic neuromodulation (USNM), and case studies Modern brain imaging technology can seem like magic. The ability to produce detailed images of the human brain without physically penetrating the skull is a technological marvel that has saved thousands of lives and allowed scientists to study the structure of the brain throughout development, disease, and aging. Furthermore, the ability to image neural activity in the brain during cognition has provided scientists the opportunity to correlate activity in distinct brain regions with specific mental operations, a truly remarkable achievement. Colorful figures depicting activity in the human brain dazzle scientists and nonscientists alike. Of course, brain imaging technology is not magic. The technology that produces detailed images of the brain depends on complex physics, expensive equipment, and skilled technicians. As with all scientific experiments, brain imaging studies must be well designed, the data accurately analyzed, and the results carefully interpreted. The purpose of this chapter is to explain the ostensible magic of whole-brain imaging technology and provide insight into how experiments are designed and interpreted. 2  Guide to Research Techniques in Neuroscience Whole-brain imaging technology can essentially be divided into two categories: structural and functional. Structural techniques produce images of the anatomical architecture of the brain, whereas functional techniques produce images of the physiological processes that underscore neural activity. This chapter will survey both classifications of techniques and describe how they can be used in modern neuroscience research. We focus on MRI and fMRI technology in humans due to the widespread use of these techniques in the neuroscience literature. After reviewing these techniques, we will survey the essential components of a functional imaging experiment: forming hypotheses, choosing appropriate task paradigms, performing experiments, acquiring and analyzing data, and producing figures for publication. STRUCTURAL BRAIN IMAGING TECHNIQUES Structural brain imaging techniques are used to resolve the anatomy of the brain in a living subject without physically penetrating the skull. These techniques can be used in combination with functional brain imaging techniques to correlate neural activity in specific anatomical regions with behavioral or cognitive functions. Structural techniques can also be used to measure anatomical changes that occur over time, such as a decrease in brain mass that occurs with aging or the progression of disease. Most often, these techniques are used in clinical neuroscience and neurology to diagnose diseases such as tumors and vascular disorders. Brain imaging technologies take advantage of the different biochemical compositions of brain regions and use these differences to form the basis of an image (Figure 1.1). Neural cell bodies contain many biomolecules, including proteins and carbohydrates. Axons and fiber tracts are relatively fatty due to the insulation provided by myelin. Cerebrospinal fluid (CSF) in the ventricles and surrounding the brain is essentially a saline solution. The microanatomy and composition of individual neural structures cause distinct regions of the brain to appear different when examined with the naked eye. For example, when examining slices of a brain, tissue mostly composed of cell bodies appears gray compared with other areas and therefore is referred to as “gray matter.” Brain tissue mostly composed of axons and fiber tracts appears white and therefore is referred to as “white matter.” Often, the most informative structural images of the brain show the contrast between gray and white matter. Thus, the ultimate goal of structural imaging technologies is to differentiate between proteins and carbohydrates (cell bodies), fat (axon tracts), and salt water (CSF), as this contrast reveals the most information about brain architecture. Until the early 1970s, there was no technology that could differentiate between these substances within the brain. Conventional X-ray technology is essentially useless for this purpose. During an X-ray procedure, an X-ray beam is passed through an object and then onto a photographic plate (Figure 1.2A). Each of the molecules through which the beam passes absorbs some of the radiation, so only the unabsorbed portions of the beam reach the photographic plate. X-ray Whole-Brain Imaging Chapter | 1  3 photography is, therefore, only effective in characterizing internal structures that differ substantially from their surroundings in the degree to which they absorb X-rays, such as bone in flesh (Figure 1.2B). By the time an X-ray beam passes through the relatively soft consistency of the brain (not to mention the relatively hard consistency of the skull!), little information about individual brain structures can be discerned (Figure 1.2C). Therefore, in the 1960s and 1970s, there was strong motivation to discover better ways of imaging the brain. The techniques described in the following sections represent 30–40 years worth of innovation in technology ultimately designed to show contrast within the soft tissue of the brain. Cerebral Angiography A cerebral angiogram is an enhanced X-ray that uses dyes to make up for the relatively poor soft-tissue contrast of conventional X-rays. A radio-opaque dye that absorbs X-rays better than surrounding tissue is injected into an artery that delivers blood to the brain. This substance heightens the contrast between the cerebral circulatory system and surrounding brain tissue during an X-ray (Figure 1.3A). Thus, the most prominent aspect of the central nervous system imaged in a cerebral angiogram is the brain vasculature. Angiograms can show vascular damage and indicate the presence of a tumor or aneurysm, an abnormal ballooning of a portion of an artery due to weakening of a blood vessel wall (Figure 1.3B). Axon terminals Cell body Dendrites Axon Myelin (A) Gray matter White matter Ventricles (with CSF) (B) FIGURE 1.1  The composition of the brain. (A) A microscopic view of a neuron. Each neuron is composed of a cell body, dendrites, and an axonal process. The cell bodies and dendrites are rich in proteins and carbohydrates. Axons are surrounded by a myelin insulation made of fats. (B) A macroscopic view of the brain. Gray matter is rich in cell bodies and, therefore, in proteins and carbohydrates. White matter is composed of axon tracts and is, therefore, rich in fatty myelin. Cerebrospinal fluid (CSF) in the ventricles is a saline solution. To produce an image, brain imaging technologies must differentiate among proteins/carbohydrates, fats, and saline. 4  Guide to Research Techniques in Neuroscience Computerized Tomography Another method that improves upon conventional X-ray technology to image the brain and body is computerized tomography (the “CT scan”—sometimes also called computerized axial tomography, or “CAT scan”).A patient or subject lies with his or her head positioned in the center of a cylinder (Figure 1.4A). A narrow beam of X-rays is aimed through the person’s head and hits a detector on the opposite side. The beam and detector rotate in a slow arc, taking many individual X-ray scans at the same axial plane. As mentioned previously, a single X-ray scan would supply little information about the structure of the brain. However, multiple scans taken from different angles can be combined to provide information about small differences in radiodensity between different brain (A) (B) (C) FIGURE 1.2  Standard X-ray technology alone cannot produce detailed images of the brain. (A) During an X-ray procedure, an X-ray beam is passed through an object and onto a photographic plate. Only the unabsorbed portions of the beam reach the plate, creating an image. (B) The contrast between the soft consistency of skin and muscle compared with the hard consistency of bone is sufficient to form an X-ray image. However, (C) the contrast between the soft consistency of different tissues within the nervous system is insufficient to form an X-ray image of the brain. Whole-Brain Imaging Chapter | 1  5 structures. These data are entered into a computer algorithm that constructs a composite picture based on the X-ray scans from all the different angles. With this information, a slice, or tomogram (tomo means cut or slice), can be generated. Typically, 8–10 images of axial brain sections are obtained for analysis. FIGURE 1.3  Cerebral angiography. (A) A cerebral angiogram depicting the vasculature of the right hemisphere of the brain. (B)A cerebral angiogram indicating the presence of a brain aneurysm. (A, B: Reprinted from Nolte, J. and Angevine, J. B. (2007). The Human Brain in Photographs and Diagrams, 3rd ed. with permission from Mosby/Elsevier: Philadelphia. Courtesy of (A) Dr Joachim F. Seeger and (B) Dr Raymond F. Carmody.) Ventricles White matterGray matter Rotating X-ray source X-ray detector CT scanner (A) (B) FIGURE 1.4  Computerized tomography (CT). (A) During an imaging session, a narrow beam of X-rays is slowly rotated around a subject’s head to hit a detector on the opposite side. Signals from around the head are combined using a computer algorithm that constructs a composite picture based on the various X-ray angles. (B) A modern CT scan can distinguish between gray and white matter, differentiate ventricles, and depict structures with a spatial resolution of millimeters. (B: Reprinted from Nolte, J. and Angevine, J. B. (2007). The Human Brain in Photographs and Diagrams, 3rd ed. with permission from Mosby/Elsevier: Philadelphia. Courtesy of Dr Raymond F. Carmody.) 6  Guide to Research Techniques in Neuroscience The quality of a CT scan depends on the width of the X-ray beam (narrower is better), the sensitivity of the X-ray detector, and the ability of the computer to construct an image from the data. Modern CT scans can distinguish between gray matter, white matter, and ventricles with a spatial resolution of millimeters (Figure 1.4B). They are particularly useful for identifying fluid boundaries in human patients, such as when blood collects on the brain surface in a hematoma, or detecting hard objects in soft tissue, such as a tumor or calcification. CT scanners are faster, cheaper to operate, and less prone to motion artifacts than MRI scanners; therefore, they tend to be the first tool used to diagnose a patient with a potential head injury. Magnetic Resonance Imaging MRI technology produces highly detailed structural images of the brain and body. The resolution of a modern MR image is far superior to a CT image, typically less than a millimeter. Thus, MRI technology has largely superseded computerized tomography as the method of imaging the brain in both clinical and research settings. The technology that makes MRI possible is complex, but necessary to understand to fully appreciate MR images of the brain. The Electromagnetic Basis of MRI Technology As the name suggests, magnetic resonance imaging takes advantage of the magnetic properties of biological tissue to produce an image. Most often, MRI utilizes the magnetic properties of hydrogen protons, as they are highly abundant in the fluids and organic compounds of the brain and body. The main function of an MRI scanner is to artificially excite these hydrogen protons and then measure their relaxation properties over time. An MRI scanner is composed of a long tube-like chamber, where a subject is placed, surrounded by electric coils hidden within the MRI apparatus (Figure 1.5). As current passes through the coils in a clockwise rotation, a magnetic field is produced longitudinal to the subject, in the direction of the feet to head. The purpose of putting the subject in a magnetic field is to affect the hydrogen protons in the subject’s tissues. Protons can be thought of as miniature magnets: they spin about an axis, and their spinning positive charge induces a tiny magnetic field (Figure 1.6A). Normally, the magnetic fields of individual protons orient in random directions (Figure 1.6B). However, when a subject is placed inside the strong magnetic field of an MRI machine, the magnetic fields of individual protons align in the axis of the field. Some protons align parallel to the magnetic field, toward the subject’s head, while others align in the opposite, antiparallel direction, toward the subject’s feet (Figure 1.6C). It is slightly more energetically favorable for the protons to orient in the parallel rather than antiparallel direction, so there is a net magnetic field vector in the parallel direction (Figure 1.6D). Whole-Brain Imaging Chapter | 1  7 There is one more important detail to know about the protons in a magnetic field: they do not simply stay stationary, aligned parallel (or antiparallel) to the magnetic field lines. Instead, they precess around their axis, spinning like a top (Figure 1.7). The frequency with which they spin is dependent on the strength of the external magnetic field (generated by the MRI machine). The larger the external magnetic field, the higher the precession frequency. The strength of a magnetic field is measured in Tesla (T). Most conventional MRI machines create external magnetic fields at 1.5–3 T. Newer, more powerful MRI scanners create fields at 7 T. Higher magnetic field strengths increase the signal-to-noise ratio and give higher contrast and spatial resolution. However, these powerful magnets are also more expensive and more likely to cause physiological discomfort in subjects, such as nausea or dizziness. The MRI scanner contains electric coils that produce a static magnetic field in the long axis and perpendicular axes of the subject A subject in an MRI scanner rests comfortably inside a narrow chamber A radiofrequency coil (not shown) is placed around the head to perturb the static magnetic fields and generate the MRI signal FIGURE 1.5  A human subject in a magnetic resonance imaging (MRI) scanner. The subject lies inside a chamber surrounded by electric coils. Current passing through these coils induces a strong magnetic field. 8  Guide to Research Techniques in Neuroscience Generating an Image Prior to the beginning of an imaging session, there is a net magnetic field ­vector in the parallel direction, longitudinal to the subject’s body (Figure 1.8A). To collect data for an MR image, the subject is briefly exposed to pulses of electromagnetic energy, referred to as radiofrequency (RF) pulses (Figure 1.8B). Applying an RF pulse that has the same frequency as the proton precession frequency causes two effects: (1) some protons in the parallel phase absorb energy, reverse polarity to the antiparallel phase, and therefore decrease the net (A) + (C) (D) (B) North South South North FIGURE 1.6  Protons align with the magnetic field. (A) As a proton spins around its axis, the rotating positive charge induces a magnetic field. Therefore, a proton can be thought of as a tiny magnet. (B) Before an external magnetic field is applied, protons orient in random directions. (C) However, in the presence of a strong magnetic field, protons align in either a parallel or antiparallel direction. (D) In a magnetic resonance imaging scanner, protons aligned in the parallel direction are directed toward the subject’s head, while protons aligned in the antiparallel direction are directed toward the subject’s feet. Slightly more protons are aligned in the parallel direction, so there is a net magnetic force toward the subject’s head. Whole-Brain Imaging Chapter | 1  9 longitudinal magnetization; (2) some protons get in sync and start to precess in phase. Their vectors now add up in a direction that is transverse to the external magnetic field (perpendicular to the subject’s body). Thus, a new transversal magnetization is established (Figure 1.8C). After the RF pulse is switched off, the high-energy nuclei begin to relax and realign (Figure 1.8D). Eventually, the longitudinal magnetization increases to its North South FIGURE 1.7  Protons precess around an axis in a pattern that can resemble a spinning top. Transversal Longitudinal RF pulse FIGURE 1.8  A radiofrequency (RF) pulse alters the net magnetic field. (A) Before the RF pulse is applied, there is a net magnetic force in the longitudinal direction toward the subject’s head. (B, C) After the RF pulse is applied, the net longitudinal field decreases, as some protons reorient to the antiparallel orientation. Also, a new net magnetic field is created in the transverse direction. (D) After the RF pulse is switched off, the longitudinal field increases to normal and the transverse field decreases until the magnetic field returns to the state before the RF pulse was applied. 10  Guide to Research Techniques in Neuroscience original value, while the transversal magnetization decreases to zero. The time (in milliseconds) required for a certain percentage of the protons to realign in the longitudinal direction is termed T1. The transversal relaxation time is termed T2. Both the information acquired in the longitudinal direction (T1) as well as the information acquired in the transversal direction (T2) are measured by an antenna inside the MRI scanner. Recall that the goal of brain imaging technology is to resolve differences in tissues made of proteins and carbohydrates, fat, and salt water. What makes MRI technology useful for producing images of the brain is that these substances exhibit different values for both T1 and T2 relative to each other. For any given point in time during the relaxation phase, the T1 white matter signal is stronger than that of gray matter, and the gray matter signal is stronger than that of CSF (Figure 1.9A). These differences in signal intensity are exactly opposite for a T2 measurement: the CSF signal is strongest, followed by gray matter and then white matter (Figure 1.9B). By examining the contrast in signal from different points in space, it is possible to differentiate between different substances and form an image. An image of the brain formed from T1 data is referred to as a T1-weighted image, while an image formed from T2 data is said to be T2-weighted. These Time T1SignalIntensity White matter CSF Gray matter Time T2SignalIntensity White matter Gray matter CSF (A) (C) (B) (D) FIGURE 1.9  Different substances exhibit different T1 and T2 time constants. (A) For any given longitudinal magnetization relaxation time (T1), the white matter signal intensity will be greater than that of cerebrospinal fluid (CSF). (B) In contrast, for any given transversal magnetization relaxation time (T2), the CSF signal intensity is greater than that for white matter. (C) Therefore, white matter will be bright and CSF dark on a T1-weighted image, while (D) white matter will be dark and CSF bright on a T2-weighted image. (C, D: Courtesy of Dr Rory Sayres.) Whole-Brain Imaging Chapter | 1  11 images appear different from one another because of the differences in signal intensity among substances in T1- vs. T2-weighted images (Figures 1.9C and D). As a rule of thumb, if the CSF is black, you are looking at a T1-weighted image, as CSF exhibits the lowest relative T1 signal intensity. If, on the other hand, the CSF appears bright white, you are looking at a T2-weighted image, as CSF exhibits the highest relative T2 signal intensity. In the literature, most anatomical data are presented as T1-weighted images, as these images usually show ­better contrast between brain structures. However, sometimes T2-weighted images are optimal—for example, lesions of white matter that occur due to the rupturing of blood vessels are more easily detectable on a T2-weighted image. Therefore, T2-weighted images may be most beneficial when examining patients following trauma or stroke. Selecting a Slice of Brain to Image How does an investigator select a slice to examine? Recall that an RF pulse will only excite protons with the same precession frequency as the frequency of the pulse. The precession frequency varies with the strength of the external magnetic field, so to select a single slice of the brain to image, an additional magnetic field is applied to the external magnetic field at a gradient (Figure 1.10A). Because the field strength is not equal at all planes in the longitudinal direction, an RF pulse at a specific frequency will only affect the protons, and thus the signal, at a specific plane. This is the slice of the brain that will be presented as a two-dimensional image. To measure signal from individual points within the slice, two additional magnetic gradients are applied in the other two axes (Figure 1.10B and C). Therefore, each point in space will have its own unique magnetic signature. Each point has a specific volume and is termed a voxel, a threedimensional version of a pixel that represents a cubic volume of brain space. The resolution of each voxel is determined by the values of the gradients applied to the subject. With greater magnetic field strengths, more dramatic gradients can be established. This is why a 3 T MRI scanner can produce an image with higher spatial resolution than a 1.5 T scanner. MRI has a number of features that have made it an especially valuable research tool for both diagnostic and research studies. l MRI is entirely noninvasive, as no substances need to be injected into human subjects. Sometimes, a contrast agent is injected that enhances the visibility of water-rich regions. However, for most brain imaging procedures, no additional contrast is necessary. l Slices of the brain can be obtained from any angle. CT scanners can only image slices of the brain based on the axis of rotation of the X-ray emitter and detector within the apparatus. l By varying the gradient and RF pulse parameters, MRI scanners can be used to generate images that highlight certain brain regions and provide contrast between specific kinds of neural tissue. l No X-ray radiation is applied to the subject in the scanner. 12  Guide to Research Techniques in Neuroscience Even with these relative advantages, MRI has not completely replaced CT imaging. CT imaging is better for visualizing bony or calcified structures in the head, and also remains the imaging technique of choice for subjects who might not be able to enter a strong magnetic field (for instance, due to a pacemaker) or subjects with claustrophobia. Additionally, CT scanners have much lower operating costs than MRI scanners, making their use relatively less expensive. Diffusion MRI Diffusion MRI is an application of MRI that is used to examine the structure of axon fiber tracts in the brain. Traditional MRI images present white matter as a homogeneous structure (as in Figures 1.9C and D). In reality, fiber tracts originate from various sources, radiate in different orientations, and travel to distinct regions of the brain. Diffusion MRI provides investigators the opportunity to Magnetic field strength Proton precession frequency RF pulse (A) Magnetic field gradient in anterior–posterior direction Magnetic field gradient in medial–lateral direction (B) (C) FIGURE 1.10  How a slice is selected in magnetic resonance imaging. (A) To image a brain slice, an external magnetic field is applied at a gradient. An RF pulse will only excite protons in a particular slice, as the pulse can only excite atoms with the same precession frequency. To measure the signal intensity for each point within the slice, two additional gradients are applied, one in the medial–lateral direction (B) and one in the anterior–posterior direction (C). Whole-Brain Imaging Chapter | 1  13 visualize these different white matter pathways and study the complexities of axonal architecture (Figure 1.11). The term diffusion refers to the fact that water molecules, like all other molecules, randomly move through a medium over time. For example, in a glass of water, any particular molecule of water will move around randomly in any direction, limited only by the walls of the container. This type of diffusion is referred to as isotropic—diffusion along all directions. However, water molecules within brain tissue diffuse quite differently due to the physical environment of the brain. They tend to diffuse most rapidly along parallel bundles of fibers with coherent orientations. This type of diffusion is referred to as anisotropic—diffusion that is not equal in all directions but instead tends to move randomly along a single axis. Diffusion MRI is able to measure the anisotropic diffusion of water along fiber bundles, highlighting the connectivity between brain regions. The physics and mathematics behind this technology are too complex for this text, but the goal is to use MRI technology to analyze the magnitude and direction of the diffusion of water molecules for each voxel of tissue, thus creating a threedimensional image of fiber tracts. There are various kinds of diffusion MRI methods, the most commonly used called diffusion tensor imaging. Diffusion MRI can provide information about which areas of the brain are connected, but it is not able to determine the direction of this connectivity (which endpoint is the source and which endpoint is the target). However, it is possible to combine diffusion MRI with functional MRI, allowing researchers to identify temporal correlations between activity in distinct regions of the brain to make conclusions about functional connectivity between brain structures. FIGURE 1.11  Diffusion tensor imaging. An example of an image of white matter tracts in the corpus callosum. (Reprinted from Van Hecke, W. et al. (2008). On the construction of an intersubject diffusion tensor magnetic resonance atlas of the healthy human brain. Neuroimage 43(1), 69–80, with permission from Elsevier.) 14  Guide to Research Techniques in Neuroscience FUNCTIONAL BRAIN IMAGING TECHNIQUES Functional brain imaging techniques are used to measure neural activity in the central nervous system without physically penetrating the skull. The ultimate goal of these techniques is to determine which neural structures are active during certain mental operations. Although these techniques cannot demonstrate that a brain region causes certain actions or is the specific structure that regulates a cognitive process, they can provide other useful information. For example, functional brain imaging techniques can show that activity in specific brain regions is often correlated with a particular stimulus, emotional state, or behavioral task. They can show that information is represented in certain places within the brain, that information may be present in the brain without being consciously perceived, and that diseased brains may process information abnormally compared to healthy brains. Importantly, these techniques have allowed neuroscientists the opportunity to study the neural basis of cognition, emotion, sensation, and behavior in humans, a feat that cannot be achieved by most other techniques in this book. Functional Magnetic Resonance Imaging Functional magnetic resonance imaging uses the same physical principles as MRI to produce high-resolution representations of neural activity over time. As with MRI technology, fMRI detects signals from excited hydrogen protons in a magnetic field. Recall that different substances within the brain exhibit different T1 and T2 values following stimulation with an RF pulse. In a structural imaging experiment, these values are used to measure differences in signal intensity between gray matter, white matter, and CSF. Functional magnetic resonance imaging technology takes advantage of the signal intensity of another substance within the brain: hemoglobin, the protein in the blood that carries oxygen to cells. On a T2 image, oxyhemoglobin (the oxygen-carrying form of hemoglobin) has a relatively stronger magnetic resonance signal than deoxyhemoglobin (the oxygen-depleted form of hemoglobin). Thus, fMRI allows an investigator the opportunity to examine changes in the oxygenation state of hemoglobin over time. The ability to examine changes in oxygen metabolism over time in the brain is useful because it serves as an indirect measure of neural activity. Active neurons will consume more oxygen relative to when they are at rest. Initially, this activity decreases the levels of oxyhemoglobin and increases the levels of deoxyhemoglobin. However, within seconds, the brain microvasculature responds to this local oxygen depletion by increasing the flow of oxygen-rich blood to the active area. This phenomenon is referred to as the blood oxygen level-­dependent (BOLD) effect (Figure 1.12), and it forms the basis of the fMRI signal. The precise physiological nature of the relationship between neural activity and the BOLD effect is the subject of much current research. Whole-Brain Imaging Chapter | 1  15 Functional magnetic resonance imaging depends on T2-weighted images because the contrast in signal intensity between deoxyhemoglobin and oxyhemoglobin is greatest on these kinds of images. In a typical experiment, a T2-weighted image of the brain is obtained prior to the presentation of a stimulus. After a stimulus is presented or the subject performs a task, additional T2-weighted images are obtained. As neurons become active due to the stimulus or task, the BOLD effect causes a relative increase in oxyhemoglobin to the microenvironment, and thus an increase in T2-weighted signal. The difference in T2 signal between the prestimulus and poststimulus time points and is measured color coded for visualization of the data. These data are typically superimposed over a T1-weighted image that more clearly depicts the underlying anatomy of the brain. The end result is a colorful statistical representation of neural activity superimposed on an anatomical image of the brain—a depiction of the BOLD response over time (Figure 1.13). It is important to understand that fMRI data in a two-dimensional figure are actually four-dimensional: the x, y, z coordinate planes for each voxel in space, as well as the fourth dimension of time (the before and after time points during which the stimulus is presented). Although fMRI technology provides a powerful tool to study the neural basis of cognition, there are important limitations. One of the biggest challenges in fMRI research is that the actual T2 signal change for a given voxel of brain space, before and after BOLD changes, can be as low as 0.2%. This change is very difficult to detect, especially given that the noise of the system can be as high as 0.3–0.4%. Therefore, fMRI stimuli must be repeated several times for a single subject, and a series of statistical tests must confirm the presence of a Hemoglobin Oxygen (A) (B) FIGURE 1.12  The blood oxygen level-dependent (BOLD) effect. (A) A group of neurons at rest are supplied by blood from capillaries. (B) When these neurons become active, they increase their metabolic demand for oxygen. The microvasculature responds by supplying more oxygenrich blood to the local area. The relative increase in the oxygenated form of hemoglobin causes an increase in the T2 signal. 16  Guide to Research Techniques in Neuroscience (A) (C) (E) (F) (B) (D) FIGURE 1.13  Example of a functional magnetic resonance imaging (fMRI) experiment. This experiment examines the difference in blood oxygen level-dependent (BOLD) signal intensity between subjects exposed to light or no light. (A) First, the investigators select a slice to examine. (B) A T1-weighted image is obtained that provides structural data. T2-weighted images are used for the actual data collection in the conditions of (C) light and (D) no light. (E) The data analysis compares BOLD signal intensity between (C) and (D). This result is superimposed on the structural data to produce an image (F) suitable for publication. (A–F: Reprinted with permission of Springer Science + Business Media and Jens Frahm from Windhorst, U. and Johansson, H. (eds.), 1999. Modern Techniques in Neuroscience Research, Chapter 38: Magnetic Resonance Imaging of Human Brain Function, p. 1064, Figure 5.) Whole-Brain Imaging Chapter | 1  17 reproducible signal. Another limitation is the temporal delay: it can take 6–10 s after the presentation of a stimulus for oxygenated blood to flow to an active region, so there can be a long time delay between the stimulus or task and the measurement of neural activity. The temporal resolution, the ability to resolve neural activity into discrete events, is about 6 to 10 s, relatively poor compared to other techniques such as electroencephalography or magnetoencephalography (described below). Finally, fMRI cannot identify the neurochemistry of neural events, such as which neurotransmitters or neuromodulators mediate a change in neural activity. Inferences about the neurochemical makeup of neural activity must be based on prior knowledge of brain anatomy or other forms of whole-brain imaging, such as positron emission tomography (PET) (see the ­following section). Even with these limitations, fMRI remains a powerful technique for correlating neural activity with specific mental operations. A well-designed experiment can reveal much about the human brain and animal brains (Box 1.1), allowing investigators to noninvasively examine the physiology of cognition. See the last part of this chapter for a thorough discussion of the design of a functional imaging experiment. Positron Emission Tomography PET provides a representation of neural activity but no information about brain structure. This technology was developed in the 1970s and 1980s as a novel method of functional imaging, but has largely been superseded by fMRI technology for most cognitive experiments. In a PET experiment, an unstable positron-emitting isotope is injected into a subject’s carotid artery (a neck artery that feeds the ipsilateral cerebral hemisphere). As the isotope decays, it emits a positron, an antimatter counterpart of an electron. When a positron comes into contact with an electron, an annihilation event occurs, resulting in a pair of gamma photons that move in opposite directions (Figure 1.14A). These photons pass through the body and can be measured by a gamma-detecting device that circles the subject’s head. The detector identifies a pair of gamma photons that arrive at opposite sides of the subject’s head at the same time (within a few nanoseconds; Figure 1.14B). As the detector rotates around the subject’s head, these signals can be used to derive the source of the annihilation events within the subject (Figure 1.14C). A PET experiment can utilize a variety of positron-emitting isotopes. One of the most commonly used is fluorodeoxyglucose (FDG), a radioactive form of glucose. As metabolically active neurons require an increase in the uptake of glucose from the blood, the presence of FDG can be used as an indirect marker of neural activity. In addition to FDG, radioactive water can be injected into the brain’s circulatory system. Because there is an increase in blood flow to active areas of the brain, the PET scan will indicate the areas where blood flow is increased during activity. 18  Guide to Research Techniques in Neuroscience A particularly useful aspect of PET imaging is the ability to use positronemitting isotopes that can bind to specific receptors in the brain. For example, a radioactive ligand that binds to serotonin receptors can indicate the locations and binding potential of these receptors in the brain, providing information about the relative metabolism of serotonin in human subjects. The ability to image the metabolism of specific bioactive molecules makes PET imaging unique among functional imaging techniques and provides a utility that fMRI cannot. There are a number of limitations to PET. Compared to fMRI, PET has about the same temporal resolution (6–10 s), a lower spatial resolution, and BOX 1.1  Functional Magnetic Resonance Imaging (fMRI) Experiments in Animals A major benefit of functional brain imaging technologies is the ability to noninvasively study neural activity in humans. However, this technology can also be used with other animals, such as primates and rodents. The major benefits of using animal subjects include the ability to validate the use of animal models and to bridge human fMRI experiments and animal electrophysiology experiments. The ability to use both fMRI and electrophysiology in the same animal (and even at the same time) greatly aids our understanding of the functional activity of individual neurons and entire brain regions in the same experiment. Furthermore, fMRI studies in primates that screen neural activity across the entire brain can inform future electrophysiological studies about brain regions containing neurons of interest to an investigator. Functional magnetic resonance imaging studies using primates present additional limitations and challenges compared to traditional human studies. The horizontal position of most MRI scanners is not ideal for primate studies, so vertical scanners have been created to accommodate the special chambers that support and brace a conscious primate. The animal’s head must be fixed in place with a head post so there is no head movement during an experiment. The animal must also be acclimated to these experimental conditions so it is comfortable in its environment and can perform the task. A further challenge is that during the actual experimental sessions, a primate may lose motivation to perform a task or attend to a stimulus. Unlike an electrophysiology experiment in which an animal occasionally receives a juice reward (Chapter 2), juice must be consistently provided so that the animal stays attentive and completes the scanning session. MRI scanners with higher magnetic field strengths allow high-resolution, detailed imaging in small animals such as rats and mice. Obviously, these animals cannot perform complicated cognitive tasks and most likely need to be anesthetized during an imaging session. However, the benefit of doing fMRI in rodents is the possibility of injecting psychoactive drugs during an imaging session or performing lesion studies. Thus, it is possible to study the global effects of a pharmacological agent, such as a receptor antagonist or the loss of a brain region over time. Furthermore, a scientist can follow up functional imaging data with histological studies after the completion of an experiment. Whole-Brain Imaging Chapter | 1  19 cannot generate anatomical data. PET images are often combined with CT or MRI images to present functional data in an anatomical context. The cost of doing PET is very expensive. Because most isotopes used for PET studies have extremely short half-lives, positron-emitting isotopes must be synthesized on site using a room-sized device called a cyclotron, which itself is expensive to purchase and maintain. The half-life of FDG is 110 min, so this compound cannot be ordered and delivered from a remote location. Finally, PET requires the injection of radioactive substances into subjects. Therefore, multiple PETimaging sessions are not recommended for a single subject. Because of these limitations, most modern functional imaging experiments utilize fMRI technology instead of PET, with the exception of those experiments that focus on the metabolism of bioactive substances, such as neurotransmitters. Single-Proton Emission Computerized Tomography Single-proton emission computerized tomography (SPECT) imaging is very similar to PET imaging, producing functional images of neural activity but no structural data. Like PET, a radioactive probe is injected (or inhaled) into the circulatory system. The probes bind red blood cells to be carried throughout the body. Because blood flow is increased in active brain structures, the radioactive signal is used to assess an increase in neural metabolism. As the label undergoes radioactive decay, it emits high-energy photons that can be detected using a gamma camera. The camera is rapidly moved around the head of a subject to collect photons from many different angles, permitting a three-dimensional reconstruction. Although SPECT is very similar to PET in concept, it is not as costly because the radiolabeled probes are usually commercially available and do not require (C)Annihilation event Gamma ray detectors (B) Positron Electron (A) FIGURE 1.14  Positron emission tomography (PET) imaging. (A) An unstable positron-emitting isotope will decay over time and emit a positron. When a positron comes into contact with an electron, an annihilation event occurs and a pair of gamma photons are released. (B) If this unstable isotope is injected into a subject, the annihilation event can be detected by a PET scanner. This scanner consists of a series of gamma ray detectors arranged around the subject’s head. (C) Unstable isotopes for many metabolic substances, such as glucose or neuropeptides, can be imaged in a PET experiment. The increase in signal over time can map sites of metabolic activity caused by a specific neural stimulus or task. 20  Guide to Research Techniques in Neuroscience an on-site cyclotron. Therefore, SPECT can be thought of as a cheaper alternative to PET. The disadvantages to using SPECT as an imaging technique are similar to the disadvantages of using PET: the technology has a relatively low spatial resolution compared to fMRI (about 8 mm), and radioactive substances must be injected into a subject. Electroencephalography EEG is a measure of the gross electrical activity of the surface of the brain. It is not truly a brain imaging technique, as no meaningful images of the brain can be produced using this technique alone. However, EEG is noninvasive and can be used to ascertain particular states of consciousness with a temporal resolution of milliseconds. EEG combined with other imaging techniques, such as fMRI, can provide an excellent temporal and spatial representation of neural activity. To produce an electroencephalogram, several disk-shaped electrodes, about half the size of a dime, are placed on the scalp (Figure 1.15). The scalp EEG reflects the sum of electrical events at the surface of the brain. These events include action potentials and postsynaptic potentials, as well as electrical signals from scalp muscles and skin. An interesting way to think about EEG technology is to imagine what it would be like to place a microphone above a large crowd of people, such as above Times Square during New Year’s Eve. It would be impossible to make out the signal from an individual person in the crowd; however, it would be possible to determine when a meaningful event occurred, such as when the crowd counted down to midnight. Likewise, it is impossible to record the electrical activity from a single neuron with EEG, but it is possible to ascertain when a meaningful event occurs in the brain, such as when a subject detects a salient stimulus in an experiment. Thus, combining the temporal resolution of EEG with the spatial resolution of fMRI provides a powerful method to detect the precise timing and location of neural activity within the brain. A powerful application of EEG is the study of event-related potentials (ERPs) in humans and animals. An ERP is a distinct, stereotyped waveform in the EEG that corresponds to a specific sensory, cognitive, or motor event. For example, if a human subject abruptly hears an alarm, the perception of the sound may be represented as an ERP in the EEG waveform. Thus, neuroscientists can study the moment that the brain processes a stimulus with subsecond temporal resolution, allowing a noninvasive means of evaluating brain function in experimental subjects or patients with cognitive diseases. Magnetoencephalography MEG measures changes in magnetic fields on the surface of the scalp that are produced by changes in underlying patterns of neural electrical activity (Figure 1.16). About 50000 neurons are required to produce a detectable signal with Whole-Brain Imaging Chapter | 1  21 MEG, a number that may seem large but is actually much smaller than what is required for an EEG signal, which may require millions of neurons. MEG offers relatively poor spatial resolution but excellent temporal resolution compared with PET, SPECT, and fMRI. Therefore, MEG can be thought of as a compromise technique: it offers excellent temporal resolution and much better spatial resolution compared with EEG, but not as good spatial resolution as other imaging techniques. In particular, MEG in combination with fMRI allows for excellent temporal and spatial resolution of neural activity. Unfortunately, MEG is a very expensive technique, requiring a room that can obstruct magnetic fields from outside sources; even something as small as a coffee machine in a neighboring building can be detected if the room is not adequately insulated. 1 2 3 4 5 6 FIGURE 1.15  A human subject wearing electrodes for electroencephalography recordings. The electrodes measure the global electrical activity of millions of neurons in the brain. Each electrode placed on the scalp records a unique trace of activity based on its location. Subject in MEG scanner Within scanner, the subject wears a cap with dozens of magnetic field sensors (A) (B) FIGURE 1.16  A magnetoencephalography (MEG) setup. (A) A subject wears a cap with magnetic field sensors and (B) is placed inside an imaging chamber. MEG equipment is expensive and requires a room with strong insulation. 22  Guide to Research Techniques in Neuroscience Like EEG, the temporal precision of MEG allows for the study of eventrelated changes in brain activity. The MEG equivalent of an event-related potential is referred to as an event-related field (ERF). Optical Imaging Optical imaging techniques produce images of neural activity by measuring changes in blood flow and metabolism from the surface of the brain. Rather than detecting changes in the magnetic or electrical properties of neurons, as in fMRI, EEG, or MEG, optical imaging detects changes in light reflectance from the surface of the brain due to changes in the amount of blood flowing to neural tissue. In animal preparations or during human surgery, light is shined on an exposed portion of the brain. Any light that is reflected off the surface is detected by a sensitive camera and recorded by a computer. When neurons are more active, changes in blood volume, blood oxygenation, and the lightscattering properties of neural tissue (resulting from ion and chemical movements) all cause small (0.1–3.0%) changes in the reflectance of light from the brain’s surface. For each experiment, the investigator images a baseline amount of light reflectance, and then compares this baseline parameter to changes in light reflectance that arise due to the presentation of a stimulus. Optical imaging technologies allow for spatial resolutions of <1 mm and temporal resolutions of 2–8 s. These technologies have been used to produce high-resolution functional maps of the visual cortex in both animals and humans. Earlier in this chapter, we defined functional imaging techniques as methods that allowed investigators to measure changes in neural activity over time without physically penetrating the skull. Obviously, many optical imaging techniques are an exception to this definition, as the brain surface must be exposed to allow light to penetrate and reflect back to a camera. However, diffuse optical imaging (DOI) and near-infrared spectroscopy (NIRS) are noninvasive alternatives that utilize the same basic principles as invasive optical imaging but record light reflectance through the scalp (Figure 1.17). The signal is much weaker than invasive optical imaging, as light must pass through the superficial layers of the head to the brain, and then from the brain to optical electrodes, known as optrodes or optodes, placed on the surface of the scalp. However, these techniques are sensitive enough to detect large changes in neural activity and can be useful in clinical applications as an alternative to fMRI or PET because of their low cost and portability. For example, long-term monitoring of cerebral oxygenation in a patient following a stroke is a potential use of DOI that would be practically impossible with fMRI or PET. Table 1.1 compares the spatial and temporal resolutions, cost, and invasiveness of the various functional brain imaging techniques described previously.As mentioned before, many imaging laboratories now combine multiple techniques to make up for the inadequacies of any single technique. For example, fMRI offers excellent spatial resolution and is noninvasive, but it does not offer good Whole-Brain Imaging Chapter | 1  23 temporal resolution of neural activity during experiments because the BOLD effect occurs over a period of 6–10 s. Therefore, some laboratories combine fMRI with MEG or EEG, in which signals can be detected within a fraction of a second. So far, this chapter has focused almost exclusively on the technology of structural and functional brain imaging. Of course, the utility of these technologies depends entirely on how they are used: the research hypotheses, experimental designs, and methods of data analysis. In the last part of this Subject with light sources/detectors placed above the frontal cortex Scalp Light sourceLight detector FIGURE 1.17  Optical imaging. Light is shined onto the surface of the brain. A portion of this light is reflected off the brain and detected by multiple optrodes. Changes in neural activity produce changes in the amount of light that is absorbed and reflected by the brain. Therefore, optical imaging can be used to indirectly detect changes in neural activity. These techniques can be either invasive (the skull is opened to reveal the surface of the brain) or noninvasive. TABLE 1.1  Comparison of Functional Imaging Techniques Spatial resolution Temporal resolution Cost Invasiveness fMRI 1 mm 6–10 s Expensive Noninvasive PET 4 mm 1 min Very expensive Radioactive injection SPECT 8 mm 2–8 s Expensive Radioactive injection EEG 1 cm 1 ms Moderate Noninvasive MEG 1 mm 1 ms Very expensive Noninvasive Optical imaging 1 mm 2–8 s Inexpensive Can be invasive 24  Guide to Research Techniques in Neuroscience chapter, we will examine the approaches that can be taken in the design and analysis of functional imaging experiments. FUNCTIONAL IMAGING EXPERIMENTAL DESIGN AND ANALYSIS Functional imaging technologies are particularly well suited to cognitive neuroscience, the field of neuroscience dedicated to elucidating the neural basis of thought and perception. Cognitive operations routinely studied in the literature include attention, learning and memory, executive function, language, emotion, and higher-order sensory processing, such as the enjoyment of art or music. Determining the neural basis of cognitive processes, especially processes that seem unique to humans, absolutely depends on brain imaging techniques. In addition to answering scientific questions about cognitive neuroscience, fMRI has added important information to other neuroscience fields, such as sensory or motor systems neuroscience. These studies identify regions of the human brain that are active in the presence of certain sensory stimuli or motor actions. Some studies reproduce findings in humans that were previously demonstrated in other animals, as well as add important insights that enhance our understanding of these systems. For example, fMRI research has demonstrated that visual pathways in humans are anatomically consistent with visual pathways found in other mammals. However, our understanding of the visual system has been greatly enhanced by the fact that human subjects can report experiences and receive verbal instructions that other animals cannot. For example, researchers have found that activity in the visual cortex increases not only when human subjects are exposed to visual stimuli, but also when subjects are told to imagine visual stimuli. Furthermore, fMRI research has demonstrated that different areas of the visual cortex process different aspects of visual stimuli, such as color, shape, and texture. Thus, fMRI contributes meaningfully to our understanding of both cognitive as well as systems neuroscience. Unfortunately, the conception and design of functional imaging experiments are not always well understood by scientists who do not use these methods. This is probably because these technologies are unlike most other techniques used in neuroscience, as well as the fact that they are usually only performed by the dedicated researchers who use them regularly. A common complaint among brain imaging specialists is the misconception that you can simply place a human subject into a scanner, tell them to look at some stimulus, and then publish the results. Like any other technique, whole-brain imaging experiments must be carefully designed and interpreted, more than nonspecialists may sometimes appreciate. The next section describes some of the scientific considerations in the design and execution of a functional imaging experiment. We will focus on fMRI, as this technique dominates the field, but the same principles can be applied to other functional imaging techniques as well. Whole-Brain Imaging Chapter | 1  25 Planning the Experiment Before any human subject is placed in a scanner, an investigator conceives and designs an experiment months or even years in advance. There are many practical considerations that must be taken into account in the design of the experiment that may affect the ambitions of the investigator. Once these limitations are understood, the investigator designs the experiment to answer a specific question or test a specific hypothesis. Finally, the investigator designs a proper task paradigm and tests efficacy of the stimuli so that the results are appropriate and accurate. Practical Considerations Perhaps the most important practical consideration of an fMRI experiment is its cost. The actual scanning machine, as well as its support apparatus, costs millions of dollars and is almost always shared by multiple labs. Routine maintenance of the scanner is also expensive, with specialized technicians on call to fix any potential physical problems that may arise. Most research institutions charge an individual lab based on the number of hours the scanner is used. These institutions may also charge money based on the time of day, with scanner use during the day costing more than scanner use late at night. It is not unusual for scanner time to cost between $100 and $1000 per hour, depending on the institution’s operating costs. Therefore, an fMRI experiment must be well designed so that scanner time is not wasted. fMRI is too expensive for simply playing around and putting humans in the scanner for no defined purpose. Finally, scanner time is usually reserved well in advance due to the number of scientists who wish to perform experiments. Another important practical concern of fMRI is that a subject must keep his or her head completely motionless during a scanning session. Even small movements can disrupt the magnetic field, causing artifacts in the images. Also, if a subject moves during a scan, the acquired images will not line up with each other, making them uninterpretable. This immobility means the subject will be unable to perform certain activities while inside the scanner, such as speaking, exercising, or moving facial muscles. However, subjects can be trained to perform simple hand movements to complete a task or respond to a stimulus, such as pressing a button on a hand device. Some specialized devices have been invented to deliver olfactory or gustatory stimuli to subjects. A practical concern for an investigator who wishes to study sleep, attention, or the auditory system is that an fMRI scanner can be very noisy. The RF pulse sequences are loud and last the entire duration of the experiment. Therefore, many subjects do not sleep normally in a scanner, and the changes in attention due to a surprising stimulus may actually be due to the surprise of a noisy scanner turning on and off. Auditory stimuli can also be very difficult to work with, for obvious reasons. These limitations are not insurmountable, but require extra planning by the scientist to make sure that a sporadically loud environment does not bias the results. 26  Guide to Research Techniques in Neuroscience Structure of a Functional Imaging Experiment Like any other experiment in neuroscience, functional imaging experiments examine the effect of an independent variable on a dependent variable. The independent variable is the experimental variable that is intentionally manipulated by the researcher and is hypothesized to cause a change in the dependent variable. In a functional imaging experiment, the independent variable can be a stimulus, task, or even a difference in the subjects being tested such as their age, gender, or disease state. The dependent variable is the quantifiable variable measured by the researcher to determine the effect of the independent variable. This variable is different for each brain imaging technique. In the case of fMRI research, the dependent variable is the BOLD signal intensity for a particular part of the brain. In the case of PET or SPECT technology, the dependent variable is the intensity of gamma radiation. All of these dependent variables are used as proxies for neural activity. The specific hypotheses for a functional imaging experiment are usually framed in the following manner: l Activity in brain region X is correlated with stimulus/taskY. l For a given stimulus or task, activity in brain region X precedes activity in brain regionY. l Activity in brain region X is higher under conditionY than condition Z. l Activity in brain region X is higher in human populationY than human population Z. l Activity in brain region X changes across time as a subject learns a task. These hypotheses depend upon the investigator knowing of the existence of “brain region X” and having a good rationale for studying that region. The amount of data generated during an fMRI experiment is quite large, and processing fMRI data can be easier if the investigator has an idea of where to look. Despite large amounts of data, fMRI can also be used to screen the entire brain to find brain regions that increase neural activity in response to a stimulus/ condition, or to examine distributed networks of neural activity in a resting state. The increasing availability of powerful computational resources and techniques has enabled these more data-driven, exploratory approaches to analyze the large quantities of data produced in brain imaging. Machine learning-based techniques, such as multivariate pattern analysis, can be used to extract patterns from large datasets, allowing the advancement of studies that look at brain-wide functional networks. Itisalsoimportanttorecognizewhetherafunctionalimagingstudyisbetween subjects or within subjects. In a study that is between subjects, the independent variable is the different populations of subjects, such as male vs. female, old vs. young, healthy vs. diseased, genotype A vs. genotype B, etc. These differences in subjects can potentially create problems in analysis. For example, it is known that the BOLD response can change with age; therefore, comparing a younger group of subjects with an older group of subjects using fMRI may require additional controls to ensure that changes in brain activation are not simply due to Whole-Brain Imaging Chapter | 1  27 age-related changes in the BOLD effect. In a study that is within subjects, each subject participates in all experimental conditions and the independent variable is something other than the identity of the particular subject. Task Paradigms After developing a specific research hypothesis, an investigator determines an appropriate task paradigm, a strategy for presenting stimuli to subjects during an experiment. These paradigms are usually divided into two major categories: blocked designs or event-related designs. In a blocked design, a subject is presented with two or more categories of stimuli that alternate every 1–2 min. For example, consider an experiment that features two categories of stimuli, X and Y. Category X might represent something such as smiling faces while category Y could represent neutral faces. In the first block of stimuli, the subject is presented with different images made up entirely of category X:X1, X2, X3, X4, and so on.A typical block may last between 10 and 60 s, with each individual stimulus presented from 1 to 10 s. After the end of the block, there may or may not be a brief period when no stimuli are presented. Then, a second block of stimuli is presented made up entirely of category Y:Y1, Y2, Y3, Y4, and so on. These blocks of stimuli are repeated in cycles, alternating X andY, until a sufficient amount of data is collected from a subject. Some blocked designs contain a third category, alternating between blocks of X,Y, and Z. Blocked designs are useful for obtaining relatively high signal-to-noise information from active brain regions. Because stimuli of the same experimental category are presented repeatedly for 1–2 min, the hemodynamic BOLD response has sufficient time to increase and a strong signal is produced. These paradigms are often chosen for experiments determining which brain regions are active for a given stimulus or task. They are also good for neural processes that last a relatively long time, such as changes in cognitive or emotional state. A blocked design is not as useful for examining a change in brain activity during relatively short events that last 2–5 s, such as when a subject is required to make a decision inside a scanner. In an event-related design, stimuli are presented as isolated, individual events of short duration. This paradigm is useful for discrete tasks in which a subject must recognize an event, detect a novel stimulus, or make a decision, all processes that take place in short time intervals. The hemodynamic properties of the BOLD effect may return to baseline between events, or for fast-paced stimuli, the BOLD response may not return to baseline, and so the effects of two stimuli on neural activity can be examined within a small time window. For example, stimulus A may indicate the nature of stimulus B, and the BOLD signal can be measured for B with and without the presence of A. Event-related designs are not usually used in experiments that affect a subject’s emotional state over time, as these states usually last over relatively longer time periods. Blocked designs and event-related designs are not mutually exclusive, and indeed some studies take advantage of both designs and employ what is referred 28  Guide to Research Techniques in Neuroscience to as a mixed design. The relative advantages and disadvantages of all three task paradigms are compared in Table 1.2. Pilot Experiments Because MRI scanner time is so expensive, it is often necessary to test the delivery of stimuli or characterize the behavior of subjects during a task before any imaging experiments take place. These pilot experiments take place outside of the MRI facility and demonstrate the efficacy of stimulus delivery in producing an appropriate human response. Statistical analysis of how subjects perform their tasks is sometimes necessary to gain approval before the actual experiments take place in the scanner. Only after the investigator characterizes a cognitive task in terms of the efficacy of the stimuli and behavior of human subjects will it be appropriate to begin to associate neural correlates with that behavior. Planning an Experiment that Uses Human Subjects Once an experiment is designed, a lab must be given permission to use humans as research subjects by an Institutional Review Board (IRB). An IRB is a TABLE 1.2  Relative Advantages and Disadvantages of Different Functional Magnetic Resonance Imaging (fMRI) Task Paradigms Advantages Disadvantages Blocked l More statistical power for detecting subtle differences across different conditions l Tend to be simpler to implement and analyze l Good for examining cognitive state changes l Predictability of stimuli may confound results l Information about the time course of the activation response is lost within a block l Not applicable to certain types of tasks (e.g., novelty) Event-related l Reduces confounds of predictable stimulus order, because stimuli can be presented randomly l Can sort trials after the experiment according to specific behavioral outcomes l Useful for examining temporal characteristics of responses l Flexible analysis strategies l More complex design and analyses than blocked design l Lower signal-to-noise ratio than block designs l Must perform longer scan runs to compensate for loss in statistical power Mixed l Can compare short-term transient activity with long-term sustained activity l Most complicated analyses Whole-Brain Imaging Chapter | 1  29 committee of about 10 individuals, typically composed of physicians and research faculty, as well as nonfaculty members such as a nurse, minister, graduate student, or lawyer. The purpose of the IRB is to review studies that use human subjects and determine if the study meets ethical, safety, and scientific obligations. An IRB typically meets anywhere from 2 to 12 times per year to review and vote on the submitted applications. Additionally, any research project involving human subjects must meet the standards of the Health Insurance Portability and Accountability Act’s Privacy Rule, which regulates the protection, security, and confidentiality of private health information. Conducting the Experiment After an experiment is designed, discussed, and approved, a scientist proceeds with recruiting human subjects and collecting data in the scanner. These scientists must be familiar with the scanner technology and experimental procedures before any actual data collection begins. The difficulty in recruiting human subjects and the high cost of scanner time makes it essential that experiments are conducted properly and without many mistakes. Working with Human Subjects Though the exact number varies based on the experimental design, usually about 10–20 subjects are necessary for a typical imaging experiment to reach an appropriate, statistically significant conclusion. Recruiting human subjects can be relatively easy or difficult depending on the type of subjects needed. If the only requirement for a subject is that he or she is an average, healthy adult, then recruiting subjects can be relatively straightforward. At academic institutions, undergraduates and graduate students are often targeted for recruitment, with compensation given in the form of extra credit for a course or a cash payment. When subjects must represent a particular population of individuals (e.g., individuals with posttraumatic stress disorder, elderly subjects, or individuals diagnosed with depression) it can be much harder to recruit a necessary number of subjects. Consider a study in which the hypothesis is that depressed individuals have a significantly different level of activity in a particular brain region than nondepressed individuals. In this case, the investigator must find individuals who have been formally diagnosed with depression. Additionally, because psychotropic drugs can confound the results, it is necessary that none of these subjects received medication after their diagnosis. Therefore, the investigator must identify 10–20 subjects who have been diagnosed with depression but chose not to undergo pharmacological treatment (or who have not yet started treatment). In studies like this, recruiting subjects can be the bottleneck in the time it takes to complete an experiment. Investigators might need to formally recruit individuals from other institutions, such as nursing homes, hospitals, or secondary schools. It might be necessary to formally collaborate with clinical neurologists to recruit patients with various mental health disorders. 30  Guide to Research Techniques in Neuroscience Once a subject is recruited, the investigator schedules a specific session time in the scanner. Prior to the start of an experiment, the investigator must inform the subject about the nature of the experiment (although details of the experiment can be withheld for experimental purposes). The subject must sign a release form to indicate that he or she is a willing participant in the study and that any data gathered can be used for publication. Finally, the subject must complete a brief survey about his or her personal health to ensure that there are no confounding variables such as history of mental health disease, cardiovascular disorders, or heavy use of alcohol or other drugs. The subject is also requested to remove any metal objects before entering the scanner, such as loose change, belt buckles, earrings, and so on. Often the investigators ask the subjects to change into a disposable gown to ensure that there are no metal objects placed inside the scanner. During the experiment, the investigator must ensure that the subject is as comfortable as possible inside the scanner. To reduce head movement, the subject’s head is surrounded by soft padding to hold it firmly in place inside a head coil. The investigator provides the subject with a squeeze ball that, when pressed, can stop the scans at any time if the subject becomes uncomfortable. For experiments that require the subject to make a choice or respond to a stimulus, the investigator also provides a nonmetallic handheld device with buttons that the subject can press to provide feedback. Visual stimuli can be projected through goggles or to a mirror inside the scanner to allow for video presentations from a nonmetallic surface. After the experiment is complete, an investigator usually debriefs the subject on the experiment and pays the subject for his or her time. Depending on the experiment, follow-up sessions can be necessary, and these are also scheduled well in advance. Data Acquisition In a typical experiment, it is common for about 10–20 human subjects to participate (twice that for a between-subjects study). Each subject usually participates in one to three sessions, the scheduled time when an actual experiment is conducted. Each session contains many runs in which the brain is repeatedly scanned and the hypothesis is tested. Within each run, functional data are accumulated and stored as volumes—three-dimensional constructions of brain space. A complete volume is typically acquired every 1–3 s. Each volume contains a complete set of slices, two-dimensional representations of a plane of the brain that are only one voxel thick. A voxel is the smallest functional unit of brain space that can be analyzed for changes in signal intensity over time. In fMRI, a voxel is usually 1–5 mm3. Remember that a voxel of fMRI data is essentially four-dimensional: three dimensions represent the location of the voxel in space (x, y, z) and the fourth dimension is the time in which the signal intensity was recorded. Thus, an fMRI experiment generates huge quantities of data: 10–20 subjects, each with one to three sessions, each with several Whole-Brain Imaging Chapter | 1  31 experimental runs, each with several volumes of brain space, each with several brain slices, and each with a grid of voxels that represent the data! Therefore, computers with ample storage space are required to record and store data until the results can be analyzed after the collection is complete. Manipulating Neural Activity during an Experiment One of the limitations of studying the human nervous system is that it is very difficult or even impossible to functionally perturb the brain during an experiment. For example, if activity in a specific part of the brain is correlated with performance on a task, it would be interesting to determine if a loss of that part of the brain caused a deficit in the ability of the subject to perform the task. While such a loss-of-function experiment could be performed on another model organism with a lesion study, this is, of course, impossible in humans. Likewise, it would also be interesting to stimulate neural activity in a particular brain region to determine if performance improves on a task. Such gain-of-function experiments are also possible in model organisms, with stimulating microelectrodes or overexpressed genes driving activity in a specific population of neurons. Again, this is impossible in typical, noninvasive fMRI experiments. Two technologies make it possible to reversibly modulate neural activity in humans. Transcranial magnetic stimulation (TMS) reversibly activates or inactivates regions of the human brain. In this technique, a coil that generates magnetic field pulses is positioned near a subject’s head (Figure 1.18). The changing magnetic fields induce weak electric currents at a specific focal point TMS coil Induced electrical activity FIGURE 1.18  Transcranial magnetic stimulation (TMS). A coil is held near a person’s head, and a magnetic field stimulates electrical activity in a superficial brain region. Depending on the strength of the magnetic field and the brain region, this electrical activity can cause either a temporary loss-of-function or gain-of-function in that brain region. 32  Guide to Research Techniques in Neuroscience on the surface of the brain. Depending on the brain region and the strength of the magnetic field impulse, the electrical activity may either stimulate neural activity or cause a hyperpolarized state in which neural activity is temporarily inactivated. Therefore, it is theoretically possible to perform a loss- or gain-offunction experiment using this technology. One of the limitations of this technology is that TMS can only induce electrical activity on the outer surface of the brain—it is currently impossible to stimulate or inactivate deep brain structures that may be of interest to an investigator. However, there are thousands of publications that have used TMS to selectively activate and inactivate activity in the human cerebral cortex. This technology can also potentially be used for therapeutic treatment of diseases and is being heavily researched in the treatment of depression and motor diseases. Another technology, ultrasonic neuromodulation (USNM), reversibly activates brain regions of interest through the delivery of transcranial ultrasound. These waves subtly increase the temperature of the brain, causing a transient increase in neural activity at a greater spatial resolution than TMS technologies. An additional strategy for investigators to perform loss-of-function experiments is to attempt to find patients with lesions (caused by an accident or stroke) to specific brain regions. Because many institutions that utilize fMRI scanners are associated with hospitals, identifying and recruiting patients with lesions can be coordinated between researchers and neurologists. However, these patients may present lesions that are not tightly localized to the brain area of interest, there may be compensatory mechanisms that develop over time, and it may be more difficult for these patients to perform specific tasks. All of these caveats must be considered before including these patients in experiments, as well as when evaluating these experiments in the literature. Postexperimental Data Analysis The end result of most fMRI experiments is a colorful figure depicting the activation or inactivation of a particular region of the brain that is correlated with a stimulus or task. These figures can seem deceptively simple to produce, yet require rigorous methods of data analysis and interpretation. Data Analysis What is an investigator to do with the huge quantity of data acquired from multiple research subjects? In most analyses, the data from each subject are analyzed separately first and then pooled together into a common subject pool. Because there is variability in every subject’s brain structure, data must first be fit into a common three-dimensional brain template. Structural data from each individual’s brain are therefore stretched and warped to fit specific anatomical landmarks in a precise mapping system. One of the most widely used coordinate systems for normalizing fMRI data is called Talairach space, based Whole-Brain Imaging Chapter | 1  33 on the stereotaxic measurements of a single postmortem brain. This coordinate system is ubiquitous in brain imaging research as it defines brain regions and Brodmann’s areas in stereotaxic coordinates, allowing investigators to make anatomical comparisons among different brains. Another commonly used coordinate system is called the MNI template (MNI stands for Montreal Neurological Institute, the institution that established this system). This template is a probabilistic mapping system based on the averages of hundreds of individual brain scans and scaled to match the landmarks within the Talairach atlas. After an individual subject’s brain is adjusted to either the Talairach or MNI templates, signal intensity is compared at different time points according to the task paradigm. There are two general strategies of analyzing signal intensity in the brain: voxelwise analysis or region-of-interest (ROI) analysis. Most fMRI studies present data using voxelwise analysis, in which each voxel is analyzed for significant differences in signal intensity in time between two experimental conditions. In ROI analysis, the brain is divided into a set of discrete regions assigned by the investigator. Rather than voxel-by-voxel comparisons, entire brain regions are compared for significant differences in signal intensity. ROI analysis can provide information about specific structures, but can initially be more time-consuming, as each voxel has to be assigned to a specific ROI. Sometimes an investigator will first analyze data using a voxelwise analysis and then analyze discrete ROIs. Each individual’s fMRI data are analyzed for significant differences, and then these data are pooled together for all the individuals within a certain subject pool. For a within-subjects study, the subject pool is the same and the investigator determines if there is a significant difference in signal intensity between two different stimuli or tasks. For a between-subjects study, the subject pools are different and the investigator determines if there is a significant difference in signal intensity between the two groups for a stimulus or task. Following this data analysis, the statistically significant differences in the pooled subject data are presented as figures for publication. Preparation of Figures The most common functional imaging figures presented in the literature take the form of color-coded brain activation data superimposed on a structural image of a brain slice (Figure 1.19). As mentioned previously, the brain slice is almost always a T1-weighted image because these images present the greatest contrast between brain regions. These brain slices can either be taken from one of the subjects during the experiment, or from a stock set of images from the Talairach or MNI templates. Sometimes, investigators superimpose imaging data on an inflated brain in which the sulci and gyri of the brain are expanded into a balloon-like shape to better show neural activity in the sulci (Figure 1.20). It is important to realize that these fMRI figures are statistical comparisons in signal intensity over time for a grid of voxels, and that color-coded 34  Guide to Research Techniques in Neuroscience activation does not just light up during a scanning session. The color-coded voxels superimposed over the structural brain image are essentially a grid of numbers (Figure 1.19). Each voxel is assigned a statistical value that represents the difference in signal intensity for that region of brain space between two conditions. The exact color scheme is chosen by the investigator to reflect 4 1 3 2 fMRI figure depicting an increase in neural activity in the visual cortex Increased magnification of the figure reveals statistical analysis for each voxel Degrees of significance FIGURE 1.19  Preparation of a functional magnetic resonance imaging (fMRI) figure for publication. A typical fMRI figure contains a representation of the changes of blood oxygen level-dependent signal intensity over time superimposed over a T1-weighted structural image. It is important to remember that the colored representation of signal intensity is essentially a grid of numbers, with each number representing a measure of statistical significance. (fMRI image reprinted with permission of Springer Science + Business Media and Jens Frahm from Windhorst, U. and Johansson, H. (eds.), 1999. Modern Techniques in Neuroscience Research, Chapter 38: Magnetic Resonance Imaging of Human Brain Function, p. 1064, Figure 5.) FIGURE 1.20  An inflated representation of the right hemisphere of the human brain used to show activity within sulci. The dark gray regions represent the exposed sulci of the inflated brain, while the light gray regions represent gyri. (Courtesy of Dr Rory Sayres.) Whole-Brain Imaging Chapter | 1  35 the statistical magnitude of signal intensity: usually bright colors represent high differences in intensity, while darker colors represent more subtle differences in intensity. CONCLUSION The goal of this chapter was to provide a thorough introduction to the technologies that currently exist to image the structure and function of the brain, as well as the approaches an investigator may take when designing and analyzing functional imaging experiments (Box 1.2). These technologies are more complicated than the brief descriptions found here, and readers interested in more detailed explanations should consult the Suggested Readings section at the end of this chapter. However, this survey of techniques hopefully provides an appreciation for the ingenuity of scientists and engineers in developing and improving these techniques over the past 10–20 years. Imaging neural function can be performed based on the electrical, magnetic, metabolic, and light-scattering BOX 1.2  Walkthrough of a Functional Magnetic Resonance Imaging (fMRI) Experiment Let us consider a hypothetical situation: Suppose you work in a laboratory that uses fMRI to study cognition in humans. Your research group becomes interested in the neural basis of how a person recognizes and identifies familiar people. One way to frame this question is “Are there specific brain regions that are active in response to familiar individuals versus unfamiliar individuals?” How would you go about designing an fMRI study to answer this question? A good starting point would be to identify the independent and dependent variables in this experiment. The independent variable, the variable that varies from trial to trial, is a person who is familiar or unfamiliar to the subject. This immediately leads to another decision to make as an investigator: the choice of stimuli. Will the stimulus be a picture of an individual? The sound of the individual’s voice? The individual’s name? In this example, let us make the independent variable a person’s name that is visually displayed on a screen. Therefore, in each trial, the subject will see a name on a screen. Because the independent variable is the stimulus and not the identity of the subject, this is a within-subjects study. The dependent variable is the change in BOLD signal intensity in the brain. We might hypothesize that a specific brain region will show a significant difference in signal intensity between the two conditions, but in this experiment such a specific hypothesis is unnecessary, as we can scan major regions of the brain (if not the entire brain) to identify all regions that significantly differ between the two stimulus conditions. What will be the subject’s task during the experiment? If all we ask the subjects to do is stare at the screen, they may become bored, distracted, and even fall asleep. Therefore, we must provide the subjects with a specific task so that Continued 36  Guide to Research Techniques in Neuroscience they fully attend to the stimuli. In this example, a good task may be to indicate whether the name on the screen is a man’s or a woman’s. Each subject could be presented with a handheld device with two buttons, one for man and one for woman. At the end of the imaging session, we can measure the accuracy of the responses to ensure that the subject was fully attentive throughout the experiment. Obviously, we will need to generate a list of names for both the familiar and unfamiliar categories of stimuli. Each list should contain a balanced list of males and females, old and young, and so forth, to ensure that there are no confounding variables in the study. Familiar names could include famous celebrities and politicians, while the unfamiliar names could include combinations of first and last names that are not well-established celebrities. How can we ensure that the subject knows all of the familiar names and none of the unfamiliar names? This is impossible to confirm before the experiment begins, but after the experiment is over, each subject could complete a brief survey in which he or she indicates the level of familiarity of each name, perhaps on a scale of 1–10. This survey can then be utilized when analyzing the functional imaging data to ensure that each stimulus is adequately categorized. The next major decision will be the best task paradigm to present stimuli to subjects. A blocked design will allow us to present multiple stimuli of the same category (familiar or unfamiliar names) to a subject over a 1- or 2-min period, a condition that may allow for maximal signal intensity in active brain regions. However, we may wish to evaluate the degree of signal intensity for each name presented to the subject. In this case, we may wish to pursue an event-related design, which will allow us to correlate the degree of signal intensity with the degree of familiarity with a famous name, as indicated by the subject on a survey after the experiment is over. Once the IRB approves the study and our proposal to use human subjects, we are ready to reserve scanner time and start collecting data! We will need 10–20 subjects for this experiment, but we should recruit at least 20–40 because some of the data will need to be discarded. For example, some subjects may fail to show up, some may fall asleep in the scanner, or some may turn out to have medical conditions or treatments that may preclude them from participation in the study. In order to recruit subjects, our lab could offer $25/h for their time. After the imaging session ends, we will provide them with the familiarity survey and then start to analyze the data. For each subject, we morph the structural coordinates from their brain scans to match a normalized template system, such as the MNI template. Then we can proceed with a voxelwise comparison in signal intensity between the familiar name trials and the unfamiliar name trials. Individual brains can then be pooled across all subjects for each condition. The final figure will be the difference in signal intensity between the two conditions, voxel-for-voxel, superimposed on a T1 image from the MNI template. Because the template contains coordinates for discrete brain regions, we can interpret the specific regions activated in response to familiar or unfamiliar names. BOX 1.2  Walkthrough of a Functional Magnetic Resonance Imaging (fMRI) Experiment—cont’d Whole-Brain Imaging Chapter | 1  37 properties of distinct brain regions. Producing so many different methods of imaging activity in the human brain is a remarkable achievement, and it will be exciting to observe, either as fellow scientists or participants, how these methods are combined and improved over the next 10–20 years. SUGGESTED READING AND REFERENCES Books Frahm, J., Fransson, P., & Kruger, G. (1999). Magnetic resonance imaging of human brain function. In U. Windhorst & H. Johansson (Eds.), Modern techniques in neuroscience research (pp. 1055–1082). Berlin: Springer. Huettel, S. A., Song, A. W., & McCarthy, G. (2004). Functional magnetic resonance imaging. New York: Sinauer. Mori, S. (2007). Introduction to diffusion tensor imaging. Amsterdam: Elsevier. Schild, H. H. (1990). MRI made easy... Well almost. Germany: Schering AG. Review Articles Aguirre, G. K. (2014). Functional neuroimaging: technical, logical, and social perspectives. Hastings Center Report, 44, S8–S18. Friston, K. (2002). Beyond phrenology: what can neuroimaging tell us about distributed circuitry? Annual Review of Neuroscience, 25, 221–250. Logothetis, N. K. (2008). What we can do and what we cannot do with fMRI. Nature, 453, 869–878. Mezer, A., et al. (2013). Quantifying the local tissue volume and composition in individual brains with magnetic resonance imaging. Natural Medicine, 19, 1667–1672. Miller, G. (2008). Neuroimaging. Growing pains for fMRI. Science, 320, 1412–1414. Mori, S., & Zhang, J. (2006). Principles of diffusion tensor imaging and its applications to basic neuroscience research. Neuron, 51, 527–539. Savoy, R. L. (2005). Experimental design in brain activation MRI: cautionary tales. Brain Research Bullettin, 67, 361–367. Turk-Browne, N. B. (2013). Functional Interactions as Big data in the human brain. Science, 342, 580–584. Primary Research Articles—Interesting Examples from the Literature Adleman, N. E., Fromm, S. J., Razdan, V., Kayser, R., Dickstein, D. P., Brotman, M. A., et al. (2012). Cross-sectional and longitudinal abnormalities in brain structure in children with severe mood dysregulation or bipolar disorder. Journal of Child Psychology and Psychiatry, 53(11), 1149–1156. Dagher, A., Leyton, M., Gunn, R. N., Baker, G. B., Diksic, M., & Benkelfat, C. (2006). Modeling sensitization to stimulants in humans: an [11C]raclopride/positron emission tomography study in healthy men. Archives of General Psychiatry, 63, 1386–1395. deCharms, R. C., Maeda, F., Glover, G. H., Ludlow, D., Pauly, J. M., Soneji, D., Gabrieli, J. D., et al. (2005). Control over brain activation and pain learned by using real-time functional MRI. Proceedings of National Academic of Sciences of the USA, 102, 18626–18631. Gil-da-Costa, R., Martin, A., Lopes, M. A., Muñoz, M., Fritz, J. B., & Braun, A. R. (2006). ­Species-specific calls activate homologs of Broca’s and Wernicke’s areas in the macaque. Nature Neuroscience, 9, 1064–1070. Glasser, M. F., & Rilling, J. K. (2008). DTI tractography of the human brain’s language pathways. Cerebral Cortex, 18, 2471–2482. 38  Guide to Research Techniques in Neuroscience Gonsalves, B. D., Kahn, I., Curran, T., Norman, K. A., & Wagner, A. D. (2005). Memory strength and repetition suppression: multimodal imaging of medial temporal cortical contributions to recognition. Neuron, 47, 751–761. Hariri, A. R., Mattay, V. S., Tessitore, A., Kolachana, B., Fera, F., Goldman, D., et al. (2002). ­Serotonin transporter genetic variation and the response of the human amygdala. Science, 297, 400–403. Honey, C. J., Sporns, O., Cammoun, L., Gigandet, X., Thiran, J. P., Meuli, R., et al. (2009). Predicting human resting-state functional connectivity from structural connectivity. PNAS, 106(6), 2035–2040. Kuhnen, C. M., & Knutson, B. (2005). The neural basis of financial risk taking. Neuron, 47, 763–770. Logothetis, N. K., Pauls, J., Augath, M., Trinath, T., & Oeltermann, A. (2001). Neurophysiological investigation of the basis of the fMRI signal. Nature, 412, 150–157. McClure, S. M., Li, J., Tomlin, D., Cypert, K. S., Montague, L. M., & Montague, P. R. (2004). Neural correlates of behavioral preference for culturally familiar drinks. Neuron, 44, 379–387. Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., & Gallant, J. L. (2011). Reconstructing visual experiences from brain activity Evoked by Natural Movies. Curr Biology, 21, 1641–1646. Pascual-Leone, A., Meyer, K., Treyer, V., & Fehr, E. (2006). Diminishing reciprocal fairness by disrupting the right prefrontal cortex. Science, 314, 829–832. Parks, N. A. (2013). Concurrent applications of TMS and near-infrared optical imaging: ­methodological considerations and potential artifacts. Frontiers in Human Neuroscience, 7, 592. Thomason, M. E., Brown, J. A., Dassanayake, M. T., Shastri, R., Marusak, H. A., et al. (2014). Intrinsic functional brain architecture Derived from Graph Theoretical analysis in the human Fetus. PLoS ONE, 9(5), e94423. Whalen, P. J., Kagan, J., Cook, R. G., Davis, F. C., Kim, H., Polis, S., et al. (2004). Human amygdala responsivity to masked fearful eye whites. Science, 306, 2061. Protocols Ashburner, J., & Friston, K. J. (2000). Voxel-based morphometry—the methods. Neuroimage, 11(6 Pt 1), 805–821. Ferrera, V., Grinband, J., Teichert, T., Pestilli, F., Dashnaw, S., & Hirsch, J. (2008). Functional ­imaging with reinforcement, eyetracking, and physiological monitoring. JoVE, 21 . http:// dx.doi.org/10.3791/992. http://www.jove.com/index/details.stp?id=992. Haacke, E. M., & Lin, W. (2007). Current protocols in magnetic resonance imaging. In T. Downey (Ed.), Current Protocols. New York: John Wiley & Sons. Websites fMRI 4 Newbies: http://fmri4newbies.com/ The Basics of MRI: http://www.cis.rit.edu/htbooks/mri/ Neuroimaging Informatics Tools and Resources Clearinghouse (NITRC): http://www.nitrc.org/ fMRI Methods Wiki: http://www.fmrimethods.org/ 39 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00002-2 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 2 Animal Behavior After Reading This Chapter, You Should be Able to: l Compare and contrast the advantages and disadvantages of common model organisms used to study animal behavior l Discuss considerations for designing a behavioral assay l Describe common behavioral assays used in rodents, invertebrates, and primates Techniques Covered: l Rodent motor assays: running wheel, homecage activity, open field test l Rodent motor coordination assays: rotarod test, footprint pattern assay, hanging wire assay, vertical pole test l Rodent sensory assays: visual cliff assay, startle response, taste assays, olfactory assays l Rodent nociception assays: tail flick assay, Hargreaves assay, hot plate assay, Von Frey assay, formalin assay l Rodent spatial learning and memory assays: Morris water maze, Barnes maze, radial arm maze l Rodent nonspatial learning and memory assays: classical conditioning, operant conditioning, novel object recognition, delayed match to sample and nonmatch to sample assays l Attention and impulsivity assays: go/no-go task, five choice serial reaction time task l Rodent reward-related assays: self-administration, progressive ratio, conditioned place preference/aversion l Rodent social assays: resident–intruder assay, social approach/avoidance assay l Rodent anxiety assays: open field test, elevated plus maze, defensive marble burying, Geller–Seifter conflict test l Rodent depression assays: forced swim test (Porsolt test), tail suspension test, sucrose preference test l Drosophila behavioral assays: locomotion, flight, sensory assays (vision, olfaction, taste), learning and memory assays, courtship, aggression l Caenorhabditis elegans behavioral assays: locomotion, sensory assays (mechanosensation, thermosensation, chemosensation) l Nonhuman primate behavioral paradigms Humans are the only animals capable of verbally reporting what they feel, what they perceive, and what they know. To gain insight into the emotional, perceptual, and cognitive processes of other animals, a scientist can only observe their behavior. 40  Guide to Research Techniques in Neuroscience Over the past several decades, scientists have developed and standardized a variety of tests that probe animals for clues about their mental status, from their ability to perceive a stimulus to whether or not they are emotionally depressed.There are now dozens of assays commonly described in the literature, with novel assays continually being developed to study a growing list of behaviors and psychiatric diseases. The goal of the behavioral neuroscientist is not only to characterize an animal’s behavior, but also to identify and describe the genetic, biochemical, and cellular correlates of that behavior. Therefore, behavioral neuroscience does not exist in isolation, but rather in combination with systems and molecular techniques. To correlate electrical events in the brain with specific behavioral outputs, a scientist can monitor neural activity during a behavioral task using electrophysiology or optical methods (Chapters 4 and 7). A scientist can also determine which genes, proteins, and/or neurons are necessary or sufficient for a behavior to occur by investigating the effects of their perturbation (Chapters 3, 4, 8, 12, and 13). Thus, the behavioral paradigms described in this chapter are used in conjunction with many other techniques described in this book. The goal of this chapter is to describe the general strategies scientists use to measure behavior in rodents, invertebrates, and primates. First, we will discuss some of the issues a scientist must consider when choosing and performing behavioral assays. Then we will survey the behavioral assays that are the most commonly used in the literature. CONSIDERATIONS FOR CHOOSING AND PERFORMING A BEHAVIORAL ASSAY There are many factors involved in the design of a behavioral assay. These factors include choosing a model organism, choosing a behavioral paradigm, reducing variability between individuals, and validating animal behavior as a useful model for human behavior and disease. Choosing an Appropriate Model Organism A scientist should consider two major factors when choosing an animal model to study behavior: the natural ethological capabilities of an animal and the appropriateness of the animal for other experiments. Ethology refers to the natural capability of an animal to perform specific behaviors. Worms and flies can be good models for studying behaviors in which animals interact with the environment, such as sensory transduction, food seeking, and motor control. Flies have also been used to study addiction, learning and memory, courtship, circadian rhythms, and other behaviors. Mice and rats are able to perform more complicated cognitive tasks that underlie emotion, cognition, and social behavior. Primates are capable of highly intelligent behaviors, such as complex decision making, identification of faces, and use of tools. In addition to these standard model organisms, scientists can choose to study other animals with specialized adaptations that make them an optimal model for certain behavioral tasks (Box 2.1). Animal Behavior Chapter | 2  41 A scientist must also consider the model organism in the context of other experiments. If the goal of the study is to identify novel genes that regulate a behavior, the best choice of animal model will be a genetically tractable organism, such as a worm, fly, or mouse. If the goal is to record from several neurons in the brain while the animal performs a behavior, a larger animal is necessary, such as a large rodent or primate. Mice are often used in behavioral assays because they are amenable to genetic, molecular, and electrophysiological approaches. It is also important to consider the ethical and moral implications of using a specific animal model in an experiment. The use of animals in research is both a necessity and a privilege that a scientist should never fail to appreciate (Box 2.2). Choosing an Appropriate Behavioral Paradigm Whether investigating a commonly studied behavior or developing a new behavioral paradigm from scratch, a scientist should consider many factors when preparing to use an assay. First and foremost, the assay must be quantifiable, with the ability to measure discrete, easily observable variables. For example, consider an experiment in which a scientist wishes to measure drinking behavior between two groups of rodents. The scientist has many potential variables to measure: the total volume of liquid consumed, the rate of water consumption, the number of licks applied to a liquid delivery system, etc. In the literature, data are usually presented as bar graphs or scatter plots, with different graphs representing different quantifiable aspects of each behavioral assay. A scientist must also consider that animals, especially animals in which an investigator has manipulated gene or protein expression or neural activity, may exhibit abnormal behaviors for reasons the scientist may not expect. For BOX 2.1  Neuroethology Neuroethology is the study of the neural basis of an animal’s natural behaviors. This term can be applied to traditional research animals, like rodents and primates. However, it is more commonly used to describe nontraditional research animals that have evolved unique adaptations to their ecological niches, making them particularly good model organisms for specific research goals. For example: l Echolocating bats have enlarged brainstem structures useful for studying the auditory system. l Various species of birds learn and practice song, allowing scientists to study developmental learning, consolidation of learning, and language. l Barn owls hunt in the dark and have an amazing capacity for sound localization, as well as coordinating visual and auditory information. l Some species of prairie voles exhibit monogamous behavior and an increased tendency, compared with other rodents, to stay in stable male/female pairs. Studying these animal models results in not only a greater understanding of the neurobiological basis of ethologically relevant behaviors, but also a greater understanding of human biology. For example, much of our understanding about the human auditory system has come from research on echolocating bats. 42  Guide to Research Techniques in Neuroscience example, animals injected with a pharmacological agent may take a statistically longer time to complete a maze than control animals injected with saline. However, this effect may not be due to problems in spatial navigation or memory, but altered motor behavior that indirectly increases the amount of time spent in the maze. Therefore, when choosing a behavioral assay, scientists must consider alternate explanations for why experimental groups may differ, adding additional behavioral experiments as necessary. It is also important to examine the overall health of the animal (body weight, heart rate, etc.) to ensure that animals do not perform abnormally because they are sick. Variability in Individuals There are many factors that can increase the variability of a model organism’s performance on a behavioral task, making it difficult to achieve consistent BOX 2.2  Ethical Considerations for Animal Use Using animal models in biological research is necessary to make conclusions about a fully functional, living organism. However, using animals is also a privilege, and scientists deciding to use animal models must consider many ethical considerations. These ethical concerns are not solely the responsibility of the individual scientist; the use of animals must also be justified to animal oversight committees that exist to maintain the welfare of research animals. Each institution has an Institutional Animal Care and Use Committee (IACUC) that must approve protocols involving animals and ensure that the research is justified. Furthermore, granting agencies and many scientific journals now require that scientists state that all animal protocols were approved by their institution’s IACUC. When using animals in experiments, distress must be minimized, and the potential benefits to humans must be maximized. Animal welfare guidelines throughout the world are based on the principles proposed by Russell and Burch in The Principles of Humane Experimental Research (1959), commonly referred to as the three Rs: Replacement: If there is any way to do the research without using animals, it should be done. This includes using computational modeling and cell culture methods. In the case of behavioral research, modeling is obviously impossible without actually performing the task. However, another component of replacement is the use of less-sentient animal species: choosing to study invertebrates rather than vertebrates, or mice rather than monkeys, etc. Reduction: Use the minimum number of animals to obtain scientifically valid data. If possible, use the same animals for multiple experiments. Also, investigators should perform mathematical analyses to determine the minimum number of animals required to provide enough statistical power for scientifically significant results. Refinement: Procedures must minimize distress and pain and enhance animal well-being. This is also important for the consistency and interpretability of experimental results; stress has numerous effects on physiology that can confound results. Following these guidelines is standard practice for developing animal protocols and justifying research goals. Animal Behavior Chapter | 2  43 results. One factor is an animal’s genetic background, which can have profound effects on behavior. Consider mice as an example: there are over 10 strains of mice commonly studied in research laboratories. Some strains behave very differently than others. For example, C57/BL6 mice are known to be much more aggressive than the FVB strain. An investigator performing social experiments on C57/BL6 mice might, therefore, observe different results than another investigator performing the same experiments using FVB mice. Strain-specific differences are also present in other model organisms, such as worms, flies, rats, and even primates. Therefore, it is important to always perform behavioral assays in the same strain and, if possible, the same generation of animal. Comparisons between genetically modified animals and wild-type animals should be performed on individuals from the same litter. Environmental variables can also impact behavior. These variables include animal handling, diet, circadian rhythms, seasonal changes, social interactions, etc. For example, an investigator may observe different behaviors on a Monday compared to a Thursday if it turns out that an animal technician changes the animal cages (and thus handles the animals) each Monday. Alternatively, an individual animal may perform differently on a behavioral assay if it is housed with other animals compared to if it is housed alone. These factors require a scientist to foresee and minimize any aspects of an animal’s environment that may introduce variability in behavior. Additionally, scientists may need to perform behavioral assays on large numbers of animals to produce reliable measurements and results. Using Animal Behavior as a Model for Human Behavior One of the primary goals of biomedical research is to learn more about humans by studying animal models. Therefore, it is worthwhile to ask the following: Can scientists extrapolate and generalize the results of an animal behavioral assay to people? Is it possible to model human neurological and psychiatric disorders, such as Alzheimer’s disease and depression, using rodents? The answers to these questions depend on the exact behavior and disease studied, but scientists generally gauge the ability of an assay to provide information about human behavior by examining the assay’s validity. There are three components to the validity of an animal experiment as a model of human behavior: face validity, construct validity, and predictive validity. A model has face validity if the model’s behavior/phenotype is similar to the analogous human behavior/phenotype. For example, mouse models of autism spend statistically less time engaging in social behaviors than wild-type mice, just as many humans diagnosed with autism have difficulty interacting socially with others. Construct validity refers to a state in which the animal model and human model have the same underlying genetic or cellular mechanism that may result in a certain behavior, such as a transgenic mouse engineered to have the same mutation that causes Huntington’s disease in humans. A model has predictive validity if treatments used in human patients have the same effect on the animal model. 44  Guide to Research Techniques in Neuroscience For example, some antidepressants prescribed to humans decrease behavioral measures of depression in rodent models. It is possible for animal models to exhibit one form of validity but not others. The more forms of validity that an animal model exhibits, the greater its relevance and value to humans. Note that validity is not something that can be objectively measured, but a concept judged by scientists and their colleagues within various subfields of neuroscience. Now that we have discussed some general features of behavioral paradigms, we describe many commonly used assays in modern neuroscience research. RODENT BEHAVIORAL PARADIGMS Rodents are perhaps the most commonly used mammalian model organisms in behavioral neuroscience because they can be simultaneously used for molecular, genetic, and electrophysiology experiments. The following assays represent common approaches used by investigators to measure motor function, sensation, learning and memory, reward-related behaviors, social interactions, and emotion. Locomotor Activity Locomotion assays are used to determine the net motor activity of an animal over a given time period. These assays are useful for determining if two cohorts of animals (e.g., knockout animals compared to wild-type littermates) have the same baseline activity. If measurements are recorded continuously over days and weeks, these assays are also useful for studying circadian rhythms—the regular, roughly 24-h cycle of stereotyped biochemical, physiological, and behavioral activity. Running Wheel A running wheel is the simplest way to measure the locomotor activity of an animal over time. The wheel is coupled to a device that measures the total number of revolutions and the speed of each revolution. If the animal is free to enter or exit the running wheel voluntarily (as occurs in most running wheel assays), this assay cannot measure locomotor activity when the animal is not on the wheel. However, a running wheel exerts significantly more energy than simple movements a rodent may perform around its cage, and therefore running wheels can serve as reliable measurements of deliberate activity. Home Cage Activity Locomotor activity can be measured in specially designed cages that project infrared beams from one side of the cage to the other. Each time an animal moves around the cage, it breaks the beam, and a computer records the time and position (Figure 2.1). Alternatively, if cages with infrared beams are not available, a scientist can place a camera above the cage and record activity. Computer Animal Behavior Chapter | 2  45 tracking programs can be used to statistically analyze the total locomotor activity over time. Depending on the specific setup, a scientist can examine horizontal activity, vertical activity, time spent in various regions of the cage, and total distance traveled. Open Field Locomotion Test An open field test utilizes a large cubic box, usually measuring 1 m long × 1 m wide × 1 m high (Figure 2.2). The top of the cube is typically left uncovered. An animal is placed in the middle of the bottom surface, and its movements are recorded over the course of minutes to hours as it moves around and explores its environment. After the experiment is completed, computer tracking programs analyze the movements of the animal over time. This assay can measure horizontal activity, time spent in various regions of the open field, and the total distance traveled. This assay can also be used to measure anxiety (described later in assays of stress). Motor Coordination and Balance Assays for motor coordination and balance are used to test the strength of an animal’s motor and vestibular systems. These tests can be used in isolation or to exclude motor deficits as a reason for poor performance in other behavioral experiments. For example, if pharmacological or genetic manipulation of an animal results in abnormal performance on a learning and memory assay, a scientist could ensure that this abnormal behavior was due to impairments other than general motor deficits. FIGURE 2.1  Locomotor activity can be monitored in a specially designed home cage. Courtesy of the Stanford Behavioral and Functional Neuroscience Laboratory. 46  Guide to Research Techniques in Neuroscience Rotarod A rotarod is a device consisting of a cylindrical rod suspended above a padded base (Figure 2.3). A scientist places a rodent on the rod and gradually accelerates the rotation until the rodent falls to the bottom of the chamber. Most animals are able to maintain balance for many minutes, but rodents with deficits in motor coordination or balance fall more quickly. This assay can also be used as a measure of motor learning, as most animals perform better on the rotarod assay after consecutive training days. Footprint Pattern Assay A scientist performs a footprint pattern assay by dipping an animal’s paws in inkandallowingittowalkdownapaper-linedtunnel.Thesefootprintscanreveal gait abnormalities and ataxia—abnormal muscle coordination (Figure 2.4). Dipping the forepaws and hind paws in different colors of ink allows a scientist to measure a variety of gait parameters: distance between each stride, changes in stride length, variability around a linear axis, width between left and right hind paws, regularity of steps, and overlap between forepaws and hind paws. Modern automated versions of this task can also track pressure and walking speed. Ataxic gait is characterized by a highly variable stride length and path. Hanging Wire Assay The hanging wire assay is used to measure neuromuscular deficits. An investigator places a rodent on a cage lid made of a wire grid, shaking the lid, so the FIGURE 2.2  Exploratory locomotor activity can be monitored in an open field testing chamber. Courtesy of the Stanford Behavioral and Functional Neuroscience Laboratory. Animal Behavior Chapter | 2  47 rodent grips the wire. The investigator then gently flips the lid, so the rodent hangs upside down. To prevent falling, the rodent grasps the wire, requiring both balance and grip strength. The investigator measures the time it takes before the rodent falls to a padded surface. A normal rodent can hang upside down for several minutes, but a 60-s cutoff is usually used for experimental purposes. Vertical Pole Test In a vertical pole test, a rodent is placed on the center of a horizontal pole wrapped in cloth to provide traction. The pole is gradually lifted toward a FIGURE 2.3  A rotarod tests balance and motor coordination.  Animals are placed on a rotating rod (red arrow) until they eventually fall onto a padded floor (blue arrow). Multiple animals can be tested at one time with a device that has individual lanes to record each animal’s latency to fall. Courtesy of the Stanford Behavioral and Functional Neuroscience Laboratory. 48  Guide to Research Techniques in Neuroscience vertical position, and the investigator records the amount of time and the angle at which the animal falls off. Normally, rodents can move up and down the pole without falling before the pole reaches a 45° angle. Sensory Function Abnormal sensory function is usually more difficult to observe than abnormal motor function, as it is impossible to directly measure what any animal perceives. In many sensory assays, a rodent uses motor output to report detection of a sensory stimulus; thus, these assays are sensitive to confounding effects due to deficits in motor function. Visual Cliff Assay A visual cliff assay evaluates the ability of a rodent to see the drop-off at the edge of a horizontal surface. The edge of a box or table can serve as a horizontal plane with a vertical drop to a lower horizontal plane, such as the floor. Each surface is coated with a black-and-white checkerboard pattern to enhance the detection of visual planes (Figure 2.5). A clear sheet of Plexiglas is placed over the drop-off, extending across the cliff so an animal cannot actually fall. Normally, rodents visually detect the drop-off and approach the edge cautiously, stopping at the cliff. Blind rodents approach the cliff but continue walking, unable to see the edge but able to feel the Plexiglas sheet. Rodents with normal vision can also use their whiskers to feel the Plexiglas extension, so the whiskers are often cut prior to an experiment. FIGURE 2.4  Footprint pattern analysis can reveal gait abnormalities. Footprint patterns of a wild-type (+/+) mouse compared to a mutant (stg/ stg) showing a wider stance (proximity of blue forelimb and black hindlimb footprints) and stumbling (red arrow). Reprinted from Meng, H., et al. (2007). BDNF transgene improves ataxic and motor behaviors in stargazer mice. Brain Research, 1160, 47–57, with permission from Elsevier. +/+ stg/stg Animal Behavior Chapter | 2  49 Startle Response Assay In a startle response assay, a scientist exposes a rodent to an unexpected and disruptive sensory stimulus and measures the degree to which the animal responds, typically exhibited as an eye blink, body flinch, or overall muscle contraction. These stimuli can be visual (a bright light), auditory (loud and disrupting noise), or tactile (unexpected touch or air puff). Investigators can either observe and score the number of times a startle response occurs or use specialized sensors, such as an electromyogram, to more quantitatively measure the degree of the response. This assay can be confounded by deficits in emotional and motor components of the startle reflex circuitry. This assay can also be used to measure animal learning by measuring extinction of the startle response over successive trials (Box 2.3). Taste Investigating an animal’s ability to taste usually involves a choice assay, in which different gustatory compounds are placed in different solutions in water bottles. The scientist measures the volume of water in each bottle to determine the animal’s taste preferences. It is possible to investigate whether an animal can detect bitter tastes by placing quinine, an extremely bitter compound, in a water solution and determining if the animal consumes less water than a solution with no quinine. Likewise, it is possible to investigate whether an animal can detect sweet tastes by placing sucrose or saccharine in a solution and determining if an animal drinks more water than a solution with no sweet compounds. FIGURE 2.5  Visual cliff assay. A checkerboard pattern enhances the perception of a drop-off. Rodents that can see this drop-off will pause at the edge, while blind rodents will feel the Plexiglas surface and continue walking without pause. 50  Guide to Research Techniques in Neuroscience Olfaction A scientist can test for normal olfactory processing by measuring how long it takes for an animal to find a piece of familiar, palatable food (like a cookie, cheese, or chocolate chip) hidden within the cage. Nociception Nociception is the ability to detect a noxious stimulus, usually perceived as pain. Nociceptive assays rely on physical indicators of discomfort, such as withdrawal reflexes, licking, and vocalizations. These assays are usually used to investigate either the neural basis of pain or the therapeutic potential of analgesic drugs. Genetically or pharmacologically manipulated animals that exhibit decreased sensitivity to pain must be monitored closely during these assays to prevent tissue damage induced by nociceptive stimuli. Tail Flick Assay In the tail flick assay, a high-intensity beam of light is aimed at a rodent’s tail. In normal animals, the beam of light produces a painful heat sensation, causing a reflex that moves the tail. The investigator measures the amount of time it takes before the tail flicks to the side. Sex, age, and body weight can affect a rodent’s response, so all animals used for these experiments (experimental and control animals) must be similar to avoid confounding results. Hargreaves Assay The Hargreaves assay is similar to the tail flick assay, but the high-intensity beam of light is aimed at the rodent’s hind paw rather than the tail. The investigator measures the time it takes for the rodent to withdraw its paw. Hot Plate Assay In the hot plate assay, a scientist places a rodent on a heated surface and prevents the animal from leaving by enclosing it within a tall cylinder (Figure 2.6). BOX 2.3  Prepulse Inhibition Interestingly, a startle response can be reduced if a nonthreatening stimulus is presented immediately before the disruptive sensory stimulus. This phenomenon is known as prepulse inhibition (PPI) and reflects the nervous system’s ability to prepare for a strong sensory stimulus after a small warning (the prepulse). Human patients with attention deficit disorder, Alzheimer’s disease, and schizophrenia show deficits in PPI. Therefore, PPI can be a useful paradigm to test animal models of psychiatric disease and establish face validity for animal experiments. Furthermore, PPI has been a useful behavioral assay to test potential antipsychotic medications that improve sensory processing. Animal Behavior Chapter | 2  51 The surface is calibrated so that a normal animal will react within about 10 s of exposure (usually 52–53 °C). The investigator records the latency and amount of time an animal reacts to the heat stimuli. Individual reactions vary: licking a rear paw is a reliable indicator of discomfort, though some animals will jump or vocalize. Von Frey Assay A Von Frey assay is used to examine the sensitivity to pinch and mechanical stimuli. Von Frey hairs are fine-gauge metal wires. The investigator pokes the hind paw of an animal standing on an elevated mesh platform by inserting a Von Frey hair through the mesh from below. Normal rodents usually react by withdrawing or licking their paws and possibly vocalizing. Formalin Assay A formalin assay is used to examine an animal’s sensitivity to noxious chemical stimuli.An investigator injects a small volume of formalin, a noxious chemical, into a rodent’s hind paw and records the total amount of time the rodent spends licking, biting, or withdrawing the affected limb. This activity is usually assessed in two phases: one phase starting immediately after the injection and lasting about 10 min and another phase starting 20 min after the injection and lasting about an hour. The second phase is a response to tissue damage caused by inflammation. FIGURE 2.6  Hot plate test of pain perception. Courtesy of the Stanford Behavioral and Functional Neuroscience Laboratory. 52  Guide to Research Techniques in Neuroscience Spatial Learning and Memory Spatial learning and memory are ethologically relevant behaviors to a mouse or rat. In the wild, these animals spend most of their lives locating good nesting spots, searching for food, and avoiding predation. Therefore, rodents are great model species to assay for spatial navigation and memory. Morris Water Maze The Morris water maze is one of the most frequently used spatial learning and memory paradigms. Rodents are placed in a large circular pool of opaque water (Figure 2.7). Their natural dislike of water makes rodents highly motivated to escape from the pool. In initial trials, a visible platform is placed within the pool so that an animal can emerge from the water. A scientist places visual cues around the pool so the animal can relate the spatial environment with the location of the platform. In subsequent trials, the platform is hidden just below the surface of the water, and a scientist measures the time required for the rodent to swim to the platform to escape the water. In the first few trials, the rodents become progressively better at swimming directly toward the platform, demonstrating their ability to learn the platform’s location. To test memory at the end of training, the investigator completely removes the platform to measure the amount of time the animal spends searching in each quadrant of the pool. This experiment measures the ability of the animal to identify a spatial location based on the visual cues placed around the chamber. Because water is stressful to rodents (in mice FIGURE 2.7  The Morris water maze is one of the most common rodent tests for learning and memory. Courtesy of the Stanford Behavioral and Functional Neuroscience Laboratory. Animal Behavior Chapter | 2  53 more than rats), it is important for a scientist to account for stress-induced performance effects. Barnes Maze The Barnes maze consists of a circular table with holes around the circumference, placed in a room with visual cues in the periphery (Figure 2.8). Most of these holes lead to an open drop to the floor, but a single hole leads to a drop box, a dark box in which the animal can hide. A rodent is naturally motivated to avoid open spaces and bright lights, and therefore attempts to find the drop box. In initial trials, the scientist gently leads the animal to the drop box. In subsequent trials, the animal is placed in the center of the table and must find the drop box on its own. After a few trials, rodents typically remember which hole contains the drop box and quickly proceed in a direct path toward the hole. Investigators can measure the amount of time to find the correct hole, the number of incorrect holes explored, and the length of the exploratory path. The Barnes maze is considered to be less stressful than the Morris water maze. Radial Arm Maze The radial arm maze consists of an array of arms, usually eight or more, which radiate from a central starting point (Figure 2.9). At the end of each arm is a cup that may or may not contain a food reward. Animals are trained to recognize that only one of the arms will contain food. Investigators measure the amount of time it takes for animals to find the arm leading to food, as well as the number of times it traverses an arm it has previously visited. Exploring a previously visited arm FIGURE 2.8  The Barnes maze. Rodents learn which one of the holes contains a hidden drop box they can use to escape from light. Courtesy of the Stanford Behavioral and Functional Neuroscience Laboratory. 54  Guide to Research Techniques in Neuroscience indicates that the animal did not remember previously choosing that spatial path. This is a relatively difficult task for rodents, requiring several days or weeks to train rats and many weeks to train mice. Unlike the Morris water maze and the Barnes maze, the radial arm maze does not use distant visual cues to aid spatial learning. Nonspatial Learning and Memory The neural circuitry that mediates nonspatial learning and memory differs from circuitry that mediates spatial learning and memory. Thus, multiple assays using both nonspatial- and spatial-based tasks are used to evaluate learning and memory as a whole. Nonspatial learning and memory assays include innate learning, such as motor performance improvement over time (such as the rotarod assay), novel object recognition, as well as learned associations through conditioning. While classical (Box 2.4) and operant (Box 2.5) conditioning can be used to directly assay an animal’s ability to learn, these paradigms are often used to train and test animals on more sophisticated tasks. For example, the delayed match to sample and nonmatch to sample assays use operant conditioning techniques to examine an animal’s working memory. These techniques are described below. Novel Object Recognition The novel object recognition paradigm assesses an animal’s innate ability to distinguish an old from a new object. Rodents are naturally curious and will take time to explore novel objects. When presented with both a new object and FIGURE 2.9  Radial arm maze. Courtesy of the Stanford Behavioral and Functional Neuroscience Laboratory. Animal Behavior Chapter | 2  55 a previously explored object, normal rodents remember the old object and spend relatively more time exploring the new object. A scientist measures the amount of time spent interacting with both objects. Delayed Match to Sample/Nonmatch to Sample Operant conditioning (Box 2.5) can be used to train an animal to make a response such as poking its nose into a hole or pressing a lever for a reward in response to a stimulus, often a light appearing above the lever. To test working memory, the investigator trains the rodent to perform an operant response under specific conditions—for example, pressing a specific lever when multiple levers are present. In a match to sample task, a light appears above one lever, BOX 2.4  Classical Conditioning In classical conditioning (also called Pavlovian conditioning), the investigator couples an initially neutral stimulus, such as a tone or light, with a salient stimulus, such as food or an electric shock. The neutral stimulus is referred to as the conditioned stimulus (CS), while the salient stimulus is referred to as the unconditioned stimulus (US). The unconditioned stimulus elicits a reflexive response, such as salivation or fear behaviors. By pairing the conditioned and unconditioned stimuli, the conditioned stimulus alone is eventually able to elicit the reflexive response of the unconditioned stimulus. Examples of Pavlovian conditioning are cued fear conditioning and contextual fear conditioning. Fear conditioning often utilizes a foot shock (US) to induce freezing behavior. By associating an auditory or visual cue (CS) with the foot shock, animals will freeze in response to the sensory cue alone (i.e., without any foot shock) demonstrating a learned association. Contextual fear conditioning measures freezing behavior when the animal is returned to the training chamber without foot shock—exhibiting a response without the unconditioned stimulus shows that the animal has learned the association between a location and the shock. Cued conditioning should be measured in a new testing chamber with a very different context (shape of the box, visual cues, lighting, odors, textures) but with the same auditory or visual cue from training to test the specific learned association between the cue and the shock. BOX 2.5  Operant Conditioning In operant conditioning, a voluntary response (e.g., a lever press or a nose poke) is induced by positive reinforcement or suppressed by punishment. Positive rewards often include food or water (especially in food- or water-deprived animals) or drugs of abuse. Punishing or aversive stimuli often include foot shocks or an unpleasant sensory stimulus such as a loud tone. Several training sessions are often required for animals to learn the association between performing a task and receiving reinforcement or punishment. While operant conditioning is a form of learning, this paradigm is often used to study other behaviors, such as reward, attention, and impulsivity. 56  Guide to Research Techniques in Neuroscience and the animal must choose to press the lever beneath that light. In a nonmatch to sample task, a light appears, and the animal must choose to press the opposite lever. After learning these tasks, the scientist adds artificial delays, such as withdrawing the lever, between the time when a cue appears and the time that an animal responds. The longer the delay, the longer the animal must remember the light cue that was presented, thus testing the animal’s working memory. Attention and Impulsivity Rodent models of attention and impulsivity test an animal’s ability to learn a relatively difficult task in which success is dependent on sustained attention to a cue and reduced desire to impulsively engage in behavioral responses. These tasks are performed using rats more often than mice because often mice have difficulty learning the task paradigms. Go/No-Go Task A go/no-go task is a form of operant conditioning (Box 2.5) in which a rodent is trained to produce a response, such as a lever press, in the presence of a specific go cue (such as a green light), yet withhold a response in the presence of a separate no-go cue (such as a red light). Rodents are relatively successful on the go trials but often have difficulty withholding responses on the no-go trials. These success rates are exacerbated in animal models of impulsivity or low attention span. Five Choice Serial Reaction Time Task In the five choice serial reaction time task, an animal is placed in an operant chamber (Box 2.5) with five lights coupled with five trays capable of delivering a reward, such as a food pellet. At the beginning of a trial, one of the lights is briefly illuminated, and the rodent must indicate via nose poke or lever press which light was on. Controlling the length of time the light is on prior to requiring the animal to produce an action can vary the difficulty of the task. Furthermore, there is an intertrial interval during which an animal must withhold all responses, demonstrating inhibitory control and reduced impulsivity. Reward-Related Behaviors Scientists can measure the rewarding aspects of a stimulus, such as an addictive drug or food, by measuring how much an animal is willing to work to obtain the reward or the degree to which the animal associates the stimulus as a positive reinforcement. These assays are frequently used to study the neurobiology of addiction, as well as motivational behaviors such as hunger and thirst. Self-Administration Self-administration is a form of operant conditioning (Box 2.5) in which animals are trained to perform a particular action, such as pressing a lever, to Animal Behavior Chapter | 2  57 receive a rewarding stimulus, such as an addictive drug. Animals are free to perform the action freely and repetitively, demonstrating the intrinsically rewarding properties of the stimulus. In a continuous reinforcement paradigm, an animal performs a single action to receive a rewarding stimulus, whereas in a fixed ratio paradigm, an animal performs a set amount of actions (such as five lever presses). In a progressive ratio paradigm, the animal must perform an action more and more to receive the rewarding stimulus (for example, pressing a lever 1 time, then 2 times, then 4 times, then 8 times, etc.), and the scientist measures the breakpoint at which the animal no longer works to receive the reward. Conditioned Place Preference The conditioned place preference assay is a form of classical conditioning (Box 2.4) in which a rodent associates a particular environment with a rewarding experience. An animal is placed inside a test chamber (the conditioned stimulus) and allowed access to an unconditioned stimulus (a drug, food, odor, etc.) for minutes each day for multiple days. The animal is also placed in a control chamber without any stimulus present for minutes each day for multiple days. On the test day, the animal is allowed free access to both chambers, and the scientist measures the amount of time the animal spends in one chamber relative to the other. If animals spend significantly more time in the chamber with the unconditioned stimulus, they are said to have developed a conditioned place preference, associating that chamber with the reward. Alternatively, the unconditioned stimulus could be relatively unpleasant for the animals and could lead to conditioned place aversion. Social Behaviors Rodents display a variety of innate social behaviors. Mating, parenting, nesting, and grooming are all innate behaviors that can be observed within an individual’s cage. Rodents also exhibit specific behaviors based on social recognition when multiple animals are exposed to one another. Exploratory sniffing allows animals to decide on an appropriate social response, such as defending territory or attempting copulation. Two standard assays are well represented in the literature: the resident–intruder assay and social approach/avoidance assays. Resident–Intruder Assay The resident–intruder assay measures territorial behavior in males. A scientist adds an intruder animal to the cage of a resident animal and measures specific aggressive behaviors: tail rattling, quivering or thrashing of the tail, physical attacks such as biting or clawing, wrestling, and chasing.The investigator can then compare the amount of time each animal spends investigating the other (following, sniffing, grooming) compared to the amount of time displaying aggressive behavior. 58  Guide to Research Techniques in Neuroscience Social Approach/Avoidance If a scientist presents a rodent with a choice between interacting with another (unfamiliar) rodent or interacting with an inanimate object, most animals will spend more time interacting with the other rodent, a phenomenon referred to as social approach. Animals with abnormal social behaviors may spend relatively little time with the other animal, referred to as social avoidance. For example, in some mouse models of autism, there is no difference in the amount of time spent exploring the object and exploring the other mouse. These behaviors are thought to be analogous to the asocial behaviors exhibited by humans diagnosed with autism. Anxiety Paradigms that measure anxiety generally involve placing rodents in naturally stressful environments, such as brightly lit, open spaces. These assays are useful for investigating brain regions thought to modulate anxiety, as well as testing novel anxiolytic drugs hypothesized to reduce anxiety. In addition to the common assays described here, investigators can also measure physiological correlates of anxiety, such as heart rate, defecation and freezing behavior. Open Field Test The open field test (Figure 2.2), previously described as an assay of locomotor activity, can also be used to measure anxiety. A rodent placed in a bare, open chamber will initially stay near the walls and avoid the center (Figure 2.10). Normal animals typically acclimate to the chamber and eventually explore the center area. More anxious animals spend significantly less time in the open area and more time closer to the walls. The investigator can monitor the animal’s activity with a camera placed above the open field, and computer software measures time spent in the center compared with the periphery. Anxiolytic drugs increase the amount of time rodents spend in the open area. (A) (B) FIGURE 2.10  The open field test as a measure of anxiety.  (A) After being acclimated to the chamber, rodents will normally explore the entire area, but (B) a rodent model of anxiety will stay near the perimeter of the chamber. Animal Behavior Chapter | 2  59 Elevated Plus Maze An elevated plus maze is a four-armed platform resembling the shape of a plus sign positioned two to three feet above the ground (Figure 2.11). Two arms of the maze have sides, and two do not. A scientist places an animal in the center of the maze and allows it to move freely in any of the arms. Normal animals avoid the open arms, spending relatively more time in the protected, closed arms. Anxiolytic drugs increase the proportion of time rodents spend in the exposed arms. Defensive Marble Burying Rodents tend to bury objects present in their environment, such as glass marbles, within the bedding of their cages. Animals that are more anxious tend to bury more objects in the 30–60 min after they are introduced. To perform a defensive marble burying assay, a scientist places 10–20 marbles in an animal’s cage and quantifies the number of marbles buried after 30 min. Anxiolytic drugs tend to decrease the number of marbles buried over time. Geller–Seifter Conflict Test The Geller–Seifter conflict test is one of the oldest assays to study anxiety and remains a robust test of anxiolytic drug effects. A scientist uses operant conditioning (Box 2.5) to train an animal to press a lever for a food reward. During the experiment, the investigator pairs a lever press with an unpleasant electrical shock. Thus, the animal must decide whether to receive food while getting FIGURE 2.11  The elevated plus maze. Courtesy of the Stanford Behavioral and Functional Neuroscience Laboratory. 60  Guide to Research Techniques in Neuroscience shocked or not receive food at all. Anxious animals tend to press the lever significantly fewer times than nonanxious animals. Anxiolytic drugs (but not other psychoactive drugs) increase the number of lever presses. Depression How is it possible to measure depression in a mouse or rat? Paradigms that investigate depression in animal models assay for the feelings of hopelessness and despair experienced by humans diagnosed with depression. Scientists can induce depression in a rodent by applying conditions of learned helplessness and chronic mild stress. The learned helplessness paradigm exposes an animal to aversive stimuli at random intervals. Theoretically, this treatment creates a condition in which the animal experiences a lack of control, exhibiting symptoms of behavioral despair. Applying conditions of chronic mild stress (leaving the lights on all the time, tilting the animal’s cage, wetting the animal’s bedding, applying intermittent air puffs) over a period of days to weeks can also theoretically lead to conditions of behavioral despair and helplessness. Both learned helplessness and chronic mild stress cause differences in behavior on the assays of depression described here. Importantly, antidepressants reverse these behaviors, demonstrating the predictive validity of these animal models. Forced Swim Test (Porsolt Test) In the forced swim test (also called the Porsolt test), rodents are placed in a small, confined space, such as a large graduated cylinder filled halfway with water (Figure 2.12). Initially, there is a period of vigorous activity during which the animal triestoescape.Eventually,theanimalceasesvigorousactivityandexhibitsacharac- teristicimmobilityinwhichitonlymovestomaintainitsheadabovewater.Thisphysical immobility is thought to be an indication of behavioral despair. Investigators measure the amount of time between when the animal is placed in the chamber and the onset of immobility. Rodent models of depression exhibit a decrease in the time spent trying to escape, but this decrease is reversed with antidepressants. Tail Suspension Assay In the tail suspension assay, either a device or the investigator holds the rodent in the air by its tail. The animal’s natural reaction is to move vigorously to escape. Depressed animals give up more quickly than nondepressed animals. Antidepressants increase the amount of time the animals struggle. An advantage of this paradigm over the forced swim test is that tail suspension is less stressful to the animal because there is no exposure to water. Sucrose Preference Test The sucrose preference test is hypothesized to model anhedonia—the loss of desire to pursue an activity that was once enjoyable. Normally, rodents that Animal Behavior Chapter | 2  61 are allowed to choose between drinking tap water or water with sucrose will choose the water with sucrose. Rodent models of depression have a significantly reduced preference for water with sucrose. Importantly, this effect is reversed by administration of antidepressants. DROSOPHILA BEHAVIORAL PARADIGMS One of the powerful advantages of using Drosophila melanogaster, the fruit fly, as a model organism is its genetic tractability. Drosophila has proven to be very useful for studying the genetic basis of behavior due to the coupling of many genetic tools with a variety of fly-specific behavioral assays. The lab of Seymour Benzer famously used Drosophila to study genes necessary for learning, memory, vision, circadian rhythms, nociception, and many other behaviors. Many other labs have used Drosophila to study sleep, addiction, courtship, and aggression. As with any animal model used to study behavior, age, diet, and environmental factors must be strictly controlled when performing behavioral experiments to attain consistent results. If experiments are performed over multiple days, the scientist should perform assays at the same time each day to avoid FIGURE 2.12  Performance on the forced swim (Porsolt) test is used as a test of behavioral despair to model depression in rodents. 62  Guide to Research Techniques in Neuroscience potentially confounding results due to circadian rhythms. Applying anesthesia or handling the flies just prior to behavioral tests can also affect the outcome. Locomotor Behavior Locomotor behavior in flies can be studied similarly to locomotor behavior in rodents. A scientist can project infrared beams of light through a housing chamber, allowing flies to break the beams and thus indicate their movement over time. Alternatively, a scientist can record a housing chamber with a video camera, and tracking software can record the activity of many flies over time. Locomotor behavior is useful when studying motor function, circadian rhythms, and the effects of alcohol and other drugs. Flight Traditional locomotor assays measure movement in an enclosed container. Another ethologically relevant movement for a fly is flight. Therefore, scientists have developed specialized flight simulators that suspend a fly from a thin pin and project visual stimuli to depict the external, visual environment (Figure 2.13). These simulators contain torque meters to record turning responses to various stimuli, allowing analysis of sensory and motor reflexes. IR laser diode Torque meter Light source Light guides Diffusor Electric motor Yaw torque signal FIGURE 2.13  A Drosophila flight simulator. This device can be used to measure free-flight behaviors or flight responses to visual stimuli, as well as to train flies by using operant conditioning methods that use a laser to generate heat reinforcement. Courtesy of Dr Björn Brembs. Animal Behavior Chapter | 2  63 These flight simulators can also be used to study sensorimotor processing and spatial learning and memory. Sensory Function Sensory modalities commonly studied in the fly include vision, olfaction, and taste. These modalities are all ethologically relevant for a fly, which spends most of its life navigating through space using visual and chemical cues in search of food or responding to chemical signals produced by other flies. Vision Flies exhibit a natural tendency to move toward light, a phenomenon known as phototaxis. To test proper visual function, scientists can use a device called a countercurrent apparatus, made up of a series of test tubes (Figure 2.14). Flies are placed in one test tube, which is gently tapped so that the flies fall to the bottom. The test tube is then laid horizontally across from a second test tube. This second tube is illuminated with a fluorescent light. The scientist waits approximately 15 s before preventing the flies from crossing between the two tubes. The flies trapped within the second tube are considered to have demonstrated more of a phototaxic response. This process is repeated several times, with flies demonstrating a phototaxic response being used for the next trial. The end result is a fractionated population of flies that vary in their phototaxic responses. Vision can also be assayed using a device called a T-maze (Figure 2.15). This apparatus consists of two arms oriented in opposite directions—a “T” shape—with an additional loading arm. Flies are placed into the loading arm and moved to the center of the other two arms of the T-maze, where they can travel to either arm. Each arm contains a different visual stimulus—for example, two different wavelengths of light, bright light versus darkness, flashing light versus constant light, and so forth. Thus, a T-maze allows scientists to understand the visual preferences of a fly, as well as investigate the genes necessary and/or sufficient for these preferences to occur. FIGURE 2.14  The countercurrent apparatus can be used to isolate a population of phototaxing flies. 64  Guide to Research Techniques in Neuroscience Olfaction Olfaction assays can also take place in a T-maze. Each arm of the maze contains a different odor, and flies decide which odor to move toward. Investigators can use this assay to determine whether an odor is attractive, repulsive, or neutral. In an olfactory trap assay, odors absorbed on a cotton swab or filter paper are placed at the bottom of a chamber with a pipette tip that allows flies to get in but not get out. By simply waiting a certain amount of time for flies to experience the baited odor, an attraction index can be calculated from the ratio of flies trapped in the odor vial compared to a water control. An attraction index of 1 indicates complete attraction to the odor, while an attraction index of 0 indicates either no preference for an odor or anosmic flies. There are many simple avoidance assays that assess behavioral responses to olfactory cues. In the olfactory avoidance (also called dipstick) assay, a scientist inserts a piece of filter paper or cotton swab containing a specific odor into a fly vial, then measures the distance that the fly maintains from the offending odorant. A similar assay is the chemosensory jump assay or olfactory jump response, which exploits the tendency of flies to exhibit a startle response when encountering a novel odor. Taste Hungry flies exhibit an unconditioned reflex called the proboscis extension response. A scientist can elicit this response by applying taste ligands to FIGURE 2.15  A Drosophila T-maze can be used to test fly sensory preferences or to train and test flies on conditioning training. Animal Behavior Chapter | 2  65 gustatory receptor neurons on the leg or proboscis. This reflex can be used to test specific gustatory receptor responses and sensitivity to different taste ligands. A scientist can also assay taste using a feeding acceptance assay. Starved flies are provided a choice between appetitive, aversive, or neutral stimuli, each dyed a different color. The amount of each stimulus ingested can be scored by examining the color of the fly’s stomach. Learning and Memory Learning and memory assays can be performed in Drosophila using a T-maze and classical conditioning. An odor is paired with an electric shock. This odor is then presented in one arm of a T-maze. Flies that avoid this arm are considered to have learned the association. This assay has been widely used in genetic screens that assay both learning and memory, as well as olfaction. Investigators can also assay learning and memory on individual flies in the flight simulator through classical or operant conditioning. For example, flies can be trained to navigate a virtual scene to find an appetitive sensory stimulus or reward. Social Behaviors There are many social behaviors that a scientist can study using Drosophila. Many of these behaviors are innate (as opposed to learned), so they can be used to investigate the specific genes necessary and/or sufficient for these behaviors to occur. Perhaps the two most commonly studied social behaviors in the fly are courtship and aggression. Courtship Flies exhibit a stereotyped courtship ritual composed of several distinct stages: (1) orientation, (2) tapping, (3) wing song, (4) licking, (5) attempted copulation, and (6) copulation. Because these behaviors are innate, scientists can use them to study the genetic basis of a hard-wired behavior. Groups of male and/or female flies can be placed into tiny chambers and recorded with a video camera. Later, an investigator scores the behavior, measuring a variety of parameters related to the timing, performance, and progression of these different stages. Aggression To study aggression, a scientist places two male flies in a chamber. The flies naturally begin to fight, and the scientist can record many different parameters: which fly initiated the fight, the outcome, the interfight interval, the order of events of each fight, etc. CAENORHABDITIS ELEGANS BEHAVIORAL PARADIGMS The relatively simple and well-understood neural circuitry of Caenorhabditis elegans, coupled with its genetic tractability, make this species a powerful 66  Guide to Research Techniques in Neuroscience animal model to study the genetic basis of behavior. In fact, many genes in the worm are named after the behavioral phenotypes that led to their discovery: Unc (uncoordinated) mutants have disrupted locomotion, Egl (egg-laying) mutants have disrupted egg-laying, and Mec (mechanosensation) mutants have disrupted responses to touch. Most behavioral assays in C. elegans examine motor and sensory functions. Locomotor Behavior Locomotor behavior in the worm can be automatically tracked using computer software, such as Worm Tracker (Figure 2.16). This software allows for a detailed analysis of the sinusoidal movement of worms, allowing scientists to determine the genetic and cellular components that make these movements possible. Sensory Behavior Caenorhabditis elegans has proved very useful for investigating sensory transduction, as worms depend on detecting mechanical, thermal, and chemical stimuli to stay alive and find food. Mechanosensation A gentle touch with a fine hair to the body, head, or tail of a worm results in different behavioral responses. For example, if touched on the nose, worms move Digital video camera To PC Zoom lens Worm arena Transmitted-light source Time (s) Speed(mm/s) FIGURE 2.16  The Worm Tracker provides detailed information about the movements of individual worms.  Representative tracks generated by the worm tracker can indicate starting position (x), path length, speed, and turning events (circles). Reprinted from Ramot, D., et al. (2008). The parallel worm tracker: a platform for measuring average speed and drug-induced paralysis in nematodes. PLoS ONE, 3(5), e2208, under the Creative Commons Attribution License. Animal Behavior Chapter | 2  67 backward. If touched on the body, worms immediately stop and sometimes move away. Because the complete circuitry diagram of C. elegans is known, it is possible to use these behavioral assays to map functional connections between different neurons in different locations of the worm’s body. Thermosensation Worm behavior is strongly influenced by environmental conditions—temperature, population density, and feeding status. Animals avoid temperatures at which they have been starved, and accumulate at temperatures associated with food. This requires thermosensation, as well as a memory of the associations. There are various methods for creating thermal gradients on which to test worms: an example of a simple method is to place frozen glacial acetic acid on one end of a plate and an incubator on the other end. A more sophisticated method is to use a thermoelectric device to maintain a steady heat gradient across a plate. Investigators can then measure thermotactic migration by placing worms on the plate and identifying the temperature where migration stops. Chemosensation Chemosensation can be thought of as a worm’s sense of taste or smell. A scientist can assay attraction or avoidance of specific compounds by dissolving these compounds in agar and placing them on opposite sides of a Petri dish. Worms are placed at the intersection of the two chemosensory cues, and the scientist measures the number of worms that migrate to each compound. NONHUMAN PRIMATE BEHAVIORAL PARADIGMS Nonhuman primates can be used to study the neural basis of complex motor actions, sensory perception, and cognitive processes such as decision making, attention, and complex learning. Experiments almost always couple electrophysiological recordings with behavior so that scientists can correlate the firing of neurons with distinct behavioral events. In addition to simply measuring neural activity, it is possible to manipulate neural activity using electrical and pharmacological methods (Chapter 8). One of the greatest strengths of primates as research models is their genetic, anatomical, and behavioral similarities to humans, allowing scientists to extend results to theories about how the human brain works. Primates are also capable of more complex behaviors and tasks than rodents and other animal models. In fact, many of the specific tasks performed by nonhuman primates are the same or adapted versions of tasks performed by humans in cognitive neuroscience or psychology studies. The most commonly used monkeys in research labs include rhesus macaques (Macaca mulatta) and cynomolgus (also known as crab-eating or long-tailed) monkeys (Macaca fascicularis). These animals can be very expensive to purchase and maintain. Therefore, each scientist typically only uses one to three 68  Guide to Research Techniques in Neuroscience monkeys over a 3- to 5-year period. While other behavioral paradigms described in this chapter typically require large numbers of individuals to achieve statistical significance, experiments involving nonhuman primates typically only need to be performed on a minimum of two animals. This is primarily due to differences in what the scientist measures: in rodent assays, scientists typically measure quantifiable behaviors. In primate assays, scientists typically measure response properties of neurons. Therefore, statistical power is achieved not by the number of animals studied but by the number of neurons recorded. Primate behavioral/electrophysiological experiments usually follow a specific structure: an investigator designs a behavioral task, trains an animal to perform the task, completes a surgical procedure on the animal to allow access to the brain, conducts many experimental trials while recording from neurons in the brain, and analyzes data outside of the experiment to correlate activity with specific, behavioral events. There are no standard behavioral tasks for primates as there are for rodents and invertebrates. Each investigator designs tasks specific to his or her research question, creating stimuli to maximize the response properties of the neurons to be studied. However, there are many common elements to primate experiments: they usually involve a single primate, working alone in a chamber for a juice reward, staring at a visual stimulus on a screen (Figure 2.17). Sometimes animals are required to make eye movements to follow moving dots on a screen; other experiments require that monkeys follow a path with their fingers; while still others require a monkey to press a lever in response to a learned cue. Like different levels of a video game, each trial within a task is slightly different, with some trials more difficult than others. Once the animal learns how to play the game, an investigator can design dozens or hundreds of individual trials with different conditions. Training a monkey to perform a task is often the most time-consuming aspect of an experiment. Much of this time is dedicated to simply acclimating the monkey to research life: sitting in a chair, learning how to accept a juice reward, becoming comfortable working alone in a chamber, and so forth. Most training involves operant conditioning (Box 2.5), in which successful completion of a task is reinforced with a juice reward. Depending on the task, monkeys are trained to perform at an 80–90% success rate. To increase success, monkeys are often given restricted amounts of food and water before the task begins, so they are motivated to work for juice. Fixate Targets On Motion On Delay Period Go Cue Hold Target FIGURE 2.17  A primate behavioral task.  Here, the monkey has been trained to make saccadic eye movements in the direction indicated by overall movement of a random dot stimulus. The monkey fixates at a point on the screen where the motion stimulus will appear. The fixation point disappears indicating the monkey can saccade to the correct target to receive a reward. Courtesy of Rachel Kalmar. Animal Behavior Chapter | 2  69 During the actual experiments, a monkey performs many repetitions (trials), completing a task for juice rewards. The number of trials performed in a day is primarily determined by the monkey, which will only work for a certain amount of time. Like people, monkeys differ in the amount of work they are willing to do in one sitting. After the completion of the daily experiments, scientists return monkeys to their cages and analyze the data. Many investigators only record from primates every other day to allow time for data analysis. CONCLUSION This chapter has provided a framework for thinking about behaviors in commonly used animal models, surveying many specific assays frequently found in the literature. These techniques are only a fraction of the assays that investigators can use to elucidate the neural basis of behavior. Most published studies, especially studies using rodents, utilize many different behavioral assays in combination to confirm hypotheses and strengthen conclusions. SUGGESTED READING AND REFERENCES Books and Review Articles Bailey, K. R., Rustay, N. R., & Crawley, J. N. (2006). Behavioral phenotyping of transgenic and knockout mice: practical concerns and potential pitfalls. ILAR Journal, 47(2), 124–131. Buccafusco, J. J. (Ed.). (2009). Methods of behavior analysis in neuroscience (2nd ed.) Boca Raton, FL: CRC Press. Conn, P. M. (1993). Paradigms for the study of behavior. San Diego: Academic Press. Conn, P. M. (2008). Sourcebook of models for biomedical research. [Digital]. Totowa, NJ: Humana Press. Connolly, J. B., & Tully, T. (1998). Behaviour, learning, and memory. In D. B. Roberts (Ed.), Drosophila: A practical approach (Chapter 9) (2nd ed.) (p. 389). Oxford: IRL Press at Oxford University Press. pp. xxiv. Crawley, J. N. (2007). What’s wrong with my mouse? Behavioral phenotyping of transgenic and knockout mice (2nd ed.). Hoboken, NJ: Wiley-Interscience. Russell, W. M. S., & Burch, R. L. (1959). The principles of humane experimental technique. [Print]. London: Methuen. Silverman, J. L., Yang, M., Lord, C., & Crawley, J. N. (2010). Behavioural phenotyping assays for mouse models of autism. Nature Reviews Neurology, 11, 490–502. Whishaw, I. Q., & Kolb, B. (2005). The behavior of the laboratory rat: A handbook with tests. New York: Oxford University Press. Primary Research Articles—Papers First Describing Standard Behavioral Assays Barnes, C. A. (1979). Memory deficits associated with senescence: a neurophysiological and behavioral study in the rat. Journal of Comparative Physiological Psychology, 93, 74–104. Geller, I., & Seifter, J. (1960). The effect of meprobamate, barbiturates, d-amphetamine and promazine on experimentally induced conflict in the rat. Psychopharmacologia, 1, 491–492. Hargreaves, K., Dubner, R., Brown, F., Flores, C., & Joris, J. (1988).A new and sensitive method for measuring thermal nociception in cutaneous hyperalgesia. Pain, 32, 77–88. 70  Guide to Research Techniques in Neuroscience McKenna, M., Monte, P., Helfand, S. L., Woodard, C., & Carlson, J. (1989). A simple chemosensory response in Drosophila and the isolation of acj mutants in which it is affected. Proceedings of the National Academy of Sciences of the United States of America, 86, 8118–8122. Morris, R. G. M. (1981). Spatial localisation does not depend on the presence of local cues. Learning and Motivation, 12, 239–260. Olton, D. S., & Samuelson, R. J. (1976). Remembrance of places passed: spatial memory in rats. Journal of Experimental Psychology: Animal Behavior Processes, 2, 97–116. Ramot, D., Johnson, B. E., Berry, T. L., Jr., Carnell, L., & Goodman, M. B. (2008). The parallel worm tracker: a platform for measuring average speed and drug-induced paralysis in nematodes. PLoS ONE, 3, e2208. Wicks, S. R., de Vries, C. J., van Luenen, H. G., & Plasterk, R. H. (2000). CHE-3, a cytosolic dynein heavy chain, is required for sensory cilia structure and function in Caenorhabditis elegans. Developmental Biology, 221, 295–307. Woodard, C., Huang, T., Sun, H., Helfand, S. L., & Carlson, J. (1989). Genetic analysis of olfactory behavior in Drosophila: a new screen yields the ota mutants. Genetics, 123, 315–326. Primary Research Articles—Interesting Examples from the Literature de Bono, M., & Bargmann, C. I. (1998). Natural variation in a neuropeptide Y receptor homolog modifies social behavior and food response in C. elegans. Cell, 94, 679–689. Moretti, P., Bouwknecht, J. A., Teague, R., Paylor, R., & Zoghbi, H. Y. (2005). Abnormalities of social interactions and home-cage behavior in a mouse model of Rett syndrome. Human Molecular Genetics, 14, 205–220. Moy, S. S., Nadler, J. J., Perez,A., Barbaro, R. P., Johns, J. M., Magnuson, T. R., et al. (2004). Sociability and preference for social novelty in five inbred strains: an approach to assess autistic-like behavior in mice. Genes Brain and Behavior, 3(5), 287–302. Ramot, D., MacInnis, B. L., Lee, H. C., & Goodman, M. B. (2008). Thermotaxis is a robust mechanism for thermoregulation in Caenorhabditis elegans nematodes. Journal of Neuroscience, 28, 12546–12557. Richter, S. H., Garner, J. P., & Wurbel, H. (2009). Environmental standardization: cure or cause of poor reproducibility in animal experiments? Nature Methods, 6, 257–261. Rothenfluh, A., & Heberlein, U. (2002). Drugs, flies, and videotape: the effects of ethanol and cocaine on Drosophila locomotion. Current Opinion in Neurobiology, 12, 639–645. Shahbazian, M.,Young, J.,Yuva-Paylor, L., Spencer, C.,Antalffy, B., Noebels, J., et al. (2002). Mice with truncated MeCP2 recapitulate many Rett syndrome features and display hyperacetylation of histone H3. Neuron, 35, 243–254. Protocols Brembs, B. (2008). Operant learning of Drosophila at the torque meter. JoVE, 16. http://dx.doi. org/10.3791/731. http://www.jove.com/index/details.stp?id=731. Crawley, J. N. (2007). Short protocols in neuroscience: Systems and behavioral methods: A compendium of methods from current protocols in neuroscience. Hoboken, NJ: John Wiley. Cunningham, C. L., Gremel, C. M., & Groblewski, P. A. (2006). Drug-induced conditioned place preference and aversion in mice. Nature Protocols, 1(4), 1662–1670. Current Protocols in Neuroscience. (2005). Behavioral neuroscience. (Chapter 8). Hoboken, NJ: John Wiley and Sons, Inc. Current Protocols in Neuroscience. (2007). Preclinical models of neurologic and psychiatric disorders. (Chapter 9). Hoboken, NJ: John Wiley and Sons, Inc. Animal Behavior Chapter | 2  71 Duistermars, B. J., & Frye, M. (2008). A magnetic tether system to investigate visual and olfactory mediated flight control in Drosophila. JoVE, 21. http://dx.doi.org/10.3791/1063. http://www. jove.com/index/details.stp?id=1063. Hart, A. C. (Ed.). (2006). WormBook. The C. elegans Research Community, WormBook. http:// dx.doi.org/10.1895/wormbook.1.87.1. http://www.wormbook.org. Komada, M., Takao, K., & Miyakawa, T. (2008). Elevated plus maze for mice. JoVE, 22. http:// dx.doi.org/10.3791/1088. http://www.jove.com/index/details.stp?id=1088. Mundiyanapurath, S., Certel, S., & Kravitz, E. A. (2007). Studying aggression in Drosophila (fruit flies). JoVE, 2. http://dx.doi.org/10.3791/155. pii.155 http://www.jove.com/index/details. stp?id=155. Nunez, J. (2008). Morris water maze experiment. JoVE, 19. http://dx.doi.org/10.3791/897. http:// www.jove.com/index/details.stp?id=897. Shiraiwa, T., & Carlson, J. R. (2007). Proboscis extension response (PER) assay in Drosophila. JoVE, 3. http://dx.doi.org/10.3791/193. http://www.jove.com/index/details.stp?id=193. Whishaw, I. Q., Li, K., Whishaw, P. A., Gorny, B., & Metz, G. A. (2008). Use of rotorod as a method for the qualitative analysis of walking in rat. JoVE, 22. http://dx.doi.org/10.3791/1030. http:// www.jove.com/index/details.stp?id=1030. Witt, R. M., Galligan, M. M., Despinoy, J. R., & Segal, R. (2009). Olfactory behavioral testing in the adult mouse. JoVE, 23. http://dx.doi.org/10.3791/949. http://www.jove.com/index/details. stp?id=949. 73 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00003-4 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 3 Stereotaxic Surgeries and In Vivo Techniques After Reading This Chapter, You Should be Able to: l Describe the process of performing stereotaxic surgeries in rodents and primates, from applying anesthetic agents to monitoring recovery l Describe common implants that allow permanent access to the brain l Discuss methods of measuring neural activity and neurochemistry in vivo Techniques covered: l Stereotaxic surgeries in animals: sterile fields, anesthesia, stereotaxic positioning of animals, accessing the brain, implantation, recovery l Implants that allow access to the brain: sealable chambers, cannulae, cranial windows l Measuring activity: electrophysiology, fluorescent activity indicators l Measuring neurochemistry: microdialysis, voltammetry, amperometry An in vivo experiment is any procedure that takes place in a whole, living organism. This type of experiment is in contrast with an in vitro experiment that takes place in a controlled environment outside a living organism, such as a cell culture dish or test tube. There is a great trade-off between these two conditions: although in vivo experiments accurately reflect the physiology and behavior of a living animal, measuring or manipulating activity in intact organs and animals can be relatively difficult. In an in vitro experiment, a scientist has a high degree of control over the extracellular environment and can perform accurate measurements of physiological activity or biochemical expression; however, these results may not correspond to conditions inside a living organism. Chapter 14 discusses in vitro methods in more detail, including their advantages and limitations. The purpose of this chapter is to describe the rationale for and process of penetrating the brain in a living animal. Accessing the brain allows a scientist to measure the activity (Chapter 4) and neurochemistry of cells in vivo, as well as to manipulate their activity (Chapter 8) to determine effects on behavior. To gain access to an animal’s brain, scientists must perform a stereotaxic 74  Guide to Research Techniques in Neuroscience surgery, a survivable procedure in which a substance is injected into a discrete brain region and/or a device is implanted into the brain or onto the skull. After describing the process of performing these surgeries in vertebrate animals, we will survey methods of measuring and manipulating physiological activity in vivo. STEREOTAXIC SURGERIES A stereotaxic surgery is an invasive procedure used to precisely target specific regions of the brain. The word stereotaxic is derived from the roots stereos, meaning “three-dimensional,” and taxic, meaning “having an arrangement.” Thus, a stereotaxic surgery places an animal brain within a three-dimensional coordinate system so an investigator can place neurobiological probes and reagents in discrete brain regions. These tools can be used to measure neural activity, sample bioactive substances within the extracellular environment, or manipulate neural function. Just as explorers use maps and astronomical reference points to navigate the globe, scientists use brain atlases and anatomical landmarks to navigate the brain. Published atlases provide three-dimensional coordinates of brain structures for a variety of animals, including rodents, primates, and even birds and bats (Figure 3.1A). The coordinates of brain structures are defined in terms of distances to anatomical landmarks visible as seams on the skull: bregma and lambda. Bregma is defined as the intersection between the sagittal and coronal sutures of the skull; lambda is defined as the intersection between the lines of best fit through the sagittal and lambdoid sutures (Figure 3.1B). To properly align an animal brain so that its structures are consistent with the coordinates in a brain atlas, a scientist must correctly position the animal’s FIGURE 3.1  Stereotaxic atlases and anatomical landmarks.  (A) A stereotaxic atlas presents the locations of distinct brain structures in three-dimensional coordinates. To use these coordinates, a scientist must properly align an animal onto a stereotaxic instrument and identify landmarks on the skull. (B) Bregma and lambda are commonly used landmarks for stereotaxic targeting of brain structures. (A: Reprinted from Paxinos, G. and Franklin, K. (2001). The Mouse Brain in Stereotaxic Coordinates, 2nd ed. With permission from Academic Press/Elsevier: New York.) Stereotaxic Surgeries and In Vivo Techniques Chapter | 3  75 head on a stereotaxic instrument (Figure 3.2). This specialized equipment holds an animal in place and uses a fine-scale micromanipulator to precisely measure and target distances in 1-μm increments. Brain atlases provide a useful three-dimensional guide for targeting discrete brain regions. However, because different animal strains and individuals vary, scientists must empirically determine the actual coordinates for a region of interest before any experiments take place. Scientists can verify the correct targeting of brain regions by examining histological sections of the brain after experiments have ended. In larger animals such as primates, it is also possible to examine the position of brain structures and implants using magnetic resonance imaging (Chapter 1). As with any procedures involving vertebrate animals, an Institutional Animal Care and Use Committee must approve procedures before any experiments take place. This committee evaluates survival surgery procedures to ensure that pain and distress are minimized. These considerations are important for the well-being of the animals, as well as for the experiments themselves, as pain and distress can alter physiology and behavior. See Box 2.2, Chapter 2 for ethical considerations for animal use. In the following section, we describe general procedures for performing stereotaxic surgeries in animals. We first describe procedures in rodents, as the FIGURE 3.2  A rodent stereotaxic instrument.  This equipment is used to precisely hold an animal’s head in place and target specific brain structures according to three-dimensional coordinates. 76  Guide to Research Techniques in Neuroscience majority of stereotaxic surgeries in neuroscience are performed on mice and rats. Next, we describe surgeries in nonhuman primates. These procedures can theoretically be adapted for use in any mammalian organism, as well as birds and reptiles. Stereotaxic Surgeries in Rodents Scientists routinely perform stereotaxic surgeries on mice and rats because of the variety of useful studies that can be performed using these model organisms. The rodent nervous system is more similar to the human nervous system in structure and function than the nervous system of invertebrates. Furthermore, rodents can be used in larger sample sizes than larger animals (such as cats or primates) to increase the statistical power of experimental results. Creating a Sterile Environment The first step in any invasive surgical procedure is to prepare sterile working conditions. Sterile does not simply mean clean, but also aseptic—free from microorganisms that can invade the brain and lead to subsequent infection. Some aspects of the surgical environment can be made clean, but not sterile. For example, a scientist’s lab coat, scrubs, face mask, and gloves can decrease the possibility of contamination but can never be 100% sterile. Cleaning the surgical environment with alcohols and specialized disinfectant solutions can also decrease the possibility of contamination but is also not 100% sterile. The only aspects of the surgical environment that must be kept completely sterile are the instruments and probes that will come in contact with the brain, as well as a dedicated surface to place these tools when they are not in use. In general, there are two ways to sterilize implants and tools: the use of an autoclave, or a tabletop heat sterilizer. An autoclave is a large appliance that increases pressure to heat water above its natural boiling point (Figure 3.3A). The high-temperature steam kills any microorganisms growing on tools and reagents placed inside the autoclave chamber. A heat sterilizer is a smaller device that can sterilize tools or fine instruments when an autoclave is not convenient, such as sterilization between surgeries (Figure 3.3B). These sterilizers are often filled with thousands of tiny glass beads that allow the heat to evenly surround the instruments. Just before a surgery takes place, a scientist cleans the surgical environment and sterilizes tools and probes in an autoclave or tabletop sterilizer. Some probes cannot handle the high heat of sterilization and can be soaked in 70% ethanol or another disinfectant prior to use. It is usually necessary to create a sterile field, a dedicated surface that can serve as a resting place for tools when they are not in use. To create a sterile field, a scientist can purchase sterile cloths/papers commercially or can autoclave a large strip of gauze. If the tips of probes or instruments come in contact with gloves or other nonsterile surfaces, they must be considered non-sterile. Stereotaxic Surgeries and In Vivo Techniques Chapter | 3  77 Anesthesia Once the environment is clean and ready for a surgery to take place, a scientist can anesthetize a rodent with either pharmacological agents or gas anesthetics. A widely used pharmacological agent is a cocktail of ketamine/xylazine. Ketamine is an anesthetic that acts by inhibiting N-methyl d-aspartate and hyperpolarization-activated cyclic nucleotidemodulated ion channels. Xylazine is a sedative and analgesic that serves as an agonist for β2 adrenergic receptors. A ketamine/xylazine cocktail typically anesthetizes an animal for 1–2 h. Reinjection of the cocktail is not recommended, as long-term exposure can lead to bradycardia (abnormally slow heart rate), blindness, seizures, or death. As an alternative to pharmacological agents, scientists can administer gas anesthetics, which are especially useful for long surgeries. A common gas anesthetic is isoflurane, an inhalable ether that likely works by disrupting synaptic transmission. To anesthetize the animal, the scientist places the rodent in a small, ventilated chamber. Gas is released into the chamber, and the scientist waits 1–3 min until the rodent stops all movements. Then, the animal is quickly placed on a bite bar or provided with a nose mask that receives a steady supply of oxygen mixed with the gas anesthetic. After an animal is anesthetized, a scientist should verify that the animal is indeed unconscious by testing for muscle responses to stimuli. For example, FIGURE 3.3  Equipment for sterilization. (A) An autoclave uses high-temperature steam to sterilize contents. (B) A bead sterilizer uses high temperatures to sterilize small instruments and surgical tools before and between surgeries. 78  Guide to Research Techniques in Neuroscience pinching the animal’s tail produces a leg withdrawal response in an animal that is not fully unconscious. Only after the animal shows no signs of motor responses should the scientist begin a surgical procedure. Investigators who wish to perform stereotaxic surgeries may consult the animal facility at their institution for recommended anesthetic agents, doses, and protocols. Positioning a Rodent on a Stereotaxic Instrument After cleaning the surgical area and properly anesthetizing an animal, the scientist places the animal on a stereotaxic instrument. The apparatus is designed to position the animal’s head in a precise orientation so that three-dimensional stereotaxic coordinates remain consistent from animal to animal (Figure 3.2). First, the scientist inserts ear bars into the animal’s ear canals. This placement allows the head to be tilted up and down but not side to side. To restrict movement up and down, the animal’s teeth are positioned on a bite bar and the nose is held firmly in a nose brace. It is crucial that the head is precisely fixed in place so that the scientist can accurately traverse the three-dimensional brain. To ensure proper orientation of the brain, the scientist must place the head in the flat skull position, such that the top surface of the skull is flat from bregma to lambda in the rostral–caudal direction. To achieve this position, the skull is exposed by a careful incision down the midline of the skin. The skin is gently moved aside, and the skull is cleaned. A micromanipulator on the stereotax is used to measure three-dimensional coordinates in the x, y, and z axes. The scientist uses the micromanipulator to ensure the top surface of the skull is flat. If the skull is not flat, the scientist can raise or lower the nose brace/bite bar until the flat skull position is achieved. Accessing the Brain Once the animal’s head is in the flat skull position, the investigator can target a discrete brain region using three-dimensional coordinates. Starting from bregma or lambda, the scientist orients the micromanipulator over the appropriate area of the skull. A dental drill is used to perform a craniotomy, a small hole in the skull that exposes the brain. Sometimes bleeding occurs at this stage, and the investigator must remove excess blood with a cotton swab or thin tissue. If a scientist intends to implant a delicate probe, such as a glass electrode, it is often necessary to also remove a section of dura, the thin meningeal layer of tissue surrounding the brain. Penetrating the Brain Once the brain is exposed, the scientist can use the micromanipulator to deliver a bioactive substance or neurobiological probe into the brain. For example, the scientist can deliver a pharmacological or viral agent into a brain region, or implant a device such as a cannula or optical window (described later in this Stereotaxic Surgeries and In Vivo Techniques Chapter | 3  79 chapter). Permanent implants can be held in place with dental acrylic and/or cement that is specifically designed to adhere small objects to the surface of bone. If properly applied, implants can remain securely adhered to the skull throughout the life of the animal. Finishing the Surgery After the adhesive has fully hardened, the final tasks are to seal the incision, clean the affected area, and help the animal recover. Depending on the length of the incision, a scientist can use sterile sutures or glues designed for tissues to seal the skin around an implant. After the skin is sealed, the top of the animal’s head can be lightly treated with an iodine or Betadine® solution to clean the affected area and help prevent infection. During surgery, an animal can lose much body heat due to blood loss and low heart rate. Therefore, recovering animals should be placed in a recovery cage on a heating pad or under a warm light. As the animal recovers, a scientist should inject further analgesic and/or antibiotic agents and should always monitor the animal until it becomes conscious before returning it to an animal storage room. It will take the animal from a few days to a full week to recover from the surgery. During this time, the animal should be inspected for any signs of pain or discomfort, including little movement, hunched posture, and a lack of eating or drinking. Further injections of an analgesic and/or antibiotic may be necessary to aid in the recovery of the animal. Stereotaxic Surgeries in Nonhuman Primates Stereotaxic surgeries in primates are conceptually similar to surgeries in rodents: the investigator anesthetizes the animal, exposes the skull, performs a craniotomy, applies dental acrylic to implant probes, sutures the skin, and carefully monitors the animal throughout the recovery process. The major difference between primate and rodent surgeries is that primates are more limited resources, and individual investigators may only use one to three monkeys over the course of single studies lasting 4–6 years. Therefore, investigators have no room for error and take extra steps to ensure correct targeting of brain regions. For example, although primate brain atlases exist, scientists often produce structural MR images (Chapter 1) for each individual weeks before the actual surgery takes place so they can target specific regions more accurately. The purpose of most primate brain surgeries is to implant devices for future electrophysiology experiments. In these experiments, the scientist trains the monkey to perform a task and then records neural activity in a specific part of the brain. To perform these experiments, a scientist usually implants one of three devices: l Sealable chamber.After performing a craniotomy, a scientist implants a chamber around the hole with a screw cap. This allows the surface of the brain to be 80  Guide to Research Techniques in Neuroscience exposed at any time after the surgery for the insertion of a recording electrode, but protected by the cap at other times. l Head post. The purpose of this post is to keep the primate’s head fixed in place during experiments. It is especially important to stabilize a monkey’s head for the presentation of visual stimuli when the monkey must fix its gaze upon a target and make decisions about complicated visual scenes.A head post can be as simple as a bolt that attaches to the top of the apparatus where the monkey sits during experiments. l Eye coil. The eye coil is also used to ensure that the monkey correctly fixes its gaze during the presentation of visual stimuli. A coil is basically a wire loop that is implanted around the outer circumference of the eye. This coil is usually hidden beneath the eyelids after the surgery. The wire is threaded beneath the skin from the eye to the top of the head, where it can be attached to additional cables to measure eye movements during experiments. An eye coil is not necessary for experiments in which there is no need to monitor the animal’s eye movements. Many laboratories implant and adjust these devices over two or three surgical sessions so that their placement is accurate, firm, and precise. IMPLANTS FOR LONG-TERM ACCESS TO THE BRAIN After a scientist completes a stereotaxic surgery, access to the brain is lost. However, there are a variety of implantable devices that allow long-term access to the brain for chronic injections or physiological recordings. The most common implants include sealable chambers, cannulae, and cranial windows, each described below. Sealable Chambers As described above, a sealable chamber is a round, hollow well with one end affixed to the skull and the other end capped with a screw top. This implant allows an investigator chronic access to a specific region of the brain for electrophysiology. Care must be taken to keep the chamber clean and as sterile as possible. Whenever not in use, the chamber must always be closed to prevent outside contaminants from penetrating the brain. Cannulae As described above, a cannula is a common device used to access deep structures within the brain. It is composed of a narrow, cylindrical tube that a scientist can insert into the brain and permanently affix to the skull. Cannulae can be made of plastic, glass, or steel of various sizes, though thinner cannulae are less damaging to neural tissue. Once implanted, a scientist can deliver multiple substances or probes: pharmacological agents, viral vectors, narrow electrodes, and Stereotaxic Surgeries and In Vivo Techniques Chapter | 3  81 even optical probes to deliver light stimulation or image deep brain structures. At the end of experiments, scientists verify the correct placement of cannulae by histological examination of brain sections. Cranial Windows There are many neuroscience techniques that allow imaging the structure and function of living cells (Chapters 6 and 7), such as labeling neurons with fluorescent probes. In order to visualize these probes in vivo, it is necessary to create an observation window through the skull. To create such a cranial window, a scientist performs a craniotomy and replaces the missing section of the skull with a glass coverslip. Alternatively, scientists can thin and polish an area of the skull so that it becomes translucent. These cranial windows allow imaging of the same brain region over long periods of time. It is usually necessary to implant a head post in addition to a window so that an animal’s head can be properly aligned, and a scientist can locate the same cells and structures over weeks and months. Now that we have described stereotaxic surgeries and implants allowing continued access to the brain, we briefly survey various techniques that allow scientists to measure and manipulate neural activity in vivo. Measuring Neural Activity In vivo Neural activity can be measured in vivo using electrophysiology or visualization of fluorescent proteins. These methods are described in greater detail in other chapters, but we mention them here in the context of in vivo approaches to studying the brain after stereotaxic surgeries. Electrophysiology Electrophysiology techniques (Chapter 4) are used to record the electrical activity of neurons. In brief, there are three basic types of electrophysiological probes that can measure activity in vivo: l Single electrodes.An electrode can be inserted through an implanted chamber on the surface of an animal brain. These electrodes are often mounted onto a microdrive, which itself can be mounted on the chamber on the skull. The microdrive precisely raises or lowers the electrodes to specific depths in the brain. l Multielectrode array. A multielectrode array is a grid of dozens of electrodes capable of recording the activity of multiple neurons on the surface of the brain. Although extremely small, the size and mass of arrays are often too large to be inserted into deep brain structures. Therefore, they are usually implanted on the outer surface of the brain. 82  Guide to Research Techniques in Neuroscience l Tetrodes. A tetrode is a type of multielectrode array consisting of four active electrodes. Unlike a microelectrode array, a tetrode is narrow and can be easily inserted into relatively deep brain structures. Fluorescent Activity Indicators Fluorescent dyes and genetically targeted activity sensors, such as calcium indicators, can be used as indirect measures of neural activity. Cranial windows are used to observe these indicators in vivo. We describe fluorescent indicators in much more detail in Chapter 7. Measuring Neural Chemistry In vivo It is often desirable to know how specific neurochemicals, such as neurotransmitters, hormones, or peptides, change in concentration over time. Two techniques are commonly used to identify and measure neurochemicals in vivo: microdialysis and voltammetry. These techniques can also be used in vitro, but their use in living animals allows scientists the opportunity to correlate changes in neurochemical expression in discrete brain regions with specific behaviors or physiological functions. Microdialysis Microdialysis is a technique used to sample chemical substances from the extracellular fluid in a specific location in the brain. Based on the principles of diffusion, a microdialysis probe creates a concentration gradient between the extracellular fluid in the probed brain region and a continuously perfused physiological solution (Figure 3.4). The probe is made of a semipermeable membrane connected to two thin cannulae, with perfusion Perfusate pumped into the brain Dialysate pumped out of the brain FIGURE 3.4  In vivo microdialysis. (A) A microdialysis probe can be implanted through a cannula into an awake, behaving rodent. Scientists perfuse a physiological solution through the probe and collect the dialysate containing neurochemicals. (B) A microdialysis probe placed in a specific brain region. Neurochemicals in the probed brain region pass through the semipermeable membrane of the probe through diffusion. Stereotaxic Surgeries and In Vivo Techniques Chapter | 3  83 solution flowing into and out of the probed brain region. Because of the concentration gradient, substances present in the extracellular fluid will passively diffuse through the probe’s membrane into the collected solution, the dialysate. This dialysate is collected continuously into a vial for an increment of 1–10 min for subsequent analysis using sensitive chemical methods, such as high-performance liquid chromatography, radioimmunoassay, or mass spectrometry. Numerous factors affect the molecules that can be collected into the dialysis solution: the characteristics of the neurochemicals (size, charge, solubility), properties of the dialysis membrane itself (material, pore diameter), flow speed of perfusion, composition of perfusion solution, and the density of the brain tissue surrounding the microdialysis probe. These factors affect both the identity and the amount of substance that can be collected. Thus, the total amount of substance collected in the dialysate over a set amount of time, known as absolute recovery, is not the true extracellular concentration. Instead, scientists present the concentration as relative recovery, the relative concentration of the substance in the dialysate from the probed brain region compared to the concentration in the perfusion solution. It is possible to determine the actual concentration of a substance using at least two different methods: (1) measuring the relative recovery for different flow rates and extrapolating the information to a zero-flow rate, where recovery should theoretically be 100%; and (2) perfusing with known concentrations of the substance to determine the equilibrium point at which the collected concentration does not change due to the lack of a concentration gradient. Some investigators calculate extracellular concentrations by calibrating probes in vitro, immersing the probes in beakers with known concentrations of the substance. However, because of the differences in the diffusion properties in a simple solution compared to the complex environment of the brain, it is not possible to equate in vitro with in vivo recovery to determine the true extracellular concentration of substances. There are a number of advantages to using microdialysis to measure neurochemicals in the brain. Microdialysis collection of extracellular fluid can be followed by sensitive chemical assays that clearly identify most types of small molecules. In addition, microdialysis can be combined with simultaneous extracellular electrophysiological recordings and stimulation, even in freely moving animals. Another use of the microdialysis probe is reverse microdialysis: the delivery of drugs through the perfusion fluid. This method of drug delivery allows for greater control over the flow of the pharmacological agent compared with pressure injection through a cannula. The greatest disadvantage of microdialysis is low temporal resolution. The concentration of neurochemicals that can be collected in the dialysate is typically very low, requiring a high volume of fluid to be collected for analysis. This low yield can require long sampling periods (1–10 min), which is an adequate timescale for measuring long-term changes in neurochemical 84  Guide to Research Techniques in Neuroscience concentrations but inadequate for detection of neurotransmitter release at a physiological timescale. Voltammetry and Amperometry Voltammetry is a technique used to detect neurochemicals capable of undergoing oxidation reactions. These neurochemicals include neurotransmitters such as serotonin and the catecholamines (e.g., epinephrine, norepinephrine, and dopamine). A scientist inserts a carbon fiber microelectrode into the brain and applies a specific voltage. When the chemicals encounter the surface of the electrode, they undergo an oxidation reaction, releasing electrons that produce a measurable change in current. The magnitude of the current is proportional to the number of molecules oxidized. Therefore, it is possible to detect the presence and relative concentration of these neurochemicals at a physiological timescale. A commonly used form of voltammetry is fast-scan cyclic voltammetry (FCV), which has a high temporal resolution (fractions of a second). In FCV, the voltage of the electrode is shifted back and forth from a nonoxidizing potential to an oxidizing potential within milliseconds. The scientist can then plot the amount of measured current versus the applied voltage, producing a cyclic voltammogram (Box 3.1). Because electroactive neurochemicals are oxidized and reduced at different potentials, a cyclic voltammogram can be used as a fingerprint to identify the specific neurotransmitter detected at the electrode. Another subtype of voltammetry is amperometry. Unlike FCV, amperometry holds the electrode at a specific, constant voltage. All molecules that can Box 3.1 The Cyclic Voltammogram In cyclic voltammetry, the carbon fiber microelectrode measures current as it cycles between oxidizing and nonoxidizing potentials (Figure 3.5A). A scientist can use these data to plot the current (I) versus voltage (V), an I/V curve similar to those found in some electrophysiology experiments (Figure 3.5B, Chapter 4, Box 4.2). The rapid, large changes in potential produce a high background current. However, these background currents are stable over a period of minutes, and thus can be subtracted from the experimental I/V curve (Figure 3.5C). Subtracting this background current produces the cyclic voltammogram, the fingerprint that identifies an electroactive substance. Another way scientists present voltammetry data is to show current traces over time to indicate changes in the substance’s concentration, demonstrating the responses to electrical, pharmacological, or behavioral stimuli (Figure 3.5D). Three-dimensional graphs can be used to display changes in both voltage and current over time, with the current encoded in color (Figure 3.5E). Stereotaxic Surgeries and In Vivo Techniques Chapter | 3  85 be oxidized at that potential will be detected. Thus, one limitation of amperometry compared with FCV is that it is difficult to identify the specific compound detected by the electrode solely based on the shape of the current. However, in amperometry, measured currents can be averaged over longer time periods, allowing more precise measurements of the relative concentrations of neurochemicals. Compared to microdialysis, voltammetric techniques offer a much higher spatial and temporal resolution for measuring neurochemicals. The higher spatial resolution is due to the small size of the carbon fiber microelectrode compared with the larger size of the microdialysis membrane. The temporal resolution of voltammetry (fractions of a second) is much higher than the standard sampling rate of microdialysis, which occurs every 1–10 min. However, voltammetric techniques can only be applied to electroactive chemicals and cannot be used to assay other bioactive molecules present in the extracellular environment. -0.4 V 1.0 V Current (nA) Time (s) Potential(V) -0.8 nA 12 nA -0.4 V -0.4 V +1.3 V -5 0 5 10 5 s 2 nA -0.4 0 1.0 125 nA -125 nA -0.4 V 1.0 V 8 nA - 8 nATime (s) Stimulus Stimulus Oxidation Reduction Oxidation Reduction Oxidation Reduction FIGURE 3.5  Analyzing and depicting cyclic voltammetry data.  (A) A voltage is applied to the microelectrode that causes oxidation and reduction of nearby neurochemicals. (B) The current (I) is recorded at these oxidizing and reducing potentials. Large background currents are generated from the rapid changes in potential, but they can be subtracted to produce the (C) cyclic voltammogram signature of a neurochemical. Notice the difference in scale between (B) and (C). Plotting current over time (D) can reveal changes in the neurochemical’s concentration in response to a stimulus. (E) Three-dimensional plots can be used to show changes in both potential and current over time. 86  Guide to Research Techniques in Neuroscience MANIPULATING NEURAL ACTIVITY IN VIVO The ability to penetrate the brain and stereotaxically target specific brain regions provides scientists with many experimental opportunities for determining the role of these regions in animal behavior. In a loss-of-function experiment, a scientist can inactivate a brain region to determine if it is necessary for a behavior to occur. Alternatively, a scientist can perform a gain-of-function experiment and stimulate neural activity to determine whether activity in that region is sufficient to cause a behavior. To perform these experiments, scientists have a diverse toolkit of physical, pharmacological, and electrical methods, along with cutting-edge genetically encoded neural actuators. These techniques are described in Chapter 8. CONCLUSION Stereotaxic surgeries provide the ability to target specific brain regions for observation or manipulation. Following a surgery, it is possible to measure the neural activity and neurochemistry of a small population of cells, as well as to manipulate activity to determine the role of these cells in physiology and behavior. The techniques described in this chapter greatly aid in the study of behavior (Chapter 2), as well as the study of individual neurons and neural circuits, as will be described in the next few chapters. These techniques are also complemented by in vitro assays, described in Chapter 14. SUGGESTED READING AND REFERENCES Books Boulton, A. A., Baker, G. B., & Bateson, A. N. (1998). In vivo neuromethods. Totowa, N.J.: Humana Press. Institute of Laboratory Animal Resources (U.S.). (1996). Guide for the care and use of laboratory animals (7th ed.). Washington, D.C.: National Academy Press. Paxinos, G., & Watson, C. (2009). The rat brain in stereotaxic coordinates (Compact 6th ed.). Amsterdam; Burlington, MA: Elsevier/Academic Press. Review Articles Du, C., & Pan, Y. (2011). Optical detection of brain function: simultaneous imaging of cerebral vascular response, tissue metabolism, and cellular activity in vivo. Reviews in the Neurosciences, 22, 695–709. Fillenz, M. (2005). In vivo neurochemical monitoring and the study of behaviour. Neuroscience & Biobehavioral Reviews, 29, 949–962. Hutchinson, P. J., O’Connell, M. T., Kirkpatrick, P. J., & Pickard, J. D. (2002). How can we measure substrate, metabolite and neurotransmitter concentrations in the human brain? Physiological Measurement, 23, R75–R109. Isshiki, M., & Okabe, S. (2014). Evaluation of cranial window types for in vivo two-photon imaging of brain microstructures. Microscopy (Oxford, England), 63, 53–63. Kim, S. A., & Jun, S. B. (2013). In-vivo optical measurement of neural activity in the brain. Experimental Neurobiology, 22, 158–166. Stereotaxic Surgeries and In Vivo Techniques Chapter | 3  87 van Rooy, I., Cakir-Tascioglu, S., Hennink, W. E., Storm, G., Schiffelers, R. M., & Mastrobattista, E. (2011). In vivo methods to study uptake of nanoparticles into the brain. Pharmaceutical Research, 28, 456–471. Wotjak, C. T., Landgraf, R., & Engelmann, M. (2008). Listening to neuropeptides by microdialysis: echoes and new sounds? Pharmacology Biochemistry and Behavior, 90, 125–134. Wright, N. J. (2014). Evolution of the techniques used in studying associative olfactory learning and memory in adult Drosophila in vivo: a historical and technical perspective. Invertebrate Neuroscience, 14, 1–11. Primary Research Articles—Interesting Examples from the Literature Chang, S.Y., Kim, I., Marsh, M. P., Jang, D. P., Hwang, S. C., Van Gompel, J. J., et al. (2012). Wireless fast-scan cyclic voltammetry to monitor adenosine in patients with essential tremor during deep brain stimulation. Mayo Clinic Proceedings, 87, 760–765. Day, J. J., Roitman, M. F., Wightman, R. M., & Carelli, R. M. (2007). Associative learning mediates dynamic shifts in dopamine signaling in the nucleus accumbens. Nature Neuroscience, 10, 1020–1028. Grutzendler, J., Kasthuri, N., & Gan, W. B. (2002). Long-term dendritic spine stability in the adult cortex. Nature, 420, 812–816. LaLumiere, R. T., & Kalivas, P. W. (2008). Glutamate release in the nucleus accumbens core is necessary for heroin seeking. Journal of Neuroscience, 28, 3170–3177. Lomber, S. G., & Malhotra, S. (2008). Double dissociation of “what” and “where” processing in auditory cortex. Natural Neuroscience, 11, 609–616. Park, J., Takmakov, P., & Wightman, R. M. (2011). In vivo comparison of norepinephrine and dopamine release in rat brain by simultaneous measurements with fast-scan cyclic voltammetry. Journal of Neurochemistry, 119, 932–944. Roitman, M. F., Wheeler, R. A., Wightman, R. M., & Carelli, R. M. (2008). Real-time chemical responses in the nucleus accumbens differentiate rewarding and aversive stimuli. Natural Neuroscience, 11, 1376–1377. Shou, M., Ferrario, C. R., Schultz, K. N., Robinson, T. E., & Kennedy, R. T. (2006). Monitoring dopamine in vivo by microdialysis sampling and on-line CE-laser-induced fluorescence. Analytical Chemistry, 78, 6717–6725. Trachtenberg, J. T., Chen, B. E., Knott, G. W., Feng, G., Sanes, J. R., Welker, E., et al. (2002). Long-term in vivo imaging of experience-dependent synaptic plasticity in adult cortex. Nature, 420, 788–794. Ungerstedt, U., & Hallstrom, A. (1987). In vivo microdialysis—a new approach to the analysis of neurotransmitters in the brain. Life Science, 41, 861–864. Xu, H. T., Pan, F., Yang, G., & Gan, W. B. (2007). Choice of cranial window type for in vivo imaging affects dendritic spine turnover in the cortex. Natural Neuroscience, 10, 549–551. Zhou, F. M., Liang, Y., Salas, R., Zhang, L., De Biasi, M., & Dani, J. A. (2005). Corelease of dopamine and serotonin from striatal dopamine terminals. Neuron, 46, 65–74. Protocols Athos, J., & Storm, D. R. (2001). High precision stereotaxic surgery in mice. Current Protocols in Neuroscience Appendix 4, Appendix 4A. Chefer, V. I., Thompson, A. C., Zapata, A., & Shippenberg, T. S. (2009). Overview of brain microdialysis. Current Protocols in Neuroscience Chapter 7, Unit 7.1. Cunningham, M. G., Ames, H. M., Donalds, R. A., & Benes, F. M. (2008). Construction and implantation of a microinfusion system for sustained delivery of neuroactive agents. Journal of Neuroscience Methods, 167, 213–220. 88  Guide to Research Techniques in Neuroscience Geiger, B. M., Frank, L. E., Caldera-Siu, A. D., & Pothos, E. N. (2008). Survivable stereotaxic surgery in rodents. Journal of Visualized Experiments, 20. http://dx.doi.org/10.3791/880. http:// www.jove.com/index/details.stp?id=880. Mostany, R., & Portera-Cailliau, C. (2008).A craniotomy surgery procedure for chronic brain imaging. Journal of Visualized Experiments, 12. http://dx.doi.org/10.3791/680. http://www.jove. com/index/details.stp?id=680. Mundroff, M. L., & Wightman, R. M. (2002). Amperometry and cyclic voltammetry with carbon fiber microelectrodes at single cells. Current Protocols in Neuroscience Chapter 6, Unit 6. 14. Saunders, R. C., Kolachana, B. S., & Weinberger, D. R. (2001). Microdialysis in nonhuman primates. Current Protocols in Neuroscience Chapter 7, Unit 7.3. Zapata, A., Chefer, V. I., & Shippenberg, T. S. (2009). Microdialysis in rodents. Current Protocols in Neuroscience Chapter 7, Unit 7.2. 89 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00004-6 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 4 Electrophysiology After Reading This Chapter, You Should be Able to: l Understand the fundamental electrical properties of neurons and the ionic basis of membrane potentials l Describe the basic components of an electrophysiology rig l Compare different categories of electrophysiological recording techniques, including which research questions each technique can best address l Compare different types of tissue preparations used in electrophysiology experiments Techniques covered: l Common electrophysiological approaches: extracellular recordings, intracellular recordings, patch clamp techniques l Common electrophysiological preparations: heterologous expression systems, primary cultures, brain slice cultures, anesthetized or awake animals Electrophysiology is the branch of neuroscience that explores the electrical activity of living neurons and investigates the molecular and cellular processes that govern their signaling. Neurons communicate using electrical and chemical signals. Electrophysiology techniques listen in on these signals by measuring electrical activity, allowing scientists to decode intercellular and intracellular messages. These techniques can answer systems-level questions, such as the role of a neuron in a neural circuit or behavior. Alternatively, they can be used to investigate the specific ion channels, membrane potentials, and molecules that give each neuron its physiological characteristics. The three main categories of electrophysiology techniques are defined by where the recording instrument, the electrode, is placed in the neural specimen. In an extracellular recording experiment, the electrode is placed just outside the neuron of interest. In an intracellular recording experiment, the electrode is inserted into the neuron of interest. Finally, using patch clamp techniques, the electrode is closely apposed to the neuronal membrane, forming a tight seal with a patch of the membrane. These different recording techniques are used to examine the electrical properties of neurons both in vitro and in vivo. In vitro cell cultures and brain slices allow for detailed investigations of the molecules 90  Guide to Research Techniques in Neuroscience responsible for electrical signals, while in vivo preparations can demonstrate the role of these electrical signals in animal behavior. The purpose of this chapter is to differentiate between the major categories of electrophysiological techniques and preparations, comparing the relative advantages, disadvantages, and common uses of each. We will start with a brief review of the physical principles that provide neurons their electrical characteristics. To explain how these techniques are performed, we will survey the instruments that make up an electrophysiology setup (the rig). We will also examine some common methods of data analysis and presentation used in the literature. A BRIEF REVIEW OF THE ELECTRICAL PROPERTIES OF NEURONS The electrical activity of a neuron is based on the relative concentration gradients and electrostatic gradients of ions within the cell and in the extracellular fluid, as well as the types of ion channels present within the neuron. The difference in charge between the intracellular and extracellular sides of the membrane creates an electrical potential, measured in units of Voltage (V). A neuron membrane at resting potential is about −70 mV. This is due to differences in the permeability of various inorganic ions, particularly sodium (Na+), potassium (K+), and chloride (Cl−), as well as to the active contributions of a sodium–potassium pump (Figure 4.1). Ions moving across the membrane generate a measurable current (I), the movement of charge over time. The movement of ions across the membrane is limited by the membrane resistance (R). This resistance is generated by properties of the membrane, such as how many channels are open or closed. The relationship among the membrane potential, the current flow, and the membrane resistance is described by Ohm’s law: V = I × R. This relationship is the fundamental basis of electrophysiological techniques. Neurons communicate by causing changes in membrane potential in other neurons. For example, a presynaptic neuron releases neurotransmitters that bind to receptors on the postsynaptic neuron. These receptors are often ligand-gated ion channels that open to allow more ions to flow through the membrane. Relative to the membrane’s resting potential, this current flow can make the membrane potential more positive, an effect called depolarization. Alternatively, current flow can make the membrane potential more negative, an effect called hyperpolarization. Whether a depolarization or hyperpolarization effect occurs depends on the charge of the flowing ions. This local voltage change is called a graded potential or localized potential, and its magnitude is proportional to the strength of the stimulus. A local voltage change that makes the membrane potential more positive is called an excitatory postsynaptic potential (EPSP), while a local voltage change that makes the membrane potential more negative is called an inhibitory postsynaptic potential (IPSP). Different EPSP and IPSP events combine to form an overall signal in the postsynaptic neuron. These localized potentials can add up in space (called spatial summation) and time (called temporal summation). If enough localized potentials sum to depolarize Electrophysiology Chapter | 4  91 the membrane to a threshold point, usually around −55 mV, an action potential will occur. An action potential, also referred to as a spike, is an all-or-none, rapid, transient depolarization of the neuron’s membrane. A local depolarization to the threshold potential opens voltage-gated sodium channels, and the rapid influx of sodium ions brings the membrane potential to a positive value (Figure 4.2). The membrane potential is restored to its normal resting value by the delayed opening of voltage-gated potassium channels and by the closing of the sodium channels. A refractory period follows an action potential, corresponding to the period when the voltage-gated sodium channels are inactivated. The all-or-none generation of an action potential initiates a wave of depolarization that preserves the amplitude of the voltage change as it propagates down the axon’s membrane. In a chemical synapse, depolarization stimulates the fusion of synaptic vesicles with the presynaptic membrane and the release of neurotransmitter molecules into the synaptic cleft. The neurotransmitters bind to receptor proteins associated with particular ion channels on the postsynaptic membrane. These ion channels then generate EPSP and IPSP events in the postsynaptic neuron, which can, in turn, sum to generate an action potential in that neuron. Sodium– potassium pump 90 mV pressure from concentration gradient 70 mV pressure from concentration gradient 50 mV pressure from concentration gradient Na+ Na+ Na+ Na+ Na+ Na+ Na+ Na+Na+ K+ K+ K+ K+ K+ K+K+ K+ K+ K+ K+ Cl- Cl- Cl- Cl- Cl- Cl- Cl- Cl- Cl- Na+ Na+ Na+ K+K+ 70 mV pressure from electrostatic gradient 70 mV pressure from electrostatic gradient 70 mV pressure from electrostatic gradient FIGURE 4.1  The ionic basis of the resting potential. The resting potential of neurons, about −70 mV, is caused by the permeability of various inorganic ions. These ions experience pressure to move in or out of the cell based on concentration gradients (differences in concentration of the ion per unit distance in the local environment) and electrostatic gradients (differences in electrical charge per distance in the local environment). In addition to passive diffusion, a sodium–potassium pump continually pumps sodium ions out of the cell and potassium ions into the cell. 92  Guide to Research Techniques in Neuroscience Many more details on the electrical properties of neurons can be found elsewhere. What is important in the context of discussing electrophysiology methods is to understand that scientists can study these properties at many different levels of investigation. For example, an investigator may want to know the frequency of action potentials in a specific neuron over time to decipher how a neuron encodes a particular stimulus or action. This kind of experiment can be performed using an extracellular recording, either in vitro or in an awake, behaving animal. Alternatively, an investigator may want to know how the presence of a drug in the extracellular fluid affects the ability of a specific ion channel to pass current. This experiment could be performed using a patch clamp technique in a heterologous expression system. Whether in the context of circuit analysis or the molecular basis of the membrane potential, nearly any aspect of neuronal physiology can be investigated with the electrophysiology methods described in this chapter. Before we describe these methods, let us examine the tools and equipment necessary to perform an electrophysiological recording. THE ELECTROPHYSIOLOGY RIG Each electrophysiology lab setup is different, reflecting the questions being addressed, the requirements of the experiment, and the personal preferences of the investigators. There is, however, a standard set of equipment necessary and desirable to record electrical signals from neurons (Figure 4.3). In general, a signal is detected by a microelectrode, which transmits that signal to an -70 -50 -30 -10 10 30 50 70 1 2 3 4 MembranePotential(mV) Time (ms) Sodium channels open Potassium channels open Sodium channels close Potassium channels close FIGURE 4.2  The ionic basis of an action potential. Localized potentials within the neuron sum to bring the membrane voltage to its threshold potential—around −55 mV. This potential causes voltage-gated sodium channels to open, further depolarizing the membrane. Potassium channels open as the membrane potential becomes more positive. At about 25 mV, sodium channels close, and the membrane potential decreases until it becomes hyperpolarized. Finally, potassium channels close, and the membrane potential returns to a resting state. Electrophysiology Chapter | 4  93 amplifier, an oscilloscope, and a computer. The oscilloscope presents a visual display of the membrane potential over time, which can also be heard using a loudspeaker system. In modern electrophysiology rigs, the oscilloscope is now completely digital and part of a software program on the computer. An electrophysiology setup also includes equipment to stabilize the microelectrode and correctly position it to record from a neuron of interest. A microscope is used to ensure proper placement of the electrode for in vitro recordings, while stereotaxic equipment (Chapter 3) is used to position electrodes for in vivo recordings. Box 4.1 details various components of a standard electrophysiology rig. Now that we have reviewed the electrical properties of neurons and surveyed the equipment needed to study these properties, we describe various electrophysiology techniques and how they are used. CATEGORIES OF ELECTROPHYSIOLOGY RECORDINGS Electrophysiological recordings can be categorized into three main types based on the placement of the electrode in relation to the cell: (1) extracellular recordings; (2) intracellular recordings; and (3) patch clamp techniques (Figure 4.4). Each technique can be used to address specific questions concerning Ground electrode Amplifier Oscilloscope (can be digitized and displayed on the computer) Computer 10x Microscope Microelectrode Headstage/ Micromanipulator/ Microdrive FIGURE 4.3  Some of the fundamental components of an electrophysiology rig. In general, an electrical signal is recorded by a microelectrode and passed along to an amplifier. The amplifier compares the recording to a ground electrode, then transmits the signal to an oscilloscope and/ or computer. Various other equipment is necessary and desirable depending on the nature of the experiment. 94  Guide to Research Techniques in Neuroscience BOX 4.1 The Electrophysiology Rig The following is a brief description of some of the equipment that may or may not be part of a scientist’s electrophysiology rig. Some of this equipment is mandatory, such as a microelectrode and amplifier. Some equipment is not, such as a loudspeaker system. Microelectrode There are two main types of electrodes: (1) glass micropipettes filled with an electrolyte solution (2 or 3 M sodium chloride or potassium chloride); and (2) metal electrodes (usually tungsten, steel, or platinum–iridium). An important characteristic of both kinds of electrodes is their electrical resistance, which is related to the exposed tip size. Smaller tips have higher resistances, and they restrict the area from which potentials can be recorded, thus permitting the isolation of the activity of either a fiber or a cell. Large tips and low resistances pick up the activity from a number of neurons and are of limited use in efforts to identify the functional properties of single cells. Tips with very high resistances are also of little use, as they cannot record neural activity unless they are very close to the cell membrane or actually inside a cell. Electrodes often also have significant capacitances at the metal/fluid interface. Usually, the impedance (a measure of electrical resistance plus capacitance) of the most successful electrodes is in the range of 5–20 MΩ when measured by an AC current at 50 Hz. Electrodes with impedances of less than 3 or 4 MΩ tend to record from more than one cell simultaneously. The spikes are smaller, and usually only a multiunit recording can be made. Glass micropipettes are necessary for patch clamp recordings because of the way the pipette must make a tight seal with the cell membrane (as discussed later in the chapter). For intracellular or extracellular electrophysiology experiments, metal electrodes are more commonly used. These metal electrodes not only provide more stable isolation of single units than micropipettes, but they tend to sample from a larger morphological variety of cells and also help in better localization of electrode tracks to identify where recordings took place in whole brains. The main advantage of using a glass electrode in extracellular or intracellular experiments is that the pipette can be filled with a dye or other materials that can subsequently be injected into the cell or cellular environment for staining and/ or pharmacological experiments. A pipette puller is used to create glass electrodes. A glass capillary is loaded into the machine, which heats the glass and pulls it apart to end in a fine tip. For glass micropipettes that will be used in patch clamp recordings, the micropipette is also fire-polished, rounding and smoothing the electrode tip. Glass capillaries and pipette pullers come in many varieties and are commercially available. Metal electrodes are generally purchased ready-for-use. It is important to note that two electrodes are necessary for any electrophysiological recording: the recording electrode itself and a reference electrode (also called a “ground” electrode) placed outside the cell of interest. A reference electrode is necessary because an electrophysiological measurement is a comparison— for example, a comparison of the potential difference across the membrane of a neuron. In extracellular electrophysiology, both electrodes are located outside the neuron but are placed in different locations in the extracellular environment. Electrophysiology Chapter | 4  95 Headstage The headstage is the central hub that connects the electronic equipment to the tissue preparation. It contains an electrode holder that stabilizes the microelectrode during recordings and also directly connects the microelectrode to the first stage of amplifier electronics needed to detect the electrical signals. The headstage passes the signal to the main amplifier for the main signal processing. The headstage is carefully positioned by the micromanipulator and is also attached to the microdrive. Micromanipulator The micromanipulator is a device that allows fine movements in the X, Y, and Z axes, permitting precise positioning of the microelectrode in tissue. Good micromanipulators have fine-scale units of measurement (usually μm) and can be used to stereotaxically place the microelectrode in specific regions of brain or tissue. Microdrive A microdrive is used to lower or raise the microelectrode to a specific depth in tissue in very fine steps. It is usually preferable to use remote-controlled microdrive systems to eliminate hand vibration. Thus, the headstage (and consequently the electrode) can be set into place by hand using the micromanipulator and then finely adjusted in and out of tissue using a microdrive for the final approach to the cell. Amplifier The signal is passed from the microelectrode on the headstage to the main amplifier where amplification of the signal takes place (100–1000×). A scientist needs an amplifier to enhance the relatively weak electrical signal derived from a microelectrode, just as a radio listener must amplify the signal from an FM antenna so it can drive the speakers on a radio. The amplifier also receives signal from the reference electrode, and it is here that the signals from the two electrodes are compared. The amplifier then transmits the signal to an oscilloscope and/or computer. Oscilloscope An oscilloscope receives the electrical signal from the amplifier and displays the membrane voltage over time. This is the major source of data output in electrophysiology experiments. While many computer programs now contain virtual oscilloscopes that also receive the signal from the amplifier, a physical oscilloscope is often used to detect subtle dynamics that can get accidentally filtered out by a computer. Loudspeaker System Changes in voltage over time are visualized using an oscilloscope or digital oscilloscope, but they can also be heard by connecting the output of the amplifier to a loudspeaker. Action potentials make a distinctive popping sound, so recording the activity of an active neuron can sound like popcorn popping. Loudspeakers can be helpful when trying to locate a neuron of interest, because different types of neurons have distinctive firing patterns. An investigator lowering an electrode into neural tissue using a microdrive can be alerted to a specific type of cell or group of nuclei by the characteristic sounds of neurons near the electrode tip. BOX 4.1 The Electrophysiology Rig—cont’d Continued 96  Guide to Research Techniques in Neuroscience Microscope A microscope is almost always necessary for all kinds of physiological recordings. For extracellular recordings, a low-power dissecting microscope is usually adequate to see laminae or gross morphological features of the tissue and brain. For intracellular physiology or patch clamp techniques, a microscope with high enough magnification power to see individual cells (300–400×) is necessary. These microscopes are usually equipped with optical contrast enhancement to view cells in unstained preparations. An inverted microscope is usually preferable for two reasons: (1) it allows easier electrode access to the sample, as the objective lens is below the chamber; and (2) it provides a larger, more solid platform upon which to bolt the micromanipulator. Computer Computers have greatly aided electrophysiological studies by automating stimulus delivery and electrical signal recording. Computer software makes it easy to write programs that can reproducibly introduce sensory or electrical stimuli to animals or tissue preparations and record neural responses. Computers can easily manipulate many parameters during recordings, such as the recording thresholds and stimulus delivery timing. Computers also allow simple real-time data analysis, displaying the results of an experiment, even while the experiment is occurring. Vibration Isolation System A vibration isolation system, usually an air table, is used to absorb tiny changes in vibration that can disturb the placement of the microelectrode. Antivibration tables are usually composed of a heavy surface on compressed-air supports. Faraday Cage A Faraday cage can be a simple enclosure made from conductive material that blocks external electrical interference. This is needed to eliminate noise from sensitive electrical recordings that can detect electrical activity from outside sources. Other equipment may be necessary, depending on the exact nature of the experiment. For example, in experiments that study auditory physiology, a sound booth is necessary to block environmental sounds that could influence recordings. Pharmacological injection equipment is often included in the rig to inject or perfuse drugs and other substances into neural tissue. Special stimulus delivery systems may be necessary, depending on the stimuli needed for the recordings. Each scientist’s rig is often a highly specific setup, with all of the equipment designed and calibrated for the particular needs of the experiments being performed. BOX 4.1 The Electrophysiology Rig—cont’d the electrical properties of neurons. For example, questions regarding signals from neurons in vivo are most easily addressed using extracellular methods. Questions regarding the open and closed states of a specific ion channel in the presence of neuropeptide activators are best addressed using patch clamp techniques. Table 4.1 compares some of the questions that can be addressed using various categories of electrophysiology recordings. Electrophysiologists using intracellular or patch clamp methods have the benefit of being able to make use of closed electrical systems to explore the Electrophysiology Chapter | 4  97 Extracellular Recording Patch Clamp Recording Intracellular Recording FIGURE 4.4  The three categories of electrophysiological recordings. Each type of recording is defined by where the scientist places the recording electrode: outside the neuron (extracellular recording), inside the neuron (intracellular recording), or adjacent to the membrane (patch clamp recording). TABLE 4.1  Questions That Can Be Addressed by Different Electrophysiological Techniques Extracellular recordings: Recording from the extracellular environment near a cell (or cells) of interest How does a neuron encode information in action potentials? What is the role of a neuron in a given sensory, motor, or cognitive behavior? How does the activity (or inactivity) of one neuron affect the activity of another neuron? How do pharmacological agents, neurotransmitters, and neuromodulators affect the firing of a neuron? How is the spiking activity of a group of neurons coordinated? Intracellular recordings: Recording from inside a cell of interest, piercing the membrane with an electrode How does the activity (or inactivity) of one neuron affect the local potentials and action potentials of another neuron? How do pharmacological agents, neurotransmitters, and neuromodulators affect the local potentials and action potentials of a neuron? Patch clamp recordings: Recording from the membrane of a cell of interest How do an ion channel’s open and closed times depend on the membrane potential? How do the concentrations of ions, pharmacological agents, neurotransmitters, and neuromodulators affect the current flowing into an ion channel or cell? How much current does a single ion channel carry? What contributions does a single channel provide to an entire neuron? 98  Guide to Research Techniques in Neuroscience relationship between membrane potential and current flow: voltage clamp and current clamp. Using a voltage clamp, a scientist can measure currents generated by ions moving across the membrane by holding the membrane potential at a set voltage, the holding potential. Deviations between the membrane potential and the holding potential are corrected through a feedback system that injects current to maintain the membrane voltage at the holding potential. In this way, membrane potential is clamped to a set voltage. Data collected using a voltage clamp can be plotted in I/V curves (Box 4.2) to examine the relationship between current and voltage or to measure fluctuations in current BOX 4.2  Data Analysis: I/V Curves The I/V curve (current/voltage curve) is one of the most commonly used methods of data analysis in electrophysiology. It is simply a plot of the voltage across a neuronal membrane and the associated ionic current flow through channels in the membrane. Voltage (V) is measured in units of volts, and current (I) is measured in units of amperes. I/V curves can be produced for an entire neuron (with current typically in the range of nanoamps, nA) or for an individual ion channel or class of channels within the neuron (with current typically in the range of picoamps, pA). To produce an I/V curve, an investigator uses the voltage clamp technique to hold the voltage of a neuron at a specific value. Current can be recorded using one of the patch clamp methods. The I/V curve is the curve of best fit for each data point of current for a given voltage. By convention, currents with a negative value are referred to as inward current, while those with a positive value are known as outward current. An inward current is the result of positively charged ions crossing a cell membrane from the outside to the inside, or a negatively charged ion crossing from the inside to the outside. An outward current is the result of positively charged ions crossing a cell membrane from the inside to the outside, or a negatively charged ion crossing from the outside to the inside. Electrophysiology Chapter | 4  99 The relationship between current and voltage is described by Ohm’s law: V IR V egatlov I tnerruc R ecnatsiser R/1 ecnatcudnoc The slope of the I/V curve is the conductance of the membrane/channel for all ions that can pass through. If the investigator is interested in the conductance of a specific ion, pharmacological agents are used to isolate specific channels. The conductance depends on an ion channel’s permeability for a specific ion (how easily the channel allows the ion to pass through), as well as the concentration of the ion in the extracellular/intracellular solution. In the preceding example, the I/V curves could represent the current/voltage relationships for particular channels within a membrane. Note that the I/V curve for the channel on the right is linear, meaning it acts like a simple resistor, and the conductance of the channel is independent of voltage. An electrophysiologist would say there is no voltage-dependent gating of this ion channel. In contrast, the I/V curve on the left is not linear. At a value of about −20 mV, there is a strong inward current. If the ion that passes through this channel is positively charged, it will flow into the cell when the membrane potential is −40–160 mV. This is an indication that the channel is voltage gated. To determine the type of ion that can flow through this channel, the ionic concentrations in the bath solution can be adjusted so that the investigator can determine which ionic species are required for current to flow through the channel. BOX 4.2  Data Analysis: I/V Curves—cont’d at a particular voltage over time (Box 4.3). In current clamp mode, membrane potential is free to vary, and the investigator records whatever voltage the cell generates on its own or as a result of stimulation. Extracellular Recording The dramatic changes in membrane voltage that occur during an action potential generate local, temporary differences in potential on the outer surface of an active neuron. Therefore, action potentials can be detected in the extracellular space near the membrane of an active neuron by measuring the potential difference between the tip of a recording electrode and a ground electrode placed in a distant extracellular position. In the absence of neural activity, there is no difference in potential between the extracellular recording electrode and the ground electrode. However, when an action potential arrives at the recording position, positive charges flow away from the recording electrode into the neuron. Then, as the action potential passes by, positive charges flow out across the membrane BOX 4.3  Data Analysis: Current Over Time Many questions in neurophysiology involve knowing how much current flows through a particular ion channel for a particular voltage or other environmental conditions. An I/V curve displays the relationship between the current that passes through a channel or membrane and the voltage across the membrane. For questions involving the ability of a cell or channel to pass current under conditions other than changes in voltage (such as the presence of a compound in the bath or the temperature of the bath), data are often presented as a plot of changes in current over time. These measurements are typically taken under voltage clamp to eliminate contributions to current dynamics due to changes in voltage. The size of detected current changes depends on the patch clamp technique used to collect the data. For example, in whole-cell mode, the glass pipette is continuous with the cytoplasm, and a large current trace can be obtained. Current data in whole-cell mode are typically presented in nanoamps and can be applied to timescales ranging from 1 ms to even minutes in length. In the following wholecell recording, an agonist for a ligand-gated ion channel is applied to the bath, as indicated by the blue bar. Current data obtained from other patch clamp modes are typically smaller in value and presented in picoamps.This is because the current flowing through single channels rather than an entire cell generates a smaller signal. The timescale of these current traces is also much smaller, in the range of milliseconds instead of seconds, due to the rapid and transient events of a single channel opening and closing. In these patch clamp modes, data are often presented as a long, continuous trace. For example, in the following experiment, data are obtained in a cellattached mode from a single channel before and after application of a pharmacological agent: Notice how the scale is different from the whole-cell recording. This experiment shows that the presence of the pharmacological agent reduces the probability that the channel will be open and conduct current. Electrophysiology Chapter | 4  101 toward the recording electrode. Thus, the extracellular action potential is characterized by a brief, alternating voltage difference between the recording and ground electrodes (Figure 4.5). The results of an extracellular experiment can be displayed and analyzed in a variety of ways (Box 4.4). A major advantage to performing extracellular recordings is their relative ease and simplicity. They do not require the precise and delicate electrode positioning that is necessary in intracellular or patch clamp recordings. -20 µV 0 µV 20 µV 40 µV -40 µV -60 µV -80 µV -20 mV 0 mV 20 mV 40 mV -40 mV -60 mV -80 mV Extracellular Electrode Intracellular Electrode FIGURE 4.5  Comparison of recordings from an extracellular and intracellular electrode. Each action potential exhibits a different waveform due to its placement relative to a neuron. Note the differences in the units of measurement. BOX 4.4  Data Analysis: Spikes The goal of many electrophysiology experiments is to quantify the number of action potentials, or spikes, that a neuron fires in response to a specific stimulus. Usually, these experiments use extracellular recording electrodes, though intracellular electrodes can also detect action potentials. Data displaying changes in the number of spikes that occur over time following the presentation of a stimulus are commonly presented in several ways. For example, consider the data collected during an extracellular electrophysiology experiment in which an electrode is placed in the auditory cortex of a mouse. The investigator presents tones to the mouse and records from a neuron close to the tip of the electrode to determine whether this neuron responds to tones of a particular frequency. The simplest way to present data from this experiment is to show a plot of voltage over time: Stimulus Recording Note that changes in voltage in this plot are in the magnitude of microvolts; an intracellular electrode would record changes in voltage on the scale of millivolts. 102  Guide to Research Techniques in Neuroscience This plot clearly shows that the stimulus leads to specific changes in voltage over time. However, a simple voltage over time plot does not present the effects of multiple types of stimuli—in this case multiple frequencies of sound. To show the effects of a continuous distribution of stimuli on spike counts, there are other, more efficient methods of data analysis and presentation that can combine the results of multiple experiments with different stimuli. For example, the results of multiple frequencies can be presented as a raster plot. Time (ms) Frequency(kHz) In this type of plot, each dot represents a spike as recorded by the electrode. The Y-axis shows the stimulus frequency, while the X-axis shows the time when the spike was recorded. Stimulus onset and duration are indicated by the vertical gray bar. The data are arranged so that the spikes resulting from a stimulus are aligned with the onset of the stimulus. A raster plot provides a strong visual depiction of the kinds of stimuli that give rise to the greatest number of spikes, as well as an indication of the background noise during an experiment. Data in a raster plot can also be presented as a peri-stimulus time histogram (PSTH) (also referred to as a poststimulus time histogram). In this type of plot, spikes are quantified and sorted into bins of a defined size. In the following plot, the number of spikes (X-axis) is presented for bins that are each 2 kHz in size, with the number of spikes plotted on the Y-axis. A PSTH provides an easy visualization of the stimuli that cause the greatest proportion of firing events. Different mathematical analysis tools, such as a Gaussian distribution curve, can be used to characterize the most effective stimuli and determine the tuning properties of an individual neuron. Sound Frequency (kHz) NumberofSpikes BOX 4.4  Data Analysis: Spikes—cont’d Electrophysiology Chapter | 4  103 Finally, it is possible to analyze the effects of multiple stimulus variables using a three-dimensional heat map plot. For example, if the investigator presented tones of different intensity in addition to tones of different frequency, the data could be presented using color to indicate the number of spikes. 5 10 15 20 10 20 30 40 50 Intensity (db) Frequency(kHz) 0% 100% Efficiency (%ofstimulithat evokeaspike) This plot shows that the neuron being investigated responds to a specific frequency of tones that produce a maximum firing rate but that the neuron can be made to produce more spikes if the intensity of the stimulus is increased. Remember that the raw number of spikes that the investigator includes as data is determined by the settings on the electrophysiology rig and computer. For example, changing the threshold of voltage change that determines what is considered to be a spike will alter the number of spikes included in the data analysis. BOX 4.4  Data Analysis: Spikes—cont’d A disadvantage to extracellular recordings is that it is impossible to measure localized potentials. Therefore, extracellular recordings cannot report detailed information about subthreshold potential changes, such as EPSPs or IPSPs. However, the summed synaptic activity of neurons in the general vicinity of the extracellular recording electrode produces measurable local field potentials, the sum of all postsynaptic activity within a volume of neural tissue. By using multiple electrodes, the activity of multiple cells can be recorded at the same time. A popular configuration is the use of tetrodes, a bundle of four individual electrodes carefully arranged into a single implant (Figure 4.6). Grids of individual electrodes or groups of tetrodes can be arranged into a single device called a multielectrode array (MEA) (Figure 4.6). This allows for stimulation and extracellular recording from several neighboring sites at once. The number of microelectrodes in an array varies, but can be anywhere from four (a single tetrode) to over 100. When recording from a single electrode, it can be difficult to determine whether the neural activity that is recorded is produced by a single neuron or multiple neurons in the vicinity of the electrode tip. A single unit spike will depend on the cell’s unique shape, size, and distance from the recording 104  Guide to Research Techniques in Neuroscience electrode. Identifying individual cells based on their spiking activity and assigning the waveforms to particular cells is known as spike sorting. A major advantage to using tetrode recordings over single electrode recordings is that it is easier to compare recordings made by each individual tetrode microelectrode. Signals are processed through triangulation methods based on the position of the electrodes and the strength and characteristic shape of a recording. These comparisons result in specific and reproducible waveforms for each individual neuron recorded, enabling single unit data to be obtained by recording from multiple electrodes at once. By investigating the activity of dozens of neurons at a time, it is possible to answer questions regarding connectivity and timing within a neural network. A scientist can also manipulate specific neurons within the network and monitor the effect on other neurons recorded by the array. Ultimately, using MEAs to study the simultaneous response of neural circuits provides pivotal information on neuronal interactions, as well as spatiotemporal information about neural networks. MEAs can be used to record from multiple neurons in vitro or in vivo. In in vitro preparations, a brain slice can be placed over a grid of many microelectrodes for extracellular recordings. In vivo tetrodes or MEAs can be implanted in the intact brains of live animals for multiunit recordings. Multielectrode technology has also been used in nonhuman primates for long-term recording. For example, recordings of network activity from the premotor cortex of monkeys performing reaching movements is helping to develop neural prosthetics that allow animals (including humans) to move prosthetic devices using conscious thought alone. Tetrode Electrodes Guide tube Plug (A) Multielectrode Array Electrodes Base (B) FIGURE 4.6  Two specialized types of electrodes. (A) An individual tetrode is composed of four microelectrode wires. A scientist implants a tetrode into the brain of an animal and connects the top plug to a cable attached to an amplifier. (B) A multielectrode array is composed of 25 or more electrodes (sometimes over 100) and is used to record from neurons on the surface of the brain. Multielectrode arrays can also be used to record from slices in vitro. Electrophysiology Chapter | 4  105 Intracellular Recording While extracellular recordings detect action potentials, intracellular recordings detect the small, graded changes in local membrane potential caused by synaptic events. Intracellular recording requires impaling a neuron or axon with a microelectrode to measure the potential difference between the tip of the intracellular electrode and the reference electrode positioned outside the cell. Note that the measurement of an action potential is different for extracellular and intracellular recordings (Figure 4.5). From the point of view of an extracellular electrode, the difference in potential between an electrode and ground initially decreases, then increases, and then returns to baseline. From the point of view of an intracellular electrode, the potential difference initially increases and then returns to baseline. Also notice that the units of measurement are different between the two techniques. Intracellular electrodes measure potential differences in units of millivolts (mV). Extracellular electrodes measure much smaller potential differences and are often expressed in units of microvolts (μV). In the early days of electrophysiology, intracellular recordings were used to examine the ionic basis of membrane potential. In fact, Alan Hodgkin and Andrew Huxley won the Nobel Prize in 1963 for using intracellular methods to determine the ionic basis of action potentials. However, most experiments that once used intracellular electrodes are now usually performed using patch clamp techniques due to higher signal-to-noise ratios and the ability to ask questions about the nature of single ion channels. Indeed, after developing patch clamp techniques and using them to study the function of single ion channels, Erwin Neher and Bert Sakmann were awarded the Nobel Prize in 1991. Patch Clamp Techniques In patch clamp methods, a glass micropipette is used to make tight contact with a tiny area, or patch, of neuronal membrane. After applying a small amount of suction to the back of the pipette, the seal between pipette and membrane becomes so tight that ions can no longer flow between the pipette and the membrane. Thus, all of the ions that flow when a single ion channel opens in the patched membrane must flow into the pipette. The resulting electrical current, though small, can be measured with a sensitive amplifier. This arrangement is usually referred to as the cell-attached mode. Resistance of the seal to flow between the pipette and the cell membrane must be very high to ensure that all current to be recorded flows through the pipette and does not leak out through other regions of the membrane. In practice, seal resistances in excess of 1 GΩ can be obtained by a cell-attached patch, often referred to as a gigaseal. The cell-attached method allows the investigator to study single ion channels. Minor technical modifications provide other recording possibilities. There are three other variations of the patch clamp technique: whole-cell, inside-out, and outside-out recordings (Figure 4.7). 106  Guide to Research Techniques in Neuroscience If the membrane patch within the pipette is disrupted by briefly applying strong suction, the interior of the pipette becomes continuous with the cytoplasm of the cell. This arrangement allows measurements of electrical potentials and currents from the entire cell and is, therefore, called the whole-cell recording Cell-attached mode Whole-cell mode Inside-out mode Outside-out mode Careful retraction of the pipette causes the ends of the broken plasma membrane to anneal Gentle suction is applied to the glass pipette Tight contact between the pipette and membrane forms a “gigaseal” Strong pulse of suction opens up the membrane, making the cytoplasm continuous with the pipette Careful retraction of the pipette removes patch of membrane Extracellular Intracellular FIGURE 4.7  The patch clamp techniques. The four fundamental patch clamp techniques include the cell-attached mode, whole-cell mode, inside-out mode, and outside-out mode. Electrophysiology Chapter | 4  107 method. The whole-cell configuration also allows diffusional exchange between the pipette and the cytoplasm, producing a convenient way to inject substances into the interior of a patched cell. Two other variants of the patch clamp technique originate from the finding that once a tight seal has formed between the membrane and the glass pipette, small pieces of membrane can be pulled away from the cell without disrupting the seal. Simply retracting a pipette that is in the cell-attached configuration causes a small vesicle of membrane to remain attached to the pipette. By exposing the tip of the pipette to air, the intracellular surface of a small patch of membrane is exposed. This arrangement, called the inside-out recording configuration, allows the measurement of single-channel currents with the added benefit of being able to change the medium exposed to the intracellular surface of the membrane. Thus, the inside-out configuration is particularly valuable when studying the influences of intracellular (cytoplasmic) molecules on ion channel function. Alternatively, if the pipette is retracted while it is in the whole-cell configuration, a membrane patch is produced that reseals to expose its extracellular surface to the bath solution. This arrangement, called the outside-out recording configuration, is optimal for studying how channel activity is influenced by extracellular chemical signals, such as neurotransmitters, because the bath solution can be easily manipulated during recordings. Table 4.2 compares the relative advantages and disadvantages of the four variations of the patch clamp techniques. Sometimes electrophysiologists TABLE 4.2  Relative Advantages and Disadvantages of Different Patch Clamp Configurations Advantages Disadvantages Cell attached Single ion channels can be recorded, and channel properties are not changed. Exact potential of the patch is unknown. Physiologically relevant because cell is intact. Not easy to control environment of single ion channels. Easiest patch clamp configuration to obtain. Whole cell Ability to manipulate the cell cytoplasm pharmacologically. The relatively large size of the current makes it easier to measure. The cell cytoplasm is continuous with the recording electrode so cell contents may be diluted or leak out. Excised patch (inside-out or outside-out) Recordings can be taken from individual ion channels. Excellent environmental control over both the intracellular and extracellular sides of single ion channels. There is a risk that channel properties are changed because the cellular environment is severely altered. 108  Guide to Research Techniques in Neuroscience use a fifth method: the perforated patch. The purpose of this technique is to make the glass pipette continuous with the cell, as in the whole-cell mode, but without the disadvantage of cytoplasmic contents potentially leaking into the pipette. Instead of applying suction in the cell-attached mode to transition into the whole-cell mode, a chemical (often an antibiotic, such as amphothericin-B or nystatin) is applied from within the glass pipette, causing small perforations in the membrane. Thus, the perforated patch technique can be thought of as an intermediate between the cell-attached mode and whole-cell mode; this configuration prevents the leaking of cytoplasmic contents but generates a smaller signal than whole-cell mode. ELECTROPHYSIOLOGY TISSUE PREPARATIONS Electrophysiology experiments can be categorized by the type of tissue preparation in addition to the type of recording technique. Physiological experiments in vitro usually fall into one of three main categories: recording from heterologous expression systems, primary cultured cells, or brain slices. While in vitro preparations allow probing of cell physiology in a controlled environment (Chapter 14), conditions in the cell’s natural environment in vivo may be different. It is possible to perform experiments in vivo using extracellular techniques (and more rarely, intracellular and patch clamp techniques), either in an anesthetized animal or in an awake, behaving animal. In Vitro Recordings In vitro culture preparations provide unparalleled physical and visual access to individual cells, allowing detailed studies of the molecules and proteins that affect neuronal physiology. Access to individual ion channels, functional subcellular regions, and small circuits can be gained using various in vitro preparations: heterologous expression systems, isolated primary cells, and brain slices. Control over the cellular environment through the culture or bathing media aids the ability to finely dissect cellular and subcellular physiological events. For in vitro electrophysiology experiments, the neural specimen is incubated in a well-regulated bath solution. The composition of the solutions used during the experiments is critical because the basis of neuronal electrical properties depends on the concentration gradient of ions inside and outside the cell. When deciding what ingredients should make up the bath solution, the guiding principle is to maintain an environment that allows the observation of physiologically relevant electrical activity and channel function. In patch clamp techniques, the solution inside the microelectrode should mimic the intracellular milieu. This solution is typically composed of salts, ionic buffering agents, and a pH buffer that can be maintained at biologically relevant temperatures to provide cells and tissue with Electrophysiology Chapter | 4  109 a simulated in vivo environment. One of the greatest advantages of in vitro recordings is the ability to study the electrophysiological properties of neurons by altering components of the bath. For example, to isolate the contribution of specific channels or receptors to the electrical signal, investigators frequently add pharmacological agents that block certain receptors, such as CNQX (to block AMPA receptors) or AP5 (to block NMDA receptors), to the bath solution. Heterologous Expression Systems To study the physiological properties of individual ion channels or investigate the role of structure on a channel’s function, scientists often utilize heterologous expression systems. A heterologous expression system is a type of cell culture system that can be easily transfected with a foreign gene (see Chapter 11 for more about transfection), such as a gene that encodes a specific ion channel. The ion channel is efficiently expressed within the cells, and its function can be studied with methods such as the patch clamp technique. By expressing mutant forms of a channel, scientists can observe the effect of molecular changes on a channel’s electrical properties, such as the role of a specific amino acid on a gating mechanism. Heterologous expression systems are also useful for isolating the role of individual channels because of the absence of endogenous channels that may otherwise alter overall cell physiology. Common expression systems are Xenopus laevis oocytes, Chinese hamster ovary (CHO) cells, and human embryonic kidney (HEK 293T) cells. These cells are usually easy to culture, readily available, have few endogenous channels, and can easily take up and express foreign DNA. The ability to combine molecular and physiological methods in a single cell system has made heterologous expression systems a powerful experimental tool for understanding the structure–function relationships of ion channels (Box 4.5). Primary Cultures Primary cultures of cells isolated from a region of interest in the nervous system can reveal properties close to those found in vivo while still providing excellent access to and control of individual channels. For example, if an investigator is interested in the physiological properties of spinal cord neurons, he or she can culture these neurons for electrophysiology experiments. Such cultures are useful for comparing the electrophysiological properties of different populations of neurons, as well as the effects of different bath solutions. For example, an investigator may want to know what happens to neuronal physiology when the extracellular concentration of a particular ion or neuropeptide increases or decreases. An investigator can also compare the physiological properties of neurons from different populations of animals—for instance, to examine changes in neuronal physiology between a genetic knockout animal and a wild-type animal. 110  Guide to Research Techniques in Neuroscience BOX 4.5  Walkthrough of an In Vitro Electrophysiology Experiment Let’s say that you are interested in examining the properties of a newly discovered ion channel. Your research group has cloned the channel and created an expression construct (Chapter 10) and introduced this channel into Xenopus laevis oocytes (Chapter 11). We can now use this heterologous expression system, the large (∼1 mm) Xenopus oocytes expressing our channel of interest, to answer questions about the electrical properties of the channel. You might be interested in a number of properties: What ions pass through the channel? How is the channel gated, and what stimuli open the channel to let ions pass? You might first look at the similarity of the amino acids to known channels. This can provide hypotheses about the properties of the channel, such as whether the channel is selective for specific ions or what types of compounds may affect gating properties. To examine single-channel properties, patch clamp recordings will be the most appropriate technique. Assuming the lab has established rigs on which to work, the first step in running the experiment will typically be to make the appropriate solutions and prepare fresh glass micropipettes. The composition of both the bath solution and the solution that will fill the micropipette are important because ionic concentration gradients affect the electrical properties of the membrane. The solution’s osmolarity and pH will also affect pressure on the cell membrane, which can cause a cell to shrink or swell and burst if not properly regulated. In a whole-cell recording, the pipettefilling solution will be continuous with the inside of the cell, so it should closely match the cell cytoplasm. The bath solution can contain pharmacological agents that block the activity of known channels so the signal from the channel under investigation is clearer. The currents of endogenous Xenopus ion channels, however, are typically much smaller than those of heterologously expressed channels; thus, it is not always necessary to add blocking agents to the bath solution. For cell-attached recordings, activity of the predominant endogenous Ca2+-activated Cl− channels can be reduced by keeping the Ca2+ concentration low in the bath solution, by adding chelators such as EGTA and using impermeable ions instead of Cl−, or by adding a pharmacological blocker. Patch pipettes are generally used immediately after they are made to ensure that the tip of the micropipette is not contaminated, which can cause a poor seal to form between the membrane and the pipette. A pulled glass pipette should be firepolished to round and smooth the tip. Now it is ready to be filled with solution and positioned in the electrode holder. Using the microscope and a micromanipulator, you can bring the electrode tip in a targeted cell into the field of view. By applying gentle, positive pressure to the electrode through a tube to prevent dirtying the electrode tip, you can insert the tip into the bath solution. By applying a test voltage pulse, the patch electrode will generate a current, and you can calculate the electrode’s resistance. Observing the current response to the test pulse can help guide the electrode position. As the tip nears the cell, resistance should increase. You can now gently touch the electrode tip against the surface of the cell, checking the computer output of data to see that resistance has increased, and then release the positive pressure. This release can be observed as a small dimple against the cell surface as the tip is pressed against the cell. You can now apply Electrophysiology Chapter | 4  111 gentle suction to the patch pipette, using resistance and microscopic appearance as evidence that contact has been made between the cell surface and the patch electrode tip. If you have a gigaseal, you should see the resistance rise to at least 1 GΩ. Establishing the gigaseal is one of the trickiest parts of patch clamping—perhaps the step that makes this technique an art. At this point, you are in cell-attached mode and can switch the amplifier to voltage clamp. Cell-attached mode can provide single-channel resolution, and you may be able to see characteristic steps in the current that demonstrate transitions between the opened and closed state of a channel. Now you can test whether changes in voltage can activate the patched channel (or channels) by applying voltage steps. Alternatively, you can test whether the addition of ligands affects the patched channel by adding various agonists or antagonists to the solution filling the micropipette (as this is the extracellular environment of the patched channel). Monitoring the effect of experimental manipulations on current will reveal open/closed times and current amplitude. Current amplitude can be analyzed to determine the conductivity and ion specificity of the channel. I/V curves (Box 4.2) reveal characteristics of voltage-activated conductance, including the reversal potential, the voltage at which there is no overall flow of ions across the membrane. The reversal potential can be used to identify the ion species that pass through the channel. Analyzing current flow through individual channels in this heterologous expression system exposes many of the channel’s properties. Further studies can use mutant versions of the channel to dissect the requirements of specific aspects of the protein’s structure on its function. This knowledge can then be used to investigate the contributions of this channel of interest in intact or natural preparations: isolated neurons, brain slices, and whole animals. BOX 4.5  Walkthrough of an In Vitro Electrophysiology Experiment—cont’d Slice Cultures The controlled in vitro environment in cell culture conditions can be substantially different from the in vivo environment. Brain slices, which preserve some endogenous connections while still providing the level of access available through cell culture conditions, can more closely mimic the in vivo environment. While it is possible to culture tissue slices for extended periods of time to capture conditions closest to those in vivo, most electrophysiological recordings from slices are from acute preparations cut the same day. At the beginning of the experiment, the brain is removed and sliced into 200–500-μm-thick sections using a vibratome. Many neurons remain healthy despite the mechanical shock and damage of slicing, though physiological responses may be slightly altered. The brain slice is placed in a chamber that is continuously bathed in a solution containing the proper proportion of inorganic ions, nutrients, and gases to allow the neurons to survive. There are several compelling advantages for recording from a brain slice rather than an intact brain. First, it is much easier to make intracellular recordings 112  Guide to Research Techniques in Neuroscience from neurons if the tissue is not subject to the periodic pulsing of blood caused by a beating heart. Second, it is also easier to study neurons from a particular region of the brain if an electrode does not have to penetrate several millimeters of cells before it can get to that region. Brain slices provide much easier access to internal brain structures. Third, it is possible to study the pharmacology of known synapses because specific drugs or other pharmacological agents can easily be applied to a brain slice. In addition, the specific role of individual neurons in circuits can be studied because it is possible to record from both pre- and post synaptic neurons in a known synaptic circuit, if they can be accessed in the same section. In Vivo Recordings Though in vitro recordings allow control of the environment and access to the brain, they may not accurately reflect neural activity in an intact organism. Most electrophysiology experiments performed in vivo submerge an electrode into the brain and record extracellularly due to the fact that it is difficult to visualize the cellular environment and extremely difficult to correctly position an electrode for intracellular or patch clamp recordings. Therefore, most in vivo recordings are extracellular, performed using electrodes, tetrodes, or multielectrode arrays. However, it is possible to perform intracellular or even patch clamp recordings with specialized, dedicated equipment. To perform intracellular recordings, a scientist can use a very sharp pipette to impale cells for sharp electrode recordings. To perform in vivo patch clamp recordings, a scientist can use twophoton microscopy (see Chapter 5) in conjunction with an animal expressing a fluorescent protein in a cell type of interest to visualize the placement of the patch pipette on the surface of the membrane. A scientist can also perform patch clamp recordings blind, but this approach is quite difficult. Acute versus Chronic Recordings While most in vitro recordings are performed during a single session, in vivo recordings can be performed either acutely (over hours) or chronically (over multiple days). An acute recording is an experiment that uses an animal only once. To perform a chronic recording, cannulae or electrodes can be surgically implanted into an animal’s brain, permitting long-term access to a recording site. Chronic recordings from implanted multielectrode arrays permit the monitoring of neural circuit activity during complex behavioral tasks over long periods of time or during a variety of different conditions and multiple tasks. Sealable chambers implanted on the skull allow an investigator to place new electrodes within the brain over periods of months in rodents or years in nonhuman primates. Anesthetized versus Awake Animals In vivo electrophysiology experiments often involve animals stabilized under anesthesia. Anesthetized animals are immobile, allowing scientists to record neuronal responses to passive stimuli and characterize tuning properties of Electrophysiology Chapter | 4  113 single neurons. Anesthesia affects the excitability and neurotransmission of normal neural activity, with varied effects depending on the anesthetic agent or neuron population of focus. Thus, it is important for an investigator to know the specific effects of the anesthetic used to be able to properly interpret the data. For some experiments, the animals are also paralyzed to eliminate motion from breathing—in these instances, artificial, controlled ventilation is provided. Recording from awake, behaving animals make it possible to correlate specific behaviors with the activity of individual cells. An animal is first trained on one or more specific behavioral tasks. After a certain level of performance has been reached, a surgical operation is performed to create a small opening in the skull over the brain area of interest. This area is then sealed by implanting a chamber with a screwable cap (Chapter 3). After the animal has recovered from the operation, electrophysiological recordings can be performed for a few hours each day with an electrode mounted on a lightweight microdrive. Single or multiunit recording during a behavioral task allows scientists to examine the neural basis of various cognitive and behavioral phenomena. MANIPULATING NEURAL ACTIVITY DURING ELECTROPHYSIOLOGY EXPERIMENTS During neurophysiological recordings, an investigator may wish to inhibit or stimulate certain neurons for a loss of function or gain of function experiment. For example, if an investigator hypothesizes that neuron A stimulates neuron B, a way to test this hypothesis would be to stimulate neuron A while recording from neuron B. Alternatively, an investigator may want to know the contribution of certain types of neuron (e.g., GABAergic, cholinergic, glycinergic) to a physiological process; in this instance, the investigator could inhibit only those neurons while examining the physiological properties of a neuron of interest. Methods of perturbing neural activity during electrophysiology experiments are covered in detail in Chapter 8: Manipulating Neural Activity. An increasingly popular technique called dynamic clamp allows scientists to simulate artificial membrane conductances or synaptic connections in living cells, even creating hybrid circuits between real and simulated neurons. In a dynamic clamp experiment, a cell’s voltage is monitored, typically using whole-cell patch clamp, and fed into a computational model, either in the computer software or implemented through hardware as an analog chip. The model calculates and then sends back a current to inject into the cell, creating a realtime feedback loop. This applied current creates an experimental condition in which a channel with the model’s properties seems as if it is embedded in the cell’s membrane, creating an artificial conductance. A dynamic clamp can be used to simulate specific channel knockins, artificial synaptic inputs, or virtual pharmacology experiments. 114  Guide to Research Techniques in Neuroscience CONCLUSION Electrophysiology remains the technique of choice for analyzing neural activity and the physiological properties that give rise to this activity. A wide range of techniques and tissue preparations make it possible to record the activity of neurons in a dish, a slice, or an awake, behaving animal. Many scientists, even nonelectrophysiologists, consider electrophysiology techniques to be the backbone of neuroscience research—they are the primary methods to precisely investigate the activity of neurons that produce cognition and behavior, the ultimate output of the nervous system. SUGGESTED READING AND REFERENCES Books The axon guide: A guide to electrophysiology & biophysics laboratory techniques (3rd ed.). (2008). Sunnyvale, CA: Molecular Devices/MDS Analytical Technologies. Boulton, A. A., Baker, G. B., & Vanderwolf, C. H. (1990). Neurophysiological techniques: Applications to neural systems. Clifton, NJ: Humana Press. Hille, B. (2001). Ion channels of excitable membranes (3rd ed.). Sunderland, MA: SinauerAssociates. Molleman, A. (2003). Patch clamping: an introductory guide to patch clamp electrophysiology. NY: Wiley. Nicholls, J. G., Wallace, B. G., Martin, A. R., & Fuchs, P. A. (2001). From neuron to brain (4th ed.). Sunderland, MA: Sinauer Associates. Nicolelis, M. A. L. (2008). Methods for neural ensemble recordings (2nd ed.). Boca Raton, FL: CRC Press. Review Articles Buzsaki, G. (2004). Large-scale recording of neuronal ensembles. Nature Neuroscience, 7, 446–4451. Cunningham, J. P., &Yu, B. M. (2014). Dimensionality reduction for large-scale neural recordings. Nature Neuroscience, 17, 1500–1509. Economo, M. N., Fernandez, F. R., & White, J. A. (2010). Dynamic clamp: alteration of response properties and creation of virual realities in neurophysiology. Journal of Neuroscience, 30, 2407–2413. Miller, E. K., & Wilson, M. A. (2008). All my circuits: using multiple electrodes to understand functioning neural networks. Neuron, 60, 483–488. Quian Quiroga, R., & Panzeri, S. (2009). Extracting information from neuronal populations: information theory and decoding approaches. Nature Reviews Neuroscience, 10, 173–185. Super, H., & Roelfsema, P. R. (2005). Chronic multiunit recordings in behaving animals: advantages and limitations. Progress in Brain Research, 147, 263–282. Windels, F. (2006). Neuronal activity: from in vitro preparation to behaving animals. Molecular Neurobiology, 34, 1–26. Primary Research Articles—Interesting Examples from the Literature Bennett, C., Arroyo, S., & Hestrin, S. (2013). Subthreshold mechanisms underlying state-dependent modulation of visual responses. Neuron, 80, 350–357. Bliss, T. V., & Lomo, T. (1973). Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. Journal of Physiology, 232, 331–356. Electrophysiology Chapter | 4  115 Evarts, E. V. (1960). Effects of sleep and waking on spontaneous and evoked discharge of single units in visual cortex. Federation Proceedings, 19, 828–837. Evarts, E. V. (1968). A technique for recording activity of subcortical neurons in moving animals. Electroencephalography and Clinical Neurophysiology, 24, 83–86. Foster, D. J., & Wilson, M. A. (2006). Reverse replay of behavioural sequences in hippocampal place cells during the awake state. Nature, 440, 603–680. Kaufman, M. T., Churchland, M. M., Ryu, S. I., & Shenoy, K. V. (2014). Cortical activity in the null space: permitting preparation without movement. Nature Neuroscience, 17, 440–448. Lin, D., Boyle, M. P., Dollar, P., Lee, H., Lein, E. S., Perona, P., et al. (2011). Functional identification of an aggression locus in the mouse hypothalamus. Nature, 470, 221–226. Mante, V., Sussillo, D., Shenoy, K. V., & Newsome, W. T. (2013). Context-dependent computation by recurrent dynamics in prefrontal cortex. Nature, 503, 78–84. Sharp, A. A., O’Neil, M. B., Abbott, L. F., & Marder, E. (1993). Dynamic clamp: computergenerated conductances in real neurons. Journal of Neurophysiology, 69, 992–995. Protocols Brown, A. L., Johnson, B. E., & Goodman, M. B. (2008). Patch clamp recording of ion channels expressed in Xenopus oocytes. Journal of Visualized Experiments, 20. http://dx.doi. org/10.3791/936. http://www.jove.com/index/details.stp?id=936. Current protocols in neuroscience, Chapter 6: neurophysiology. (2007). John Wiley & Sons, Inc. Kodandaramaiah, S. B., Franzesi, G. T., Chow, B.Y., Boyden, E. S., & Forest, C. R. (2012).Automated whole-cell patch-clamp electrophysiology of neurons. In Vivo Nature Methods, 9, 585–587. Martina, M., & Taverna, S. (Eds.). (2014). Patch-clamp methods and protocols (methods in molecular biology) (2nd ed.). Totowa, NJ: Humana Press. Murthy, M., & Turner, G. (2013). Whole-cell in vivo patch-clamp recordings in the Drosophila brain. Cold Spring Harbour Protocals, 2013(2), 140–148. Nicolelis, M. A., Dimitrov, D., Carmena, J. M., Crist, R., Lehew, G., Kralik, J. D., et al. (2003). Chronic, multisite, multielectrode recordings in macaque monkeys. Proceedings of the National Academy of Sciences of the United States of America, 100, 11041–11046. Perkins, K. L. (2006). Cell-attached voltage-clamp and current-clamp recording and stimulation techniques in brain slices. Journal of Neuroscience Methods, 154, 1–18. Tammaro, P., Shimomura, K., & Proks, P. (2008). Xenopus oocytes as a heterologous expression system for studying ion channels with the patch-clamp technique. Methods in Molecular Biology, 491, 127–139. 117 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00005-8 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 5 Microscopy After Reading This Chapter, You Should be Able to: l Define magnification and resolving power, as well as the parts of a microscope that affect these parameters l Describe the fundamental parts of a standard light microscope and how they manipulate light to magnify an image l Compare the relative strengths and limitations of common forms of microscopy l Discuss issues related to image preparation and analysis Techniques Covered: l Light microscopy: brightfield, phase contrast, darkfield, differential interference contrast (DIC/Nomarski) l Fluorescence microscopy: epifluorescence, confocal, two-photon laser scanning microscopy, total internal reflection fluorescence (TIRF) microscopy l Electron microscopy: transmission electron microscopy (TEM), scanning electron microscopy (SEM), electron tomography (ET) l Microscopy data: image processing and interpretation The first microscopes were used in the seventeenth century to expose the microscopic world of cells and single-celled organisms. Scientific pioneers such as Robert Hooke and Anton von Leeuwenhoek, often called the fathers of microscopy, used homemade microscopes to study cell types from a variety of living organisms. In the late nineteenth and early twentieth centuries, Santiago Ramón y Cajal used a microscope in combination with histology to produce highly detailed, seminal studies of nervous system structure. The microscope is now an indispensable tool in neuroscience. Because organelles, glial cells, neurons, and even populations of neurons cannot be seen by the naked eye, a microscope is essential for examining the nervous system at the cellular level. Light microscopes can enlarge the image of an object up to 1000 times greater than normal, providing access to the structure of a cell and its local environment. Fluorescent microscopes provide an even greater ability to highlight individual subcellular structures. Electron microscopes can theoretically enlarge an image 1 million times greater than normal, providing unparalleled insight into the smallest neuronal structures including synapses, surface receptors, and even individual proteins. 118  Guide to Research Techniques in Neuroscience The goal of this chapter is to provide a fundamental understanding of common forms of microscopy. First, we define the fundamental parameters and parts of a light microscope. Next, we survey the different forms of microscopy and why an investigator might choose one form over another. Finally, we examine issues related to processing and interpreting microscopy data. The information in this chapter will complement information in many other chapters, especially methods in Chapters 6 and 7 used to investigate neural structure and function. ESSENTIAL PRINCIPLES OF MICROSCOPY While modern microscopes are undoubtedly more advanced than those used a few centuries ago, the fundamental parts of a microscope and the theory behind how they work together are essentially the same. Before surveying different forms of microscopy, we review the fundamental parameters and parts of standard light microscopes. Fundamental Parameters in Microscopy There are two important values to consider in microscopy: magnification and resolution. Magnification refers to how much larger the sample appears compared to its actual size. Figure 5.1 provides a sense of the relative size of nervous system components, from an entire brain down to individual atoms. The naked eye is able to perceive objects about 0.2 mm in size and larger. Therefore, it is possible to view entire brains and large neural structures, but not individual neurons or axons without additional magnification. Resolution (or resolving power) refers to the minimum distance by which two points can be separated yet still be distinguished as two separate points. For example, imagine looking at the side of a building with a mosaic image made up of thousands of colored tiles. If you were standing a block away from the building, you would see the entire image but would be unlikely to make out the individual tiles themselves because the resolution would be too low to see that level of detail. If you move closer, the resolution becomes better and you can see the distinct tiles making up the image. Likewise, in microscopy, it can be difficult to perceive two structures as separate from one another—only by increasing the resolving power is it possible to make out distinct points in a specimen. Figure 5.1 characterizes the resolution of commonly studied neural structures, demonstrating the range of objects that can be resolved by the naked eye and the increased resolving power provided by light and electron microscopy (EM). Though it may seem that the resolving power of a microscope is based on the ability of the microscope to magnify an object, this is not the case. Magnifying an object does not necessarily improve the clarity or resolution of the final image. The resolution of a magnified object depends on two factors: numerical aperture and the wavelength of the light source. The numerical aperture (NA) is a measure of the light-collecting ability of the microscope objective, the lens that gathers and focuses light from the specimen. An objective with a higher NA will Microscopy Chapter | 5  119 collect more light rays, leading to better resolving power. The NA of an objective depends on the angle that light enters into the objective and the index of refraction of the medium in which the objective is working (Figure 5.2). The index of refraction (or refractive index) indicates changes in the speed of light traveling Mouse brain Cell body Dendrites Brain nuclei µ µ µ Dendritic spines Synapses Synaptic vesicles Ion channels DNA Ions Atoms FIGURE 5.1  Resolving power in microscopy.  The resolution required for distinguishing neuroscience structures, from a rodent brain to individual atoms. 120  Guide to Research Techniques in Neuroscience through a particular medium (1.0 for air, 1.33 for water, and up to 1.55 for oils). The greater the index of refraction, the greater the NA. As light travels from one medium (e.g., air) to a medium with a higher index of refraction (e.g., oil), the angle of the light changes, allowing an objective to collect more light. This is the same principle that causes a straw in a glass of water to appear bent at the surface of the water. An objective lens immersed in oil has a higher NA than an objective in air, and thus the oil-immersed objective has a greater resolving power. The other factor that determines the resolution of a magnified object is the wavelength of light either illuminating or emanating from the specimen. In light microscopy, light illuminates the specimen; in fluorescent microscopy, light emanates from the specimen. Wavelength is the distance between two repeating units of a light wave (Figure 5.3A). Light rays with different wavelengths appear as different colors to the human eye (Figure 5.3B). The wavelength of light affects how broad a single point of light appears. Shorter wavelengths of light (UV through green) appear sharper than longer wavelengths (red through infrared), which tend to appear more spread out, decreasing resolving power. Therefore, the smallest object that a microscope can display is a function of the magnification and resolving power used to image a specimen. Interestingly, while the ability of a microscope to magnify an object is essentially unlimited, the resolution is finite. A light microscope cannot resolve details of a specimen finer than 0.2 μm—about the size of a neural synapse—no matter what the magnification. This limitation is due to the minimum wavelength of visible light (Box 5.1); wavelengths shorter than the UV to green portion of the spectrum are invisible to the human eye. This limitation causes the resolving power of any light microscope to have a maximal value. The unaided human eye is limited to seeing details about 0.2 mm in size. A light microscope can only resolve details of about 0.2 μm, so magnifying those details more than 1000× (1000 × 0.2 μm =  200 μm = 0.2 mm) would not be useful. Increasing the magnification higher α1 α2 1 2 FIGURE 5.2  Numerical aperture (NA) and the angle of light entering the objective. An objective with a higher NA collects more light rays, leading to a higher resolving power. The NA depends on the angle that light enters the objective, α, as well as the refractive index of the medium, n. When light travels through material with a different n, such as between glass and air (n1), it refracts more and yields a smaller α. Immersion, such as water or oil (n2), reduces the difference and allows for a greater α. Microscopy Chapter | 5  121 than 1000× is like using the zoom feature on a computer to enlarge a digital picture of a fixed size: you won’t get any finer details with a digital zoom—you simply make the pixels bigger. However, technologies that do not use visible light to image a specimen, such as EM, can have much higher resolving powers and can image much smaller structures than light microscopes. Now that we have discussed magnification and resolution, letus examine the fundamental parts and design of two categories of microscopes: the compound microscope and the stereomicroscope. The Design of a Compound Microscope The simplest form of microscopy, one that has been used for centuries, is the magnifying glass. A magnifying glass is a convex lens that enlarges nearby objects (Figure 5.4). It works by refracting the light scattered by an object to a focal point on the other side of the glass. What appears is a larger image of the object behind the lens. A single lens has limited magnification ability, but multiple lenses can be arranged, one behind the other to multiply the magnification of each lens. This is the basis of the compound microscope. The term compound refers to the two or more lenses used in concert to magnify an object. λ λ FIGURE 5.3  The visible spectrum of light. (A) The wavelength of light, λ, is the distance between two repeating units of a propagating light wave. (B) Light rays of different wavelengths appear as different colors to the human eye. Shorter wavelengths appear violet-blue while longer wavelengths appear redder. (C) Approximate peak excitation (black) and emission (blue) wavelengths of commonly used fluorophores. 122  Guide to Research Techniques in Neuroscience Most light microscopes use at least two lenses. The first lens is called the objective lens and is placed adjacent to the specimen (Figure 5.5). Investigators can choose from multiple objective lenses for a desired degree of magnification, usually 4×, 10×, 20×, 40×, or 100×. The second lens is located in the eyepiece and is referred to as the ocular lens. This lens is often set at 10×. The total magnification of the microscope is multiplicative, so if the objective lens is set at 40× and the ocular lens is set at 10×, the total magnification achieved is 400×. A condenser focuses light from a light source onto a specimen (Figure 5.6). Light transmitted through the specimen is then collected and magnified by the objective lens. This magnified image is focused onto the ocular lens, which essentially acts like a magnifying glass to further enlarge the image. Although each manufacturer designs light microscopes with unique features, the conceptual design of a compound microscope remains the same. BOX 5.1 The Diffraction Limit of Resolution and Ways around It Why does the wavelength of visible light ultimately set the limit of resolution for light microscopes? The reason is due to a phenomenon known as diffraction. Diffraction can be thought of as the tendency for a wave to spread out and bend around obstacles as it propagates. Light behaves like a wave, so it too can bend around obstacles, just like water ripples moving past a boulder in a river. When a single point of light passes through the opening of a microscope objective, it does not appear as an infinitely small single point of light; it has a finite size that looks more like a bright point with concentric rings of decreasing brightness, like the rings of ripples you see in a puddle when a raindrop hits. This ring of light ripples is known as an Airy disk. Because each point of light is really a ring of light ripples, the distance between two points needs to be large enough so the ripples do not interfere with each other. Diffraction is what causes the points of light to create these ripples. Because the ripples are the factor that limit how close together two points of light can be and still be seen as two distinct points of light, this is known as the diffraction limit of resolution. The minimum wavelength for light visible to the human eye (∼400 nm) sets this diffraction limit. New forms of superresolution microscopy attempt to get beyond the diffraction limit of resolution to image nanoscale molecular interactions. In fact, work on superresolution fluorescent microscopy received the Nobel Prize in 2014. Stimulated emission depletion (STED), ground-state depletion, and saturated structuredillumination microscopy break the diffraction limit by controlling the spatial and/ or temporal transition between a fluorophore’s molecular states (e.g., dark vs. bright state).This is done by modifying the excitation light pattern to effectively shrink the size of the Airy disk. Demonstrating its importance in cellular and molecular neuroscience, STED was used to produce video-rate images of synaptic vesicles in live hippocampal neurons at a resolution of 60 nm. Photoactivated localization microscopy, fluorescence photoactivation localization microscopy, and stochastic optical reconstruction microscopy break the diffraction limit by detecting the positions of single molecules. By using photo-switchable fluorescent probes, overlapping individual fluorophores can be separated in time to construct a superresolution image. Microscopy Chapter | 5  123 FIGURE 5.4  The magnifying glass. A magnifying glass is a convex lens that refracts light to make an object appear larger. Eyepieces (ocular lens) Camera Objective Lenses Excitation and Emission Filters (inside) Shutter Stage Mercury lamp source Transmitted light source Focus knob Condenser FIGURE 5.5  Components of a compound microscope. In brightfield microscopy, the light comes from the transmitted light source. For fluorescent microscopy, the light comes from a mercury lamp (green arrow). Light is transmitted through the specimen or excited from the specimen and travels to the eyepieces or camera (red arrow). 124  Guide to Research Techniques in Neuroscience Compound microscopes are classified as being either upright or inverted. In an upright microscope, the specimen is placed just below the objective (Figure 5.7A). This type of microscope is ideal for examining a specimen mounted on a glass slide. However, there is little space between the objective and specimen, Ocular lens Objective lens Condenser lens Specimen Magnified Specimen FIGURE 5.6  Compound microscopes use multiple lenses to magnify a specimen. A condenser lens focuses light onto a specimen, which then transmits (or emits) light through the objective and ocular lenses that magnify the image that reaches the eye. Microscopy Chapter | 5  125 making it impractical for manipulating brain sections for electrophysiology or for viewing live cells in a thick cell culture dish. In such situations, it is better to use an inverted microscope, which positions the objective below the specimen with the light source and condenser located relatively far above the sample (Figure 5.7B). The other parts of a basic compound microscope (Figure 5.5) allow the investigator to adjust the magnification, focus, and light level of the microscope, as well as adjust the precise placement of the specimen. The Design of a Stereomicroscope A stereomicroscope, also called a dissecting microscope, serves a different purpose than a compound microscope and works in a different way (Figure 5.8). The main purpose of a stereomicroscope in neuroscience is to examine the surface of brains, tissue slices, or large neural structures. It is especially useful for the fine manipulation needed during dissections, surgeries, or the fabrication of small tools such as electrodes or implants. Rather than passing through a single objective and ocular lens system, light in a stereomicroscope goes through two separate lens systems. While a compound microscope directs light to both eyes from a single lightpath, a stereomicroscope directs light to each eye from two independent lightpaths. Because light from a single point on the specimen travels independently through two different paths to reach each eye, the specimen appears three dimensional. In addition to the appearance of the specimen, there are at least two other major differences between a stereomicroscope and a compound microscope. First, stereomicroscopes typically use light that is reflected off a specimen, while compound microscopes often use light transmitted through a specimen. This makes stereomicroscopes useful for looking at specimens too thick for FIGURE 5.7  Upright versus inverted microscopes.  (A) In an upright microscope, the specimen is placed just below the objective. (B) In an inverted microscope, the specimen is placed just above the objective, allowing for larger cell culture plates and access from above. 126  Guide to Research Techniques in Neuroscience light to pass through. Second, the magnification power of a stereomicroscope is often less than that of a compound microscope. The ocular lenses of a stereomicroscope are usually fixed at 10× and the objective lenses typically range from 0.1 to 8×. Therefore, a stereomicroscope is useful for examining gross neural structures, while a compound microscope is better for examining microscopic structures such as single neurons or fiber tracts. Now that we have examined the basic concepts of microscopy, we survey the various categories of light microscopy and why each may be particularly useful for a given experiment. LIGHT MICROSCOPY A light microscope is any microscope that uses visible light to illuminate and image a specimen. This includes white light composed of all wavelengths, as well as the light of a specific wavelength used in fluorescent microscopy. Usually when people refer to light microscopy, they refer to nonfluorescent microscopy, even though fluorescent microscopy does, of course, use light. The most common and general form of light microscopy is brightfield microscopy, in which light passes directly through or is reflected off a specimen. Most cells and tissues are transparent due to their high water content. Unless naturally pigmented or artificially stained, distinct structures are FIGURE 5.8  A standard stereomicroscope. Microscopy Chapter | 5  127 difficult to differentiate using brightfield microscopy. Thus, a variety of histological procedures have been developed to preserve and stain specimens to enhance contrast among different microscopic structures (Chapter 6). However, most of these procedures result in the death of cells to preserve the specimen. For many experiments, working with preserved cells is not a problem. However, when an investigator wishes to magnify living, unpreserved cells or tissues, as in cell culture experiments, it is necessary to provide contrast without killing the specimen. Therefore, methods of manipulating light to enhance contrast have been developed to make resolvable details stand out to the eye without requiring the application of special dyes. Variations in density within cells cause tiny differences in the way different regions scatter light; therefore, different cellular structures have different indices of refraction. Phase-contrast microscopy takes advantage of these slight differences, amplifying them into larger intensity differences with high contrast that can be easily seen. Because this form of microscopy does not require special treatment of the specimen to see many details, phase contrast is often used to examine cultured cells. Darkfield (or darkground) microscopy illuminates a specimen using oblique rays of light directed from the side so that only scattered light enters the objective lens. Most regions of the specimen (e.g., cytoplasm) do not scatter much light so they appear dark, while highly refractive organelles that scatter large amounts of light will appear bright. Differential interference contrast (DIC) microscopy, also known as Nomarski microscopy, exaggerates changes in the light-scattering properties of cellular structures, highlighting the edges around structures to create a textured, three-dimensional appearance. Figure 5.9 demonstrates the different appearances of a single cell when viewed using these various forms of light microscopy. (A) (B) (C) (D) FIGURE 5.9  Four types of light microscopy used to analyze the same cell. Various methods of light microscopy are used to enhance contrast and visualize details in unstained tissue preparations. (A) Brightfield microscopy. (B) Phase-contrast microscopy. (C) Differential interference contrast (DIC/ Nomarski) microscopy. (D) Darkfield microscopy. Reprinted from Alberts, B., et al. (2002). Molecular Biology of the Cell, (4th ed.). Fig. 9–8, with permission from Garland Science: NewYork, NY. 128  Guide to Research Techniques in Neuroscience FLUORESCENCE MICROSCOPY Fluorescence microscopy takes advantage of specialized molecules called fluorophores that have the property of absorbing light at a specific wavelength and then emitting light at a different (typically longer) wavelength. Fluorophores that absorb blue light typically emit green light; fluorophores that absorb green light emit red; fluorophores that absorb red emit infrared, and so on. There are dozens of commercially available and commonly used fluorophore molecules each with their own characteristic absorption (excitation) and emission wavelengths (Figure 5.3C). These molecules can be linked to antibodies or other molecular probes to indicate the presence of specific proteins, organelles, or other structures within a cell. The principal use of fluorescence microscopy is to examine specimens labeled with these fluorescent reagents or expressing genetically encoded fluorescent proteins (Chapter 6). Light from an extremely bright source passes through an excitation filter that allows only light in a specific range of wavelengths through the objective to illuminate the specimen. That light is absorbed by and excites the fluorophores in the specimen, which react by emitting a longer wavelength of light. The emitted light is collected back through the objective and passed through a second filter, an emission filter. The emission filter blocks extraneous wavelengths of light, including the light used to illuminate the specimen, but allows emitted light wavelengths to pass through to a detector (i.e., your eye or a camera). The optical filters make it appear as if structures tagged with the fluorophores light up against a dark background. Because the background is dark, even a tiny amount of the glowing fluorescent reagent is visible, making fluorescence microscopy a very sensitive technique. A major appeal of fluorescent microscopy over conventional light microscopy is that structures and proteins within a cell can be labeled with a wide range of fluorophores. Because each fluorophore has its own characteristic excitation and emission spectra (the wavelengths of light they absorb and emit), it is possible to label different structures in the same sample and investigate their relative locations. For example, an investigator can use a fluorophore that emits light in the green range of the spectrum to label one protein, and a fluorophore that emits light in the red range of the spectrum to label another. Because the excitation and emission spectra of the green and red fluorophores do not overlap, they distinctly mark the locations of these different proteins within the same sample. This technique can even be used in living specimens to observe the dynamic interactions of multiple fluorescently labeled proteins and molecules. Despite the advantages of fluorescence microscopy, there are some disadvantages. One major limitation is that fluorescent reagents cannot be illuminated indefinitely. The intensity of the light emitted from a fluorophore will decrease over time as it is continuously exposed to light, a process called photobleaching. This makes it necessary to limit the light exposure of fluorescently tagged specimens and capture images of them before fluorescence becomes too dim.Another limitation in live cell imaging is phototoxicity, in which illumination leads Microscopy Chapter | 5  129 to the death of cells expressing the fluorophore due to free-radical generation. Finally, while fluorescently labeled structures can be detected easily, in practice, background noise can mask the actual signal of interest. Background noise is nonspecific fluorescence that does not indicate true specific signal. One form of background can be autofluorescence, in which certain structures, chemicals, and organisms naturally fluoresce without the addition of a fluorophore label. If this fluorescence has the same emission wavelength as the fluorescent label, it can be difficult to determine true signal from noise. Fluorophores have advanced every area of neuroscience. Fluorescent molecules have made breakthroughs possible in studying neural structure and function, identifying the spatial relationships of proteins within a cell, examining the fine branching of neural structures, and detailing the time course of a cell’s functional response to chemical or electrical stimulation. Various forms of microscopy attempt to maximize the advantages and minimize the disadvantages of fluorescent reagents. We now survey the main categories of fluorescence microscopy. Epifluorescent Microscopy The most elementary form of fluorescent microscopy is epifluorescent microscopy, also known as wide-field fluorescent microscopy. Epifluorescent microscopes work as just described: specimens labeled with fluorescent probes are illuminated by light of the excitation wavelength (Figure 5.10A). The specimen is then viewed using a second filter that is opaque to the excitation wavelength but transmits the longer wavelength of the emitted light. Thus, the only light transmitted through the eyepiece is the light emitted by the specimen. Epifluorescence is an excellent tool, but has one major disadvantage: light excites fluorophores throughout the entire depth of the specimen. Thus, fluorescence signals are collected not only from the plane of focus but also from areas above and below this plane. Such background fluorescence can lead to hazy, out-of-focus images that appear blurry and lack contrast. Thus, the value of epifluorescence in relatively thick specimens (>15–30 μm) is limited. Alternative forms of fluorescent microscopy attempt to limit out-of-focus fluorescence to produce sharper, clearer fluorescent images. Confocal Microscopy Confocal microscopes produce clear images of structures within relatively thick specimens by selectively collecting light from thin (<1 μm) regions of the specimen. It is the tool of choice for examining fluorescently stained neurons in brain slices or small, intact organisms such as Drosophila or embryonic zebrafish. With epifluorescence, the entire specimen is illuminated at once so all the fluorophores emit light at the same time, making it difficult to observe signals from a specific focal plane within a thick specimen. In a confocal system, light from 130  Guide to Research Techniques in Neuroscience FIGURE 5.10  Comparison of different forms of fluorescence microscopy: epifluorescence, confocal, and two-photon.  (A) Basic strategy of an epifluorescent microscope. Light from a bright lamp passes through a filter that only allows light of the excitation wavelength through to the specimen. Light emanates from the specimen and passes through a second filter that is opaque to the excitation wavelength of light but that transmits the longer wavelength of the emitted light. (B) Basic strategy of a confocal microscope. Light is focused from a pinhole aperture onto a single point on the specimen. Fluorescent light emanating from the specimen is transmitted through a second pinhole aperture before hitting a detector. The two apertures are confocal with each other, so out-of-focus fluorescence is avoided by the detector pinhole. (C) Basic strategy of a two-photon microscope. Pulsed laser light illuminates the sample with light about twice as long as the wavelength required for excitation. When focused on a specific plane of section, two photons can arrive near-simultaneously at one fluorophore in that plane and deliver enough energy to excite the fluorophore. Above and below the plane in focus, light is too low-energy to excite fluorescence, eliminating out-of-focus fluorescence. Microscopy Chapter | 5  131 a laser passes through a pinhole aperture that focuses the light at a specific depth in the specimen (Figure 5.10B).Although this pinhole aperture restricts focused illumination to one plane, other regions receive dispersed, less intense illumination and emit out-of-focus fluorescence. However, all the emitted light passes through a second pinhole aperture before reaching a detector. The key aspect of confocal microscopy is that the pinhole aperture in front of the detector is at a position that is confocal with the illuminating pinhole—that is, the only light to reach the detector comes from the same plane in the specimen where the illuminating light comes to a focus, minimizing background fluorescence and maintaining sharp focus on a single plane. The detected light is digitized and sent to a computer for display and storage. To generate a two-dimensional image, data from each point in the plane of focus are collected sequentially by scanning across the field. Confocal microscopes can focus the laser beam at precise intervals throughout the thickness of the specimen, producing optical sections at multiple focal planes. Computer image analysis can then be used to stack these sections to reconstruct a threedimensional volume with optimal contrast and resolution. A major advantage of confocal microscopy to standard epifluorescence is the ability to produce sharp images of cells and cellular structures without background fluorescence (Figure 5.11). However, there are still limitations. Confocal lasers scan specimens point by point to form a complete image of the specimen. Unfortunately, this scanning requires that confocal microscopes expose samples to intense light for a longer time period than epifluorescent microscopes, making photobleaching and phototoxicity more problematic. The longer time necessary to capture an image also makes it less desirable for live cell imaging of extremely fast events. In addition, though it is a great advantage to be able to create threedimensional reconstructions by stacking multiple optical sections, the lateral X-Y resolution is finer than resolution in the Z (depth) dimension, so threedimensional reconstructions may appear warped. Despite these disadvantages, (A) (B) FIGURE 5.11  Comparison of images from epifluorescence and confocal microscopy.  (A) A standard epifluorescent microscope captures fluorescence emitted from all planes, not just those in focus, leading to a hazy background. (B) A confocal microscopy image of the same specimen appears sharper and in focus. 132  Guide to Research Techniques in Neuroscience confocal microscopy is the gold standard for imaging, especially for examining brain sections or small invertebrate organisms. Two-Photon Microscopy Two-photon microscopy, also referred to as two-photon laser scanning microscopy, is a further refinement of precision fluorescence microscopy. In epifluorescent or confocal microscopy, a single photon of light at a specific wavelength excites a fluorophore to emit light. In two-photon microscopy, a fluorophore can be excited by effectively absorbing two photons simultaneously. In this case, each photon by itself has a wavelength that is twice the usual excitation wavelength, and thus has half the energy level per photon. When a fluorophore absorbs the two lower-energy photons at the same time, each contributes half the needed excitation energy to interact with the fluorophore as if a single photon with more energy and a shorter wavelength had struck it (Figure 5.10C). The main advantage of this method is that the near-simultaneous arrival of two photons is an extremely rare event, so fluorescence excitation is restricted to a narrow plane of focus within the specimen. This provides clearer images than traditional confocal microscopy, even in cases where relatively few fluorophores are present, because photons are only emitted from the excited fluorophores located at the focal plane. Thus, all emitted light can be collected. There is not enough energy to excite fluorophores located above and below the focal plane to produce any background fluorescence. Two-photon microscopy can penetrate deeper into tissue (500 μm−1 mm) than confocal or epifluorescence microscopy because the excitation wavelength is about twice as long as the usual wavelengths used. Light with longer wavelengths scatters less, allowing deeper penetration into tissue. Because excitation is restricted to a focal point, it also causes less photobleaching and phototoxicity, making two-photon microscopy particularly useful for extended observation of cells in undissected, living brain tissue in whole animals or fine-scale neural structures in brain slices. The primary limitation of this method is that a twophoton microscope setup requires expensive, specialized lasers and equipment. Total Internal Reflection Fluorescence Microscopy Total internal reflection fluorescent (TIRF) microscopy is another method used to eliminate hazy images with out-of-focus fluorescence by restricting the excitation and detection of fluorophores to a thin section of the specimen. In TIRF microscopy, the thin section of detection is fixed at the interface between the specimen and the surface on which the specimen is mounted, usually a glass coverslip (Figure 5.12). TIRF microscopes excite fluorophores with an evanescent wave of light exclusively at the interface between the specimen and its adjacent surface. The evanescent wave exponentially decays with distance and is typically lost at a depth of 100 nm into the sample. Therefore, it only Microscopy Chapter | 5  133 effectively excites the plasma membrane (∼7.5 nm thick) and the immediately adjacent cytoplasmic zone. This method is useful for investigating molecules and biochemical events that occur at the cell surface. For example, a fluorophore-tagged molecule of interest may be located on the plasma membrane as well as in the cytoplasm of the cell. Using an epifluorescent microscope, fluorescence emitted from the membrane can be overwhelmed by the fluorescence coming from the much larger population of molecules in the cytoplasm. A TIRF microscope allows an investigator to selectively excite and image the membrane-bound population of this molecule. This kind of microscopy is often used to selectively visualize processes that occur at the plasma membrane in living cells with high resolution. ELECTRON MICROSCOPY Although there is no theoretical limit to the magnification power of a light microscope, resolving power has a maximal value of about 200 nm due to the fixed minimal wavelength of visible light (Box 5.1). However, an electron has a much shorter wavelength than a photon. By focusing a beam of electrons rather than a beam of photons to image a specimen, electron microscopy (EM) can increase the resolving power as much as 1000-fold to 0.2 nm, about the radius of a glutamate molecule. EM is essential for examining small neural structures such as synapses, synaptic vesicles, and ion channels. The major limitation to using EM is that specimens must be fixed, dehydrated, and chemically treated. Therefore, EM cannot be applied to living samples. Furthermore, artifacts could appear that do not exist in the living cell. To reduce these artifacts, investigators can rapidly freeze specimens so that water and other components in cells do not have time to rearrange themselves or even crystallize into ice. There are two major categories of EM: transmission electron microscopy (TEM) and scanning electron microscopy (SEM). FIGURE 5.12  The basic strategy of TIRF microscopy. Reflected laser light causes an evanescent wave of excitation at the interface between a specimen and a glass coverslip. The light penetrates only 100 nm, ensuring that all fluorescent excitation and emission occurs only at the surface of the specimen, usually the plasma membrane. 134  Guide to Research Techniques in Neuroscience Transmission Electron Microscopy In transmission electron microscopy (TEM), a beam of electrons is aimed through a thin section of a specimen that has been chemically treated to enhance contrast. Extremely thin (<100 nm) sections of preserved tissue are stained with atoms of heavy metals, which are preferentially attracted to certain cellular components. These heavy metal stains are electron dense, so when the electron beam hits the heavy metal atoms in the specimen, the electrons are absorbed or scattered, making electron-dense areas appear dark. Electromagnets focus and magnify the image by bending charged electron trajectories. Since we cannot see electrons, the variations in electron intensity are converted into photons by projecting the electrons into a special detector or onto a screen that fluoresces at intensities relative to the amount of beam electrons hitting it. TEM produces two-dimensional images of thin tissue sections that are useful for studying the internal structure of cells (Figure 5.13). TEM is frequently used to define synapses, as the postsynaptic density appears clearly as an electron-dense region at close contact with a presynaptic membrane containing neurotransmitter vesicles. Scanning Electron Microscopy Scanning electron microscopy (SEM) is useful for detailed study of a specimen’s surface.A high-energy electron beam scans across the surface of a specimen, usually coated with a thin film of gold or platinum to improve contrast and the signal-to-noise µ µ FIGURE 5.13  Comparison of images taken by different electron microscope techniques. (A) A differential interference contrast image of stereocilia projecting from a hair cell in the inner ear. (B) An image of the same structure taken with a transmission electron microscope and (C) a scanning electron microscope Courtesy of Dr. A.J. Hudspeth and Dr. R. Jacobs. Microscopy Chapter | 5  135 ratio.As the beam scans across the sample’s surface, interactions between the sample and the electron beam result in different types of electron signals emitted at or near the specimen surface. These electronic signals are collected, processed, and eventually translated as pixels on a monitor to form an image of the specimen’s surface topography that appears three dimensional (Figure 5.13). Low-energy secondary electrons excited on the sample’s surface are the most common signal detected. High-energy backscattered electrons and X-rays are emitted from below the specimen surface, providing information on specimen composition. Electron Tomography Electron tomography (ET), also known as electron microscope tomography, is similar to computerized tomography described in Chapter 1. In a CT scan, the imaging equipment is moved around a patient to generate different images of a slice of the brain. In ET, the specimen is tilted within the electron microscope to produce TEM images from many different perspectives to reconstruct a three-dimensional view. With a resolution of 2–20 nm, ET can be used for determining both molecular structures and the three-dimensional ultrastructure of organelles. ET has been especially informative for elucidating the organization of presynaptic vesicles in axon terminals (Figure 5.14). FIGURE 5.14  Example of an electron tomography image: synaptic vesicles docking at release sites in the neuromuscular junction.  (A) Two-dimensional TEM images used in the ET volume reconstruction of an active zone at a frog neuromuscular junction. AZM, active zone material, denoted by arrows; sv, docked synaptic vesicles; pre, presynaptic membrane. Scale bar, 50 nm. (B) Three-dimensional surface-rendered AZM, docked vesicles, and presynaptic membrane that allow the shape, size, and associations of different components to be examined. Reprinted by permission from Harlow, M.L., et al. (2001). The architecture of active zone material at the frog’s neuromuscular junction. Nature: Macmillan Publishers Ltd. 136  Guide to Research Techniques in Neuroscience PREPARING AND INTERPRETING MICROSCOPY DATA Microscopy data are probably the most frequently presented form of data in neuroscience, and possibly all life sciences. They may also be the most subjective form of data to interpret. Therefore, it is important to understand standard practices used to prepare and display data to know what to look for in published results. Image Processing The digital images taken at the microscope are generally not the exact same images that appear in publications. Raw images are usually processed by computer software accompanying the microscope or by secondary programs such as Adobe Photoshop® or NIH ImageJ. Image processing is appropriate and sometimes necessary to extract meaningful information: to facilitate quantitative measurements and accurate analysis, to highlight meaningful aspects of an image, and to improve aesthetic appearance when it does not interfere with the results. The key is to faithfully represent reproducible results and not alter the inherent information contained in the image. The most common form of processing is creating pseudocolor representations of fluorescent microscopy images taken in different channels (different emission wavelengths) to overlay them in a single image. Most microscope cameras capture light intensity in grayscale rather than color. If multiple fluorophores are imaged in a single specimen, different channels can be selected by changing the excitation and emission filters on the microscope. This process allows different fluorophores to be imaged independently, but they will all appear in grayscale. The investigator can then assign a different color to each signal and merge the separate images of each fluorophore into a single image with multiple colors. This helps visualize relative intensity differences and spatial relationships for each fluorescently labeled molecule (Figure 5.15). Other common forms of image processing include digitally enhancing contrast or removing background fluorescence. Although this processing can aid in elucidating useful information from an image, it must not introduce false elements into the image to provide support for what an investigator wishes to see. Many journals provide guidelines for the types of image processing allowed and require that all processing be reported in the Materials and Methods section. In general, there are two important guidelines to follow when processing microscopy images for analysis and publication: (1) Any processing applied to one image of a set should be applied to all images within the set. For example, when removing background fluorescence from one image, it is important that all remaining images receive the exact same digital enhancements. This is especially important when comparing images between groups. (2) An investigator should process and analyze images blind to the identity of the samples between Microscopy Chapter | 5  137 experimental and control groups. Even with the best of intentions, scientists can subconsciously make mistakes in analysis or introduce subconscious bias. When working with microscopy data, as with most data, blind processing and scoring is critical to prevent potential bias. It is important to properly represent reproducible features contained in the data itself rather than in interpretations of the data. Interpreting Images Some general guidelines are useful when evaluating your own or published microscope images. When evaluating images, it can be helpful to ask some critical questions: l Does it look right? Familiarity with other work in the field can provide an intuition and expectation for standard images. If the image is supposed to be of neurons, do the structures look like neurons? If the image is supposed to represent a specific brain region, can you identify that region based on the image presented? Is the image taken at an appropriate scale to make a judgment? If the author claims that expression of a gene, protein, or lesion is localized to a specific structure, does the image show that structure in the context of neighboring regions? l Does the signal-to-noise ratio allow for proper examination of the data? Nonspecific background staining can potentially mask a specific signal. Is the contrast between the background and the signal substantial enough to produce a consistent, observable result? l Are the exposure and brightness optimal? An overexposed image (Figure 5.16A) can mask details, enhance nonspecific background, or improperly FIGURE 5.15  Merging images of different fluorophores into a single image. A sample is viewed for fluorescence using filters for (from left to right) rhodamine (red), FITC (green), and DAPI (blue). Because these spectra do not overlap, it is possible to overlay these images, assigning each channel a distinct color, to compare the spatial relationship of fluorescent signals. 138  Guide to Research Techniques in Neuroscience suggest a high concentration of the measured signal. An underexposed image (Figure 5.16B) can make it appear as if the measured signal is not present when it is. An investigator can control the proper exposure of an image during both the capture of an image and the secondary image processing.When evaluating microscopy data, ensure that the conclusions are not drawn from images with inappropriate exposure conditions or inappropriate image processing. l If this is the most convincing example, what do the other samples look like? For obvious reasons, authors tend to publish images that best represent and support their conclusions. Usually, these images represent one of many individual data points. Therefore, a critical review of the data should assume that the other data points are of an equal or lesser quality than the images presented in a publication. If the image is suboptimal or difficult to interpret, the validity of conclusions reached from unpublished images should be questioned. l Is there quantitative analysis of the images? Though a picture is worth a thousand words, subjective interpretation of data is not as strong as objective, quantifiable results. Therefore, the strongest evidence to support a conclusion presents images alongside complementary quantification of the data that demonstrates the reproducibility of the results from image to image. Common types of quantitative information calculated from images include relative cell counts, size and shape distributions, variation in fluorescence intensity to indicate relative concentrations of the measured signal, and colocalization of independently labeled proteins. l Does the illumination look uniform? Improper microscope alignment can cause differences in signal intensity that might lead to inappropriate conclusions. If FIGURE 5.16  Over- and underexposure of an image.  (A) An overexposed image of neurons. (B) The same cells underexposed and (C) with the proper level of exposure. The yellow arrows point to a neural process that is lost when underexposed and difficult to see under a properly balanced exposure level. It can be difficult to find the proper exposure to see both intensely and weakly fluorescent structures in the same field of view. Microscopy Chapter | 5  139 the entire field of view is not evenly illuminated, important information and resolution can be lost. l Could fluorescent images represent poor quality control issues, such as photobleaching or autofluorescence? Photobleaching could lead an investigator to believe there is no signal in a sample when the signal has actually faded. Alternatively, autofluorescent substances in a sample might lead an investigator to believe there is signal where there is none. Investigators can avoid inappropriate conclusions caused by photobleaching or autofluorescence by processing all samples in the same way and by performing the appropriate positive and negative control experiments. l Could fluorescent images represent bleed-through of different emission spectra? Investigators usually assume that signals from fluorophores of different wavelengths are independent of each other. However, some fluorophores exhibit overlapping excitation or emission spectra and may be detected when not expected. Bleed-through, also known as cross-talk, occurs when the signal for one fluorophore is present when examining the signal for a different fluorophore(i.e. the signal of one fluorophore has bled through the filters that are set for the other fluorophore). To control for this situation, specimens can be labeled with a single fluorophore and the signal recorded in each channel. Bleed-through of fluorophores becomes especially pertinent when an investigator claims that two signals overlap, as when two proteins colocalize. It is rare that the signal from two different fluorophores will be exactly identical, no matter how strong the colocalization may be. Therefore, if two different channels appear exactly the same, despite use of different detection agents, the image could indicate an artifact. CONCLUSION Whether using photons or electrons, microscopes provide the ability to explore structures much smaller than can be viewed by the naked eye. Although microscopes are one of the most frequently used tools in neuroscience, many people do not understand the essential concepts necessary to produce quality, scientifically meaningful images. This chapter has served as an introduction to microscopy and surveyed multiple techniques (Table 5.1). Most microscope manufacturers (Nikon, Leica, Zeiss) provide detailed instructions in the proper use and theory behind their products. These guides can provide further information about the specific microscopes used in your laboratory. While optical methods can be used to provide enhanced contrast to make out individual structures within cells, histological preparations are utilized to examine specific processes and structures. The next two chapters present methods on preparing samples to visualize neural structure and function that depend on good microscopes and, just as important, good microscopy skills. 140 GuidetoResearchTechniquesinNeuroscience TABLE 5.1 Comparison of Different Forms of Microscopy Description Advantages Disadvantages Common Use in Neuroscience Brightfield Light transmitted through specimens. Contrast generated by natural pigmentation or added dyes. Simple and inexpensive. Different colored dyes can be used. Most cells and tissues are transparent, so they can be difficult to see. Adding dyes to visualize structures usually requires fixation and sectioning. Fixed samples processed using detection reagents that produce colored byproducts (see Chapter 6). Phase-contrast Contrast generated by changes in index of refraction because of variation in organelle densities. Can be used on live cells. Does not require added chemicals or tissue processing. Out-of-focus signals. Tissue culture cells. Darkfield Specimen illuminated from the side at an oblique angle. Only scattered light is imaged. Can be used on live cells. Sensitive. High signal-to-noise ratio. No added chemicals necessary. Cannot see structures that do not scatter light. Samples processed for radioactive in situ hybridization (see Chapter 6). DIC/Nomarski Optical methods used to exaggerate changes in index of refraction, enhancing contrast at the edges of objects. Has a 3-D appearance. Can be used on live cells. Can create thin optical sections. Can create 3-D reconstructions from thin optical sections. Can only image a single, thin focal plane at a time. Tissue culture cells and tissues. Microscopy Chapter|5 141 Epifluorescent Specimen is illuminated by excitation wavelengths and emits light from excited fluorophores throughout the entire thickness. (Advantages of all forms of fluorescent microscopy) Can be used to detect specifically labeled fluorescent molecules. Multiple fluorophores can be imaged in same sample. High signal-to-noise ratio. Sensitive. (Advantage of epifluorescence over other forms) Simpler and faster image collection methods than other forms of fluorescent microscopy (confocal, two-photon). Out-of-focus fluorescence can cause blurry pictures, making structures difficult to resolve. Thin physical sections of fluorescently stained tissue or cells. Time-lapse fluorescence imaging in cells. Confocal Out-of-focus illumination is blocked by the use of pinhole apertures, so only in-focus emitted light is collected. Optical sectioning ability creates sharp, in-focus images. Intense laser illumination can cause photobleaching and phototoxicity. Long scan time. Detecting fluorescence in thick tissues or small organisms (e.g., Drosophila). Time-lapse fluorescence imaging in thicker tissues (e.g., slice cultures) or for small structures (e.g., synaptic vesicles). Two-photon Fluorophores in a thin focal plane are selectively excited by effectively absorbing the combined energy of two photons that cannot excite fluorophores on their own. Longer wavelengths of laser illumination allow deeper penetration of fluorescence excitation. Reduced photobleaching and phototoxicity. Expensive equipment. In vivo imaging of intact organisms. Long-term fluorescence imaging. Continued 142 GuidetoResearchTechniquesinNeuroscience Total internal reflection fluorescence (TIRF) An evanescent wave of illumination rapidly decays after passing the interface between a cell membrane and the surface it is on. Images a thin optical section. Eliminates out-of-focus fluorescence from regions more than 100 nm away from the surface. High resolution. Only fluorophores at the interface between the membrane surface and the glass coverslip are illuminated; cannot image an arbitrary region within the cell. Imaging protein dynamics at the plasma membrane in live cells. Transmission electron microscopy (TEM) Electron beams transmitted through ultrathin sections. Nanometer resolution. Cannot be used on live cells. Harsh processing conditions can cause artifacts. Requires specialized equipment. Ultrastructure of cells. Synapse structure. Scanning electron microscopy Detect secondary electrons scattered off the surface of the sample. Provides 3-D topological information. Cannot be used on live cells. Harsh processing conditions can cause artifacts. Requires specialized equipment. Topography of cells and tissues. Electron tomography Rotate specimen to take TEM images from multiple perspectives and create 3-D reconstruction. Provides 3-D organization information. Cannot be used on live cells. Requires intensive computation to reconstruct TEM views. 3-D cellular ultrastructure and organization. TABLE 5.1  Comparison of Different Forms of Microscopy—cont’d Description Advantages Disadvantages Common Use in Neuroscience Microscopy Chapter | 5  143 SUGGESTED READING AND REFERENCES Books Bradbury, S., Bracegirdle, B., & Royal Microscopical Society (Great Britain). (1998). Introduction to light microscopy. Oxford: Bios Scientific Publishers, Springer. Chandler, D. E., & Roberson, R. W. (2009). Bioimaging: Current concepts in light and electron microscopy. Sudbury, MA: Jones and Bartlett Publishers. Herman, B. (1998). Fluorescence microscopy (2nd ed.). Abingdon, UK; New York: Taylor & Francis in association with the Royal Microscopical Society. Murphy, D. B. (2012). Fundamentals of light microscopy (2nd ed.). Hoboken, NJ: John Wiley & Sons, Inc. Pawley, J. B. (2006). Handbook of biological confocal microscopy (3rd ed.). NY: Springer. Russ, J. C. (2011). The image processing handbook (6th ed.). Boca Raton, FL: CRC Press. Review Articles Brown, C. M. (2007). Fluorescence microscopy—avoiding the pitfalls. Journal of Cell Science, 120, 1703–1705. Kapitza, H. G. (1994). Microscopy from the very beginning (2nd revised ed.). Oberkochen, Germany: Carl Zeiss. McEwen, B. F., & Marko, M. (2001). The emergence of electron tomography as an important tool for investigating cellular ultrastructure. Journal of Histochemistry and Cytochemistry, 49, 553–564. Mancuso, J. J., Chen, Y., Li, X., Xue, Z., & Wong, S. T. (2013). Methods of dendritic spine detection: from Golgi to high-resolution optical imaging. Neuroscience, 251, 129–140. Osten, P., & Margrie, T. W. (2013). Mapping brain circuitry with a light microscope. Nature Methods, 10, 515–523. Pearson, H. (2007). The good, the bad and the ugly. Nature, 447, 138–140. Saibil, H. R. (2007). How to read papers on three-dimensional structure determination by electron microscopy. In D. Zuk (Ed.), Evaluating techniques in biochemical research. Cambridge, MA: Cell Press. Schneckenburger, H. (2005). Total internal reflection fluorescence microscopy: technical innovations and novel applications. Current Opinion in Biotechnology, 16, 13–18. Waters, J. C., & Swedlow, J. R. (2007). Interpreting fluorescence microscopy images and measurements. In D. Zuk (Ed.), Evaluating techniques in biochemical research. Cambridge, MA: Cell Press. http://www.cellpress.com/misc/page?page5ETBR. Willig, K. I., & Barrantes, F. J. (2014). Recent applications of superresolution microscopy in neurobiology. Current Opinion in Chemical Biology, 20, 16–21. Primary Research Articles—Interesting Examples from the Literature Ahrens, M. B., Orger, M. B., Robson, D. N., Li, J. M., & Keller, P. J. (2013). Whole-brain functional imaging at cellular resolution using light-sheet microscopy. Nature Methods, 10, 413–420. Harlow, M. L., Ress, D., Stoschek, A., Marshall, R. M., & McMahan, U. J. (2001). The architecture of active zone material at the frog’s neuromuscular junction. Nature, 409, 479–484. Jung, J. C., Mehta, A. D., Aksay, E., Stepnoski, R., & Schnitzer, M. J. (2004). In vivo mammalian brain imaging using one- and two-photon fluorescence microendoscopy. Journal of Neurophysiology, 92, 3121–3133. 144  Guide to Research Techniques in Neuroscience Lu, H. E., MacGillavry, H. D., Frost, N. A., & Blanpied, T. A. (2014). Multiple spatial and kinetic subpopulations of CaMKII in spines and dendrites as resolved by single-molecule tracking PALM. Journal of Neuroscience, 34, 7600–7610. Micheva, K. D., Busse, B., Weiler, N. C., O’Rourke, N., & Smith, S. J. (2010). Single-synapse analysis of a diverse synapse population: proteomic imaging methods and markers. Neuron, 68, 639–653. Westphal, V., Rizzoli, S. O., Lauterbach, M. A., Kamin, D., Jahn, R., & Hell, S. W. (2008). Video-rate far-field optical nanoscopy dissects synaptic vesicle movement. Science, 320, 246–249. Protocols Coling, D., & Kachar, B. (2001). Theory and application of fluorescence microscopy. Current Protocols in Neuroscience (Chapter 2), Unit 2.1. http://dx.doi.org/10.1002/0471142301. ns0201s00. Smith, C. L. (2011). Basic confocal microscopy. Current Protocols in Neuroscience (Chapter 2), Unit 2.2. http://dx.doi.org/10.1002/0471142301.ns0202s56. Spector, D. L., & Goldman, R. D. (2006). Basic methods in microscopy: Protocols and concepts from cells: A laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Websites Davidson, M. W. Molecular Expressions: http://micro.magnet.fsu.edu/. Nikon Microscopy U: http://www.microscopyu.com. Olympus Microscopy Resource Center: http://www.olympusmicro.com/. Zeiss MicroImaging: http://www.zeiss.com/us/micro/home.nsf. 145 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00006-X Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 6 Visualizing Neural Structure After Reading This Chapter, You Should be Able to: l Explain the process of preparing neural tissue for histological procedures l Describe methods for visualizing gross cellular morphology l Describe methods for visualizing gene and protein expression l Describe methods for visualizing neural circuitry and connections between different brain regions Techniques covered: l Tissue preparation: fixation, embedding, sectioning l Visualizing structure: fundamental histological methods (basophilic stains, fiber stains), Golgi stain, intracellular/juxtacellular labeling l Visualizing gene and protein expression: in situ hybridization, immunohistochemistry, array tomography, enzymatic histochemistry, reporter genes l Visualizing circuitry: anterograde tracers, retrograde tracers, transsynaptic tracers A person can learn a lot about a complicated system simply by examining its structure. For example, imagine a satellite image of a city taken from space. How might you learn about what it would be like to live in the city just by examining the city from that vantage point?You could start by separating the image into component parts—for example, distinguishing buildings from roads. Larger buildings clustered together could reveal popular hubs of activity, whereas smaller, scattered buildings might indicate residential communities. You could learn even more by identifying the different categories of buildings and classifying different structures as schools, grocery stores, gas stations, shopping centers, and so forth. Major highways and freeways would indicate routes for long-distance travel; narrower roads would show local areas of transit. By simply examining the structure of the city, one could form educated guesses about how the city functions. Similarly, one can learn a lot about the nervous system simply by examining its structure. Just like distinguishing between buildings and roads in a city, a neuroscientist can begin to examine the brain by discriminating between cells and fiber tracts. Cells, like buildings, can be classified into different structural and functional groups, with each cell type expressing a unique combination of genes and proteins that specify its role in the brain. Thick white-matter tracts show long-distance sites of neural communication, whereas smaller fiber tracts 146  Guide to Research Techniques in Neuroscience show local communication networks between populations of neurons. Learning about the structure of distinct brain regions can lead to educated hypotheses about the function of different neurons and how they influence neural circuits and behavior. Indeed, in the early twentieth century, Santiago Ramón y Cajal formed many hypotheses (many of which turned out to be correct) about the function of neural systems simply by examining their structures. The goal of this chapter is to survey techniques used to investigate the structure and connectivity of the nervous system. Using these techniques, it is possible to classify cells based on location, morphology, gene/protein expression profiles, and connections with other cells. These methods can be used in combination with techniques described in Chapter 7 to investigate neural function. TISSUE PREPARATION Neural tissue is soft, delicate, and degradable. Therefore, to accurately study neural structure, an investigator’s first goal is to keep the tissue as close to its living state as possible. Investigators process tissue by (1) fixing, (2) embedding, and (3) sectioning before proceeding with visualization methods. Fixing and embedding methods stabilize tissue to capture its current state, while sectioning makes the tissue thin enough for light to pass through so that internal structures can be examined under a microscope. Fixation Fixation is the process of using chemical methods to preserve, stabilize, and strengthen a biological specimen for subsequent histological procedures and microscopic analysis. This process preserves cells and tissues by strengthening molecular interactions, disabling endogenous proteolytic enzymes, and killing microorganisms that might otherwise degrade the specimen. Thus, fixation terminates any ongoing biochemical reactions by “fixing” proteins into place—a process that kills cells that are not already dead. There are two different categories of chemical fixatives: cross-linking fixatives and dehydrating fixatives. Which fixative an investigator chooses to use depends on the type of subsequent histology to be performed. Cross-linking fixatives create covalent chemical bonds between proteins in tissue and include organic compounds with aldehyde groups, such as formaldehyde, paraformaldehyde, and glutaraldehyde. These fixatives are good at preserving structure and are often used when processing cells and tissues for light microscopy. Another cross-linking fixative, osmium tetroxide, causes molecules to oxidize and is often used as a secondary fixative for electron microscopy. Dehydrating fixatives disrupt lipids and reduce the solubility of protein molecules, precipitating them out of the cytoplasmic and extracellular solutions. Common dehydrating fixatives include methanol and acetone. A scientist can fix cells in culture simply by adding a chemical fixative to the culture chamber. Whole brains can be fixed either by immersion or Visualizing Neural Structure Chapter | 6  147 perfusion. Immersion refers to placing small brains, or even entire animals, in fixative solutions. The amount of time necessary for the fixative to penetrate the entire tissue is variable, depending on the size of the brain: a fly brain may need to be fixed for only a few minutes, while a mouse brain may require days. Perfusion is the process of delivering a fixative through an animal’s cardiovascular system. To perfuse an animal, a scientist anesthetizes the animal, carefully opens the rib cage to expose the heart, and then penetrates the left ventricle with a needle connected to a perfusion pump. The left ventricle pumps blood throughout the entire body and can therefore deliver a fixative to the nervous system. A scientist initially pumps a buffered saline solution to rinse the blood out of the veins and arteries, and then pumps a cross-linking reagent to fix all the tissues in the body. Perfusion fixation is thorough, quick (can be performed in 5–15 min, depending on the animal), and adequately fixes all cells in the brain if properly performed. Some histological procedures do not require fixation or even require that a brain not be fixed. For example, the use of radioactively labeled ligands to localize neurotransmitter receptor-binding sites depends on those sites remaining intact, and fixation could disrupt their structure. Each histological method determines whether fixation is necessary, which fixatives to use, and when the fixation process should occur. Embedding Embedding is the process of surrounding a brain or tissue section with a substance that infiltrates and forms a hard shell around the tissue. This process stabilizes the tissue’s structure and makes it easier to cut. Standard embedding materials include gelatin, paraffin wax, and plastic. Sectioning Sectioning is the process of cutting a brain into thin slices for subsequent histological procedures or microscopic examination. This process is almost always necessary to examine the structure of neurons within the brains of vertebrate model organisms. Sectioning the brain allows histological reagents access to cells within the slice, as well as allows a scientist to examine brain structures with a microscope. There are three common machines used to section tissue (Figure 6.1, Table 6.1): l A microtome (micro =  “small,” tome = “cut”) is an instrument used to section tissue using a retractable blade. A specialized microtome called a freezing microtome is commonly used to section frozen tissue. Before cutting, the tissue must be cryoprotected by soaking in a sucrose solution. This minimizes artifacts caused by freezing. Using a freezing microtome, a scientist can collect 25- to 100-μm sections off the blade for subsequent histochemical processing. These sections are usually stored in a buffered saline solution 148  Guide to Research Techniques in Neuroscience until use. Other types of microtomes are used to section paraffin-embedded tissue or to create the ultrathin sections required for electron microscopy. l A cryostat is essentially a microtome housed in a freezing chamber that allows sectioning at temperatures of −20 to −30 °C. Maintaining the tissue and blade at low temperatures allows a scientist to section nonfixed, frozen tissue. Cryostats are able to cut thin, 10- to 50-μm sections. A scientist can mount sections onto slides within the cryostat or store sections in a buffered saline solution until use. l A vibratome cuts tissue with a vibrating knife, somewhat similar to an electric vibrating toothbrush. The main function of a vibratome is to section tissues that are not frozen. The knife vibrates side to side so quickly that a soft brain can be cut into 100- to 400-μm sections. Thus, vibratomes are useful for avoiding artifacts, changes in morphology, or interruption of biochemical activities caused by freezing. A vibratome is also necessary for experiments in which a slice of brain tissue must be kept alive, as in tissue culture or electrophysiology studies. FIGURE 6.1  Equipment used to section tissue.  (A)A microtome. (B)A cryostat. (C)A vibratome. TABLE 6.1  Comparison of Different Brain Sectioning Methods Essential Features Standard Section Thickness Microtome Brains are frozen Preserves structures better than cryostat Medium: 25–100 μm Cryostat Brains are frozen Sections are cut in a cold chamber Thin: 5–50 μm Vibratome Brains do not need to be frozen Vibrating knife Can be used to cut tissue that will be kept alive Thick: 100–400 μm Visualizing Neural Structure Chapter | 6  149 Once a brain is sectioned, a scientist can either mount the sections directly onto slides or can keep the sections floating in buffered saline for subsequent histological analysis. Brains are typically cut in one of three standard orientations: coronal, sagittal, or horizontal (Figure 6.2). The coronal plane divides the brain in the anterior-to-posterior direction, revealing a section of the brain that is complete from top-to-bottom and left-to-right (ear to ear). The sagittal plane shows the complete top-to-bottom of the brain, cutting it into left and right portions. A midsagittal cut is a section that perfectly divides the left and right hemispheres of the brain. The horizontal plane divides the brain into superior (top) and inferior (bottom) sections, such that an area of the brain is complete from left-to-right and anterior-to-posterior. There are some preparations for which brain sectioning is not necessary. For example, scientists may wish to examine whole-mount preparations of intact brains or animals. Also, the brains of Drosophila and other invertebrates are usually small enough such that confocal or two-photon microscopy (Chapter 5) can provide good images of individual neurons within neural structures. A new technique called clear lipid-exchanged acrylamide-hybridized rigid imaging/immunostaining/in situ hybridization-compatible tissue hydrogel (CLARITY) fixes tissue using a hydrogel to preserve the physical structure of proteins and nucleic acids, but allows light-scattering lipids to be removed. This process effectively makes intact tissue transparent and amenable to the structural visualization methods FIGURE 6.2  Neuroanatomical planes. Standard orientations for sectioning brain tissue. (A) Coronal. (B) Sagittal. (C) Horizontal/transverse. 150  Guide to Research Techniques in Neuroscience described in this chapter. Therefore, CLARITY allows for visualization of nervous system structure without the need to section the tissue. Now that we have discussed some common methods of preparing neural tissue for histological analysis, we describe fundamental methods of visualizing the gross morphology of the brain. VISUALIZING MORPHOLOGY Although lipids, proteins, carbohydrates, and other organic molecules provide the nervous system with its structure and durability, more than three-quarters of the mass of the brain is water. This composition imparts significant consequences for any scientist interested in studying the structure of the brain— most notably that unprocessed brain sections are almost completely transparent. Unless treated by histochemical stains, the structure of the nervous system will remain invisible and appear as a wafer-thin slice of gelatin mounted on a glass slide. Treating brain sections with various dyes enhances the contrast between different brain structures and allows an investigator to visualize cells, fibers, or other features of neural systems (Table 6.2). These methods each provide contrast to specific structures in the brain so that they stand out from other aqueous structures in a neural specimen. Cell Body Stains A basophilic stain is used to visualize cell bodies. The term basophilic is derived from baso-, meaning “base” and -philic, meaning “loving” and conveys the fact that these stains are basic (pH > 8) and good for staining acidic molecules. As DNA and RNA molecules are acidic, basophilic dyes label structures enriched in nucleic acids, such as cell nuclei and ribosomes. A scientist can use these stains to examine the cytoarchitecture of individual neurons with high magnification or to visualize the macroscopic features of distinct brain regions, such as layers of cerebral cortex or subdivisions of hypothalamic nuclei (Figure 6.3A). Common basophilic stains used in bright field microscopy include hematoxylin and thionine. A special category of basophilic stains, referred to as Nissl stains, specifically label RNA within cells, providing contrast to ribosomes and rough endoplasmic reticulum, which are enriched in neurons. Cresyl violet is one of the most commonly used Nissl stains. Redistribution of Nissl-stained structures in injured or regenerating neurons allows Nissl stains to reveal the physiological state of some neurons. In fluorescent microscopy, common markers of cell nuclei include DAPI (4’,6-diamidino-2-phenylindole), Hoechst (bis-benzamide), and propidium iodide (PI). These dyes intercalate within the helical spirals of DNA in the cell nucleus. Both DAPI and Hoechst emit blue light when excited by ultraviolet (UV) light, whereas PI emits red light when excited by green light. PI is useful Visualizing Neural Structure Chapter | 6  151 TABLE 6.2  Histochemical Stains for Studying Neuroanatomy Stain Use Appearance Comments Cresyl violet Cell nuclei, Nissl stain Blue to purple Useful for examining cytoarchitecture; stains each type of neuron slightly differently Hematoxylin Cell nuclei Blue to blue-black Often used in combination with eosin; known as H&E Eosin Y Cytoplasm Pink to red Counterstain with hematoxylin; acidophilic stain Thionine Cell nuclei, Nissl stain Blue to purple Methylene blue Cell nuclei Blue Can be perfused through the brain before fixation Toluidine blue Cell nuclei Nucleus is stained blue; cytoplasm light blue Often used to stain frozen sections DAPI Cell nuclei Fluorescent blue Fluorescent DNA intercalating agent; excited by UV illumination Hoechst (bis-benzamide) Cell nuclei Fluorescent blue Fluorescent DNA intercalating agent; excited by UV illumination Propidium iodide (PI) Cell nuclei Fluorescent red Fluorescent DNA intercalating agent; excited by green light illumination Weigert Myelin Normal myelin is deep blue; degenerated myelin is light yellow Combines hematoxylin with other chemicals to selectively stain myelin Weil Myelin Black Combines hematoxylin with other chemicals to selectively stain myelin Luxol fast blue (LFB) Myelin Blue Golgi stain Fills neuron cell bodies and processes Black Stains individual neurons at random 152  Guide to Research Techniques in Neuroscience in assays of cell death because it is membrane impermeant; therefore, it labels dead but not healthy cells. Fiber Stains Fiber stains label white-matter tracts by staining myelin, the fatty substance that insulates and provides electrical resistance to axons. These stains are useful for visualizing the major fiber tracts that project throughout the brain (Figure 6.3B). However, the high density of fibers in most brain regions makes it impossible to trace individual axons from cell body to synapse. When used in combination with basophilic stains in adjacent tissues sections, a scientist can identify more anatomical structures than if either stain is used alone. There are various protocols useful for labeling myelin, such as the Weigert or Weil methods, which stain myelin dark blue or black. A scientist can also stain myelin using lipophilic fluorescent stains. Golgi Stain The Golgi stain is a classic technique used to completely label individual neurons and their processes (Box 6.1). There are two major advantages to using a Golgi stain: (1) neurons are labeled in their entirety, including cell bodies, processes, and even microscopic structures such as dendritic spines; (2) only 5–10% of the total number of cells are labeled. Therefore, individual neurons stand out in a background of numerous unlabeled cells (Figure 6.4). A scientist cannot control which neurons in a tissue slice become labeled, but newer techniques allow a scientist to duplicate the effects of a Golgi stain in a genetically defined population of neurons (Chapter 12). However, this method is still used to define and trace the elaborate morphology of neurons, especially in neurodegeneration studies. (A) (B) FIGURE 6.3  Comparison of basophilic and fiber stains.  (A) Cresyl violet Nissl stain of a coronal mouse brain section. (Reprinted with permission from Rosen, G.D., et al. (2000). The Mouse Brain Library. International Mouse Genome Conference 14:166.) (B) Coronal section stained for myelin. (Courtesy of Dr Pushkar Joshi.) Visualizing Neural Structure Chapter | 6  153 Intracellular and Juxtacellular Labeling The preceding methods label cell bodies and/or axons after a brain has already been fixed and sectioned. It is also possible to label individual neurons during an experiment for later histological analysis so that a neuron’s activity or function can be coupled to the neuron’s structure and location in the brain. During electrophysiology experiments, a scientist can use a glass pipette to fill a neuron of interest with a chemical that can later be visualized using histological methods. This chemical is usually a variant of a molecule called biotin, such as biocytin or neurobiotin, which has properties that make it useful for subsequent histology (Box 6.2). A scientist can intracellularly label cells in vivo or in slices. During an extracellular recording session, the electrode tip can be apposed to the neuron’s membrane for juxtacellular labeling. This labeling can completely fill a neuron’s dendrites or axonal arborizations. BOX 6.1  Historical Use of the Golgi Stain The Golgi stain has an important place in the history of neuroscience. In the late nineteenth and early twentieth centuries, Santiago Ramón y Cajal used this technique for his seminal studies of the cellular morphology of the nervous system. Among his many achievements was the discovery that neurons are distinct, independent cells that communicate via synapses. His work heavily contributed to the acceptance of the neuron doctrine—the hypothesis that neurons are the basic structural and functional units of the nervous system. Cajal also used the Golgi stain to reveal the morphological diversity of neurons, as well as to trace neural circuits in the retina, hippocampus, cerebellum, and other brain regions. In 1906, Cajal shared the Nobel Prize with Camillo Golgi, the scientist who developed the Golgi stain, for elucidating the structure of the nervous system. FIGURE 6.4  The Golgi stain reveals neuronal structure in fine detail.  Random neurons are selectively stained in the (A) 100x view of a Golgi-stained mouse cortex and (B) a higher magnification. Courtesy of Dr Jocelyn Krey. 154  Guide to Research Techniques in Neuroscience BOX 6.2  Methods of Labeling Molecular Probes A probe, such as an antibody in immunohistochemistry (IHC) or antisense strand in in situ hybridization (ISH), is not visible by itself. To be visualized, the probe must be conjugated to a label that is usually visible on its own or that reacts with other chemicals to create a visible product. The use of a particular label depends on a number of factors, including the degree of specificity needed, whether other methods will be used in combination, and what types of microscopes are available. This box describes common types of labels (fluorescent, chromogenic, radioactive, and colloidal gold) and how they can be detected. Label Type Comments Visualization Method Fluorescent dyes Green dyes: FITC, Alexa 488, Cy2 Fluorescence microscopy Red dyes: Rhodamine, Texas red, Alexa 594, Cy3 Infrared dyes: Cy5 Quantum dots Spectral selectivity based on size Fluorescence microscopy Alkaline phosphatase (AP) NBT/BCIP substrate produces blue/purple product Usually light microscopy, though fluorescent substrates are available Horseradish peroxidase (HRP) DAB substrate produces brown/ black product Light or electron microscopy of colored DAB product β-galactosidase (encoded by lacZ gene) X-gal or IPTG substrate produces blue/purple product Usually light microscopy, though fluorescent substrates are available Biotin Biotinylated antibodies or proteins are detected using avidin conjugated to another label described (fluorescent, chromogenic, or gold) Light, fluorescent, or electron microscopy depending on type of label conjugated to avidin Digoxygenin Used to label ISH probes, detected with antidigoxygenin probe conjugated to another label, frequently AP Light or fluorescence microscopy, depending on substrate for label Radioactive isotope (35S,33P,3H,125I) ISH probes:35S,33P,3HProteins: 3H,125I Light microscopy Gold Electron-dense label used for detecting immunostaining in EM Electron microscopy Fluorescent Labels Fluorescent molecules are ubiquitous in biological research. Fluorescent proteins, such as GFP, can be genetically encoded and serve as reporter proteins. Other nonprotein fluorophores can be attached to a variety of molecular probes. Because these fluorescent probes have different excitation and emission spectra (Chapter Visualizing Neural Structure Chapter | 6  155 5), scientists can use multiple fluorescent labels to visualize distinct genes, proteins, and structures in the same neural specimen. There are a variety of fluorophores available for labeling probes. Commonly used fluorescent labels include the organic cyanine and Alexa fluor dyes, which are commercially available and often used to label antibodies. Quantum dots are semiconductor nanocrystals that are extremely bright and resistant to photobleaching, making them particularly useful for live-cell imaging over long time periods. Fluorescence is a sensitive method of detection, as bound fluorophores will emit light against a dark background. Attaching multiple fluorophores to a probe can amplify the signal to enhance detection. However, this amplification also makes it trickier to quantify exact levels of the probed substance because the degree of amplification does not necessarily vary linearly with the amount of substance initially present. Fluorescently labeled structures can be made visible to light and electron microscopes through a process called photoconversion. Fluorophores can photoconvert the chemical diaminobenzidine (DAB), a substrate commonly used for chromogenic enzymes. Exciting certain fluorophores with the appropriate wavelength of light in the presence of DAB forms a dark-colored precipitate. Chromogenic/Colorimetric Labels Chromogenic or colorimetric labels are enzymes that react with a substrate to produce a colored product that is visible under a light microscope.The most commonly used combinations are alkaline phosphatase (AP) with the substrates NBT/BCIP to produce a blue to purple stain; horseradish peroxidase (HRP) with the substrate DAB to produce a brown to black stain; and β-galactosidase (genetically encoded by lacZ) with X-gal or IPTG substrate to produce a blue to purple stain. Commercially available substrates that produce fluorescent rather than colored products are also available, though colorimetric substrates are more frequently used. Although not a chromogenic label itself, biotin is commonly conjugated to antibodies or incorporated into ISH oligonucleotide probes. The biotinylated probe can then be detected using the biotin-specific binding partner avidin/streptavidin conjugated to a chromogenic label such as AP or HRP. Avidin and biotin can form large complexes and greatly enhance the signal compared to using a probe alone. Digoxigenin (abbreviated dig, sounds like “dij”) is another common molecule that is incorporated into ISH probes as an alternative to radioactive labels. Antidigoxigenin antibodies conjugated to chromogenic enzymes are used to detect the presence of the dig-labeled probe. Radioactive Labels Radioactive isotopes, or radioisotopes, have an unstable nucleus that randomly disintegrates to produce a different atom. In the course of disintegration, these isotopes emit either energetic subatomic particles, such as electrons, or radiation, such as gamma rays. By using chemical synthesis to incorporate one or more radioactive atoms into a small molecule of interest such as a sugar, amino acid, or neurotransmitter, the fate of that molecule can be traced during any biological reaction. BOX 6.2  Methods of Labeling Molecular Probes—cont’d Continued 156  Guide to Research Techniques in Neuroscience VISUALIZING GENE AND PROTEIN EXPRESSION A scientist can classify a cell using its morphology, location in the brain, or its gene and protein expression profile. The genes expressed by a neuron determine whether it is excitatory or inhibitory, what kinds of neurotransmitters it releases and responds to, its physiological characteristics, and many other functional properties. Therefore, determining the genes and proteins expressed by a cell provides vital clues to its role in the brain. Conversely, a scientist may identify a gene or protein of interest, and then want to know where in the brain it is expressed. The following methods provide tools an investigator can use to visualize gene and protein expression in the nervous system. In Situ Hybridization In situ hybridization (ISH) is used to visualize the expression of nucleic acids, usually mRNA, allowing a scientist to determine when and where a specific gene is expressed in the nervous system. This technique is good for identifying which neurons express a gene, although it does not reveal where the functional protein product is expressed within the cell. To perform an ISH experiment, a scientist must first identify the genetic sequence of an mRNA to be studied. Next, the scientist creates a single-stranded Radioisotopes are visualized using autoradiography. Specimens are coated with a photographic emulsion and left in the dark while the radioisotope decays. Each decay event causes a silver grain to precipitate in the emulsion, so the position of the radioactive probe is indicated by the position of the developed silver grains. Because it is a direct chemical reaction, each silver grain represents one decay event. Therefore, radioactive labels allow quantitative studies of biological probes. Primarily used to label ISH oligonucleotide probes, 35S and 33P are radioisotopes incorporated into the probe sequence, allowing sensitive detection, even for genes expressed at low levels. However, it can be tricky to calibrate the amount of time needed for exposing the emulsion to get the right signal-to-noise ratio, and the exposure may require weeks to months to see a signal. 3H and 125I are radioisotopes that are used to label proteins and nucleic acids to investigate a number of processes, such as receptor localization, proliferation, and protein trafficking. Radioactively labeled ligands added to a specimen can bind to active receptors, allowing autoradiographic detection for the presence of those receptors. See Chapter 7 for a description of autoradiography in proliferation and protein trafficking. Gold Labels Colloidal gold labels are conjugated to antibodies for probe detection in electron microscopes (EM; Chapter 5). Gold provides an electron-dense material that can be visualized as a dark spot in EM. Avidin/streptavidin can also be conjugated to gold for detection of biotinylated probes in EM. Gold particles of different sizes can be used to detect multiple targets in the same sample. BOX 6.2  Methods of Labeling Molecular Probes—cont’d Visualizing Neural Structure Chapter | 6  157 nucleic acid probe that has a complementary base-pair sequence. The probe is tagged with radioactive nucleotides or other molecules that allow detection (Box 6.2). A tissue sample is exposed to the tagged probes such that the complementary strands can form a probe-mRNA duplex, marking the location of neurons that produce the mRNA sequence (Figure 6.5). This tissue sample is almost always a brain section, although it is possible to perform ISH on cultured cells or even unsectioned, whole-mount brains and animals. ISH techniques require specific control experiments. Because an investigator uses a complementary (“antisense”) strand of mRNA as a probe, a common negative control is a probe with the exact same (“sense”) sequence of the gene of interest. This control probe should not produce any signal, as no hybridization should occur. Another good control to ensure that the antisense strand of mRNA is specific to the gene of interest is to perform additional experiments with other antisense strands that hybridize to a different region of the same mRNA. The Allen Brain Institute in Seattle has provided researchers with a searchable, high-throughput collection of ISH images for every gene expressed in the mouse brain. Using this online resource (www.brain-map.org), a scientist can view a series of brain sections with an ISH reaction for a gene of interest. Alternatively, a scientist can type in the name of a brain region and receive a list of the most highly expressed genes in that area of the brain. This resource allows scientists to quickly obtain information about gene expression throughout the mouse brain without having to personally perform in situ reactions. The only limitation to this resource is that it does not examine gene expression for conditions other than baseline conditions, so gene expression could be different depending on environmental stimuli. FIGURE 6.5  In situ hybridization shows where a gene is expressed.  (A) Labeled oligonucleotide probes are synthesized with a complementary sequence to a gene of interest so they hybridize to the mRNA when applied to a brain section. (B) Example of an in situ hybridization reaction performed on a sagittal mouse brain section. The probe was conjugated to digoxigenin and visualized with a colorimetric reaction. 158  Guide to Research Techniques in Neuroscience Although it is possible to use fluorescent molecules to label probes for ISH, fluorescent in situ hybridization generally refers to a specific technique used to examine the location of DNA within chromosomes rather than tissue. A scientist can use this method to visualize chromosomal abnormalities, such as missing, mutated, or dislocated genetic segments. Immunohistochemistry Immunohistochemistry (IHC; histo = “tissue”) is used to visualize the expression of proteins. IHC applied to cells is often referred to as immunocytochemistry (cyto = “cells”), and IHC using fluorescent reagents is referred to as immunofluorescence. The root immuno- refers to the fact that these techniques depend on antibodies to recognize and bind to specific proteins (antigens). By exploiting the immune system of other animals, such as mice, sheep, rabbits, and goats, scientists and commercial suppliers have generated a wide number of antibodies for many proteins. Table 6.3 lists a few of the many antibodies that are commonly used in neuroscience. See Chapter 15 for details about antibody production. To perform an IHC experiment, a scientist incubates a sample with an antibody that recognizes a specific antigen. This primary antibody, the antibody that binds to the protein, can be conjugated directly with a fluorescent molecule or chromogenic enzyme. This procedure is known as direct IHC (Figure 6.6A). Alternatively, in indirect IHC, a scientist adds a secondary antibody that recognizes the primary antibody (Figure 6.6B). Because multiple secondary antibodies TABLE 6.3  Commonly Used Antibodies that Label Specific Neural Cell Types Cell Type Antibodies Progenitors/radial glia Nestin; Pax6; RC2; vimentin; NF (neurofilament) Young neurons Doublecortin (DCX); NeuroD Neurons Tuj1 (neuron-specific β-tubulin); NeuN Dendrites MAP2 Axons Tau-1, L1, Tag-1 Synapses PSD95, synapsin Neuronal subtypes GAD (GABAergic neurons); vGLUT (glutamatergic); TH (dopaminergic); 5-HT (serotonergic); AChE (cholinergic) Glia GFAP (astrocytes); MBP (oligodendrocytes, myelin); PLP Oligodendrocyte progenitor cells (OPC) NG2, A2B5, O4 (late progenitor) Visualizing Neural Structure Chapter | 6  159 can bind to the primary antibody, the signal is amplified. The secondary antibody will only react with the primary antibody if it is directed against the correct immunoglobulin molecule of the animal species in which the primary antibody was raised. For example, if the primary antibody was produced in a rabbit, then the secondary antibody must be antirabbit and produced in a different animal. The secondary antibody is conjugated with a fluorescent molecule or chromogenic enzyme that allows the scientist to visualize the expression of the protein. It is critical that an investigator determines whether the result of an IHC experiment is specific for the protein of interest. Using both a positive control (a specimen known to contain the antigen of interest) and a negative control (a specimen known not to contain the antigen of interest) is useful for determining whether the antibody works properly. Another good control experiment for primary antibody specificity is to preincubate the primary antibody with a known solution of its antigen; this process should form antibody/antigen complexes so there is no free antibody to bind to a sample. In indirect IHC, incubating the specimen with the secondary antibody but no primary antibody is important to test for nonspecific binding of the labeled secondary antibody. While IHC is straightforward in theory, there are a number of steps that must be optimized for IHC to work in practice. A scientist may need to adjust the method of fixation, antibody concentration, length of time the antibody is incubated with tissue, and accessibility to the antigen for good results. Each batch of antibody is different, even antibodies directed against the same antigen. Therefore, an investigator should optimize experimental conditions for each new antibody before performing IHC experiments on multiple samples. A technique called array tomography uses IHC on ultrathin serial sections to improve spatial resolution and eliminate out-of-focus fluorescence during imaging. This method allows investigators to create highly detailed three-dimensional images of protein expression in the nervous system. An ultramicrotome is used FIGURE 6.6  Immunohistochemistry (IHC). (A) In direct IHC, primary antibodies are conjugated directly to a label that can be visualized. (B) In indirect IHC, primary antibodies attract labeled secondary antibodies, further amplifying the signal. 160  Guide to Research Techniques in Neuroscience to cut and collect ribbons of ultrathin (50–200 nm) serial sections on glass microscope slides. A scientist can perform IHC to detect the presence of an antigen, but then the antibody can be removed and another IHC reaction performed. As many as 10 IHC reactions can be performed on the same section. Because the sections are so thin, antibodies have greater access to antigens in the section. The serial sections can then be reconstructed to provide high-resolution, three-dimensional information about the spatial relationships of these proteins. In addition, the sections can be used for electron microscopy, allowing an unprecedented investigation of protein expression combined with powerful magnification and resolution. Enzymatic Histochemistry An enzyme is a biomolecule capable of catalyzing a biological reaction. These proteins can be detected using IHC; however, another method to detect enzymes is to use the enzyme’s endogenous activity to create a visible reaction product— a process known as enymatic histochemistry. The scientist incubates a brain section in a solution containing a chromogenic chemical capable of serving as a substrate for the enzyme. A colored byproduct indicates the presence of the enzyme. For example, endogenous acetylcholinesterase (the enzyme that breaks down acetylcholine) is often used as a marker for acetylcholine in brain sections. Another commonly detected enzyme is cytochrome oxidase, which aids in visualizing metabolically active neurons, such as whisker barrels in rodent sensory cortex (Figure 6.7). Reporter Genes A reporter gene is a nonendogenous gene encoding an enzyme or fluorescent protein whose expression is controlled by a promoter for a separate gene of interest. Therefore, it is possible to examine the spatial and temporal expression of the gene by measuring the expression of the reporter. For example, imagine that a scientist is interested in studying the expression pattern of Gene X. The scientist can create a reporter construct (using recombinant DNA technology methods described in Chapter 10) in which the reporter gene is placed under the control of the promoter for Gene X. Alternatively, the scientist can fuse the DNA sequence of the reporter with the DNA sequence for Gene X and place both under the control of the promoter for Gene X. In either case, the reporter gene provides the scientist the ability to study the spatial and temporal expression of Gene X by measuring expression of the reporter (Figure 6.8). Common reporters include green fluorescent protein (GFP) and its derivatives, lacZ (which encodes the enzyme β-galactosidase), and luciferase (the firefly protein). The expression pattern of reporter genes can reveal information about the gene on multiple levels, from subcellular localization within a neuron to patterns of organization in a whole animal. Visualizing Neural Structure Chapter | 6  161 FIGURE 6.7  Enzymatic histochemistry. (A) In this example, rodent whisker barrels are enriched in the enzyme cytochrome oxidase, so incubating sections with the substrate (cytochrome c) and a colorimetric indicator (e.g., DAB) allows the whisker barrels to be visualized. Courtesy of Dr Pushkar Joshi. FIGURE 6.8  Using reporter genes to examine gene expression patterns. (A) In this BAC transgenic mouse line (Chapter 12), the promoter for neurotensin was used to drive expression of the GFP reporter gene. This mouse allows for visualization of the specific neurons in which neurotensin is normally expressed. (Reprinted with permission from The Gene Expression Nervous System Atlas Project.) (B) Reporter genes can also be used to replace a gene when creating a knockout animal (Chapter 13). Here, LacZ gene expression (shown by β-galactosidase staining) replaces the gene Satb2, showing where Satb2 is normally expressed. Mutants have altered axonal projections (arrowheads). (Reprinted from Alcamo E.A. et al. (2008) Satb2 regulates callosal projection neuron identity in the developing cerebral cortex, Neuron 57(3): 364–377, with permission from Elsevier.) 162  Guide to Research Techniques in Neuroscience VISUALIZING CIRCUITRY So far, this chapter has reviewed methods of investigating cells based on their morphology, location, and gene/protein expression profiles. Of course, a neuron never functions in isolation; its role in the brain is governed by its unique combination of inputs and outputs and the neural circuits to which it belongs. Therefore, scientists often want to know the specific cells in the brain that provide synaptic input to a neuron of interest, as well as the specific cells to which a neuron projects. As just described, a Golgi stain can provide initial insight into the connections formed by neurons, but this method is limited by the random staining pattern of cells. Other methods allow scientists to more specifically investigate a neuron’s direct and indirect connections with other neurons in the nervous system. Anterograde and Retrograde Tracers A neuronal tracer is a chemical probe that labels axon paths to illuminate connectivity in the nervous system. Tracers are described by the direction they travel in a neuron. Anterograde refers to transport from the cell body through the axon to the presynaptic terminal; retrograde refers to tracing in the opposite direction— transport from the synaptic terminal back to the cell body (Figure 6.9). Some tracers are strictly capable of only anterograde or retrograde transport. Others work in both directions. Classical tracers include horseradish peroxidase, biotinylated dextran amine, diamidino yellow, Fluoro-Gold, and fast blue (Table 6.4). These tracers may be fluorescent or produce a colorimetric product that can be visualized using light microscopy (Box 6.2). Many scientists use viruses to deliver a genetically encoded fluorescent protein, such as GFP, to a brain region of interest for use as an anterograde tracer. The fluorescent protein fills the soma, but also diffuses down the length of an axon to show anterograde targets. To use an anterograde or retrograde tracing technique, a scientist almost always injects the tracer into the brain of a live animal. After a waiting period (1–7 days), the brain can be removed, processed, and examined for presence FIGURE 6.9  Anterograde and retrograde tracers.  (A) Anterograde tracers show efferent projections, revealing regions that receive projections from cells in the labeled area. (B) Retrograde tracers show afferent projections, indicating regions that project to the labeled area. Visualizing Neural Structure Chapter | 6  163 TABLE 6.4  Classical Anterograde and Retrograde Tracers Tracer Direction Comments Horseradish peroxidase (HRP) Retrograde Produces brown precipitate after reaction with hydrogen peroxide and DAB (diaminobenzidine) Fluorescent microspheres Retrograde Available in many different colors; nontoxic Fluoro-gold Retrograde Widely used, rapid labeling Diamidino yellow Retrograde Produces yellow fluorescence Fast blue Retrograde Stable, rapid labeling; produces blue fluorescence Cholera toxin, subunit B (CTB) Retrograde May also be anterograde DiI, DiO Anterograde and retrograde Lipophilic dye crystals Biotinylated dextran amine (BDA) Anterograde and retrograde Widely used; direction of transport depends on molecular weight and pH; can be visualized by EM Phaseolus vulgaris- leucoagglutinin (PHA-L) Anterograde Plant lectin; can be visualized by EM Tritiated amino acids (3H-proline, 3H-leucine) Transsynaptic (anterograde) Detected using autoradiography Wheat germ agglutinin (WGA) Transsynaptic Plant lectin; anterograde and retrograde transport possible; often conjugated to HRP for detection; transgene encoding WGA can be used to label genetically defined neural circuits Tetanus toxin, fragment C (TTC) Transsynaptic (retrograde) Transgene encoding TTC can be used to label genetically defined neural circuits; nontoxic fragment Pseudorabies virus (PRV) Transsynaptic Does not infect primates, including humans; bartha strain most commonly used for tracing studies; less virulent, only retrograde transport Herpes simplex virus (HSV) Transsynaptic Broad host range, including humans 164  Guide to Research Techniques in Neuroscience of the tracer. An anterograde tracer should be present in the cell bodies and presynaptic terminals of the injected neurons, highlighting areas to which those neurons project. A retrograde tracer should be present in the cell bodies of neurons that project to the injection site. Tracing experiments are often combined with other techniques, such as electrophysiology or IHC. For example, an investigator might wish to know which cells in the brain send synaptic input to a neuron immediately following an in vivo electrophysiology recording. The scientist could inject a retrograde tracer, such as horseradish peroxidase, wait 3–7 days for transport to take place, and then perfuse the animal. By performing histology and surveying brain sections for labeled cells, the scientist could identify which cells projected to the cell(s) at the recording site. Transsynaptic Labeling Traditional anterograde or retrograde tracing experiments label only direct connections in a neural circuit. However, functional brain circuits are not simply composed of direct one-to-one connections, but also contain complicated pathways involving second- or third-order neurons. Scientists wishing to study these multisynaptic circuits can use a variety of transsynaptic tracers—tracers that can cross synapses and label multiple neurons in a circuit. Radioactive Amino Acids Radioactive amino acids, such as 3H-proline or 3H-leucine (3H is pronounced “tritiated”), can be used for anterograde transsynaptic labeling. A scientist injects the amino acids into the local environment of the primary neurons, some of which are taken up by cells via amino acid transporters. Once inside the cell, they are incorporated into proteins, some of which are shuttled down axons by endogenous anterograde transport. Some of these proteins are secreted at the synaptic terminal and received by a second-order neuron. This process may repeat itself so that a third-order neuron is labeled. The labeled tracts can be detected using autoradiography (Box 6.2). Plant Lectins Lectins are proteins that exhibit extremely high binding affinities for specific sugars. Some plant lectins have affinities for specific glycoproteins on neuronal plasma membranes. After binding, they can be internalized and transported within the neuron. Some examples of plant lectins that can also be transported across synapses include wheat germ agglutinin and tetanus toxin fragment C. These lectins can be detected using antibodies against the lectin itself or by conjugation with a chromogenic enzyme (Box 6.2). In addition to the standard method of injecting the tracer into an area of interest, researchers can create transgenic animals (Chapter 12) that express transneuronal lectin tracers under the control of neuron-specific promoters to examine cell-type specific circuits. Visualizing Neural Structure Chapter | 6  165 Viruses The use of viruses as transsynaptic tracers exploits the ability of viruses (especially viruses that naturally infect neural tissue) to invade neurons and then produce infectious progeny. Some strains of virus release progeny capable of crossing synaptic junctions and subsequently infecting downstream neurons in a circuit. The virus acts as a self-amplifying tracer that progressively labels neurons in a circuit as they become infected. The most common viruses used to label transneuronal circuits are pseudorabies virus and herpes simplex virus. The major limitation to using viruses as transsynaptic tracers is that these viruses can actually kill neurons and lead to generalized infection in the animals. The longer the animal is infected, the more likely that higher-order neurons in a circuit will be labeled, but the greater the likelihood that the first- and second-order neurons will die as the infection proceeds. CONCLUSION The organization and structure of the nervous system provide the first clues about function and how particular structures or classes of cells can give rise to behaviors. This chapter has surveyed methods of investigating the morphology, expression patterns, and connectivity of cells. In the next chapter, we will survey methods that visualize the electrical and biochemical activity within neurons. SUGGESTED READING AND REFERENCES Books Blurry, R. W. (2010). Immunocytochemistry: A practical guide for biomedical research. New York, NY: Springer. Chandler, D. E., & Roberson, R. W. (2009). Bioimaging: Current concepts in light and electron microscopy. Sudbury, MA: Jones and Bartlett Publishers. Kiernan, J. A. (2008). Histological and histochemical methods: Theory and practice (4th ed.). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Polak, J. M., & Van Noorden, S. (2003). Introduction to immunocytochemistry (3rd ed.). Oxford: BIOS Scientific Publishers. Yuste, R., & Konnerth, A. (2005). Imaging in neuroscience and development: A laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Yuste, R., Lanni, F., & Konnerth, A. (2000). Imaging neurons. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Review Articles Enquist, L. W., & Card, J. P. (2003). Recent advances in the use of neurotropic viruses for circuit analysis. Current Opinion in Neurobiology, 13, 603–606. Jones, E. G. (2007). Neuroanatomy: cajal and after cajal. Brain Research Reviews, 55, 248–255. Kobbert, C., Apps, R., Bechmann, I., Lanciego, J. L., Mey, J., & Thanos, S. (2000). Current concepts in neuroanatomical tracing. Progress in Neurobiology, 62, 327–351. Massoud, T. F., Singh,A., & Gambhir, S. S. (2008). Noninvasive molecular neuroimaging using reporter genes: part I, principles revisited. AJNR American Journal of Neuroradiology, 29, 229–234. 166  Guide to Research Techniques in Neuroscience Pollack, J. D.,Wu, D.Y., & Satterlee, J. S. (2014). Molecular neuroanatomy: a generation of progress. Trends in Neurosciences, 37, 106–123. Primary Research Articles—Interesting Examples from the Literature Alcamo, E. A., Chirivella, L., Dautzenberg, M., Dobreva, G., Farinas, I., Grosschedl, R., et al. (2008). Satb2 regulates callosal projection neuron identity in the developing cerebral cortex. Neuron, 57, 364–377. Braz, J. M., & Basbaum, A. I. (2008). Genetically expressed transneuronal tracer reveals direct and indirect serotonergic descending control circuits. Journal of Comparative Neurology, 507, 1990–2003. Chen, B., Schaevitz, L. R., & McConnell, S. K. (2005). Fezl regulates the differentiation and axon targeting of layer 5 subcortical projection neurons in cerebral cortex. Proceedings of the National Academy of Sciences of the U.S.A, 102, 17184–17189. Chung, K., Wallace, J., Kim, S.-Y., Kalyanasundaram, S., Andalman, A. S., Davison, T. J., et al. (2013). Structural and molecular interrogation of intact biological systems. Nature, 497, 332–337. Fakhry, A., & Ji, S. (2014). High-resolution prediction of mouse brain connectivity using gene expression patterns. Methods http://dx.doi.org/10.1016/j.ymeth.2014.07.011. Faucherre, A., & Lopez-Schier, H. (2014). Dynamic neuroanatomy at subcellular resolution in the zebrafish. Methods in Molecular Biology, 1082, 187–195. Micheva, K. D., & Smith, S. J. (2007). Array tomography: a new tool for imaging the molecular architecture and ultrastructure of neural circuits. Neuron, 55, 25–36. Takeuchi, A., Hamasaki, T., Litwack, E. D., & O’Leary, D. D. (2007). Novel IgCAM, MDGA1, expressed in unique cortical area- and layer-specific patterns and transiently by distinct forebrain populations of Cajal-Retzius neurons. Cerebral Cortex, 17, 1531–1541. Protocols Card, J. P., & Enquist, L. W. (2001). Transneuronal circuit analysis with pseudorabies viruses. Current Protocols in Neuroscience (Chapter 1), Unit 1.5 http://dx.doi.org/10.1002/0471142301. ns0105s09. Gerfen, C. R. (2003). Basic neuroanatomical methods. Current Protocols in Neuroscience (Chapter 1), Unit 1.1 http://dx.doi.org/10.1002/0471142301.ns0101s23. Helmchen, F., Konnerth, A., & Yuste, R. (Eds.). (2011). Imaging in neuroscience: A laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Kay, K. R., Smith, C., Wright, A. K., Serrano-Pozo, A., Pooler, A. M., Koffie, R., et al. (2003). Studying synapses in human brain with array tomography and electron microscopy. Nature Protocols, 8, 1366–1380. Tomer, R., Ye, L., Hsueh, B., & Deisseroth, K. (2014). Advanced CLARITY for rapid and highresolution imaging of intact tissues. Nature Protocols, 9, 1682–1697. Volpicelli-Daley, L. A., & Levey, A. (2004). Immunohistochemical localization of proteins in the nervous system. Current Protocols in Neuroscience (Chapter 1), Unit 1.2 http://dx.doi. org/10.1002/0471142301.ns0102s25. Wilson, C. J., & Sachdev, R. N. (2004). Intracellular and juxtacellular staining with biocytin. CurrentProtocolsinNeuroscience(Chapter1),Unit1.12http://dx.doi.org/10.1002/0471142301. ns0112s26. Websites IHC World: http://www.ihcworld.com. The Antibody Resource: http://www.antibodyresource.com. Allen Brain Atlas: http://www.brain-map.org. Mouse Brain Architecture: http://brainarchitecture.org. 167 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00007-1 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 7 Visualizing Neural Function After Reading This Chapter, You Should be Able to: l Describe techniques for visualizing activity and function in fixed tissue l Describe nonelectrophysiological methods of measuring neural activity l Describe techniques for visualizing protein function Techniques Covered: l Static measures of activity and function: measuring neural activity using immediate early genes, measuring cell proliferation with thymidine analogs, measuring protein trafficking with pulse-chase labeling l Visualizing dynamic neural activity: voltage sensors (voltage-sensitive dyes, genetically encoded voltage indicators), calcium sensors (calcium-sensitive dyes, genetically encoded calcium indicators), synaptic transmission sensors (FM dyes, synaptopHluorin) l Visualizing protein function: reporter genes, fluorescence resonance energy transfer (FRET), bimolecular fluorescence complementation (BiFC), fluorescence recovery after photobleaching (FRAP), photoactivation/photoconverstion Photographs in textbooks and other publications can sometimes create the impression that an individual cell is a small, static blob—a microscopic collection of organelles in a cytoplasmic soup that sits silently within a tissue. These individual snapshots are extremely misleading. The living cell is a dynamic wonder—an exquisitely complicated machine full of activity that a scientist cannot see with the naked eye, especially not in a single photograph. A neuron can fire electrical impulses in a steady tonic rhythm or phasic bursts, depending on the activity in other cells. Genes are constantly transcribed, with mRNA shuttling out of the nucleus and ribosomes quick to translate genetic sequences into functional proteins. Many proteins dash about the cell, changing their function and activity depending on the other proteins they encounter and the cellular signals transduced from the plasma membrane. Methods to visualize the structure and localization of cells and proteins, as described in the last chapter, offer a good introduction to studying the activity that takes place in a neuron. However, these methods do not fully reveal the dynamic electrical and biochemical activity found within the cell. As a 168  Guide to Research Techniques in Neuroscience metaphor, pretend you know nothing about soccer (or football if you are outside the United States). To learn more about the sport, you find several photographs of people playing soccer, mostly photographs taken from a bird’s-eye view of the field. While these photographs would no doubt teach you some aspects of the game, the only way to determine the rules, how individual players move, and how the players work as a team is to watch movies of the game. The same is true for biological systems; while photographs reveal the structure and localization of proteins and cells at a snapshot in time, a scientist can learn much more by observing the dynamic events that take place within neurons. The purpose of this chapter is to survey methods used to visualize electrical and biochemical activity in neurons. We begin with a description of how we can visualize functional processes in fixed tissue. Next, we will survey the major methods of characterizing electrophysiological properties of neurons using optical methods. Finally, we survey the growing collection of methods that can be used to visualize and manipulate biochemical dynamics. Many of these methods are based on the use of fluorescent dyes and proteins that make specific components of living cells visible under the microscope. STATIC MARKERS OF ACTIVITY Although it is ideal to measure neural activity as it occurs, it is not always possible or practical. Some cellular functions can be examined through the use of activity markers in fixed, histological brain sections. There are two fundamental approaches to examining neural activity after it has already occurred: (1) measuring activity indirectly by measuring byproducts that accumulate during specific processes in active neurons and (2) incorporating a marker into cells that can indicate the presence of activity during subsequent histological examination. In both methods, a scientist examines a snapshot of activity, a fixed representation of one moment in a dynamic process. Assaying Neural Activity in Fixed Tissue Neural activity leads to the transient and rapid (within minutes) transcription of a group of genes known as immediate early genes (IEGs). These genes encode a diverse range of proteins, including transcription factors (Fos), cytoskeletal-interacting proteins (Arc), and phosphorylated ribosomal subunits (pS6). By using histological methods to stain for these markers, a scientist can indirectly determine which regions were active just prior to the time the brain was fixed (Figure 7.1).A scientist can use in situ hybridization (ISH) to show expression of the IEGs or immunohistochemistry (IHC) to show expression of their protein products (Chapter 6). Because IEG expression patterns and levels vary due to a number of uncontrollable factors, investigators cannot examine IEGs as stand-alone indicators of activity. The most reliable evidence compares IEG expression between a group Visualizing Neural Function Chapter | 7  169 of experimental animals and a control group—for example, demonstrating that there is significantly higher expression of IEGs in a specific neuron population in animals that received a stimulus compared to control animals that did not receive the stimulus. IEG patterns can also be used to screen neurons for the presence of activities that correlate with specific behaviors. For example, some brain regions thought to promote sleep states in the brain were discovered by examining IEGs in animals that were asleep prior to fixation. In these types of studies, IEGs only provide the first clue as to the function of a brain region. It is necessary to follow up these studies with other, more reliable measurements of neural activity to obtain meaningful results. Assaying Cellular Function in Fixed Tissue To measure some functional processes in fixed tissue, a scientist can introduce a marker while the process occurs in live tissue, then detect this marker in subsequent histological experiments. Two biological functions commonly assayed using these methods include cell proliferation and protein trafficking. Assaying Cell Proliferation with Thymidine Analogs As a cell divides, it progresses through different stages of the cell cycle. Cells in the DNA synthesis phase of mitosis can be marked by exposing them to tagged DNA base-pair analogs. BrdU (bromo-deoxyuridine), a synthetic analog of the DNA base thymidine, or radioactive tritiated thymidine (3H-thymidine), can be injected into an animal or introduced into cells or tissue in culture media. Cells that are synthesizing DNA will incorporate BrdU or 3H-thymidine into their (A) (B) FIGURE 7.1  Measuring activity in a fixed brain sample.  Overall activity levels can be detected in fixed tissue by labeling for IEGs using immunohistochemistry or in situ hybridization techniques. Here, Fos immunoreactivity demonstrates that patterned visual stimulation evokes more activity in the (A) contralateral visual cortex compared to the (B) ipsilateral visual cortex. Reprinted from Dotigny, F., et al. (2008). Neuromodulatory role of acetylcholine in visually induced cortical activation: behavioral and neuroanatomical correlates. Neuroscience, 154(4), 1607–1618, with permission from Elsevier. 170  Guide to Research Techniques in Neuroscience newly made DNA in place of endogenous thymidine molecules. In subsequent histology experiments, BrdU can be detected using IHC, and 3H-thymidine can be detected using autoradiography. The presence of BrdU or 3H-thymidine indicates that a cell was dividing around the time of injection. However, these markers do not indicate whether the cell continued to proliferate or stopped dividing to become a functional, differentiated cell. To answer these questions, a scientist performs additional IHC experiments for proteins that are present only during specific stages of the cell cycle or in certain types of cells. For example, Ki67 and PCNA are proteins that are present during the active, proliferating stages of the cell cycle but not during the resting stage. Therefore, investigators can compare the number of proliferating cells that differentiated into postmitotic cells to the number of proliferating cells that continued to divide by combining BrdU/3H-thymidine pulses with IHC for Ki67 (Figure 7.2). Assaying Protein Trafficking with Pulse-Chase Labeling Pulse-chase labeling is used to observe the movement of a substance through a biochemical or cellular pathway. A labeled probe is injected into animals or added to cultured cells for a brief period (the pulse) and then washed away and replaced by unlabeled molecules (the chase). By following changes in the localization of a marker in different subcellular compartments over time, a scientist can observe protein trafficking pathways. For example, multiple proteins can be pulsed at the same time with a radioactively labeled amino acid. By fixing specimens at different time points, a scientist can use autoradiography to determine the location of labeled probes at regular intervals to observe changes in the localization of the incorporated label. Pulse-chase experiments have been important in elucidating many biochemical pathways, including synthesis and release of neurotransmitters. FIGURE 7.2  Measuring cell proliferation in a fixed brain sample. (A) In this example, cells that were actively dividing during the time of BrdU injection are identified using immunohistochemistry against BrdU. (B) Immunohistochemistry against Ki67 in the same sample identifies cells that were actively proliferating prior to perfusion. Courtesy of Dr Sandra Wilson. BrdU Ki67 Visualizing Neural Function Chapter | 7  171 VISUALIZING NEURAL ACTIVITY In Chapter 1, we discussed methods of visualizing activity in the whole brain with a spatial resolution of millimeters (about 100,000 neurons) and a temporal resolution of several seconds. In Chapter 4, we discussed electrophysiological methods, which can record neural activity with the spatial resolution of a single neuron (depending on the type of electrode and location in neural tissue) and a temporal resolution of milliseconds. The methods of visualizing neural activity described here combine advantages of both whole-brain imaging and electrophysiology techniques: it is possible to visualize neural activity with single-neuron resolution or in thousands of neurons at once while simultaneously examining the spatial relationships of those neurons. The temporal resolution is typically on the order of milliseconds to seconds, depending on the specific technique. These methods can be used to visualize activity in dissociated cells in culture, tissue slices in bath media, and even intact brains in living organisms. Visualizing neural activity depends on specialized fluorescent probes that report changes in membrane potential, calcium concentration, or synaptic vesicle fusion. These probes can be organic dyes that a scientist adds to a neural system prior to performing an experiment or a genetically encoded fluorescent protein stably expressed in a transgenic animal. Dyes tend to exhibit better temporal properties and signal-to-noise characteristics than proteins. However, genetically encoded proteins can be targeted to specific cell types in the brain, allowing a scientist to observe activity in molecularly defined circuits. Table 7.1 compares various methods of visualizing neural activity. Imaging Voltage Techniques that visualize changes in membrane potential are the closest analogs to electrophysiological recordings, as they report voltage changes in neuronal membranes with a fine temporal resolution. Voltage-sensitive dye imaging (VSDI) is currently the primary method by which scientists visualize changes in voltage. However, new genetically encoded voltage indicators (GEVIs) are now making it possible to image the membrane potential in genetically defined cell types. Voltage-Sensitive Dyes Voltage-sensitive dyes shift their absorption or emission fluorescence based on the membrane potential, allowing a scientist to gauge the global electrical state of a neuron. Unlike extracellular electrophysiology techniques, it is possible to detect subthreshold synaptic potentials in addition to spiking activity. These dyes allow activity measurements in large populations of neurons at once, a task that would otherwise require impractically large electrode arrays. There are a variety of voltage-sensitive dyes available, each differing in signal duration, intensity, signal-to-noise ratio, and toxicity. Most dyes exhibit small signal changes; the fractional intensity change (ΔI/I or ΔF/F, I = intensity and 172 GuidetoResearchTechniquesinNeuroscience TABLE 7.1  Commonly Used Neural Activity Sensors Category Examples Advantages Disadvantages Voltage-sensitive dyes Di-8-ANEPPS, RH 414 High temporal resolution (microsecond response time). Low signal intensity changes (order of 0.1%). Promiscuous labeling. Genetically encoded voltage sensors ArcLight Targeting specificity. Lower temporal resolution than dyes (milliseconds). High background. Nonratiometric calcium indicator dyes Fluo-3, Fluo-4, calcium Green-2 Direct fluorescence intensity changes correlated with Ca2+ concentration. Prone to artifacts from loading concentration, photobleaching, and other experiment-specific conditions. Cannot be calibrated. Buffering of intracellular Ca2+ can alter signaling pathways if present at high concentration. Ratiometric calcium indicator dyes Fura-2, Indo-1 Large signal changes (10%). Resistant to experiment-specific artifacts, such as loading concentration and photobleaching. Enables calibration. More complicated data acquisition and measurement than nonratiometric dyes. Buffering of intracellular Ca2+ can alter signaling pathways if present at high concentrations. Genetically encoded calcium sensors GCaMP Targeting specificity. Can have long signal decay time, giving low temporal resolution. Buffering of intracellular Ca2+ can alter signaling pathways if present at high concentration. Dye-based synaptic vesicle markers FM dyes (FM1- 43, AM4-64) Can vary stimulation and exposure to dye to control number of vesicles labeled. Labels all membrane surfaces so background can be high and difficult to wash out of brain slices. pH-sensitive fluorescent proteins SynaptopHluorin Allows study of presynaptic release. Many active release sites can make imaging of specific synapses difficult. Vesicle is only visible while fused for exocytosis, but becomes invisible after endocytosis. Visualizing Neural Function Chapter | 7  173 F = fluorescence) is in the range of 10−4–10−3 (0.1%), although some dyes may reach 6% for a large 100-mV change in potential. Thus, there must be very little noise to detect a reliable signal. Furthermore, activity-dependent changes in the intrinsic optical absorption and reflection properties of the brain itself can interfere with the voltage-sensitive dye measurements. Results fromVSDI tend to look like electrophysiology data, showing changes in ΔI/I over a certain period of time to report membrane potential changes. This information can then be encoded to show the magnitude of changes in pseudocolor maps overlaying the specimen, providing spatial information about the signal (Figure 7.3). Genetically Encoded Voltage Indicators To examine molecularly defined subsets of neurons, a scientist may use a GEVI. Despite much effort, historically these tools have been inadequate for precise measurements of membrane potential. However, new tools that have recently been developed are making GEVIs a reality. For example, ArcLight is a relatively new GEVI that features improved signal size and signal-to-noise ratios over previous probes. Imaging Calcium Dynamics Intracellular calcium is central to many physiological processes, including neurotransmitter release, ion channel gating, and second messenger pathways. In neurons, calcium dynamics link electrical activity and biochemical events. • • ∆ FIGURE 7.3  Example of a voltage-sensitive dye imaging experiment.  (A) A craniotomy was performed over somatosensory cortex (S1) and mortor cortex (M1) in anesthetized mice to deliver the voltage sensitive dye RH1691. (B) Moving the C2 whiskers evokes transient increases in fluorescence, first in S1 (black), then in M1 (blue). (C) The spatiotemporal dynamics of voltage changes across the brain are observed after deflection of the C2 whisker. Reprinted from Ferezou, I., et al. (2007). Spatiotemporal dynamics of cortical sensorimotor integration in behaving mice. Neuron, 56, 907–923, with permission from Elsevier. 174  Guide to Research Techniques in Neuroscience Thus, changes in calcium concentration can indirectly indicate changes in electrical activity. As with voltage-sensitive probes, fluorescent calcium indicators exist as dyes and genetically encoded calcium indicators. Like voltage sensitive imaging, data for calcium imaging experiments typically show changes in fluorescence intensity or the ratio of fluorescence intensity over time, normalized to initial levels of fluorescence (F0). These traces are then used to create pseudo-colored maps to show spatial information about the changes, with hot colors (yellow–red) representing large changes and cool colors (blue–purple) representing small changes. Compared to VSDI, calcium imaging produces larger, more robust visible signal changes (ΔF/F = 1–20%) but with lower temporal resolution (typically milliseconds rather than microseconds). Calcium Indicator Dyes Calcium indicator dyes—organic molecules that change their spectral properties when bound to Ca2+—can be categorized as ratiometric or nonratiometric dyes. Ratiometric dyes are excited by or emit at slightly different wavelengths when they are free of Ca2+ compared to when they are bound to Ca2+. Thus, they can report changes in Ca2+ through changes in the ratio of their fluorescence intensity at distinct wavelengths. A common ratiometric indicator is Fura-2, which emits at 510 nm but can change the wavelength at which it is excited from 340 to 380 nm in response to calcium binding. Thus, monitoring the ratio of the fluorescence intensity emitted by Fura-2 at 340 nm to the fluorescence emitted at 380 nm reveals changes in calcium concentration (Figure 7.4). Ratiometric dyes allow investigators to correct for background changes in fluorescence unrelated to calcium dynamics, such as artifacts related to photobleaching, variations in illumination intensity, or differences in dye concentration. The primary disadvantage of using ratiometric dyes compared to nonratiometric dyes is that data acquisition and measurements are more complicated. Nonratiometric dyes report changes in Ca2+ directly with changes in either excitation or emission fluorescence intensity. The common nonratiometric indicators Fluo-3 and Fluo-4 exhibit predictable increases in emission fluorescence intensity with increases in calcium concentration. While the direct relationship between fluorescence intensity and calcium concentration is sensitive for detecting changes due to calcium binding, the measurement is also prone to detecting changes based on dye concentration and experiment-specific conditions. However, nonratiometric dyes tend to be easier to use and quantify. Because dye-indicators bind to Ca2+, they contribute to buffering of cellular Ca2+ and can alter Ca2+ homeostasis. Thus, quantitative measurements of Ca2+-signaling require careful understanding of the indicator properties, experimental conditions, and potentially interfering binding partners (e.g., Mg2+). Visualizing Neural Function Chapter | 7  175 Genetically Encoded Calcium Indicators The bioluminescence of the jellyfish Aequorea victoria is due to the presence of both green fluorescent protein (GFP) and a chemiluminescent protein called aequorin. Although GFP has received much attention and is ubiquitous in biological research aequorin has been used for calcium imaging since the late 1960s. This protein emits light when bound to calcium, without requiring illumination for excitation. Unfortunately, the measurable signal is extremely low and requires amplification. Thus, it has mostly been superseded by the use of fluorescent dyes and more robust genetically encoded calcium sensors. Genetically encoded calcium indicators (GECIs) take advantage of the conformational changes that occur in certain endogenous calcium-binding proteins when they bind to calcium. For example, the nonratiometric GECI GCaMP reports changes in calcium based on a fusion protein between GFP and the calcium binding protein calmodulin (Figure 7.5). When calmodulin binds to Ca2+, a conformational change causes an increase in GFP fluorescence. Modern iterations of the GCaMP protein can report neural activity at the resolution of single action potentials and reveal subthreshold activity. These probes can also reveal activity in neural processes, including axons and synaptic terminals. µ µ FIGURE 7.4  Imaging calcium dynamics using the ratiometric dye Fura-2. Fura-2 changes its emission spectra when bound to calcium.  To examine calcium dynamics, images are collected after stimulating cells loaded with Fura-2 at (A) 340 nm, dim at resting levels of calcium and (B) 380 nm, brighter at resting levels of calcium, and collecting emission data at 510 nm. The ratio of 340/380 indicates cells with high (bright) or low (dim) concentrations of calcium (C). The ratio can be converted into known concentrations of free calcium levels (D). Courtesy of Dr Jocelyn Krey. 176  Guide to Research Techniques in Neuroscience Imaging Synaptic Transmission A critical component of neural activity is the release of neurotransmitters from vesicles into the synaptic junction. Electrophysiology first revealed the fusion and reuptake of synaptic vesicles, but imaging fluorescent dyes and proteins that report synaptic vesicle activity provides additional details about the dynamic process of synaptic transmission. FM Dyes FM dyes are lipophilic styryl dyes that fluoresce when bound to membranes. The FM stands for Fei Mao, the chemist who developed these dyes. They are useful for reporting the exocytosis and endocytosis events that occur during synaptic vesicle recycling and can stain nerve terminals in an activity-dependent manner. Typically performed in cell culture preparations, a scientist adds dye to the medium to label the membrane surface. For activity-dependent labeling, neurons are stimulated so that all synaptic vesicles undergo exocytosis and endocytosis with the stained membrane. The stained membranes are internalized during endocytosis. The rest of the dye is washed away, leaving behind clusters of stained recycled synaptic vesicles. During an experiment, the investigator observes the destaining process that occurs as the stained synaptic vesicles release dye during further exocytosis. FM dyes have been used to characterize neurotransmitter release mechanisms, measure the kinetics of vesicle recycling, and observe vesicle movements. pH-Sensitive Fluorescent Proteins Neurotransmitter release can also be examined using the genetically encoded synaptic transmission reporter synapto-pHluorin, which uses pH-sensitive mutants of GFP. The interiors of synaptic vesicles are acidified to a pH of approximately 5.7, an environment that keeps the pHluorins in an off state. When the vesicle fuses with the plasma membrane during exocytosis, the pH rises to extracellular levels, switching the pHluorin on and causing it to fluoresce (Figure 7.6). One of the advantages of using synapto-pHluorins is that the signal regenerates through FIGURE 7.5  Mechanism behind modern genetically encoded calcium indicators (GECI).  Modern GECIs, such as GCaMPs, bind green fluorescent protein (GFP) to a calmodulin (CaM) protein and a peptide sequence called M13. In the presence of Ca2+, a conformational change causes an increase in GFP fluorescence intensity. Visualizing Neural Function Chapter | 7  177 multiple rounds of vesicle release and recycling, which permits vesicle recycling to be imaged in addition to synaptic transmission. VISUALIZING PROTEIN FUNCTION Scientists can use fluorescent probes to visualize the activity of specific proteins within a cell. Using these methods, it is possible to track the subcellular localization of proteins over time, as well as to detect the interaction between two or more proteins. These techniques are summarized in Table 7.2. Time-Lapse Imaging with Reporter Genes Reporter genes were described in the previous chapter as useful markers to localize protein expression. Fusing a fluorescent reporter gene, such as GFP or its variants, to a protein of interest allows a scientist to observe the location and trafficking of the protein in live cells and tissues (Figure 7.7). In most cases, the GFP fusion protein behaves in the same way as the original protein. The GFP fusion protein strategy has become a standard way to determine the distribution and dynamics of many proteins of interest in living cells. For example, a transcription factor fused to GFP can reveal changes in localization in response to growth factors, moving from cytoplasm to nucleus. Fluorescence Resonance Energy Transfer Fluorescence (or Förster) resonance energy transfer (FRET) is used to monitor interactions between proteins (Figure 7.8A). Two molecules of interest are each labeled with a different fluorophore, chosen so the emission spectrum of one fluorophore (donor) overlaps with the absorption spectrum of the other (acceptor). If the two proteins bind, bringing their fluorophores into close proximity (closer than about 2 nm), the energy of the absorbed light can be transferred directly from the donor to the acceptor. Thus, when the complex is illuminated FIGURE 7.6  SynaptopHluorin reports synaptic transmission dynamics. SynaptopHluorin does not fluoresce when in acidified synaptic vesicles (pH ∼ 5.7), but does fluoresce when the vesicles exocytose and expose the reporter to physiological pH (pH ∼ 7.4). pH ~ 5.7 pH ~ 7.4 178  Guide to Research Techniques in Neuroscience at the excitation wavelength of the first fluorophore, light is emitted at the emission wavelength of the second. FRET is measured as the ratio of fluorescence intensity of the donor to fluorescence intensity of the acceptor. This method can be used with two different spectral variants of GFP as fluorophores to monitor processes such as the interaction of signaling molecules with their receptors. FRET has traditionally been exploited to assess distances and orientations TABLE 7.2  Comparison of Techniques to Optically Investigate Protein Function Description Reporter genes Proteins or organelles can be fluorescently tagged and observed over time. Useful for observing behavior, movement and dynamics of tagged proteins. Fluorescence/Förster resonance energy transfer (FRET) A donor fluorophore with an emission spectra that overlaps with the excitation spectra of an acceptor fluorophore transfers energy to the acceptor when the donor and acceptor are in close proximity. Measures the shift in the ratio of donor emission intensity to acceptor emission intensity. Useful for observing interactions between two tagged proteins or two parts of a single protein. Bimolecular fluorescence complementation (BiFC) Two different proteins can be tagged with half a fluorophore. The fluorophore will fluoresce only if the two halves come near enough, reporting a close interaction between the tagged proteins. Measures the change in fluorescence intensity to indicate whether the two proteins are near each other. Fluorescence recovery after photobleaching (FRAP) High-intensity light is used to photobleach a region of fluorescence. Measures the kinetics of fluorescence intensity returning to photobleached region. Useful for observing dynamics of trafficking, diffusion, or binding/dissociation of fluorescently tagged proteins. Photoactivation/ Photoconversion Photoactivation causes a dramatic increase in fluorescence intensity after stimulation by light in specialized fluorophores. PA-GFP (photoactivatable GFP) becomes fluorescent after stimulation with UV light. Photoconversion uses light to change the emission spectra of a fluorophore. Useful for tracking localization changes of small populations or individual cells, organelles, or proteins. Visualizing Neural Function Chapter | 7  179 ′ ′ ′ ′ ′ ′ FIGURE 7.7  Example of time-lapse imaging with reporter proteins.  This migrating neuron expresses two different reporter proteins with different excitation and emission spectra: a cytoplasmic fluorescent protein that allows visualization of the cell’s morphology (white) and a GFP fusion protein that labels early endosomes (green). Observing changes in the position of the endosomes over time can reveal intracellular trafficking dynamics. FIGURE 7.8  Fluorescence resonance energy transfer (FRET) and bimolecular fluorescence complementation (BiFC) are used to determine if and when two proteins interact.  (A) In FRET, one protein is fused with cyan fluorescent protein (CFP), the other with yellow fluorescent protein (YFP). CFP is excited by light at ∼436 nm, causing it to emit light at ∼480 nm.YFP can be excited by ∼480 nm light, so if n emitting CFP-fused protein is nearby,YFP will emit at ∼530nm, indicating that the two proteins are in close proximity. (B) In BiFC, one fluorescent protein molecule is split in two, and each are fused with a separate protein. If excited by light when the two proteins are not in proximity, no emission occurs; however, if excited by light when the two proteins are in proximity, the two subunits can interact to emit light. 180  Guide to Research Techniques in Neuroscience between separate molecules or different sites within a single macromolecule. This technique can also be used to report any substrate modification that leads to a conformational change resulting in the interaction of donor and acceptor fluorophores. Bimolecular Fluorescence Complementation BiFC is similar to FRET in that it relies on the interactions and proximity of two proteins or parts of a protein to produce a visible change. Rather than two separate fluorophores, however, two halves of a single fluorophore (e.g., GFP) are split. Each half of the fluorophore is fused to one protein and does not fluoresce. When the two proteins are close in proximity, the two halves of the fluorophore can fold and form the functional fluorophore (Figure 7.8B). Fluorescence Recovery after Photobleaching One limitation to using fluorophores is the phenomenon of photobleaching and decrease in fluorescence intensity over time. However, investigators can exploit this phenomenon to examine protein turnover rates or trafficking. FRAP uses strong laser illumination to bleach the fluorescent molecules in a particular region. The investigator then measures the time course of fluorescence returning to the bleached region. This can reveal the kinetics of a fluorescently tagged protein’s diffusion, binding, or dissociation, or active transport into the bleached region (Figure 7.9). Photoactivation and Photoconversion Some fluorescent proteins have been engineered so they are not fluorescent until hit with a specific wavelength of light (often ultraviolet (UV)) that photoactivates the fluorophore. Photoactivatable fluorescent proteins can be fused to a protein and later activated, allowing selected subpopulations of the protein to be labeled and followed as they move around a cell. This is particularly useful for observing the behavior and dynamics of a highly abundant protein that may be difficult to track in a large population of fully fluorescent molecules. Some sophisticated photoactivatable proteins allow precise control over their fluorescence. For example, the photoactivatable protein Dronpa can be reversibly photoactivated; UV light increases fluorescence intensity, while blue light will quench the fluorescence. Photoconversion serves a similar purpose as photoactivation, but rather than activating a previously nonfluorescent molecule, a pulse of light causes a fluorophore to change its emission spectra from one color to another. This can be used to examine the timing of protein behaviors. For example, Kaede is a photoconvertible fluorescent protein that initially emits green light, but emits red light after UV illumination. Visualizing Neural Function Chapter | 7  181 CONCLUSION The development of fluorescent probes to observe neural activity and protein function allows scientists to view dynamic processes in living cells, organs, and whole animals. Taken together, these techniques, along with complementary methods to visualize structure (Chapter 6), record from neurons (Chapter 4), or investigate intracellular signaling cascades (Chapter 15), constitute a fantastic toolbox of methods to investigate how proteins determine the properties of neurons, as well as how neurons determine the properties of circuits and behavior. SUGGESTED READING AND REFERENCES Books Yuste, R., & Konnerth, A. (2005). Imaging in neuroscience and development: A laboratory manual. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Yuste, R., Lanni, F., & Konnerth, A. (2000). Imaging neurons. Cold Spring Harbor, NewYork: Cold Spring Harbor Laboratory Press. Review Articles Barth, A. L. (2007). Visualizing circuits and systems using transgenic reporters of neural activity. Current Opinion in Neurobiology, 17, 567–571. Ellis-Davies, G. C. (2007). Caged compounds: photorelease technology for control of cellular chemistry and physiology. Nature Methods, 4, 619–628. Giepmans, B. N., Adams, S. R., Ellisman, M. H., & Tsien, R.Y. (2006). The fluorescent toolbox for assessing protein location and function. Science, 312, 217–224. Guzowski, J. F., Timlin, J. A., Roysam, B., McNaughton, B. L., Worley, P. F., & Barnes, C. A. (2005). Mapping behaviorally relevant neural circuits with immediate-early gene expression. Current Opinion in Neurobiology, 15, 599–606. Lippincott-Schwartz, J., Altan-Bonnet, N., & Patterson, G. H. (2003). Photobleaching and photoactivation: following protein dynamics in living cells. Nature Cell Biology, Suppl. S7–S14. Miesenbock, G., & Kevrekidis, I. G. (2005). Optical imaging and control of genetically designated neurons in functioning circuits. Annual Review of Neuroscience, 28, 533–563. FIGURE 7.9  Fluorescence Recovery after Photobleaching reveals information about reporter protein kinetics.  Intense illumination is applied to a small region of a cell expressing a fluorescent reporter, and the recovery of fluorescent signal in the photobleached area is observed over time. The recovery time can be used to measure the kinetics of protein trafficking or turnover. 182  Guide to Research Techniques in Neuroscience Tian, L., Akerboom, J., Schreiter, E. R., & Looger, L. L. (2012). Neural activity imaging with genetically encoded calcium indicators. Progress in Brain Research, 196, 79–94. Zhang, F., Aravanis, A. M., Adamantidis, A., de Lecea, L., & Deisseroth, K. (2007). Circuitbreakers: optical technologies for probing neural signals and systems. Nature Reviews Neuroscience, 8, 577–581. Zhang, J., Campbell, R. E., Ting, A. Y., & Tsien, R. Y. (2002). Creating new fluorescent probes for cell biology. Nature Reviews Molecular Cell Biology, 3, 906–918. Primary Research Articles—Interesting Examples from the Literature Airan, R. D., Meltzer, L. A., Roy, M., Gong, Y., Chen, H., & Deisseroth, K. (2007). High-speed imaging reveals neurophysiological links to behavior in an animal model of depression. Science, 317, 819–823. Angevine, J. B., Jr., & Sidman, R. L. (1961). Autoradiographic study of cell migration during histogenesis of cerebral cortex in the mouse. Nature, 192, 766–768. Cao, G., Platisa, J., Pieribone, V. A., Raccuglia, D., Kunst, M., & Nitabach, M. N. (2013). Genetically targeted optical electrophysiology in intact neural circuits. Cell, 154, 904–913. Chenn, A., & Walsh, C. A. (2002). Regulation of cerebral cortical size by control of cell cycle exit in neural precursors. Science, 297, 365–369. Dombeck, D. A., Khabbaz, A. N., Collman, F., Adelman, T. L., & Tank, D. W. (2007). Imaging large-scale neural activity with cellular resolution in awake, mobile mice. Neuron, 56, 43–57. Dulla, C., Tani, H., Okumoto, S., Frommer, W. B., Reimer, R. J., & Huguenard, J. R. (2008). Imaging of glutamate in brain slices using FRET sensors. Journal of Neuroscience Methods, 168, 306–319. Goldsmith, C. J., Stadele, C., & Stein, W. (2014). Optical imaging of neuronal activity and visualization of fine neural structures in non-desheathed nervous systems. PlOS One, 9, e103459. Gunaydin, L. A., Grosenick, L., Finkelstein, J. C., Kauvar, I. V., Fenno, L. E., Adhikari, A., et al. (2014). Natural neural projection dynamics underlying social behavior. Cell, 157, 1535–1551. Marek, K. W., & Davis, G. W. (2002). Transgenically encoded protein photoinactivation (FlAsHFALI): acute inactivation of synaptotagmin I. Neuron, 36, 805–813. Marvin, J. S., Borghuis, B. G., Tian, L., Cichon, J., Harnett, M. T., Akerboom, J., et al. (2013). An optimized fluorescent probe for visualizing glutamate neurotransmission. Nature Methods, 10, 162–170. Matsuzaki, M., Honkura, N., Ellis-Davies, G. C., & Kasai, H. (2004). Structural basis of long-term potentiation in single dendritic spines. Nature, 429, 761–766. Murphy, L. O., Smith, S., Chen, R. H., Fingar, D. C., & Blenis, J. (2002). Molecular interpretation of ERK signal duration by immediate early gene products. Nature Cell Biology, 4, 556–564. Ohki, K., Chung, S., Ch’ng, Y. H., Kara, P., & Reid, R. C. (2005). Functional imaging with cellular resolution reveals precise micro-architecture in visual cortex. Nature, 433, 597–603. Outeiro, T. F., Putcha, P., Tetzlaff, J. E., Spoelgen, R., Koker, M., Carvalho, F., et al. (2008). Formation of toxic oligomeric alpha-synuclein species in living cells. PLoS ONE, 3, e1867. Patterson, G. H., & Lippincott-Schwartz, J. (2002). A photoactivatable GFP for selective photolabeling of proteins and cells. Science, 297, 1873–1877. Plath, N., Ohana, O., Dammermann, B., Errington, M. L., Schmitz, D., Gross, C., et al. (2006). Arc/ Arg3.1 is essential for the consolidation of synaptic plasticity and memories. Neuron, 52, 437–444. Reiff, D. F., Ihring, A., Guerrero, G., Isacoff, E. Y., Joesch, M., Nakai, J., et al. (2005). In vivo performance of genetically encoded indicators of neural activity in flies. Journal of Neuroscience, 25, 4766–4778. Reijmers, L. G., Perkins, B. L., Matsuo, N., & Mayford, M. (2007). Localization of a stable neural correlate of associative memory. Science, 317, 1230–1233. Visualizing Neural Function Chapter | 7  183 Schroder-Lang, S., Schwarzel, M., Seifert, R., Strunker, T., Kateriya, S., Looser, J., et al. (2007). Fast manipulation of cellular cAMP level by light in vivo. Nature Methods, 4, 39–42. Teather, L. A., Packard, M. G., Smith, D. E., Ellis-Behnke, R. G., & Bazan, N. G. (2005). Differential induction of c-Jun and Fos-like proteins in rat hippocampus and dorsal striatum after training in two water maze tasks. Neurobiology of Learning and Memory, 84, 75–84. Wang, J. W., Wong, A. M., Flores, J., Vosshall, L. B., & Axel, R. (2003). Two-photon calcium imaging reveals an odor-evoked map of activity in the fly brain. Cell, 112, 271–282. Protocols Barreto-Chang, O. L., & Dolmetsch, R. E. (2009). Calcium imaging of cortical neurons using Fura-2 AM. JoVE, 23. http://dx.doi.org/10.3791/1067. http://www.jove.com/index/details. stp?id51067. Carlson, G. C., & Coulter, D. A. (2008). In vitro functional imaging in brain slices using fast voltagesensitive dye imaging combined with whole-cell patch recording. Nature Protocols, 3, 249–255. Fuger, P., Behrends, L. B., Mertel, S., Sigrist, S. J., & Rasse, T. M. (2007). Live imaging of synapse development and measuring protein dynamics using two-color fluorescence recovery after photo-bleaching at Drosophila synapses. Nature Protocols, 2, 3285–3298. Gaffield, M. A., & Betz, W. J. (2007). Imaging synaptic vesicle exocytosis and endocytosis with FM dyes. Nature Protocols, 1, 2916–2921. Greenberg, D. S., Wallace, J. J., & Kerr, J. N. (2014). Imaging neuronal population activity in awake and anesthetized rodents. Cold Spring Harbor Protocols, 9, 912–22. Hatta, K., Tsujii, H., & Omura, T. (2006). Cell tracking using a photoconvertible fluorescent protein. Nature Protocols, 1, 960–967. Royle, S. J., Granseth, B., Odermatt, B., Derevier, A., & Lagnado, L. (2008). Imaging phluorinbased probes at hippocampal synapses. Methods in Molecular Biology, 457, 293–303. Shyu,Y. J., Suarez, C. D., & Hu, C.-D. (2008). Visualization of ternary complexes in living cells by using a BiFC-based FRET assay. Nature Protocols, 3, 1693–1702. 185 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00008-3 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 8 Manipulating Neural Activity After Reading This Chapter, You Should be Able to: l Describe the experimental logic behind gain-of-function and loss-of-function experiments l Compare different approaches for manipulating neural activity in the nervous system l Discuss approaches to manipulating neural activity in various organisms, including humans Techniques Covered: l Physical manipulation: ablation, thermal cooling l Electrical manipulation: microstimulation, electrolytic lesions l Pharmacological manipulation: agonists, antagonists, drug delivery methods l Genetic manipulation: perturbation of endogenous genes, expression of genes to ablate neurons l Pharmacogenetics: transgenes from other species, designer receptors exclusively activated by designer drugs (DREADDs: hM3Dq, hM4Di) l Optogenetics: channelrhodopsin-2 (ChR2), halorhodopsin (NpHR), archaerhodopsin (Arch), light delivery methods l Neuromodulation in humans: pharmacology, microstimulation, transcranial magnetic stimulation (TMS), ultrasonic neuromodulation (USNM) One of the ultimate experimental goals in neuroscience is to perturb a neural system to test hypotheses about the role of that system for a behavior or phenotype. For example, if a scientist hypothesizes that a population of neurons in the brain regulates thirst, a key experiment would be to artificially inhibit or stimulate those neurons to determine their necessity or sufficiency for water-seeking. The ability to manipulate neural activity in vivo or in vitro has never been more feasible, and many techniques have been developed and optimized over just the past 5–10 years in a variety of model organisms. Importantly, many of these tools have proven to be relatively cheap and easy to apply in a wide variety of laboratories. This chapter surveys approaches to artificially inducing or suppressing electrical activity in neurons (Figure 8.1). Physical methods ablate neural tissue to determine their role in behavior. Thermal cooling suppresses neural activity by decreasing the kinetics of cell physiology. Electrical methods apply current to a tissue preparation to stimulate neural activity or cause a lesion. Pharmacological 186  Guide to Research Techniques in Neuroscience methods employ chemical compounds to alter the function of ion channels and cell receptors. Genetically encoded transgenes can be used to ablate neurons or alter baseline physiological firing properties. A combination of multiple techniques, such as with pharmacogenetic tools, allows the targeting of nonendogenous receptors to control neurons with chemical compounds that are normally inert. Finally, optogenetic methods target light-sensitive ion channels and proteins to neurons that allow for neuromodulation with unprecedented spatial and temporal precision. EXPERIMENTAL LOGIC OF MANIPULATING NEURAL ACTIVITY Before performing a neuromodulation experiment or choosing any specific technique, it is important to clearly articulate the research goal and comprehend the meaning of experimental outcomes. In general, neural manipulation experiments can be classified in two ways: gain-of-function experiments and loss-of-function experiments. In the context of manipulating neural activity, a gain-of-function experiment stimulates activity in neurons and tests their -- - - - CNO Physical Electrical Pharmacological Genetic Pharmacogenetic Optogenetic FIGURE 8.1  Methods of manipulating neural activity. Neurons in the brain typically reside within heterogeneous populations composed of many cell types (here denoted by different colors). Physical lesioning techniques can ablate cells indiscriminately. Electrical techniques can ­microstimulate brain regions or cause electrolytic lesions. Pharmacological techniques inject a psychoactive drug either centrally or systemically. Genetic methods are capable of affecting very specific cell types by taking advantage of endogenous promoter regions of genes. Pharmacogenetic techniques use genetic targeting methods to deliver receptors capable of affecting neural activity to specific cell types. Optogenetic techniques deliver light-activated neuronal actuators to specific cell types. Manipulating Neural Activity Chapter | 8  187 sufficiency in causing a phenotype. In contrast, a loss-of-function experiment inhibits activity in neurons and tests their necessity for a phenotype. For example, consider a population of neurons that are hypothesized to regulate hunger. Perhaps these neurons exhibit increased activity following food deprivation or during conditions of hunger. If artificially stimulating these neurons in freely moving mice causes an increase in food seeking, the scientist can fairly conclude that activity in these neurons is sufficient to increase food-seeking behavior. Alternatively, if artificially inhibiting these neurons decreases food seeking, especially when an animal is normally hungry, the scientist can fairly conclude that activity in these neurons is necessary for normal food-seeking behavior. Note that the concepts of sufficiency and necessity are completely separable, and it is possible for a neural system to be sufficient but not necessary (or vice versa). For example, perhaps stimulating neurons could increase food-seeking behavior while inhibiting neurons would have no significant effect. This result could suggest that there are other neurons in the brain that also regulate food seeking, and no single neural population is necessary for food seeking to occur. Scientists must be careful not to make inappropriate conclusions from gainof-function and loss-of-function experiments. For example, if inhibiting a population of neurons decreases food intake, a scientist would be correct that these neurons are necessary for food-seeking behavior, but it would be unwise to conclude that these neurons function as “hunger neurons” or “food intake neurons.” Perhaps these neurons are also necessary for wakefulness or motor activity, and animals eat less simply because they move less. Thus, concluding that a neural system is necessary for a behavior might not, by itself, be informative. Such a result should impel a scientist to perform other experiments to rule out other potential roles for that neural system in other biological processes. PHYSICAL MANIPULATION A physical manipulation is any method that physically disrupts the activity of neurons due to direct physical contact. Two of the most common methods of physically manipulating neural activity are through ablation experiments and thermal cooling. Ablation Ablation, or creating a brain lesion, is the oldest technique used to study the necessity of a particular brain region for a biological function. A scientist can physically lesion the brain by excising or aspirating a piece of brain tissue during a stereotaxic surgery (Chapter 3). Despite their place in the history of neuroscience research, producing physical lesion is a relatively crude method for performing loss-of-function experiments in the brain. It is often difficult to control the precision of the lesion or the impact of removing tissue on the remaining brain structures. Because ablating 188  Guide to Research Techniques in Neuroscience an area damages not only cell bodies but also fibers of passage (the axons traveling through that region of the brain), it is difficult to know if the effects of the lesion are due to the neurons in the ablated area or neurons in a distant area whose fibers have been damaged. Sometimes ablating tissue can also affect the local circulatory system and cause systemic changes in the entire organism. Finally, an obvious limitation to physical ablation is that it is completely irreversible (see Table 8.1 for a comparison of neural lesioning techniques). Physical ablation techniques can also be used in invertebrate model organisms. For example, individual cells in Caenorhabditis elegans can be eliminated by lasers under a dissection microscope. These lesions are cleaner and more precise than physical ablations in vertebrates; however, they are also irreversible. Thermal Cooling Neurons in the brain function at an optimal temperature, and precise homeostatic mechanisms ensure that the brain remains at an optimal 37 °C. Cooling neurons a few degrees can temporarily slow down the kinetics of ion channels and cell physiology, effectively decreasing neural activity. Neurons can be easily cooled in in vitro preparations by slightly cooling the temperature of the bath solution. Neurons can be cooled in vivo by placing a piezoelectric implant in the brain. In contrast to physically removing brain tissue, cooling a brain region is reversible—assuming that the altered temperatures do not cause longlasting damage. Therefore, each animal can serve as its own internal control, and behavior can be compared during cooled and noncooled conditions. TABLE 8.1  Methods of Lesioning the Nervous System Category Tool Precision Physical Surgical blade Not very precise. Cannot discriminate between cell types or fibers of passage Electrical Microelectrode More spatially precise than surgical (physical) lesions, but cannot discriminate between cell types or fibers of passage Pharmacological Toxins (e.g., ibotenic acid) Spatial precision determined by quality and specificity of injection. Fibers of passage are unaffected Genetic Genetically encoded proteins (e.g., ataxin, diphtheria toxin receptor) Cell-type specificity Manipulating Neural Activity Chapter | 8  189 Although reversible, cooling the brain is not very spatially precise because it is not possible to discriminate among neurons in a given area. Cooling is also not temporally precise, as it takes a variable amount of time for the neurons to cool down and then warm up when the experiment is complete. Finally, thermal manipulation techniques are much more feasible for loss-of-function experiments than gain-of-function experiments because heating the brain more often causes damage rather than an increase in neural activity. ELECTRICAL MANIPULATION Microelectrodes are used during electrophysiology experiments to record electrical activity from neurons, but they can also be used to deliver electrical current into the brain or to neurons in culture in a process called microstimulation. A microelectrode is inserted into the brain or next to a neuron of interest and current is applied at a fixed frequency and time. The electrode elicits action potentials by changing the extracellular environment such that voltage-gated ion channels open, depolarizing the neuron. If too much current is applied to the electrode, the scientist can kill nearby cells. Often, this is done deliberately to create an electrolytic lesion that permanently ablates neurons. Therefore, microelectrodes can be used to either reversibly stimulate neural activity or irreversibly lesion neural tissue. The spatial resolution of electrical manipulation is limited, as injection of current affects all neurons in a local area, regardless of cell type, and can affect fibers passing through the area of stimulation as well. Microstimulation exhibits very precise temporal resolution because the frequency and timing of current delivery is under the control of the investigator. PHARMACOLOGICAL MANIPULATION A scientist can influence neural activity pharmacologically using one of hundreds of compounds that selectively bind to membrane-bound receptors (Figure 8.2). An agonist is a compound that can bind to and cause activation of a receptor, thus mimicking an endogenous ligand or neurotransmitter. A partial agonist is a compound that can produce an agonist-like effect, but not to the maximal extent of the full agonist or endogenous ligand.An antagonist binds to the receptor but does not cause activation. Therefore, it prevents the endogenous ligand from binding, inhibiting its biological activity. Where do pharmacological compounds come from? Some drugs are engineered in pharmaceutical laboratories to target specific receptors or other proteins. However, many plants and animals naturally produce pharmacological compounds and neurotoxins for self-defense. Scientists have purified several of these compounds to manipulate neural activity in vitro and in vivo. For example, many fish of the order Tetraodontiformes (the most famous being the pufferfish) produce a compound called tetrodotoxin. This compound is an antagonist of voltage-gated sodium channels, and therefore can be used to inhibit 190  Guide to Research Techniques in Neuroscience action potentials. Another commonly used agent to inhibit action potentials is tetraethylammonium, a potassium-selective ion channel antagonist. Many commonly used pharmacological agents target inhibitory GABA receptors. GABA-receptor antagonists include bicuculline and metrazol; they have a net effect of increasing neural excitability, as they prevent the normal inhibitory effects of GABA signaling. GABA-receptor agonists decrease neuronal excitability. For example, muscimol is a common GABA-receptor agonist that is often used during behavioral and/or in vivo electrophysiology experiments to inhibit neural activity. Some pharmacological agents can be used to irreversibly ablate neurons. For example, ibotenic acid causes excitotoxic effects when directly injected into the brain. Pharmacological ablation of neurons is preferable over physical or electrical ablation methods because it spares fiber tracts crossing through the target area. Therefore, if an injection of ibotenic acid is precise, a scientist can make better conclusions about any resulting effects on behavior or physiology. For in vitro experiments, drugs can be delivered in cell culture media or perfused in the bath solution incubating a brain slice. For in vivo experiments, some drugs can be delivered to the brain via the bloodstream, allowing a scientist to perform an intraperitoneal (i.p.) injection. However, many drugs do not cross the blood–brain barrier. In this case, a scientist can perform an intracerebroventricular (i.c.v.) injection into the lateral ventricles of the brain. Substances in the cerebrospinal fluid in the ventricles can then diffuse throughout the extracellular milieu of the brain. Alternatively, a scientist can perform a small, local injection into a discrete brain region. To inject a drug into the brain just before a behavioral experiment, a scientist can first implant a cannula above the brain region of interest during a surgery (Chapter 3), then subsequently inject the drug through the cannula. Injections into Ligand Agonist Partial Agonist Antagonist Binding of ligand to its receptor causes receptor activation Binding of agonist to a receptor causes receptor activation Binding of partial agonist to a receptor causes some activation, but not full effect Binding of antagonist to a receptor does not cause activation and prevents ligand from binding FIGURE 8.2  Compounds used in pharmacology: ligands, agonists, partial agonists, and antagonists. Manipulating Neural Activity Chapter | 8  191 the brain can be performed either by pressure injection or microiontophoresis, the process of using a small electrical current to drive substances out of a glass pipette. The advantage to microiontophoresis is that it allows for careful delivery of small volumes to discrete, local areas of the brain. For sustained, chronic delivery of a drug over multiple days or weeks, a scientist can use specially implanted osmotic minipumps that infuse the drug through an implanted cannula. Manipulating neural activity using pharmacological agents is usually reversible (except in the case of neurotoxins such as ibotenic acid). These agents have varying degrees of spatial precision depending on the specific receptors they target and the receptor expression patterns in the area of injection. They are typically not very temporally precise in vivo because, once injected, they require several minutes or even hours to clear from the system. GENETIC MANIPULATION Unlike physical, electrical, and pharmacological techniques, genetic methods of modulating neural activity can target specific cell types in the brain by taking advantage of endogenous promoters and other genetic regulatory elements that limit expression of genes to specific populations of neurons (Chapter 12). Therefore, if a cell type of interest exists in a heterogeneous population of neurons with one or more other cell types, it is possible to selectively target these neurons and not surrounding cell types using modern transgenic and gene targeting technologies (Chapters 12 and 13). To permanently ablate a specific cell type, a scientist can express a transgene that causes the death of the cell. For example, the ataxin transgene efficiently causes apoptosis when expressed in neurons. One potential limitation to using the ataxin transgene is that the neurons die immediately upon expression, which typically occurs during development or even in utero. Therefore, the brain may compensate and another brain region may assume the functions of the lost neurons. To gain more temporal control over expression of the ataxin gene, it is possible to use transgenic technologies that allow for controlled expression at specific time points (Chapter 12). Alternatively, it is also possible to express the diphtheria toxin receptor to ablate neurons at a time point of the investigator’s choosing. This receptor can cause the death of neurons, but only in the presence of the diphtheria toxin. Therefore, a scientist can target the diphtheria toxin receptor to a specific cell type and ablate those neurons at any time point during an animal’s lifespan by injecting the toxin. Another method of modulating neural activity using genetics is to overexpress or knockout a naturally occurring ion channel or membrane-bound receptor that regulates cellular excitability. For example, knocking out an endogenous GABA receptor subunit might increase neuronal excitability. However, neurons have a remarkable ability to compensate for alterations in cell physiology, and other proteins affecting cell physiology may be upregulated or downregulated to return the intrinsic excitability to normal. Furthermore, even if no compensatory 192  Guide to Research Techniques in Neuroscience mechanisms occur, genetically manipulating neurons does not allow for a finetune control over the neurons’ firing properties. In the last few years, scientists have developed ingenious methods of targeting genetically encoded actuators to neurons that allow a scientist to reversibly stimulate or inhibit neural activity with exceptional temporal precision. Two major categories of genetically encoded neuromodulation tools include pharmacogenetics and optogenetics. PHARMACOGENETIC MANIPULATION Pharmacogenetics refers to a set of membrane-bound receptors that are activated by chemical compounds (“pharmaco-”) and are genetically encoded (“-genetics”). The overall strategy is to genetically target neurons with nonendogenous receptors that are capable of being activated by a normally inert ligand. Because other cell types in the brain do not naturally respond to the ligand, only the cells that express the receptor are affected. Some pharmacogenetic strategies use receptor-ligand systems from other species. For example, the mammalian capsaicin receptor (TrpV1) is a ligandgated channel that depolarizes a neuron upon exposure to capsaicin. Flies and other invertebrates do not normally express this channel, so capsaicin exposure will depolarize only those neurons that express the receptor. A similar strategy can be used to inhibit activity and hyperpolarize neurons in the presence of a ligand. For example, the insect allatostatin receptor (Alstr) is not expressed in mammals. In the presence of the insect protein allatostatin, this receptor inhibits neural activity. Thus, if Alstr is expressed in the mammalian brain, addition of allatostatin will reversibly inactivate genetically targeted populations of neurons. These strategies have proven useful in many studies, but there are substantial disadvantages to using these systems in rodents. The capsaicin receptor is endogenous in rodents and causes a painful sensation when activated, so its use has been limited to TrpV1 knockout mice. The allatostatin receptor is not endogenous in rodents, however, allatostatin does not cross the blood–brain barrier, so it must be injected centrally through a surgically implanted cannula. In recent years, a new set of pharmacogenetic tools have been specifically engineered for use in a wide range of model organisms. Referred to as designer receptors exclusively activated by designer drugs (DREADDs), these tools consist of artificially engineered G-protein coupled receptors that activate intracellular signaling pathways to affect neuronal activity. The hM3Dq receptor (human M3 muscarinic DREADD coupled to Gq) takes advantage of the endogenous Gq signaling pathway to increase neuronal excitability and induce firing, whereas the hM4Di receptor (human M4 muscarinic DREADD coupled to Gi) takes advantage of the endogenous Gi signaling pathway to decrease neuronal excitability and silence neural activity. These receptors are sensitive to the normally inert ligand clozapine-N-oxide (CNO). Importantly, CNO is capable of crossing the blood–brain barrier, so animals can be injected intraperitoneally Manipulating Neural Activity Chapter | 8  193 during in vivo experiments. The effects of CNO typically last 6–10 h, the time it takes CNO to be metabolized and neuronal signaling cascades to return to baseline. For in vitro culture experiments using cells or tissue expressing DREADDS, CNO can be directly added to the bath solution. Pharmacogenetic techniques are spatially precise because neuromodulation is restricted to the genetically targeted population of cells. However, these techniques do not offer temporal precision greater than traditional pharmacological approaches. OPTOGENETIC MANIPULATION Optogenetics refers to a set of neuromodulation tools that are activated by light (“opto-”) and are genetically encoded (“-genetics”). Like other transgenes, optogenetic tools can be delivered with high spatial precision to specific cell types in the nervous system. However, because scientists can deliver light to neurons at any duration or frequency, optogenetics allows for unprecedented spatial and temporal precision. Since their development and application in the past few years, hundreds of laboratories now routinely use optogenetics to make major advances in our understanding of neural systems and circuits in a variety of model organisms. Optogenetic Transgenes The most commonly-used optogenetic probes are engineered versions of microbial opsins, proteins that translocate ions across membranes in response to stimulation by specific frequencies of light (Figure 8.3). FIGURE 8.3  Genetically encoded actuators used in optogenetics. (A) Channelrhodopsin-2 (ChR2) is a nonspecific cation channel that opens in response to blue light. (B) Halorhodopsin (NpHR) is a chloride pump that actively translocates chloride ions from outside to inside a cell in response to yellow light. (C) Archaerhodopsin (Arch) is a hydrogen (proton) pump that actively translocates hydrogen ions from inside to outside a cell in response to yellow light. 194  Guide to Research Techniques in Neuroscience Optogenetic Stimulation Channelrhodopsin-2 (ChR2) is a nonspecific cation channel derived from the green algae Chlamydomonas reinhardtii. ChR2 absorbs blue light, causing a conformational change that allows H+, Na+, K+, and Ca+ ions to passively diffuse down their concentration gradients. When expressed in neurons, the opening of these channels causes a rapid depolarization of the plasma membrane that can cause action potentials. Importantly, the channel also closes very rapidly when the blue light is switched off. Therefore, single action potentials can be generated with a brief pulse of blue light, with no long-lasting residual effects of stimulation. ChR2 is capable of driving neuronal activity at frequencies from 1 to 40 Hz. Scientists have made slight genetic modifications to ChR2 to create other channels with desirable neuromodulatory properties. For example, a mutation in the coding sequence of ChR2 causing a single amino acid substitution allows for stimulation of frequencies up to 200 Hz (scientists have named this tool “ChETA”). To cause activation of neurons over longer time periods without the need to continuously deliver light, scientists have created amino acid substitutions in ChR2 that cause the channel to remain open for several minutes or even hours in response to a single pulse of blue light. Called “step function opsins,” these ChR2-derived tools can be closed by a single pulse of yellow light. Scientists have also engineered mutations that alter the wavelength of light used to activate ChR2. The ability to express different versions of ChR2 that react to different colors of light allow for combinatorial experiments in which neuronal activity in multiple cell types can be distinctly manipulated in the same individual. Optogenetic Inhibition There are two major transgenes used for optogenetic inhibition. Halorhodopsin (NpHR) is a chloride pump derived from the halobacterium Natronomonas pharaonis. NpHR actively pumps Cl− ions into cells in response to yellow light. Archaerhodopsin-3 (Arch) is derived from Halorubrum sodomense and actively pumps hydrogen ions outside of the cell in response to yellow light. The translocation of Cl− ions into the cell or H+ neurons out of the cell effectively hyperpolarizes neurons. One potential limitation of using optogenetic techniques to inhibit neurons is the need for continuous light illumination. Unlike ChR2, for which blue light is typically applied at regular pulses to mimic action potential trains, yellow light must be continuously delivered to neurons for sustained hyperpolarization. For experiments in which a scientist wants to inhibit neural activity for minutes or even hours, it is not practical to continuously deliver yellow light due to potential side effects due to heat that may cause neuronal damage. Therefore, a scientist must carefully consider whether NpHR or Arch are the best tools for a given experiment. For experiments in which neural inhibition is necessary over multiple hours, perhaps a pharmacogenetic strategy may be the best approach. Manipulating Neural Activity Chapter | 8  195 Methods of Delivering Light to the Nervous System Optogenetic methods require the delivery of light to targeted neurons of interest. For in vitro applications, such as recording neurons in acute brain slice preparations (Chapter 14), scientists can position a light source directly next to the bath or culture plate. Alternatively, light can also be directly coupled to the light path of a microscope, and specialized shutters can be installed to deliver light at specific frequencies. For experiments in invertebrate model organisms, light is typically delivered to the entire housing chamber and passes into the brain directly. In vivo light delivery to rodents is more challenging because of the need to implant a light delivery device capable of penetrating the skull. Light must be delivered very close to the target neurons because brain tissue scatters light exponentially, with only about 10% light intensity remaining at a distance 500 μm from the light source. Furthermore, light delivery systems must not weigh more than a freely moving animal can carry. The most common method for delivering light to freely moving rodents is implanting a guide cannula for placement of a lightweight fiber optic cable (Figure 8.4A). This strategy allows for the targeting of relatively deep brain structures. During an experiment, the exposed cannula implanted on the skull is coupled to a fiber-optic cable (usually about 2 m in length) connected on the other end to a light delivery source, often a laser. For optical modulation of superficial cortical neurons, small light emitting diodes (LEDs) can be mounted directly above the brain over a cranial glass window (Figure 8.4B). These LEDs are powerful enough to deliver light to all cortical layers. Some scientists have even devised methods of coupling a small LED to a fiber-optic implant for delivery to deep brain structures so that the animal does not require tethering to a fiber-optic cable. Fiber optic cable Light emitting diode (LED) Cannula Optical fiber Light from laser diode Optical window LED Connector (A) (B) FIGURE 8.4  Methods of targeting light to neurons in the brain.  (A) The most common form of light delivery in optogenetic experiments is to implant a cannula above a cell population of interest, then deliver light via a fiber-optic cable inserted through the cannula. The animal is connected to fiberoptic cable that itself is connected to a laser or other light source. (B) Light-emitting diodes (LEDs) are useful for illuminating greater areas of neural tissue at once. Some LED implants do not need to be tethered to a power source at all, allowing an animal to freely move about its cage with no tether. 196  Guide to Research Techniques in Neuroscience Strategies for Studying Neural Circuits Using Optogenetics In a typical optogenetic experiment (typical does not mean easy), a scientist transduces a population of neurons with an optogenetic transgene and then investigates the effects of directly stimulating or inhibiting those neurons with light. In addition to stimulating the cell bodies of transduced neurons, it is also possible to stimulate fiber projections to distinct downstream target regions (Figure 8.5). The ability to selectively stimulate individual downstream projections from a neuronal population provides scientists with an unprecedented neural dissection tool. For example, if stimulating a population of neurons is sufficient to cause a phenotype, a scientist can test the hypothesis that this effect is mediated by a specific projection to a downstream brain region. The only caveat to these experiments is that stimulating fiber projections may cause antidromic effects that propagate action potentials up axon tracts back to neuronal cell bodies; therefore, scientists should perform careful controls to ensure that their stimulation effects are truly specific to the targeted projections of interest. FIGURE 8.5  Stimulation of projections using optogenetics. (A) Optogenetic transgenes can be delivered to a cell population of interest, and scientists can directly test the effects of stimulation or inhibition of the cell soma. Alternatively, scientists can test hypotheses about specific downstream pathways by implanting a fiber-optic cable above discrete projection fields (B, C). Manipulating Neural Activity Chapter | 8  197 NEUROMODULATION TECHNIQUES USED IN HUMANS The rapid development of cutting-edge neuromodulation tools begs the question about their applications in humans. The tools described in this chapter have obvious therapeutic potential in neurodegenerative diseases and neuropsychiatric disorders. Interestingly, they also have potential for “neuroenhancement”— improving human attention, memory, and cognitive abilities in those with already normal function. These applications in humans are highly controversial and hotly debated among neuroethicists. Which neuromodulation techniques have immediate potential for use in humans? Physical methods are still in use, primarily for trying to limit epileptic electrical activity from spreading by surgically removing the seizure-generating regions or disrupting the connections through which the activity spreads. Electrical stimulation is used to treat severe symptoms in intractable human diseases, including rigidity in Parkinson’s disease and serious cases of depression. Pharmacological techniques have been used over several decades to treat pain, depression, anxiety, attention-deficit disorders, and other neurological causes of human suffering. Most psychoactive agents were developed many years ago, and new pharmacological treatments for neuropsychiatric conditions are relatively rare. Chapter 1 described two methods of manipulating neural activity in humans without penetrating the skull. Transcranial magnetic stimulation (TMS) is capable of altering neural activity by applying a magnetic field outside the skull, inducing weak electric currents at a specific focal point on the surface of the brain. Depending on the brain region and the strength of the magnetic field impulse, the electrical activity may either stimulate neural activity or cause a hyperpolarized state in which neural activity is temporarily inactivated. This technique has a spatial resolution of millimeters but can only be applied to superficial neurons near the surface of the skull, such as cortical neurons. Another technique, ultrasonic neuromodulation (USNM), uses focused sound waves to heat the brain in a defined area. Like TMS, this technique is currently only efficient in superficial neurons and also has a spatial resolution of millimeters. Both of these techniques have been applied in clinical and experimental settings. Neuromodulation tools that require genetically encoded transgenes, such as optogenetics, have not been applied to humans, mostly due to the need for delivering DNA to human cells in vivo. The only gene delivery strategy that can efficiently target transgenes to human neurons is viral gene delivery (Chapter 11). Although commonly used adeno-associated virus and lentiviral vectors are relatively safe in animal models, scientists have been reasonably reluctant to inject these vectors into human subjects. Perhaps in the future, other methods of gene delivery will be possible so that cell-type specific modulation of neural activity in humans can become a possibility. CONCLUSION The past decade has seen a dramatic expansion in the toolkit used to manipulate neural activity in model organisms. There are multiple methods of stimulating (Table 8.2) and inhibiting (Table 8.3) neurons in a wide variety of model 198 GuidetoResearchTechniquesinNeuroscience TABLE 8.2  Methods of Stimulating Neural Activity Category Tool Invasiveness Spatial precision Temporal precision Electrical Microelectrode Surgical implantation Cannot distinguish between cell types or fibers of passage Milliseconds Pharmacological Various agonists and antagonists of membrane-bound receptors Intraperitoneal injection or surgical implantation of cannula Depends on specificity of receptor expression Effects can last minutes to hours Genetic Modulation of endogenous membrane-bound receptors Not invasive unless viral gene delivery Depends on specificity of receptor expression Permanent effects Pharmacogenetic Ligand-gated receptors (e.g., hM3Dq, TrpV1) Not invasive unless viral gene delivery; need to inject drugs intravenously Cell-type specificity Effects can last minutes in vitro or hours in vivo Optogenetic Channelrhodopsin-2 (ChR2) Surgical implantation; attachment to fiber-optic cable Cell-type specificity Milliseconds ManipulatingNeuralActivity Chapter|8 199 TABLE 8.3  Methods of Inhibiting Neural Activity Category Tool Invasiveness Spatial precision Temporal precision Electrical Microelectrode (causes electrolytic lesion) Surgery necessary Cannot distinguish between cell types or fibers of passage Permanent lesion Pharmacological Various agonists and antagonists of membrane-bound receptors Intraperitoneal injection or surgical implantation of cannula Depends on specificity of receptor expression Effects can last minutes to hours Genetic Modulation of endogenous membrane-bound receptors Not invasive unless viral gene delivery Depends on specificity of receptor expression Permanent effects Pharmacogenetic Ligand-gated receptors (e.g., hM4Di, AlstR) Not invasive unless viral gene delivery; need to inject drugs intravenously Cell-type specificity Effects can last minutes in vitro or hours in vivo Optogenetic Halorhodopsin (NpHR), Archaerhodopsin (Arch) Surgical implantation; attachment to fiber-optic cable Cell-type specificity Milliseconds 200  Guide to Research Techniques in Neuroscience organisms, each with their own advantages and disadvantages. The recent development of genetically encoded tools that can reversibly affect neural activity with unprecedented spatial and temporal resolution has allowed scientists to probe many previously intractable questions about the role of specific neural systems and circuits in behavior. SUGGESTED READING AND REFERENCES Books Knopfel, T., & Boyden, E. S. (2012). Optogenetics: Tools for controlling and monitoring neuronal activity. Progress in brain research (Vol. 196). Elsevier. Meyer, J. S., & Quenzer, L. F. (2013). Psychpharmacology: Drugs, the brain, and behavior (2nd ed.). Sinauer Associates, Inc. Nestler, E., Hyman, S., & Malenka, R. (2011). Molecular neuropharmacology: A foundation for clinical neuroscience (2nd ed.). McGraw Hill. Review Articles Cheney, P. D., Griffin, D. M., & Van Acker, G. M. III. (2013). Neural hijacking: action of highfrequency electrical stimulation on cortical circuits. Neuroscientist, 19, 434–441. Luo, L., Callaway, E. M., & Svoboda, K. (2008). Genetic dissection of neural circuits. Neuron, 57, 634–660. Sternson, S. M., & Roth, B. L. (2014). Chemogenetic tools to interrogate brain functions. Annual Review of Neuroscience, 37, 387–407. Warden, M. R., Cardin, J. A., & Deisseroth, K. (2014). Optical neural interfaces. Annual Review of Biomedical Engineering, 16, 103–129. Yizhar, O., Fenno, L. E., Davidson, T. J., Mogri, M., & Deisseroth, K. (2011). Optogenetics in neural systems. Neuron, 71, 9–34. Primary Research Articles—Interesting Examples from the Literature Adamantidis,A. R., Zhang, F.,Aravanis,A. M., Deisseroth, K., & de Lecea, L. (2007). Neural substrates of awakening probed with optogenetic control of hypocretin neurons. Nature, 450, 420–424. Alon, T., & Friedman, J. M. (2006). Late-onset leanness in mice with targeted ablation of melanin concentrating hormone neurons. Journal of Neuroscience, 26, 389–397. Carter, M. E., Soden, M. E., Zweifel, L. S., & Palmiter, R. D. (2013). Genetic identification of a neural circuit that suppresses appetite. Nature, 503, 111–114. Garner, A. R., Rowland, D. C., Hwang, S. Y., Baumgaertel, K., Roth, B. L., Kentros, C., et al. (2012). Generation of a synthetic memory trace. Science, 335, 1513–1516. Gradinaru, V., Mogri, M., Thompson, K. R., Henderson, J. M., & Deisseroth, K. (2009). Optical deconstruction of parkinsonian neural circuitry. Science, 324, 354–359. Lin, D., Boyle, M. P., Dollar, P., Lee, H., Lein, E. S., Perona, P., et al. (2011). Functional identification of an aggression locus in the mouse hypothalamus. Nature, 470, 221–226. Long, M. A., & Fee, M. S. (2008). Using temperature to analyse temporal dynamics in the songbird motor pathway. Nature, 456, 189–194. Luquet, S., Perez, F. A., Hnasko, T. S., & Palmiter, R. D. (2005). NPY/AgRP neurons are essential for feeding in adult mice but can be ablated in neonates. Science, 310, 683–685. Marella, S., Fischler, W., Kong, P., Asgarian, S., Rueckert, E., & Scott, K. (2006). Imaging taste responses in the fly brain reveals a functional map of taste category and behavior. Neuron, 49, 285–295. Manipulating Neural Activity Chapter | 8  201 Noudoost, B., & Moore, T. (2011). Control of visual cortical signals by prefrontal dopamine. Nature, 474, 372–375. O’Doherty, J. E., Lebedev, M. A., Ifft, P. J., Zhuang, K. Z., Shokur, S., Bleuler, H., et al. (2011). Active tactile exploration using a brain-machine-brain interface. Nature, 479, 228–231. Salzman, C. D., Britten, K. H., & Newsome, W. T. (1990). Cortical microstimulation influences perceptual judgements of motion direction. Nature, 346, 174–177. Winkowski, D. E., & Knudsen, E. I. (2006). Top-down gain control of the auditory space map by gaze control circuitry in the barn owl. Nature, 439, 336–339. Protocols Dong, S., Rogan, S. C., & Roth, B. L. (2010). Directed molecular evolution of DREADDs: a generic approach to creating next-generation RASSLs. Nature Protocols, 5, 561–573. Zhang, F., Gradinaru, V., Adamantidis, A. R., Durand, R., Airan, R. D., de Lecea, L., et al. (2010). Optogenetic interrogation of neural circuits: technology for probing mammalian brain structures. Nature Protocols, 5, 439–456. Websites DREADD wiki: http://pdspit3.mml.unc.edu/projects/dreadd/wiki/wikistart. Optogenetics resource center: http://web.stanford.edu/group/dlab/optogenetics/index.html. Optogenetics wiki: OpenOptogenetics.org. 203 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00009-5 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 9 Identifying Genes and Proteins of Interest After Reading This Chapter, You Should be Able to: l Understand basic concepts in genetics, such as transcription and translation l Compare and contrast the strengths and limitations of commonly used genetic model organisms l Describe methods for identifying genes and proteins important to a biological phenotype, including genetic, in silico, and molecular screening techniques Techniques covered: l Genetic screens: forward and reverse genetic screens l In silico screens: BLAST, Ensembl l Molecular screens: cDNA microarray, RNAi screen Virtually every cell in your body contains a copy of your genome, the complete set of genetic information embedded in DNA that is necessary for your development, metabolism, homeostasis, and plasticity. The functional unit of the genome is the gene, a segment of DNA that codes for a functional protein product. Proteins are the molecular machines responsible for virtually all of the cell’s structural and functional properties. To understand how the brain works, a neuroscientist must understand the anatomy and physiology of neurons—and to understand the anatomy and physiology of neurons, a neuroscientist must understand the genes and proteins that give rise to their properties. One of the overarching questions in neuroscience is how an animal’s genotype, the genetic constitution of an animal, determines that animal’s phenotype, an observable trait or set of traits. In other words, what genes influence a trait, such as proper development of an axon toward its target or an appropriate behavioral response following a stimulus? Researchers regularly publish a variety of studies either identifying a novel gene and demonstrating its important role in an animal’s phenotype or showing that a previously discovered gene also contributes to a separate phenotype. How do scientists identify genes that function as the molecular substrates of behavior? The purpose of this chapter is to describe methods useful for identifying which genes or proteins are important for a specific biological phenotype. These methods 204  Guide to Research Techniques in Neuroscience usually involve screening thousands of genes to identify just a few that contribute to a neurobiological phenomenon. In previous chapters, we examined techniques that generally study whole brains, behaviors, or the activity of cells. Starting with this chapter, we begin to study genetic and molecular neuroscience—the contribution of genes and molecules to development, physiology, and behavior. Therefore, we start the chapter with a brief introduction to studying genes and proteins, as well as a survey of genetically tractable model organisms. The second half of the chapter describes various methods for identifying genes or proteins of interest: using genetic screens, in silico screens, and molecular screens. Once these genes are identified, scientists can perform other experiments to further understand their contribution to a phenotype, as described in later chapters. A BRIEF REVIEW OF GENES AND PROTEINS The number of genetic and molecular techniques used in the literature can often be intimidating to neuroscientists without formal training in genetics or molecular biology. Fortunately, the background information necessary to understand these techniques and why they are used is not complex; it can be summarized in terms of the flow of information within each cell. Highly detailed descriptions of this information flow can be found in other textbooks. Here, we provide the essential information that a neuroscientist needs to know to understand genetic and molecular neuroscience, the subject of the remaining chapters in this book. The Central Dogma of Molecular Biology The central dogma of molecular biology is a model of the flow of information within a cell. Described by Francis Crick in 1958, this model posits that information in a nucleic acid, encoded by the sequence of specific DNA or RNA bases, can be transferred to another nucleic acid or to a protein, but information in a protein, encoded by the sequence of specific amino acids, is not transferred back to another protein or to a nucleic acid. Many of us learned the simplest version of the central dogma suggesting a unidirectional flow of information—-specific sequences of DNA code for molecules of RNA—and these RNA molecules, in turn, code for proteins (Figure 9.1). This model lays the foundation for all of molecular biology research. While scientists have elaborated on this model over the past 40 years and found more complex situations that contradict the simple version of the central dogma, such as the discovery of microRNAs and other RNA-based methods of DNA regulation, it continues to serve as the framework from which all molecular neuroscientists design and execute experiments. Deoxyribonucleic Acid DNA is the ultimate source of all genetic information within a cell. Residing within the cell nucleus, a DNA molecule is a long polymer made up of repeating units called nucleotides (Figure 9.2). These nucleotides are composed of a DNA is transcribed into mRNA mRNA leaves the nucleus mRNA is translated into protein DNA Nucleus mRNA mRNA translated by ribosome into amino acid polypeptide Protein FIGURE 9.1  The flow of information within a cell.  DNA resides in the nucleus and codes for relatively short sequences of mRNA. After transcription, an mRNA molecule leaves the nucleus for the cytoplasm, where it is translated by ribosomes into proteins. Thus, the molecular flow of information starts with nucleic acids and ends with amino acids. O OH OH O O O O O O O O O O OO O O O O O O O O OO O O O OO O O OO O O O O O O O O OO O O O OO O P P P P P P P P N N N N N N N N N N N N N N N N NN N N NH2 NH2 NH2 HN HN H2N H2N H2N NH NH _ _ _ _ _ _ _ _ _ _ Adenine Thymine Guanine Cytosine FIGURE 9.2  The molecular structure of DNA.  DNA is composed of a sugar-phosphate backbone on the outside and nitrogenous bases on the inside. The four nitrogenous bases form hydrogen bonds in the interior of the DNA molecule and hold the two complementary strands together. The two strands are oriented in opposite directions, with one strand running in the 5′ to 3′ direction and the other running in the 3′ to 5′ direction. 206  Guide to Research Techniques in Neuroscience phosphate group, a sugar molecule (2-deoxyribose—what gives DNA its name), and a nitrogenous base. The phosphate and sugar molecules form the backbone of a DNA polymer. There are four types of nitrogenous bases found in DNA that define the properties of the nucleotide: adenine (abbreviated A), thymine (T), guanine (G), and cytosine (C). In all eukaryotic organisms, DNA exists as a tightly associated pair of two long strands that intertwine, like two handrails of a spiral staircase, forming the shape of a double helix. The two strands of DNA are stabilized by hydrogen bonds between the nitrogenous bases attached to the two strands. A base on one strand forms a bond with only one type of base on the opposite strand: A only bonds with T, and C only bonds with G. The arrangement of two nucleotides binding together across the double helix is referred to as a base pair. Strands of DNA that form matches among base pairs are called complementary strands. Hydrogen bonds, unlike covalent bonds, are relatively weak and can be broken and rejoined relatively easily. This allows other enzymes to unwind DNA during DNA replication or gene transcription. In addition to being complementary, the two strands of DNA are also antiparallel, oriented in opposite directions. The directionality of a DNA strand is defined by the carbon atom in the 2-deoxyribose molecule that attaches to another phosphate molecule in a nucleotide monomer. One strand is said to be oriented in the 5′ (pronounced “five prime”) to 3′ (“three prime”) direction, while the other strand is oriented in the 3′ to 5′ direction (Figure 9.2). This orientation has important consequences in DNA synthesis, as DNA can only be synthesized in the 5′ to 3′ direction. This orientation also has consequences for RNA synthesis, as RNA can also only be synthesized in the 5′ to 3′ direction. Transcription Transcription is the synthesis of a messenger RNA (mRNA) molecule from a DNA template. There are three major differences between an RNA molecule and a DNA molecule: (1) RNA molecules use the sugar ribose instead of 2-deoxyribose; (2) RNA molecules are single-stranded; and (3) RNA molecules use the nitrogenous base uracil (U) instead of thymine. In transcription, only one of the two strands of DNA, called the template strand, is transcribed. The other strand, the coding strand, has a sequence that is the same as the newly created RNA transcript (with U substituted for T). An enzyme called RNA polymerase reads the template strand in the 3′ to 5′ direction and synthesizes the new mRNA in the 5′ to 3′ direction (Figure 9.3). Transcription begins when various proteins called transcription factors interact with regions of the genome called promoters. A promoter is a specific sequence of DNA that facilitates the transcription of a specific gene. These promoters are often located adjacent and upstream of the genes they regulate, toward the 5′ region of the coding strand. When a transcription factor binds to a promoter, multiple proteins are recruited to begin the transcription process. Identifying Genes and Proteins of Interest Chapter | 9  207 Coding strand Template strand Coding strand Template strand Coding strand Template strand Coding strand Template strand TF TF TF TF RNA Polymerase RNA Polymerase RNA Polymerase FIGURE 9.3  The transcription of DNA into RNA. (A) The process of transcription begins when specific transcription factors interact with promoter sequences on the genome. These factors recruit other proteins that are necessary to attract RNA polymerase. (B) RNA polymerase binds to the complex between transcription factors and DNA. Once bound, the enzyme begins prying the two strands of DNA apart. (C) RNA polymerase proceeds down the length of the DNA, simultaneously unwinding DNA and adding nucleotides to the 3′ end of the growing RNA molecule. (D) When transcription is complete, RNA polymerase dissociates from the DNA and the two DNA strands reform a double helix conformation. A new mRNA is produced that is complementary to the DNA template strand. 208  Guide to Research Techniques in Neuroscience After transcription terminates, a newly formed RNA molecule receives further processing in the form of RNA splicing before leaving the nucleus. At this point, the RNA is composed of exons and introns flanked by untranslated regions (UTRs) at either end (Figure 9.4). An exon is a sequence of an RNA strand that remains after splicing occurs, while an intron is a sequence that is lost. The 5′ and 3′ UTRs are sequences that cap the RNA molecule and are not translated into amino acids. RNA splicing is catalyzed by a macromolecular complex known as a spliceosome, which cuts exon and intron sequences apart and then joins the exons back together to form a single strand. After RNA splicing, a mature mRNA molecule leaves the nucleus and migrates to the cytoplasm, where it can be translated into a protein. Translation Translation is the process of using an mRNA molecule as a template to produce a protein. This process is performed by specialized macromolecules in the cytoplasm called ribosomes. A ribosome reads an mRNA molecule in the 5′ to 3′ direction and uses this template to guide the synthesis of a chain of amino acids to form a protein. Amino acids are the building blocks of all proteins and are assembled as a long strand, much like nucleotides are assembled as a long strand in a nucleic acid molecule. Each sequence of three nucleotides of an mRNA molecule code for one amino acid. The three nucleotides collectively form a codon, a translation of genetic information into an amino acid monomer. There is a precise genetic code for the conversion of codons into amino acids (Figure 9.5), with each codon precisely coding for one specific amino acid. One codon, AUG, codes for the amino acid methionine, which is the start of any protein sequence. Other codons (UAA, UAG, and UGA) serve as “stop” codons and terminate translation. After translation ends, the newly formed polypeptide chain folds upon itself to assume its native protein conformation. These proteins can FIGURE 9.4  RNA splicing. Before mRNA leaves the nucleus, macromolecular complexes called spliceosomes remove noncoding intron sequences and rejoin exon sequences. Noncoding sequences remain at the 5′ and 3′ ends. Identifying Genes and Proteins of Interest Chapter | 9  209 be further processed by other organelles, such as the endoplasmic reticulum, or receive posttranslational modifications from other proteins. This incredibly brief survey of DNA, transcription, and translation is no match for much more detailed descriptions found in genetics or molecular biology textbooks. However, this information is the foundation on which most molecular neuroscience techniques are based. GENETIC SCREENS A genetic screen is a process of identifying genes that contribute to a phenotype. Genetic screens fall into two categories: forward or reverse screens. A forward genetic screen is a screen used to identify genes important for a biological phenotype. In these screens, an investigator selects a scientifically interesting phenotype and seeks to determine which genes are necessary for the generation of this phenotype. A reverse genetic screen is the opposite: an investigator selects an interesting gene and seeks to determine which phenotypes result from its absence. Forward genetic screens often test thousands of genes at once, while a reverse genetic screen tests only a single gene. Forward genetic screens have been compared to fishing because a scientist is never quite sure which genes he or she will catch; reverse genetic screens have been compared to gambling because a scientist places all of his or her hopes on that gene producing an interesting phenotype. Forward genetic screens can be considered Stop Stop Stop U C A G U C A G U C A G U C A G U C A G U C A G Firstletter Second letter Thirdletter UUU UUC UUA UUG CUU CUC CUA CUG AUU AUC AUA AUG GUU GUC GUA GUG UCU UCC UCA UCG CCU CCC CCA CCG ACU ACC ACA ACG GCU GCC GCA GCG UAU UAC UAA UAG CAU CAC CAA CAG AAU AAC AAA AAG GAU GAC GAA GAG UGU UGC UGA UGG CGU CGC CGA CGG AGU AGC AGA AGG GGU GGC GGA GGG Phe Leu Ile Leu Met Val Ser Pro Thr Ala Tyr His Gln Asn Lys Asp Glu Cys Trp Arg Ser Arg Gly FIGURE 9.5  The translation of mRNA into proteins. (A) A ribosome translates the genetic sequence of an mRNA molecule into an amino acid sequence that forms a polypeptide. mRNA sequences are processed in the 5′ to 3′ direction. Each series of three nucleotides, called codons, translates into a single amino acid. (B) The genetic code for the conversion of codons into amino acids. Each codon translates into a specific amino acid (abbreviated here in three letters, such as Phe, which is the abbreviation for phenylalanine). Because there are more possible codons than amino acids, some codons translate into the same amino acids, such as UUU and UUC. The AUG codon translates into methionine and signals the start of translation. The UAA, UAG, and UGA codons do not translate into amino acids, instead serving as a stop signal and terminating translation. 210  Guide to Research Techniques in Neuroscience “hypothesis generating,” because they typically produce multiple candidate genes that may be important for a phenotype; reverse genetic screens can be considered “hypothesis testing,” because a scientist can test specific hypotheses about the role of a gene in a phenotype. Here, we briefly describe the process of performing a forward or reverse genetic screen. Forward Genetic Screen In a forward genetic screen, a scientist mutagenizes animals to produce thousands of lines that may have a mutation in a single gene, and then screens each individual to identify animals with an abnormal phenotype. These screens are typically performed in flies and worms, although they have also rarely been performed in mice (Box 9.1). There are six main steps in performing a forward genetic screen: 1. Design an assay to measure a phenotype. The first step in determining which genes may be important for a phenotype is to design a specific, quantitative assay to discriminate between wild-type individuals and animals with an aberrant phenotype. This requires fully characterizing the wild-type phenotype and choosing measurable parameters to identify individuals with an abnormal phenotype. For example, a scientist performing a forward genetic screen to determine genes important for axon guidance might choose a specific axon and characterize its normal growth and development over time. When screening individuals, the scientist could look for axons that mistarget to abnormal locations in the brain. Note that this screen requires the scientist to identify the same axon in thousands of individuals. Alternatively, a scientist could choose a behavioral phenotype, characterizing the normal behavioral response to a stimulus and then examining individuals with an abnormal behavior. The abnormal behavior must be statistically different from the normal variation in behavior between individuals. 2. Mutagenize eggs/larvae. There are three methods of developing mutant lines of organisms. In chemical mutagenesis, a scientist applies a mutagenizing chemical, such as ethyl methane sulfonate or N-ethyl-N-nitrosourea, to thousands of eggs/larvae, which statistically creates lines of animals with mutations in a single gene in the genome. In irradiation mutagenesis, a scientist achieves the same goal with high-intensity ultraviolet light, which damages DNA at a rate such that animals statistically have a mutation in a single gene in the genome. Finally, in insertional mutagenesis, a scientist uses various methods to insert mobile genetic elements called transposons into the genomes of offspring (Chapter 12). These transposons insert at random locations in the genome, occasionally disrupting an endogenous gene. Therefore, they can be used to test thousands of animals, each with a potential loss-of-function of a single gene. 3. Screen for abnormal phenotypes. After hundreds or thousands of mutant lines are produced, a scientist screens each individual in a phenotypic assay, Identifying Genes and Proteins of Interest Chapter | 9  211 Box 9.1  Genetic Model Organisms Hundreds of studies are regularly published in peer-reviewed journals that use genetic tools to investigate the nervous system. Interestingly, nearly all of these studies use the same five species as model organisms: the worm (Caenorhabditis elegans), the fly (Drosophila melanogaster), the zebrafish (Danio rerio), the mouse (Mus musculus), and the human (Homo sapiens). There are occasional exceptions, such as the frog (Xenopus laevis) or the sea slug (Aplysia californica), but these species collectively represent less than 1% of published studies. It is obvious why neuroscientists would want to learn more about the genes of humans, as the ultimate goal of neuroscience is to better understand our own species. But why did the other four species become such tractable organisms for genetic studies? Why use roundworms and fruit flies? Why not flatworms, bees, goldfish, or gerbils? Many of these other organisms have the same short breeding time, high number of offspring, and other factors that might make them attractive model species. The answer can essentially be summarized in a single word: history. There really is not a good reason why scientists started using mice instead of gerbils, but when multiple laboratories started performing genetic experiments on these animals, it made sense for other laboratories to follow. It is useful for multiple laboratories to study the same species: when dozens or hundreds of laboratories each study the same organism, a community develops that allows researchers to better share results, reagents, and genetic tools. There is no better example of a community forming around a genetic model organism than the community of scientists that study Drosophila. Ever since Thomas Hunt Morgan began using the fly as a model organism to study genetics in 1910, a growing number of scientists learned more and more about ideal conditions and practices for using flies in experiments. As the knowledge about fly maintenance increased, so did the number of genetic tools amenable to fly genetic studies. In the late 1960s, Seymour Benzer used Drosophila to study the link between genes and behavior. In the 1980s and 1990s, scientists developed dozens of molecular genetic tools for carefully dissecting genetic circuits in flies, and in 2000, the fly became the second eukaryotic organism to have its genome sequenced. A similar history developed for the worm: the prominent scientist Sydney Brenner began using Caenorhabditis elegans as a model organism in the 1960s. After years of study, scientists have given all 302 neurons specific names and mapped out the connections of all 7000 synapses. There is now an energetic C. elegans community just as there is a vibrant Drosophila community, with many established methods for maintaining and manipulating genetic strains of worms. Table 9.1 summarizes some of the traits and benefits of these commonly used genetic organisms. We mention these species not only because they are highly represented in the literature, but also because virtually all novel genes and proteins are discovered in these species. The methods described in this chapter to identify functionally relevant genes depend on the decades-long development of these organisms for biological research. 212  Guide to Research Techniques in Neuroscience one mutant at a time. This is easily the most time-consuming aspect of a forward genetic screen. Any mutant lines that show an abnormal phenotype are maintained and bred for future experiments. The vast majority of mutant lines that do not show aberrant phenotypes are either used for other genetic screens or discarded. TABLE 9.1  Relative Advantages and Disadvantages of Different Organisms Used for Genetic Studies Species Advantages Disadvantages Worm (Caenorhabditis elegans) Simple, multicellular organism Genome sequenced Short lifespan Relatively easy to manipulate genes and screen for genes of interest through mutagenesis Can be frozen All neurons and their connections are known Invertebrate Primitive nervous system Fruit fly (Drosophila melanogaster) Complex, multicellular organism Genome sequenced Short lifespan, rapid reproduction rate Relatively easy to manipulate genes Many mutant lines available Invertebrate Primitive nervous system Zebrafish (Danio rerio) Genome sequenced Can use powerful invertebrate techniques like Gal4/UAS system Good for imaging studies during development because eggs are clear Large number of offspring Not a mammal Mouse (Mus musculus) Complex, higher-order organisms Genome sequenced Possible to manipulate genes in whole organism or specific tissues Nervous system is homologous to humans Relatively long lifespan and reproductive time Genetic manipulations are lengthy and costly Greater genetic redundancy than invertebrates Human (Homo sapiens) Complex, higher-order organism Genome sequenced Can observe naturally occurring genetic mutations and polymorphisms Long lifespan Not tractable for experimental genetic manipulation Identifying Genes and Proteins of Interest Chapter | 9  213 4. Perform complementation analysis. Once a genetic mutant is isolated, investigators often perform a complementation test to determine if the mutation is unique or has been previously described. This approach is used when a mutation results in the same phenotype as a previously described mutation. The two strains could have mutations in the same gene or in two different genes. To differentiate between these two possibilities, the two strains are mated and the phenotypes of the offspring are identified. If the offspring exhibits the wild-type phenotype, a scientist concludes that each mutation is in a separate gene. However, if the offspring does not express the wild-type phenotype but rather a mutant phenotype, the scientist concludes that both mutations occur in the same gene. 5. Map the gene. Just because a scientist has discovered a mutation in a gene that alters a phenotype does not mean that the scientist knows the molecular identity of that gene. The only way to determine the molecular sequence of the gene and its location in the genome is to map the gene. If mutagenesis was caused by chemical or irradiation methods, investigators typically perform a linkage analysis to map the gene. This technique is used to identify the relative position of a novel gene in relation to the location of known genes. To map the gene, a scientist mates the mutant animals with animals of a separate mutant phenotype. The two traits are then assessed in the offspring to determine if they are inherited together at relatively high rates. When two traits are closely linked in future offspring, they appear together more frequently and there is a greater likelihood that they are located close to each other on a chromosome. Because many genomes have been sequenced, a scientist can identify potential regions on a chromosome where the mutant gene may be found. If mutagenesis was caused by insertional mutagenesis methods, it is much easier to identify the molecular identity of the mutated gene. A scientist simply uses the transposon’s genetic sequence to design primers for DNA sequencing, and then sequences DNA in either direction of the transposon to identify the region of the genome where it landed (Chapter 10). 6. Clone the gene. The final step in the process of identifying novel genes for a phenotype is to clone the DNA encoding the gene. This allows a scientist to perform future genetic experiments that manipulate the gene in vitro or in vivo. We will discuss methods of cloning genes in Chapter 10. Reverse Genetic Screen In a reverse genetic screen, a scientist identifies a gene of interest and then perturbs the gene to determine its role in various phenotypes. For example, if a scientist hypothesizes that a certain gene is necessary for proper formation of neuromuscular junctions in the mouse, the scientist could make a knockout mouse, removing the gene from the genome and examining whether proper synapses form between peripheral nerves and motor units. We describe the process of performing these kinds of experiments in Chapters 12 and 13. 214  Guide to Research Techniques in Neuroscience IN SILICO SCREENS Using modern technology, it is possible to use bioinformatics and genomics databases to identify genes and proteins that may be relevant for a phenotype of interest. An in silico screen is a computer-based method that identifies and compares similar DNA and/or protein sequences from multiple species. Suppose that a scientist identifies a gene encoding a novel transmembrane receptor. The scientist may be able to identify and discover other receptors by looking for similar sequences in the genome. Alternatively, if a gene is discovered in one species, a scientist can use bioinformatics to identify orthologous sequences in other species. These genes may have the same function or very different functions. Two examples of publicly available genetic databases are BLAST and Ensembl. BLAST The Basic Local Alignment Search Tool (BLAST) is a tool used to compare the nucleotide sequences of DNA or the amino acid sequences of proteins. The power of BLAST is the ability to compare a DNA or protein sequence with the entire library of known sequences from all species. The user can set a threshold for the degree to which two sequences form a match. Therefore, if a scientist wants to identify genes with high sequence similarity to a gene of interest, the BLAST program can identify similar genes with high sequence similarity, and potentially similar functions. Many genes in the mouse genome have been identified due to their similarity with other structures, especially neuropeptides and ion channels. The best way to learn about BLAST is to try it: http://blast.ncbi.nlm.nih.gov/Blast.cgi. Ensembl Ensembl is a bioinformatics browser that functions much like BLAST but with a greater emphasis on genomic analysis. This tool provides the actual genomic sequences for a huge diversity of animal models, allowing comparative genomic searches to be performed. Additionally, this program can be used for analyzing sequence alignment, to investigate gene homology, and to obtain specific gene sequences to be used for subsequent experiments. Ensembl has the complete genomes and rough drafts of genomes for dozens of animals, including standard model organisms, as well as relatively esoteric organisms, such as the bush baby, the bat, and the platypus. Again, the best way to learn about this genetic tool is to try it: http://www.ensembl.org/index.html. MOLECULAR SCREENS A molecular screen uses molecular biology techniques to identify genes important for a specific biological phenotype. Unlike standard genetic screens, molecular screens use high-throughput strategies and thousands of nucleic acid probes to identify the molecular components of a phenotype. Identifying Genes and Proteins of Interest Chapter | 9  215 cDNA Microarray Screen A microarray is a device used to measure and compare gene expression profiles from biological samples. It can be used to analyze gene expression from a single sample or to compare gene expression between two different samples. Because gene expression is tightly regulated, genes are only expressed at times when they serve a useful function. Therefore, a scientist can potentially identify a gene of interest based on how its expression changes over time or if its expression changes in response to a stimulus or environmental condition. The power of a microarray is that it is possible to assay all genes in the genome at the same time in a relatively small device that can fit in the palm of your hand. A microarray consists of an array of thousands of tiny microscopic spots of DNA sequences, often sequences that are parts of genes (Figure 9.6). A scientist takes a biological sample, extracts the RNA, and then uses a special enzyme called reverse transcriptase to make complementary DNA (cDNA) FIGURE 9.6  A cDNA microarray.  A microarray is used to analyze gene expression in a biological sample and compare expression patterns with other samples. mRNA is extracted from a sample and reverse transcribed to produce cDNA. The cDNA is labeled and then hybridized to a microarray chip containing thousands of potentially complementary strands. A computer detects the amount of signal from each specific sequence. Thus, a microarray can be used to detect which specific mRNA sequences were expressed in the original sample. Obtain sample Extract RNA Use reverse transcriptase to make cDNA Hybridize cDNA to microarray Analyze hybridization data 216  Guide to Research Techniques in Neuroscience from the RNA strands. The cDNA is labeled and incubated with the surface of the DNA array. The cDNA is allowed time to hybridize with the sequences on the array. A computer subsequently scans the entire array and detects hybridization between cDNA in the sample and spots on the array. Because the computer contains a database of every spot on the array, a scientist can learn which genes were present in a sample and which genes were undetected. A microarray can be used to identify novel genes for a neurobiological phenotype. For example, suppose a scientist wishes to identify genes that respond to light/dark cycles. The scientist can collect samples from a part of the brain known to play a role in regulating light/dark cycles, such as the suprachiasmatic nucleus. Some samples can be collected from animals during the day, while other samples are collected at night. A comparison in the gene expression profile between day/night genes will produce a list of genes that show differential expression during those two periods. Most microarray experiments usually result in the identification of dozens, if not hundreds of genes. Sometimes these lists can be overwhelming to an investigator, and much follow-up is required to reduce the list to a more manageable set of candidates for future experiments. For example, a scientist can repeat the experiment several times to reduce the rate of false-positive identification of genes, or more precisely control the environmental conditions during which the original RNA was extracted. RNAi Screen RNA interference (RNAi) is a technique used to knock down the expression of mRNA molecules (Chapter 13). RNAi screens use thousands of RNAi molecules to disrupt individual gene function in a variety of organisms, most notably worms, or cultured cells. For example, in a multiwell-based RNAi screen, hundreds of unique RNAi molecules are each added to the bottom of 384-well plates, with a separate RNAi molecule in each well. Cells are added to the wells and the scientist observes changes in various parameters, such as morphology, proliferation, or death rate. Thus, an investigator can determine which RNAi molecules, and therefore which genes, are necessary for a cellular process to occur. CONCLUSION This chapter describes the first steps in identifying new genes and their protein products. These techniques are often the beginning steps of discovering the role of a gene in a biological process or phenotype. Other methods are necessary to validate this discovery, such as methods that perturb the expression of the gene to identify its necessity and/or sufficiency for a phenotype. The techniques described in the remaining chapters use molecular and genetic methods to gain insight into the roles of newly discovered genes in vitro and in vivo. Identifying Genes and Proteins of Interest Chapter | 9  217 SUGGESTED READING AND REFERENCES Books Greenspan, R. J. (2004). Fly pushing: The theory and practice of drosophila genetics (2nd ed.). Cold Spring Harbor Laboratory Press. Griffiths, A. J. F., Wessler, S. R., & Carroll, S. B. (2010). Introduction to genetic analysis (10th ed.). New York: Freeman. Krebs, J. E., Goldstein, E. S., & Kilpatrick, S. T. (2012). Lewin’s genes IX (9th ed.). Sudbury, MA: Jones and Bartlett. Review Articles Goldowitz, D., Frankel, W. N., Takahashi, J. S., et al. (2004). Large-scale mutagenesis of the mouse to understand the genetic bases of nervous system structure and function. Brain Research. Molecular Brain Research, 132, 105–115. Medland, S. E., Jahanshad, N., Neale, B. M., & Thompson, P. M. (2014). Whole-genome analyses of whole-brain data: working within an expanded search space. Nature Neuroscience, 17, 791–880. Paddison, P. J., Silva, J. M., Conklin, D. S., et al. (2004). A resource for large-scale RNA-interferencebased screens in mammals. Nature, 428, 427–431. Ryder, E., & Russell, S. (2003). Transposable elements as tools for genomics and genetics in Drosophila. Briefings in Functional Genomics & Proteomics, 2, 57–71. Wu, S.,Ying, G., Wu, Q., & Capecchi, M. R. (2007). Toward simpler and faster genome-wide mutagenesis in mice. Nature Genetics, 39, 922–930. Primary Research Articles—Interesting Examples from the Literature Crick, F. H. (1970). Central dogma of molecular biology. Nature, 227, 561–563. Devineni, A. V., Eddison, M., & Heberlein, U. (2013). The novel gene tank, a tumor suppressor homolog, regulates ethanol sensitivity in Drosophila. Journal of Neuroscience, 33, 8134–8143. Husken, U., Stickney, H. L., Gestri, G., Bianco, I. H., et al. (2014). Tcf7l2 is required for left-right asymmetric differentiation of habenular neurons. Current Biology, 24, 2217–2227. Schuldiner, O., Berdnik, D., Levy, J. M., et al. (2008). piggyBac-based mosaic screen identifies a postmitotic function for cohesin in regulating developmental axon pruning. Developmental Cell, 14, 227–238. Tracey, W. D., Jr, Wilson, R. I., Laurent, G., & Benzer, S. (2003). Painless, a Drosophila gene essential for nociception. Cell, 113, 261–273. Ule, J., Ule, A., Spencer, J., et al. (2005). Nova regulates brain-specific splicing to shape the synapse. Nature Genetics, 37, 844–852. Walsh, T., McClellan, J. M., McCarthy, S. E., et al. (2008). Rare structural variants disrupt multiple genes in neurodevelopmental pathways in schizophrenia. Science, 320, 539–543. Zarbalis, K., May, S. R., Shen,Y., Ekker, M., Rubenstein, J. L., & Peterson, A. S. (2004). A focused and efficient genetic screening strategy in the mouse: identification of mutations that disrupt cortical development. PLoS Biology, 2, E219. Protocols Bökel, C. (2008). EMS screens: from mutagenesis to screening and mapping. Methods in Molecular Biology, 420, 119–138. Lehner, B., Tischler, J., & Fraser, A. G. (2006). RNAi screens in Caenorhabditis elegans in a 96-well liquid format and their application to the systematic identification of genetic interactions. Nature Protocols, 1, 1617–1620. 218  Guide to Research Techniques in Neuroscience Nolan, P. M., Kapfhamer, D., & Bucan, M. (1997). Random mutagenesis screen for dominant behavioral mutations in mice. Methods, 13, 379–395. Ramadan, N., Flockhart, I., Booker, M., Perrimon, N., & Mathey-Prevot, B. (2007). Design and implementation of high-throughput RNAi screens in cultured Drosophila cells. Nature Protocols, 2, 2245–2264. Websites BLAST: http://blast.ncbi.nlm.nih.gov/Blast.cgi. Ensembl: http://www.ensembl.org/index.html. 219 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00010-1 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 10 Molecular Cloning and Recombinant DNA Technology After Reading This Chapter, You Should be Able to: l Describe tools used to manipulate and recombine DNA l Explain how to create novel DNA constructs Techniques covered: l Isolation of DNA fragments: restriction enzymes, polymerase chain reaction (PCR), reverse transcription PCR (RT-PCR), quantitative real time PCR (qRT-PCR), gel electrophoresis l Cloning DNA fragments: vectors, ligation, transformation, DNA purification l Identifying DNA fragments: DNA sequencing, Southern blot, northern blot Modern molecular biologists manipulate DNA much like classic filmmakers edit movies. If a filmmaker wishes to take a scene from one film reel and add it to another film reel, he or she can locate the scene, physically cut it out, splice it together into the new film reel, and run the film to ensure that the sequence has integrated correctly. Likewise, a molecular biologist may want to cut out one DNA sequence from a specialized DNA storage system, called a vector, and splice it into a new vector. A scientist can do this in the same way as the filmmaker, with specific molecular tools that take the place of scissors, tape, and a sequence identification system. Identifying and cutting a piece of DNA out of the genome for future use is referred to as molecular cloning. Cloning allows the isolation and amplification of a specific DNA sequence. This sequence can then be manipulated using the tools of recombinant DNA technology to form novel sequences, mutate endogenous genes, tag genes with reporter constructs, and much more. In other words, modern molecular techniques allow a scientist to manipulate DNA in just about any way that he or she would like. In particular, aspects of two molecular cloning have revolutionized a scientist’s ability to manipulate DNA: (1) the ability to cut, alter, and join DNA molecules in vitro and (2) the ability to copy and mass produce recombinant DNA molecules in host/vector systems. 220  Guide to Research Techniques in Neuroscience The purpose of this chapter is to describe these tools and how they are used to clone and manipulate DNA constructs. First, we will discuss methods of isolating DNA fragments from the genome and other DNA vectors. Then we will describe molecular cloning: how these isolated fragments are inserted into storage vectors and mass produced using bacteria. Next, we describe how DNA fragments or recombinant DNA constructs are purified from other DNA molecules. Finally, we describe methods of reading the molecular sequence of a DNA construct. After creating and purifying a novel DNA construct, a scientist can deliver this construct into various cell types (Chapter 11) or use the construct to produce a transgenic or knockout animal (Chapters 12 and 13). ISOLATING DNA FRAGMENTS In Chapter 9, we discussed methods of identifying genes that could be important for a neural phenotype. Once a gene has been identified, isolating the DNA sequence that encodes the gene is critical for subsequent experiments and techniques used to manipulate and recombine the DNA in useful ways. There are two primary methods used to isolate DNA sequences from their location in the genome or a storage vector: (1) restriction digests that cut DNA into fragments and (2) polymerase chain reaction (PCR), which amplifies DNA sequences of interest. Restriction Enzymes The discovery of restriction enzymes (also called restriction endonucleases) that recognize and cut specific double-stranded DNA sequences is one of the most important advances in biology that makes recombinant DNA technology possible. Restriction enzymes are like targeted molecular scissors, cutting double-stranded DNA at unique base pair sequences called recognition (or restriction) sites. These enzymes are naturally produced by various strains of bacteria, which normally produce these enzymes to cleave foreign DNA molecules in their environment. Molecular biologists have identified and purified hundreds of unique restriction enzymes from many species of bacteria, and now these enzymes can be easily ordered from a variety of commercial manufacturers. Each enzyme recognizes a unique sequence of about 4–8 nucleotides and cuts at specific locations within these recognition sites in a process called a restriction digest. Some restriction enzymes digest DNA in a way that produces blunt ends on the newly cut strands (Figure 10.1A). Others produce staggered cuts that leave sticky or cohesive ends with short single-stranded tails hanging over the ends of each fragment (Figure 10.1B). Each tail can form complementary base pairs with the tail at another end produced by the same enzyme. This allows any two DNA fragments to be easily joined together, as long as the fragments were generated with the same restriction enzyme (or another enzyme that leaves the same sticky ends). These different ends can be thought of as puzzle pieces whose shape is determined by the particular enzyme Molecular Cloning and Recombinant DNA Technology Chapter | 10  221 used. Therefore, if the same enzyme is used to cut two different pieces of DNA, then the resulting pieces will fit with each other because they have the same corresponding sticky ends that complement each other. Polymerase Chain Reaction PCR is a biochemical reaction that uses controlled heating and cooling in the presence of DNA synthesizing enzymes to exponentially amplify a small DNA fragment. It requires only tiny amounts of DNA to be used as a template from which a selected region will be amplified a billion-fold, effectively purifying this DNA fragment away from the rest of the template DNA. Short single-stranded oligonucleotides (about 20–30 nucleotides in length) called primers are essential ingredients in PCR. Primers specify what fragment will be amplified by binding to the template in the region flanking the sequence of interest (Figure 10.2). Where do these primers come from? They are synthesized in vitro using what is called the “solid-phase” method. In this method, short DNA strands are synthesized by adding activated monomers to a growing DNA chain linked to an insoluble support. DNA chains of as many as 100 nucleotides can be synthesized using this automated method. This type of chemical DNA synthesis can also be used to create labeled oligonucleotides for nucleic acid hybridization techniques. DNA sequences can also be modified using PCR either by adding restriction sites to the ends of a DNA fragment or by introducing mutations in the sequence to characterize the function of the DNA fragment. Laboratories do not often chemically synthesize DNA fragments themselves but rather order them from a company or core facility at their institution that specializes in oligonucleotide synthesis. Standard PCR Standard PCR starts with a DNA template, often from a genomic or complementary DNA (cDNA) library (Box 10.1), and uses primers to selectively amplify a specific DNA fragment of interest. The ingredients of a PCR reaction include a DNA template containing the target sequence to be amplified; a pair of oligonucleotide primers that hybridize with the sequences flanking the target sequence; all four deoxyribonucleoside triphosphates (dNTPs—A, T, G, C); FIGURE 10.1  Restriction enzymes. Acting like molecular scissors, restriction enzymes recognize specific sequences and cut at specific locations within those sequences. They can either (A) leave blunt ends, such as the enzymes SmaI and HpaI or (B) make uneven cuts that result in sticky ends, such as the enzymes EcoRI and HindIII. 222  Guide to Research Techniques in Neuroscience and a heat-stable DNA polymerase, an enzyme that adds nucleotides onto the end of a forming DNA strand. A PCR cycle consists of three steps (Figure 10.2): 1. Strand separation. The two strands of the template DNA molecule are separated by heating the solution to 95 °C for about 15 s so that each strand can serve as a template. 2. Hybridization of primers. The solution is abruptly cooled to about 50–55 °C to allow each primer to hybridize to a DNA strand. One primer hybridizes FIGURE 10.2  Polymerase chain reaction (PCR).  Double-stranded DNA is heated to separate the strands. The temperature is lowered to approximately 50–55 °C to allow primers to hybridize to their complementary sequences on the separated DNA strands. The temperature is then raised to approximately 70–75 °C so that Taq DNA polymerase can add nucleotides to the growing 3′ end of the primer. This cycle is repeated 20–30 times until the target DNA sequence between the primers is greatly amplified. Molecular Cloning and Recombinant DNA Technology Chapter | 10  223 BOX 10.1  DNA Libraries A DNA library is a comprehensive collection of cloned DNA fragments from a cell, tissue, or organism. DNA libraries can be used to isolate a specific gene of interest, as they generally include at least one fragment that contains the gene. Libraries can be constructed using any of a variety of different vectors (plasmid, phage, cosmid, or artificial chromosomes) and are generally housed in a population of bacterial cells. Nowadays, it is rare for a single scientist or laboratory to produce a DNA library. Instead, if a scientist wants to isolate a particular genomic sequence, he or she can order a vector from a DNA library repository. Plasmid DNA libraries are created using the same process used to clone genes into expression vectors (Figure 10.5). Plasmid and cellular DNA are both cut with restriction enzymes and annealed via their sticky ends to form recombinant DNA constructs. These recombinant molecules containing foreign DNA inserts are then covalently sealed with DNA ligase, and bacteria are transformed using these recombinant DNA plasmids. Each bacterial cell that was initially transformed contains, in theory, a different foreign DNA insert, replicated in all the clones of the bacterial colony. For many years, plasmids were used to clone DNA fragments of 1–30 kb. Larger DNA fragments are more difficult to handle and thus were harder to clone. Then researchers began to use artificial chromosomes—larger DNA vectors, such as yeast artificial chromosomes (YACs), which could handle very large pieces of DNA (100-1000 kb). Today, new plasmid vectors based on the naturally occurring F plasmid of the bacteria Escherichia coli are used to clone DNA fragments of 150–350 kb. Unlike smaller bacterial plasmids, the F plasmid—and its derivative, the bacterial artificial chromosome (BAC)—is present in only one or two copies per E. coli cell. Because of their stability, ability to accept large DNA inserts, and ease of handling, BACs have been the preferred vector for building DNA libraries of complex organisms—including those representing the human and mouse genomes. Genomic DNA Libraries Cleaving the entire genome of a cell with a specific restriction enzyme and cloning each fragment is sometimes called the “shotgun” approach to gene cloning. This technique can produce a very large number of DNA fragments that will generate millions of different colonies of transformed bacterial cells. Each of these colonies is composed of bacteria derived from a single ancestor cell and therefore contains a particular stretch of the fragmented genome. Such a plasmid is said to contain a genomic DNA clone, and the entire collection of plasmids is called a genomic DNA library. Because the genomic DNA is cut into fragments at random, only some fragments contain genes that encode functional protein products. Many of the genomic DNA clones obtained from mammalian DNA samples contain only noncoding DNA, which makes up most of the DNA in such genomes. By making genomic libraries based on mRNA rather than DNA, investigators can produce cDNA libraries that contain only the DNA sequences used to produce proteins. cDNA Libraries By starting with mRNA rather than genomic DNA, only DNA sequences that correspond to protein-encoding genes are cloned. These libraries are created by extracting mRNA from cells and then using RT-PCR to make a cDNA copy of each mRNA Continued 224  Guide to Research Techniques in Neuroscience to the 3′-end of the target on one strand, and the other primer hybridizes to the 3′-end on the complementary target strand so that DNA extension from the primers will be directed toward each other. This amplifies the region between the primers. Parent DNA duplexes do not form because the primers are present in great excess. 3. DNA synthesis. The solution is heated to about 70–75 °C, which is the optimal temperature for Taq DNA polymerase. This heat-stable polymerase is derived from Themus aquaticus, a thermophilic bacterium found naturally in hot springs. The polymerase elongates both primers in the direction of the target sequence because DNA synthesis is in the 5′ to 3′ direction; new dNTPs are added to the growing 3′ end of each primer. DNA synthesis takes place on both strands and extends so that the target DNA sequence is fully extended to the length between each of the two primers. These three steps—strand separation, hybridization of primers, and DNA synthesis—constitute one cycle of PCR amplification. A standard PCR protocol consists of 25–30 cycles, and each cycle effectively doubles the amount of the target DNA from the previous cycle. A single cycle requires only about 2–3 min (depending on the length of the PCR template to be amplified), and the entire procedure can be easily automated using modern PCR machines, also known as thermocyclers. The end result of a PCR reaction is a solution in which the predominant DNA species are several million copies of the target DNA sequence dictated by the position of the two primers. molecule present, and inserting this cDNA into a vector. Each clone obtained in this way is called a cDNA clone, and the entire collection of clones derived from one mRNA preparation constitutes a cDNA library. There are important differences between genomic DNA clones and cDNA clones. Genomic clones represent a random sample of all the DNA sequences in an organism and, with very rare exceptions, are the same regardless of the cell type used to prepare them. In contrast, cDNA clones contain only those regions of the genome that have been transcribed into mRNA. Because cells of different tissues produce distinct sets of mRNA molecules, a distinct cDNA library is obtained for each cell type used to prepare the library. For example, a cDNA library derived from mouse neurons will be different than a cDNA library derived from mouse fibroblasts. A major advantage to using a cDNA library over a standard genomic DNA library is that a cDNA library contains only the uninterrupted coding sequence of a gene—only the exons that code for proteins, free from noncoding intron sequences. Thus, when the aim of cloning is either to deduce the amino acid sequence of a protein from the DNA sequence or to produce a protein in high quantities by expressing the cloned gene in a bacterial or yeast cell, it is preferable to start with cDNA. However, genomic DNA libraries are useful for examination of noncoding regions in a genome. BOX 10.1  DNA Libraries—cont’d Molecular Cloning and Recombinant DNA Technology Chapter | 10  225 Reverse Transcription PCR Reverse transcription PCR (RT-PCR) uses mRNA rather than DNA as the starting template. First, the enzyme reverse transcriptase uses the mRNA template to produce a complementary single-stranded DNA strand called cDNA in a process known as reverse transcription. Next, DNA polymerase is used to convert the single-stranded cDNA into double-stranded DNA. These DNA molecules can now be used as templates for a PCR reaction as described above. The value of RT-PCR is that it can be used to determine if an mRNA species is present in a sample or to clone a cDNA sequence for a subsequent experiment. Quantitative Real-Time PCR To compare the relative amount of mRNA or DNA present in different samples, a scientist can employ a modified version of PCR called quantitative realtime PCR (qRT-PCR, not to be confused with RT-PCR described above). This technique is based on detecting a fluorescent signal produced in proportion to the PCR product during each cycle of DNA amplification. There are multiple methods of producing a fluorescent signal during a qRT-PCR reaction. In the TaqMan® assay (Figure 10.3), a fluorogenic probe within the region to be amplified contains a fluorescent reporter tag on one end and a fluorescence quencher on the other end. During each cycle of amplification, DNA polymerase cleaves the fluorescent probe molecule, removing the quencher and allowing the reporter to emit fluorescence. The increase in fluorescence is proportional to the amplification during the exponential phase of amplification. An alternate method uses a fluorescent dye called SYBR Green® that binds to double-stranded DNA; the resulting DNA-dye complex is capable of absorbing blue light and emitting green light, indicating the concentration of amplified double-stranded DNA. A sensitive probe within the PCR chamber can measure and record the fluorescent signal after each PCR cycle and produce a real-time measure of the concentration of double-stranded amplified DNA fragments. Thus, a scientist can compare the relative concentration between different samples. Nucleic acid hybridization techniques such as Southern (DNA) or northern (RNA) blots (discussed later in the chapter) are alternative methods to DNA synthesis displaces the probe Probe is cleaved by DNA polymerase Quenched fluorogenic probe binds template FIGURE 10.3  Quantitative real-time PCR (qRT-PCR) using a TaqMan® assay.  A probe containing a fluorescent reporter, R, is initially fused to a quencher, Q. During a PCR reaction, DNA synthesis of the growing strand cleaves the probe, causing removal of the quencher and increasing the fluorescent signal. Each round of amplification produces fluorescence that can be quantified to determine the amount of nucleic acid template present in the sample. 226  Guide to Research Techniques in Neuroscience quantitatively compare amounts of DNA or RNA present in samples. However, these methods are not as quantitative as qRT-PCR and typically require large concentrations of starting material. Modifying DNA Sequences Using PCR Although the primary purpose of PCR is to make exact copies of a DNA sequence, this technique can also be used to modify a DNA sequence for subsequent experiments. For example, a scientist may want to introduce a point mutation into a sequence of DNA or add base pairs to a sequence to introduce restriction sites. To modify DNA sequences using PCR, scientists design primers that not only specify the DNA fragment to be amplified, but also add or alter base pairs within the primer region itself. Although the primers may not hybridize with the template DNA strands as tightly for the first few PCR cycles, with each successive cycle, the added or mutated sequences in the primers will be copied and incorporated into the amplified PCR product. Isolation and Characterization of DNA Fragments Using Gel Electrophoresis Once DNA fragments have been generated using restriction enzymes or PCR amplification, they must be purified and separated from other DNA fragments that are not of interest. The separation and identification of DNA fragments based on their size is possible using a ubiquitous tool called gel electrophoresis. Gel electrophoresis is used to isolate, identify, and characterize properties of DNA fragments (Figure 10.4). First, a scientist uses molds to make an agarose gel with wells at one end for placing DNA samples. The gel is placed in an electrophoresis chamber and DNA samples are added to the wells. The electrophoresis apparatus produces a small electrical field, driving negatively charged DNA strands away from the cathode (the negative end) and toward the anode (the positive end). The mobility of the DNA fragments through the gel is inversely proportional to the logarithm of the number of base pairs in the fragments. Therefore, smaller DNA fragments will migrate much more quickly through the gel than large fragments. One of the lanes of the gel typically contains a DNA ladder—DNA that has been previously digested and characterized, producing fragments with known base pair sizes. Therefore, it is possible to determine the sizes of the fragments in the DNA samples by comparing their location to the location of the fragments in the ladder. The DNA bands on agarose gels are invisible unless the DNA is labeled or stained. A common method of staining agarose gels is briefly incubating them in a solution containing ethidium bromide, a DNA intercalating agent that fluoresces under ultraviolet light when bound to DNA. After identifying DNA Molecular Cloning and Recombinant DNA Technology Chapter | 10  227 fragments in a gel, it is possible to physically cut the DNA fragments out of the gel and purify the DNA from the agarose using simple, commercially available kits and protocols. After purifying the DNA, a scientist can clone the fragment into a storage vector. CLONING DNA Once a DNA fragment is isolated using restriction enzymes or PCR techniques, it is necessary to “store” the fragment into a vector for future use and amplification. The process of isolating a piece of DNA, ligating it into a storage vector, and producing enough copies of the vector for subsequent experiments is called molecular cloning (Box 10.2). The process of transferring a DNA sequence that already exists within a vector to a new vector is called subcloning. Vectors A DNA vector is a nucleic acid storage system that can hold an isolated DNA sequence of interest. The essential features of a vector are that it can replicate autonomously in a host species, usually bacteria, and that it can be combined with other pieces of DNA. Often, the inserted DNA fragment comes from a different organism than the vector DNA, and the fusion is called a recombinant DNA construct. Two commonly used vectors are plasmids, naturally occurring circles of DNA that act as independently replicating accessory chromosomes in bacteria, and bacteriophage or phage, a virus that can deliver its genetic cargo to bacteria. It is much more common for scientists performing recombinant DNA techniques in the laboratory to use plasmids over phage vectors. Many BOX 10.2 The Definition of Clone The word clone is used often in molecular biology. Unfortunately, this word has distinct definitions that can often cause confusion. In general, the precise definition of a clone is an exact copy. In reference to a piece of DNA, this term implies that a scientist will make exact copies of a DNA sequence. In reference to a bacterial cell, this term implies that a bacterium will multiply to eventually produce millions of genetically identical bacteria, each harboring the same DNA as the initial cell. Nowadays, it is possible to produce clones of larger organisms, including mammals—meaning that the animals share identical copies of genomic DNA. The term “cloning a gene” usually refers to the process of isolating a particular DNA sequence, placing the sequence in a vector, and producing many copies of the vector for use in subsequent experiments. For example, if a scientist identifies a novel gene after performing a screen (Chapter 9), he or she could clone the gene using PCR techniques, placing the gene in a vector, and using recombinant DNA techniques to modify the gene in subsequent experiments. 228  Guide to Research Techniques in Neuroscience plasmids have been ingeniously modified to enhance the delivery of recombinant DNA molecules into bacteria and to facilitate the selection of bacteria harboring these vectors. Because there is a size limit to the number of base pairs that can be stably inserted into a plasmid or phage vector (typically 1–25 kb), scientists sometimes use other vectors that can hold larger amounts of DNA. A cosmid is a plasmid that contains phage sequences that allow the vector to be packaged and transmitted to bacteria like a phage vector. Cosmids are more stable than regular plasmids, and therefore they can hold larger inserts (30–50 kb). Artificial chromosomes containing telomeric and centromeric regions can hold even larger inserts. Because cosmids and artificial chromosomes can stably hold large DNA inserts, they are often used to create DNA libraries (Box 10.1) Common Vector Features Useful vectors contain three main features: l An origin of replication. An origin of replication is a specific DNA sequence at which DNA replication is initiated. The origin is therefore absolutely essential for the amplification of the vector inside a bacterial host. l Useful restriction enzyme sites. Restriction sites are needed to place a DNA sequence of interest into a vector. Because vectors are designed to be easy storage systems for cloned DNA, a standard feature in many vectors is the presence of a multiple cloning site (MCS), a region with multiple useful restriction enzyme sites to make a compatible digest of the vector and DNA fragment more feasible. Both the vector and the DNA fragment are cut using the same restriction enzymes so that sticky ends from the DNA fragment match the sticky ends on the vector (Figure 10.5). The two molecules can then be stitched together through DNA ligation. The recombinant DNA molecule comprised of the vector and the DNA insert is also known as a DNA construct. l A selectable marker.A selectable marker is a method of allowing bacteria containing the vector to be readily identified and purified. Plasmids often confer antibiotic resistance to the bacteria, so only bacteria containing the vector will survive treatment with antibiotics. Common antibiotics used in the laboratory include ampicillin and kanamycin, and many plasmids contain genes that confer ampicillin and kanamycin resistance. Types of Vectors Cloning vectors provide a backbone for the DNA insert to be reproduced and propagated in bacteria; however, these vectors are only useful for storing a genetic sequence. By themselves, they are incapable of allowing for transcription and translation of the gene into a functional protein product. For a gene to give rise to a protein product, an expression vector must be used that contains the necessary elements for a host cell to transcribe and translate the gene. In the case of a mammalian cell, a standard mammalian expression Molecular Cloning and Recombinant DNA Technology Chapter | 10  229 vector will contain an origin of replication, MCS, and selectable marker, as described above. However, the expression vector will also need a promoter found in mammalian cells that can drive the expression of the gene. The coding DNA needs other features to be transcribed and translated, such as the polyadenylation tail that normally appears at the end of transcribed pre-mRNA and a sequence that attracts the ribosome for translation. Ligation To place a DNA fragment into a vector, both the fragment and vector are usually cut using the same restriction enzymes so that they have the same sticky end fragments. The fragment can then be “pasted” into the vector using DNA ligase, an enzyme normally used by cells for DNA repair (Figure 10.5). It is also possible to ligate DNA fragments and vectors together from restriction enzyme reactions that produce blunt end fragments. However, because the blunt ends can join in multiple orientations and configurations, there is a high chance of ligation in the wrong orientation, or even the potential for multiple DNA fragments to insert into the same vector backbone. Some vectors, such as FIGURE 10.4  Gel electrophoresis. (A) DNA samples are loaded into individual wells of an agarose gel. (B) When an electric current is passed through the gel, negatively charged DNA travels away from the cathode (−) and toward the anode (+) in a size-dependent manner. (C) A DNA ladder (left side) of DNA fragments of known sizes is usually run adjacent to the samples to identify the sizes of the sample fragments. 230  Guide to Research Techniques in Neuroscience bacterial artificial chromosomes (BACs, Box 10.1), are so large that it is nearly impossible to use standard restriction enzymes and DNA ligase protocols to produce recombinant DNA. An alternative to this molecular cutting and pasting is recombineering (Box 10.3). The vector and DNA fragment to be inserted are cut with the same restriction enzymes The “sticky” ends of the linearized vector and the DNA insert are complementary DNA ligase joins the complementary ends of the vector and insert together Competent bacteria are transformed with the recombinant DNA vector Successfully transformed bacteria have resistance to an antibiotic grown on the culture plate. These bacteria contain recombinant DNA molecules FIGURE 10.5  Steps in making a recombinant DNA plasmid. Molecular Cloning and Recombinant DNA Technology Chapter | 10  231 Transformation After a DNA fragment and vector are ligated together, the construct is introduced into competent cells, which are host cells (usually bacteria) that have been made transiently permeable to DNA (Figure 10.5). The process of introducing a DNA vector into competent cells is called transformation. Bacteria can be used to grow large quantities of a single recombinant DNA molecule because they divide and grow exponentially, doubling in number every 30 min. When the host cell divides, the recombinant plasmids also replicate such that a high concentration of the desired DNA will be produced. After transformation, the bacteria are usually grown on agar plates with an antibiotic such that only bacteria successfully carrying the vector (containing an antibiotic resistance gene) survive. Purifying DNA from Host Cells After transforming bacterial host cells to grow many copies of DNA, the recombinant DNA plasmid must be purified from the host cell DNA and proteins.A scientist typically takes a sample of a bacterial colony from an agar plate and grows the bacteria in fresh culture media so that the bacteria continue to divide and produce many copies of the desired DNA. Depending on the volume of bacterial culture, different amounts of DNA can be purified, ranging from micrograms to grams. A scientist performs a routine protocol to lyse the bacteria, extract the DNA, and separate plasmid DNA from host cell DNA. In fact, many kits are now available from commercial manufacturers that greatly simplify the process of purifying DNA from transformed bacteria. This procedure is known as the miniprep (for small amounts derived from 1 to 3 ml starting bacterial culture), maxiprep (for 200–300 ml starting culture), or gigaprep (for 1–2 l starting culture). IDENTIFYING DNA After a complete cloning or subcloning experiment (Box 10.4), it is a good idea to verify that the recombinant DNA product turned out as planned, typically by BOX 10.3  Recombineering Recombineering (“recombination-mediated genetic engineering” or “recombinogenic engineering”) is a method used to create constructs without the use of restriction enzymes and DNA ligase. This process uses a bacterium’s natural process of homologous recombination (Chapter 13) to recombine DNA without the “cutting and pasting” of traditional molecular biology techniques. These techniques can be useful when it is difficult to find unique restriction enzyme sites in cloning vectors or genomic DNA. Furthermore, recombineering allows for recombinant DNA products using large cosmids or BACs, for which finding unique restriction enzyme sites would be nearly impossible. 232  Guide to Research Techniques in Neuroscience BOX 10.4  Walkthrough of a Subcloning Experiment Imagine that you have identified an interesting gene called devX from a microarray experiment (Chapter 9) to screen for genes that decrease over the course of neural development. You identified the sequence of this gene and know that it is highly expressed in neurons.You would like to examine the subcellular trafficking of devX in neurons using time-lapse imaging, so you want to produce a reporter construct (Chapter 7) by tagging devX with green fluorescent protein (GFP) to produce devXGFP. Once created, this DNA construct can be introduced into cells or animals using the transfection methods described in Chapter 11. To create your devX-GFP sequence, you must first choose an appropriate vector to build your construct and determine the best restriction enzymes to use. Because you want to create a GFP-fusion construct, you could use a commercially available GFP-containing plasmid vector called pEGFP-N1. This vector contains a promoter that constitutively drives gene expression (called CMV); the eGFP (enhanced green fluorescent protein) coding sequence; a gene that confers antibiotic resistance to kanamycin; and a MCS, all contained within a circular plasmid. The MCS is located in front of the eGFP coding sequence, meaning the devX coding sequence will be attached to the N-terminus of eGFP (this is what the N1 in the vector’s name refers to). An MCS contains many unique restriction sites to choose from, so compatible restriction enzymes can be used on both the vector and the insert. The primary factor to consider when choosing the appropriate restriction enzyme to use is whether the site is unique. You obviously do not want to cut your DNA of interest in unexpected or undesired areas. Because you will be using the same restriction enzyme to open up the vector that you will be using to generate the ends of your insert, you must consider the sequences of both items. After using software that tells you where various restriction sites are located in your DNA plasmid, you end up choosing the common enzymes EcoRI and BamHI, because devX does not contain those recognition sites within its coding sequence. The next major step is to generate a devX fragment with the appropriate restriction sites to prepare the insert to go into the vector. There are multiple ways to generate the small fragment of DNA that represents devX. If you already had a vector containing the segment of devX, you might be able to directly cut it out of this vector to paste into a new vector. Alternatively, you could use PCR to amplify the appropriate fragment from a cDNA template, adding the recognition sequences for EcoRI and BamHI to the ends of the devX fragment. After choosing a vector and preparing the insert, you will need to make them compatible by cutting them with the same restriction enzymes. You need to insert the devX fragment in the proper direction and the same reading frame as the eGFP so that they form a single protein together. You have done this by choosing two different restriction enzymes (EcoRI and BamHI) that leave “sticky” ends after digestion and by making sure that the frame of the devX PCR product after digestion is the same as the eGFP coding sequence within the vector. To isolate your digested products, you run the digested vector and insert on an agarose gel. You should be able to identify the appropriate bands based on their size. Then, you can physically cut out these bands and purify the DNA from the gel using simple gel purification protocols or kits. Molecular Cloning and Recombinant DNA Technology Chapter | 10  233 sequencing the DNA product. DNA sequencing identifies the individual base pairs that make up the DNA strand. Another common method of DNA identification is to use nucleic acid hybridization techniques to probe whether a specific sequence is present in a sample. DNA Sequencing DNA sequencing is used to determine the exact sequence of nucleotides (A, G, C, T) in a strand of DNA. The Sanger dideoxy chain-termination method can determine the sequence of nucleotides with high fidelity for a stretch of approximately 200–500 base pairs in any purified DNA sample. Based on in vitro DNA synthesis, the Sanger method synthesizes short pieces of DNA in the presence of nucleotide bases to which other bases cannot be added: chain-terminating dideoxyribonuceoside triphosphates. These chainterminating nucleotides are labeled and mixed with regular nucleotide bases so that fragments of DNA will be created at many different lengths, each You mix the purified digested products with DNA ligase, which will ligate the fragments together. The EcoRI digest on the 5′ end of the devX insert matches the EcoRI digest on the 3′ end of the plasmid, while the BamHI digest on the 3′ end of the devX insert matches the BamHI digest on the 5′ end of the plasmid. Thus, the devX sequence should be inserted in the proper direction into the plasmid. Next, you can transform competent bacteria to grow many copies of your recombinant DNA plasmid. You mix the ligation product with competent cells and then spread the cells on plates containing nutrients—to help cells grow—and kanamycin, the antibiotic that cells containing your construct should be resistant against. The next morning, you should see clonal colonies of bacteria growing on the plate, and you can select individual colonies whose member cells contain identical copies of the construct. If you place a bit of the colony in a new tube with bacterial culture medium, you can allow the bacteria to multiply over a period of 8–12 h. Then, you can purify the plasmid DNA from each of the clones using a “miniprep” kit. You lyse the bacteria, extract the DNA from the lysed bacteria, and purify the plasmid DNA from the bacterial DNA using purification columns. The purified plasmid DNA can be examined using restriction enzyme analysis. Using restriction digests, you can ensure the presence of the insert, whether the insert was ligated in the correct direction, and whether there is more than one insert present.These questions can be answered by choosing appropriate restriction enzymes, which will not necessarily be the ones used to join the insert and vector. Finally, you verify that the sequence in your construct matches the DNA sequence of devX-GFP by performing sequencing reactions (Figure 10.6). Now that you have produced and verified your devX-GFP construct, you can introduce your vector into cells using techniques that will be described in the next chapter. BOX 10.4  Walkthrough of a Subcloning Experiment—cont’d 234  Guide to Research Techniques in Neuroscience randomly stopped by the addition of a chain-terminating nucleotide. The four chain-terminating nucleotides are each labeled with a different colored fluorescent dye. Thousands of fragments of different lengths are run on a gel, and an automated fluorescence detector can quickly scan the gel to read the identity of the last, terminating nucleotide (Figure 10.6). Modern molecular biology laboratories tend not to run sequencing reactions themselves, as it is typically faster and less expensive to send DNA samples off to a dedicated facility for sequencing. The Sanger method is the gold standard for DNA sequencing, and it is still commonly used for the day-to-day verification of molecular biology experiments. However, DNA sequencing methods have evolved rapidly over the past decade in what are referred to as next-generation sequencing (NGS) techniques, which make sequencing large stretches of DNA cheaper and faster. NGS techniques have enabled large-scale sequencing of entire genomes and allowed for the rapid development of the field of genomics. Nucleic Acid Hybridization Techniques The term hybridization refers to the phenomenon whereby two complementary nucleic acid strands bind to each other. Any fragment of DNA with a known base-pair sequence can be detected by labeling a probe strand with the complementary sequence with a radioactive label or biochemical tag that can produce a colorimetric or fluorescent signal. A C G T G G C T G C A A G A C G C C T T C T C T G C C A Signal intensity at band position FIGURE 10.6  DNA sequencing.  Each chain-terminating nucleotide is labeled with a different fluorescent dye, and many fragments of DNA are synthesized. When all of the synthesized DNA fragments are separated using gel electrophoresis, they will be separated by size and a detector can automatically detect which label is at each position by measuring the intensity of each fluorescent signal. Molecular Cloning and Recombinant DNA Technology Chapter | 10  235 Southern Blot Developed by and named after Edward Southern, a Southern blot is used to identify if a known DNA sequence is present in a sample by using a specific probe to identify the sequence in a large mixture of other DNA fragments. A DNA sample (typically genomic DNA) is digested with restriction enzymes, and the mixture of fragments is separated by gel electrophoresis. The DNA is denatured to form single-stranded DNA, and the DNA is transferred to a nitrocellulose membrane (Figure 10.7). The positions of the DNA fragments FIGURE 10.7  The Southern and northern blot methods detect specific sequences of nucleic acids.  (A) In a Southern blot, restriction enzymes are used to digest DNA into fragments. (B)A northern blot uses RNA molecules as its starting material. The nucleic acids are separated by size using gel electrophoresis, then transferred to a membrane. The membrane is incubated with labeled probes capable of hybridizing to specific nucleic acid sequences. The bands can then be visualized using autoradiography. 236  Guide to Research Techniques in Neuroscience in the gel are preserved on the nitrocellulose membrane so a scientist can identify fragments based on their size. The membrane is exposed to a radioactive 32P-labeled single-stranded DNA probe corresponding to the sequence of interest. This probe hybridizes with the restriction fragment containing the complementary sequence, and autoradiography reveals the position of the restriction fragment-probe duplex. Therefore, a specific fragment in a mixture of millions of other fragments can be identified on the membrane. Southern blots are often used to demonstrate a successful gene knockout by using a probe against the region of the gene that was deleted from the genome (Chapter 13). PCR techniques are more sensitive and a faster alternative to identifying a DNA fragment, but these techniques are often used in conjunction with Southern blots to confirm a result. Northern Blot A technique similar to a Southern blot used to identify the presence of RNA has been humorously named the northern blot. Northern blots directly compare the relative amounts of mRNA levels between different samples. Intact mRNA molecules are separated on the basis of their size using gel electrophoresis, and a procedure identical to that used in a Southern blot is performed to visualize the size and relative expression of the mRNA species. Alternative methods for examining mRNA expression levels include qRT-PCR and microarrays (Chapter 9). Performing experiments with RNA typically requires more care in protecting the samples because RNases (enzymes that degrade RNA) are ubiquitous in the environment and RNA molecules are much less stable than DNA.An analogous technique to Southern and northern blots used to detect protein expression levels using an antibody is called the western blot (Chapter 15). CONCLUSION Molecular cloning and recombinant DNA technology have provided scientists with the ability to manipulate genes and the proteins they encode. These tools allow us to alter DNA sequences, cutting and pasting them in novel ways for many subsequent experiments. The power of these methods is complemented by other techniques that allow for the delivery of DNA to cells and creation of transgenic and mutant organisms, as will be discussed in the next several chapters. SUGGESTED READING AND REFERENCES Books Dale, J. W., von Schantz, M., & Plant, N. (2011). From genes to genomes: Concepts and applications of DNA technology (3rd ed.). West Sussex, UK: Wiley. Drlica, K. (2004). Understanding DNA and gene cloning: A guide for the curious (4th ed.). Hoboken, NJ: Wiley. Griffiths, A. J. F. (2010). Introduction to genetic analysis (10th ed.). New York: Freeman. Molecular Cloning and Recombinant DNA Technology Chapter | 10  237 Nicholl, D. S. T. (2008). An introduction to genetic engineering (3rd ed.). Cambridge: Cambridge University Press. Review Articles Bras, J., Guerreiro, R., & Hardy, J. (2012). Use of next-generation sequencing and other wholegenome strategies to dissect neurological disease. Nature Reviews Neurology, 13, 453–464. Campbell, T. N., & Choy, F. Y. (2002). Approaches to library screening. Journal of Molecular Microbiology and Biotechnology, 4, 551–554. Copeland, N. G., Jenkins, N. A., & Court, D. L. (2001). Recombineering: a powerful new tool for mouse functional genomics. Nature Reviews Genetics, 2, 769–779. Koboldt, D. C., Steinberg, K. M., Larson, D. E., Wilson, R. K., & MArdis, E. R. (2013). The nextgeneration sequencing revolution and its impact on genomics. Cell, 155, 27–38. Marziali, A., & Akeson, M. (2001). New DNA sequencing methods. Annual Review of Biomedical Engineering, 3, 195–223. Primary Research Articles—Interesting Examples from the Literature Agate, R. J., Choe, M., & Arnold, A. P. (2004). Sex differences in structure and expression of the sex chromosome genes CHD1Z and CHD1W in zebra finches. Molecular Biology and Evolution, 21, 384–396. Bacon, A., Kerr, N. C., Holmes, F. E., Gaston, K., & Wynick, D. (2007). Characterization of an enhancer region of the galanin gene that directs expression to the dorsal root ganglion and confers responsiveness to axotomy. Journal of Neuroscience, 27, 6573–6580. Brenner, S., Williams, S. R., Vermaas, E. H., Storck, T., et al. (2000). In vitro cloning of complex mixtures of DNA on microbeads: physical separation of differentially expressed cDNAs. Proceedings of the National Academy of Sciences of the United States of America, 97, 1665–1670. Gong, S., Zheng, C., Doughty, M. L., Losos, K., Didkovsky, N., Schambra, U. B., et al. (2003). A gene expression atlas of the central nervous system based on bacterial artificial chromosomes. Nature, 425, 917–925. Hsu, S. Y., Kaipia, A., McGee, E., Lomeli, M., & Hsueh, A. J. (1997). Bok is a pro-apoptotic Bcl-2 protein with restricted expression in reproductive tissues and heterodimerizes with selective anti-apoptotic Bcl-2 family members. Proceedings of the National Academy of Sciences of the United States of America, 94, 12401–12406. Paisan-Ruiz, C., Jain, S., Evans, E. W., Gilks, W. P., et al. (2004). Cloning of the gene containing mutations that cause PARK8-linked Parkinson’s disease. Neuron, 44, 595–600. Valenzuela, D. M., Murphy, A. J., Frendewey, D., Gale, et al. (2003). High-throughput engineering of the mouse genome coupled with high-resolution expression analysis. Nature Biotechnology, 21, 652–659. Protocols Choi, S., & Kim, U. J. (2001). Construction of a bacterial artificial chromosome library. Methods in Molecular Biology, 175, 57–68. Lee, S. C., Wang, W., & Liu, P. (2009). Construction of gene-targeting vectors by recombineering. Methods in Molecular Biology, 530, 15–27. Sambrook, J., & Russell, D. W. (2001). Molecular cloning: A laboratory manual (3rd ed.). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Websites New England BioLabs catalog: http://www.neb.com/nebecomm/tech_reference. Primer-BLAST: primer designing tool: http://www.ncbi.nlm.nih.gov/tools/primer-blast/. Addgene plasmid repository: http://www.addgene.org/. 239 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00011-3 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 11 Gene Delivery Strategies After Reading This Chapter, You Should be Able to: l Describe common methods of delivering recombinant DNA into cells l Compare and contrast methods of DNA delivery in various in vitro and in vivo preparations Techniques Covered: l Physical gene delivery: microinjection, electroporation, biolistics l Chemical gene delivery: calcium phosphate transfection, lipid-based transfection l Viral gene delivery: adenovirus, adeno-associated virus (AAV), lentivirus, herpes simplex virus (HSV), canine adenovirus (CAV) The previous chapter discussed methods for making and manipulating DNA constructs using recombinant DNA technology. Of course, the ultimate goal of creating a DNA construct is to deliver the manipulated sequence into living cells so that the endogenous cellular machinery can transcribe and translate the sequence into functional proteins. A neuroscientist might want to deliver a DNA sequence into a population of cells in culture, in a brain slice, or in the brain of a living animal. The purpose of this chapter is to survey the common methods of delivering recombinant DNA into cells in each of these environments. There are three general categories of DNA delivery: physical, chemical, and viral. Transfection refers to nonviral methods of delivering DNA to cells, including physical and chemical methods. Infection refers to viral DNA delivery, in which viruses attach themselves to cells and inject their DNA cargo. Each method has its own relative advantages and disadvantages that make it particularly well-suited for delivering DNA in different experimental contexts (Table 11.1). Major factors that dictate the choice of DNA delivery method include cellular environment, cell type, and experimental goals. Additionally, these DNA delivery methods vary in terms of the number of cells they affect, levels of gene expression they induce, and the length of gene expression over time. PHYSICAL GENE DELIVERY Physical gene delivery methods deliver DNA into a cell by physically penetrating the cell membrane with force. These methods are often highly efficient 240  Guide to Research Techniques in Neuroscience (transfected cells express high levels of the delivered gene) and can be used with any cell type. However, they can sometimes disrupt the integrity of the cell membrane, traumatizing the cell and leading to its death. Therefore, care must be taken in moderating the amount of force applied to deliver DNA to cells. Three common physical methods include microinjection, electroporation, and biolistics. Microinjection A scientist can microinject a solution containing DNA into a cell by piercing the cell membrane with a small glass needle and then applying pressure (Figure 11.1). If the scientist uses an extremely thin needle and withdraws the needle carefully, the cell membrane remains intact, and the cell has an opportunity to incorporate the injected DNA into its genome. This process requires a microscope to view the living cell, as well as fine micromanipulators to precisely place the needle adjacent to the cell membrane. Each cell must be injected one at a time (in contrast to other methods that deliver DNA to millions of cells at once), making microinjection laborious and low throughput. This method requires technical skill and is therefore usually performed by trained technicians or laboratory personnel who use these techniques regularly. Microinjection is rarely used to transfect neurons because their small shape and sensitivity makes them more difficult to inject than other cell types. TABLE 11.1  Categories of Gene Delivery Strategies Method Advantages Disadvantages Physical High-efficiency gene transfer No limitations on construct size No cell type dependency Low throughput Requires specialized equipment Can physically harm cells Chemical High efficiency in vitro No limitations on construct size Relatively easy to perform Rapid High throughput Low immunogenicity Limited in vivo applications Efficiency depends on cell type Can be toxic to cells Transient expression Viral High-efficiency gene transfer Cell-specific targeting is possible Long-term expression Can be used in vitro and in vivo Complex cloning required More expensive than transfection methods Safety concerns regarding production of infectious viruses in humans May provoke immune response Laborious preparation Limited construct size Gene Delivery Strategies Chapter | 11  241 DNA microinjection is the method used to make transgenic animals. To make a transgenic mouse, a scientist microinjects the transgenic DNA construct into newly fertilized mouse eggs. In some of the new eggs, the transgene randomly incorporates into the mouse genome. These cells are injected into a female mouse to carry the egg to gestation. The full details of creating transgenic mice, as well as other transgenic animals, are discussed in Chapter 12. Electroporation Electroporation is the process of using an electric pulse to transfect cells with DNA (Figure 11.2). Applying an electric field to cells is thought to induce temporary pores in the cell membrane, allowing the cell to take up DNA sequences. The electric field also drives negatively charged DNA strands away from the cathode (negative end) toward the anode (positive end) of an electric field. Therefore, an electric pulse causes some of the DNA to enter the cell. After the electric field is switched off, the cell membrane reseals, trapping some of the electroporated DNA within the cell. Electroporation can be used to efficiently deliver DNA to cells for in vitro conditions using special containers with electrical contacts. In a process known as ex vivo electroporation, a scientist removes a brain from an animal, applies electroporation techniques, and then sections the brain for slice cultures. In neuroscience research, electroporation has been particularly useful for in utero preparations (Figure 11.3). In utero electroporation surgeries begin by exposing the embryonic pups of a pregnant animal, usually a rodent. DNA is injected into the pups’ ventricles, and electrode paddles direct the uptake of the injected DNA construct(s) toward the anode and away from the cathode. The pups are placed back inside the mother, who is sutured and allowed to recover. Days later, pups are removed or the mother gives birth, and the neurons of the offspring express the introduced DNA construct. Similar methods can be used in the chicken model system in a process called in ovo electroporation. These methods are useful for delivering genes to brain regions adjacent to the FIGURE 11.1  DNA microinjection.  An extremely thin glass needle punctures the plasma membrane, allowing DNA to be injected into the cell. 242  Guide to Research Techniques in Neuroscience nervous system’s fluid-filled ventricles, as the DNA-containing solution must be injected into the ventricle. Thus, they have primarily been used in studies of chick spinal cord and rodent cerebral cortex. A scientist can control both spatial and temporal specificity by controlling the position of the electrode paddles and the timing of electroporation. For example, a scientist can target gene transfection to distinct cortical layers in the mammalian cerebral cortex by electroporating at specific embryonic time periods. Electroporation has numerous advantages, such as the ability to transfect cells in many different in vivo and in vitro environments. Furthermore, investigators can vary expression levels of a transfected gene by varying the strength and pattern of the electric field pulses. FIGURE 11.2  Electroporation.  (A) Cells are placed in a special chamber with a solution containing the DNA to be transfected. (B)Applying an electric field drives the negatively charged DNA strands away from the cathode and toward the anode. (C)After the electric field is switched off, the cell membrane reseals, and some DNA remains in the cell. Gene Delivery Strategies Chapter | 11  243 Biolistics Biolistics, short for “biological ballistics” and also known as particle-mediated gene transfer, is the method of directly shooting DNA fragments into cells using a device called a gene gun. To use a gene gun, a scientist first mixes a DNA construct with particles of a heavy metal, usually tungsten or gold. These fine particles stick to the negatively charged DNA. The DNA/metal particles are loaded onto one side of a plastic bullet (Figure 11.4). A pressurized gas, usually helium, provides the force for the gun. Gas pressure builds up until a rupture disk breaks, driving the plastic bullet down a shaft. The plastic bullet is abruptly stopped at the end of the shaft, but the DNA/metal particles emerge from the gun with great speed and force. If the gun is aimed at biological tissue, some of the metal particles will penetrate the cell membranes and deliver DNA constructs to cells. A neuroscientist can use biolistic technology to cause efficient gene expression in neurons. This technology can produce a dispersed transfection pattern, similar to a Golgi stain, in which only individual cells receive the foreign DNA in a background of untransfected cells. Another advantage to using biolistic technology is that a gene gun can deliver DNA through relatively thick tissue, such as a tissue slice in culture. Therefore, it is possible to transfect cells within a slice that would be difficult to target using other gene delivery methods. This technique has not yet been successful in transfecting mammalian neurons in vivo, although it has been used in vivo to transfect liver and skin cells. The main disadvantage of using this technology is that it may cause physical damage to cells. Optimization is required to limit the amount of tissue damage caused by the force of impact of the projectiles. + _ Pregnant rodent Developing embryos Injection of DNA construct Application of electric field FIGURE 11.3  In utero electroporation. A scientist performs a surgical procedure on a pregnant female, exposing the embryonic pups. DNA is injected into the pups’ ventricular system, and paddles generate electric field pulses to introduce DNA into the cells lining the ventricles. 244  Guide to Research Techniques in Neuroscience CHEMICAL GENE DELIVERY Chemical gene delivery is the process by which a scientist uses a chemical reaction to deliver DNA into cells. Negatively charged DNA can form macromolecular complexes with positively charged chemicals. These complexes can then interact with a cell’s membrane, promoting uptake through endocytosis or fusion. Chemical gene delivery strategies are useful in that they are incredibly high-throughput and often simple to perform. However, they are generally inefficient for in vivo delivery and therefore mainly used for cell culture experiments (Chapter 14). Expression of the target gene is usually transient, lasting for days to weeks depending on the cell type. Common chemical gene delivery strategies include calcium phosphate transfection and lipid transfection. Calcium Phosphate Transfection The calcium phosphate method of DNA transfection is one of the simplest and least expensive methods of chemical gene delivery (Figure 11.5). This method requires two chemical solutions: calcium chloride, which serves as a source of Pressurized gas Rupture disk Bullet DNA-coated metal particles Filter screen Gas breaks rupture disk, driving bullet down the shaft Bullet blocked by filter screen, DNA-coated metal particles emerge Gun aimed at biological specimen FIGURE 11.4  A gene gun.  DNA-coated metal particles are placed on the front end of a bullet. High-pressured gas drives the bullet down a shaft. At the end of the shaft, the bullet is blocked, but the DNA-coated particles emerge with great speed and force. Gene Delivery Strategies Chapter | 11  245 calcium ions, and HEPES buffered saline (HBS), which serves as a source of phosphate ions. First, the calcium chloride solution is mixed with the DNA to be transfected, and then the HBS is added. When the two solutions are combined, the positively charged calcium ions and negatively charged phosphate ions form a fine precipitate. The calcium ions also cause the DNA to precipitate out of solution. After a few minutes, the solution with precipitate is directly added to cells in culture. By a process that is not entirely understood, the cells take up some of the precipitate, and with it, the DNA. It is thought that the DNA precipitate enters cells by endocytosis, but the exact mechanism remains a mystery. This method works best for a cell monolayer so that the DNA precipitate covers the cells evenly. The advantages to using the calcium phosphate method are that it is relatively easy, reliable, and cheap. It is useful for transient expression or creating stable cell lines from immortalized cells (Chapter 14). The main disadvantage is that transfection efficiencies are low in neurons (1–3%), and this method does not work at all for transfecting neurons in intact tissue. Some immortalized cell Solution with DNA and calcium chloride HEPES buffered saline solution DNA precipitates out of solution Solution with precipitate added to cells Ca2+ Ca2+ Ca2+ Ca2+ Ca2+ Ca2+Ca2+ Ca2+Ca2+ Ca2+ Ca2+ Ca2+ FIGURE 11.5  Calcium phosphate transfection.  A scientist mixes DNA, calcium chloride, and HEPES-buffered saline. The chemical reaction forms a DNA-calcium phosphate precipitate that is able to pass through the cell membrane and deliver DNA to the nucleus. 246  Guide to Research Techniques in Neuroscience lines, such as HEK 293T and HeLa cells (Chapter 14), exhibit high transfection efficiencies (90–100%), which is one reason for their ubiquitous use in the life sciences. Lipid Transfection Lipofection, also known as “lipid transfection” or “liposome-based transfection,” uses a lipid complex to deliver DNA to cells. Lipids are a broad class of fat-soluble biomolecules, such as fats, oils, and waxes. The cell membranes of animal cells are composed of a bilayer of phospholipids with hydrophilic surfaces facing the cytoplasm and extracellular environment (Figure 11.6A). Lipofection technology uses tiny vesicular structures called liposomes that have the same composition as the cell membrane (Figure 11.6B). A scientist performs a simple reaction that forms a liposome around the DNA sequence to be transfected (Figure 11.6C). Depending on the liposome and cell type, the liposome can be endocytosed (Figure 11.6D) or directly fuse with the cell membrane to release the DNA construct into cells (Figure 11.6E). The advantage to lipofection is that it works in many cell types, including cultured neurons. Commercially available kits allow transfection reactions to be performed within 30 min and gene expression to be assayed within hours. FIGURE 11.6  Lipofection. (A) The cell membrane is composed of a lipid bilayer, with a hydrophobic interior and hydrophilic exterior. (B) Liposomes are also composed of a lipid bilayer arranged as a spherical shell. (C) A scientist performs a brief reaction that allows liposomes to form around DNA. (D) Cells in culture can endocytose the liposome, digesting it within vesicles to release DNA. (E) Alternatively, liposomes can directly fuse with the plasma membrane, directly releasing DNA into cells. Gene Delivery Strategies Chapter | 11  247 However, like the calcium phosphate method, lipofection is almost exclusively used in cell culture experiments. VIRAL GENE DELIVERY Viral gene delivery uses one of many available viral vectors to deliver DNA to cells in vitro or in vivo. A virus can be thought of as a tiny molecular machine whose entire purpose is to attach to cells and inject genetic material. In nature, this genetic material encodes the proteins necessary to make more virus, so the infected cell essentially becomes a virus-making factory. In the laboratory, many viruses have been manipulated so that they can no longer replicate inside a host cell on their own. Additionally, the coding region of viral DNA is exchanged with a gene of interest to make the virus deliver a transgene to a cell without forcing that cell to produce more virus particles. This process of using a nonreplicating viral vector to deliver foreign DNA into a cell is called transduction. Scientists can either produce virus in their own laboratories or outsource viral production commercially (Figure 11.7). To begin, a scientist uses recombinant DNA technology (Chapter 10) to place a DNA sequence of interest into a plasmid containing the necessary sequences to incorporate into a virus particle. Note that this plasmid is not the same as a virus itself but is simply a piece of DNA with the necessary sequences to incorporate into a virus particle, provided the other necessary proteins are present. To produce a batch of virus, cultured cells are used to produce all the necessary viral components. A scientist transfects these cells with the recombinant plasmid, as well as additional DNA sequences coding for the necessary viral proteins. Because viral DNA has been engineered so that it cannot replicate more virus on its Transfect cells with plasmids encoding viral particles After 2 days, collect viral-containing supernatant Centrifuge to collect viral pellet Resuspend pellet in buffer FIGURE 11.7  Virus production. A scientist transfects cultured cells with DNA encoding the necessary proteins to make virus. Over 2–3 days, the cells produce virus and release it into the culture medium. The medium is collected and centrifuged to collect a viral pellet. The pellet is resuspended in buffer and frozen until use. 248  Guide to Research Techniques in Neuroscience own, these viral proteins are necessary to produce functional, infectious viral particles. Immortalized cell lines, such as HEK 293T cells, are the best cells to serve as virus packaging cells because they grow quickly, are transfected relatively easily, and can produce large amounts of virus. One to two days after transfection, the packaging cells produce many virus particles and release them into the extracellular medium. The final step of virus production is to collect the medium, filter out cell debris, and centrifuge the medium to collect the viral pellet. The pellet is dissolved in a buffer and, depending on the needs of the investigator, either used immediately to infect the target cells of interest or frozen until needed. Viral infections are robust, highly efficient, and can lead to long-term expression. Viruses have been used to deliver genes to cultured cells, brain slices, tissue explants, and brain regions in vivo. They are the tools of choice to deliver genes into the brains of rodents, as most other methods of gene delivery are incapable of working efficiently in vivo. However, there are also disadvantages and limitations to using viral gene delivery. The infected cells, especially cells in vivo, may not begin expressing the transduced gene until 7–14 days after exposure to the virus. Also, the size of the viral vector restricts the size of the DNA construct that a scientist can transduce. Finally, scientists must take great care and safety when using viruses because it is possible for scientists to accidentally infect themselves. There are a variety of viral vectors that are useful to neuroscientists (Table 11.2). They vary in terms of infection efficiency, expression levels, duration of expression, time to start of expression, host cell toxicity, and host cell preference. Adeno-associated virus and lentivirus are the most commonly used viral vectors due to long-term expression of transgenes and low toxicity in neurons. Adenovirus Adenovirus can infect both dividing and postmitotic cells in a broad range of species and cell types. It can carry DNA constructs that are 7.5–30 kb (kilobases) long. The DNA does not integrate into the genome, so expression is transient, lasting weeks or months. It is useful for high-level transient expression in neurons, but it can often cause an inflammatory response and cell death. Canine Adenovirus Canine adenovirus (CAV) is a form of adenovirus that is highly infectious in dogs. This virus has a high carrying capacity of 7.5–30 kb and is used in neuroscience research because of its retrograde properties. Upon injection into the brain, CAV can be taken up by the presynaptic membrane and retrogradely transported to cell soma. Therefore, this viral vector is useful for targeting neurons based on their connectivity. Gene Delivery Strategies Chapter | 11  249 Herpes Simplex Virus Herpes simplex virus (HSV) is a neurotropic virus that naturally targets neurons as host cells. It has a very large genome compared to other viruses (100–200 kb) that allows delivery of long DNA sequences. While the DNA does TABLE 11.2  Viral Vectors Used in Neuroscience Vector Insert size Features Adenovirus 7.5–30 kb Infects dividing and nondividing cells Transient expression—does not integrate in the genome High transduction efficiency Onset of expression in days Triggers immune response Canine adenovirus (CAV) 7.5–30 kb Exhibits retrograde transport properties Infects dividing and nondividing cells Transient expression—does not integrate in the genome High transduction efficiency Onset of expression in days Triggers immune response Herpes simplex virus (HSV) >30 kb Neuron specific Stable, long-term expression—does not integrate in the genome Low transduction efficiency Safety concerns Complex production Onset of expression in hours Adeno-associated virus (AAV) 5 kb Naturally replication-deficient—safe Infects dividing and nondividing cells Stable expression—integrates in the genome High transduction efficiency Onset of expression in weeks Minimal toxicity Labor-intensive production Lentivirus 8 kb Form of retrovirus that can infect nondividing cells in addition to dividing cells Stable expression—integrates in the genome Safety concerns 250  Guide to Research Techniques in Neuroscience not integrate into the host genome, it exists stably in the nucleus to produce long-term (months to years) expression. The disadvantages are that it can be difficult to engineer HSV because of its large size, and it can be toxic to cells. There are also safety concerns because it is highly infectious. Adeno-Associated Virus Adeno-associated virus (AAV) is naturally replication deficient, requiring a helper virus to replicate, so it is safer to use than other viral strains. AAV can infect both dividing and postmitotic cells, but, unlike adenovirus, does integrate into the host genome, permitting long-term gene expression. AAV also seems to be less toxic to neurons than adenovirus. However, AAV production is labor-intensive and the carrying capacity is much less than adenovirus, about 5 kb. Lentivirus Lentivirus belongs to a class of virus called retrovirus that has an RNA genome rather than DNA. To produce functional gene products, the virus also contains the enzyme reverse transcriptase, which produces cDNA from the RNA template (Chapter 10). When a cell endocytoses a lentivirus particle, the RNA is released and reverse transcriptase produces cDNA. The DNA migrates to the nucleus, where it integrates into the host genome. Most retroviruses only infect dividing cells, making them useful for studying neuronal development and cell fate. However, lentivirus is capable of infecting both dividing and postmitotic cells (e.g. neurons) and is therefore widely used in neuroscience experiments. Lentivirus is based on the human immunodeficiency virus and has an 8-kb carrying capacity. Because the DNA integrates into the genome, lentivirus delivery leads to long-term expression. CONCLUSION This chapter has surveyed common methods used to deliver DNA sequences into cells. A scientist chooses one method over another based on a number of factors, such as the cell type and the cell’s environment, as well as the specific goals of the experiment (Table 11.3). For cell culture experiments, chemical, electroporation, or viral gene delivery strategies all work well. For in vivo DNA delivery to adult animals, viral gene delivery is the most efficient option. Now that we have surveyed these methods, the next two chapters will focus on methods of manipulating the genomes of living organisms. Gene Delivery Strategies Chapter | 11  251 SUGGESTED READING AND REFERENCES Books Heiser, W. C. (Ed.). (2010a). Gene delivery to mammalian cells. Volume 1: Nonviral gene transfer techniques. New York, NY: Humana Press. Heiser, W. C. (Ed.). (2010a). Gene delivery to mammalian cells. Volume 2: Viral gene transfer techniques. New York, NY: Humana Press. Twyman, R. M. (2005). Gene transfer to animal cells.Abingdon, Oxon, UK: Garland Science/BIOS Scientific Publishers. Review Articles Bonetta, L. (2005). The inside scoop—evaluating gene delivery methods. Nature Methods, 2, 875–883. Callaway, E. M. (2008). Transneuronal circuit tracing with neurotropic viruses. Current Opinion in Neurobiology, 18, 617–623. Luo, L., Callaway, E. M., & Svoboda, K. (2008). Genetic dissection of neural circuits. Neuron, 57, 634–660. Washbourne, P., & McAllister, A. K. (2002). Techniques for gene transfer into neurons. Current Opinion in Neurobiology, 12, 566–573. TABLE 11.3  Choosing a Gene Delivery Strategy Environment Considerations Commonly used gene delivery methods Cells in the intact nervous system of living animals Cells within the brain are notoriously resistant to most transfection methods. Electroporation, virus Cells in brain slices Gene delivery must be fast and efficient and may need to go through thick tissue. Electroporation, biolistics Cells in dissociated cultures Gene delivery must be highly efficient and high throughput to transfect/ transduce thousands or millions of cells. Electroporation, biolistics, chemical transfection, virus Individual cells Cells are often valuable, such as extracted embryos or newly fertilized eggs, so care is taken to efficiently deliver DNA to each cell. Microinjection, electroporation 252  Guide to Research Techniques in Neuroscience Primary Research Articles—Interesting Examples from the Literature Carter, M. E., Soden, M. E., Zweifel, L. S., & Palmiter, R. D. (2013). Genetic identification of a neural circuit that suppresses appetite. Nature, 503, 111–114. Chen, J. L., Carta, S., Soldado-Magraner, J., Schneider, B. L., & Helmchen, F. (2013). Behaviourdependent recruitment of long-range projection neurons in somatosensory cortex. Nature, 499, 336–340. Krashes, M. J., Shah, B. P., Madara, J. C., Olson, D. P., Strochlic, D. E., et al. (2014). An excitatory paraventricular nucleus to AgRP neuron circuit that drives hunger. Nature, 507, 238–242. Pekarik, V., Bourikas, D., Miglino, N., Joset, P., Preiswerk, S., & Stoeckli, E. T. (2003). Screening for gene function in chicken embryo using RNAi and electroporation. Nature Biotechnology, 21, 93–96. Tsai, H. C., Zhang, F., Adamantidis, A., Stuber, G. D., Bonci, A., de Lecea, L., et al. (2009). Phasic firing in dopaminergic neurons is sufficient for behavioral conditioning. Science, 324, 1080–1084. Wilson, S. P.,Yeomans, D. C., Bender, M. A., Lu,Y., Goins, W. F., & Glorioso, J. C. (1999). Antihyperalgesic effects of infection with a preproenkephalin-encoding herpes virus. Proceedings of the National Academy of Sciences of the United States of America, 96, 3211–3216. Protocols Lappe-Siefke, C., Maas, C., & Kneussel, M. (2008). Microinjection into cultured hippocampal neurons: a straightforward approach for controlled cellular delivery of nucleic acids, peptides and antibodies. Journal of Neuroscience Methods, 175, 88–95. O’Brien, J., & Lummis, S. C. (2004). Biolistic and diolistic transfection: using the gene gun to deliver DNA and lipophilic dyes into mammalian cells. Methods, 33, 121–125. Walantus, W., Castaneda, D., Elias, L., & Kriegstein, A. (2007). In utero intraventricular injection and electroporation of E15 mouse embryos. Journal of Visualized Experiments, 6. http://dx.doi. org/10.3791/239. http://www.jove.com/index/details.stp?id=239. Websites University of North Carolina Vector Core: http://www.med.unc.edu/genetherapy/vectorcore. University of Pennsylvania Vector Core: http://www.med.upenn.edu/gtp/vectorcore/. 253 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00012-5 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 12 Making and Using Transgenic Organisms After Reading This Chapter, You Should be Able to: l Describe and categorize commonly used transgenes l Describe the necessary components of a transgenic DNA construct l Understand the benefits and utility of binary transgenic expression systems l Explain the process of creating viable transgenic organisms Techniques covered: l Common transgenes in neuroscience: reporter genes, genes used to measure neural activity, genes used to manipulate neural activity, genes used to ablate neurons, genes used to disrupt endogenous genes l Binary transgenic systems: Gal4/UAS, Cre/lox, Flp/Frt, Tet-off/Tet-on l Making transgenic organisms: mice, flies, worms One of the modern marvels of genetic engineering technology is the ability to take a gene from one species and place it in the genome of another species. The rich diversity of life on Earth provides an extraordinary variety of genes that not only help organisms survive unique niches on the planet but also help neuroscientists probe the structure and function of neural systems. For example, the jellyfish Aequorea victoria evolved a gene capable of emitting photons of light for bioluminescence. This gene, known as green fluorescent protein (GFP), has been used in thousands of studies to examine the structure and function of the nervous system. A variety of species have evolved other genes useful for visualizing neurons and manipulating their function, such as luciferase from fireflies, channelrhodopsin-2 from single-celled algae, lacZ from Escherichia coli, or allatostatin receptor from insects. Genetic engineering technology provides the ability to express these genes in specific cell types at specific times in an organism’s lifespan. Thus, modern genetic engineering technology allows scientists to stably express genetically encoded tools for the precise study of specific neural systems and circuits. Any gene expressed in an animal that does not normally carry this gene is called a transgene, and any animal that carries this foreign DNA is called a transgenic organism. The purpose of this chapter is to survey common transgenes used in modern neuroscience research and to describe how these genes 254  Guide to Research Techniques in Neuroscience can be restricted to specific cell types at specific times. Furthermore, we will examine the process of creating a transgenic animal in commonly used model organisms including mice, flies, and worms. Chapter 13 will provide additional information on gene targeting technologies used to remove genes from a genome or insert genes in precise locations in the genome. TRANSGENES There are now dozens of commonly used transgenes that can be used to answer a variety of experimental questions. Scientists can inject these transgenes into embryos to create transgenic lines of animals that pass on the transgene to their offspring. Alternatively, these transgenes can be introduced directly into cells or brain regions, as described in Chapter 11. We survey some of the major categories of transgenes here and in Table 12.1, although this is not an exhaustive list. Reporter Genes A reporter gene is any gene whose protein product can visually report the location of a particular protein, cell type, or circuit. The most common reporter genes are GFP and its variants. When placed under the control of a cell-specific promoter, GFP can mark the location of these specific cells so they can be used in subsequent experiments (e.g., electrophysiology, co-localization experiments, identification of cells in culture). In addition to GFP, there are dozens of fluorescent proteins with different spectral properties that appear as different colors, such as red fluorescent protein, cyan fluorescent protein, and yellow fluorescent protein. Fluorescent proteins can be manipulated so that they tag proteins or organelles within a cell. For example, a scientist can use recombinant DNA methods (Chapter 10) to attach GFP to a gene that encodes a protein of interest so that the location, expression, and trafficking of the protein can be studied in real time. It is also possible to add specific sequence tags to fluorescent proteins so that they localize to discrete regions of the cell, such as the cell membrane. Membrane-bound fluorescent reporters can be useful for tracing axons and dendrites in addition to cell bodies. LacZ is another common reporter gene. Originally derived from bacteria, lacZ encodes the enzyme β-galactosidase. This enzyme reacts with a substrate, X-gal, to create a dark blue product. Therefore, lacZ can serve as a faithful reporter of gene expression after histological processing with X-gal. Genes Used to Measure Neural Activity Chapter 7 described methods of indirectly measuring neural activity by imaging calcium dynamics and the membrane potential. Genetically encoded calcium indicators (GECIs), such as GCaMPs, or genetically encoded voltage indicators (GEVIs), such as ArcLight, allow targeted observation of neural activity in specific cell types. This specificity is an advantage over injectable activity-sensitive dyes that label all neurons at the injection site. Making and Using Transgenic Organisms Chapter | 12  255 TABLE 12.1  Examples of Commonly Used Transgenes Type of transgene Utility Examples Reporter genes Fluorescent proteins Marks the location of cells, subcellular structures, and organelles Green fluorescent protein (GFP) and multicolored variants Fluorescent proteins to trace neural connections Addition of membrane tags to fluorescent proteins so they are bound to the membrane Farnesylated-GFP, Myristolated-GFP Bioluminescent proteins Proteins that produce light as a byproduct of a biochemical reaction Luciferase Nonluminescent proteins Proteins that mark expressing neurons after histological processing LacZ Alkaline phosphatase (AP) Genes for measuring neural activity Calcium indicators Sensors that change fluorescence in response to different concentrations of calcium ions GCaMPs Voltage sensors Sensors that change fluorescence in response to changes in membrane potential ArcLight Synaptic activity sensors Sensors that change fluorescence in response to changes in synaptic vesicle fusion Synapto-pHluorin Genes for inducible activation of neural activity Optogenetics Membrane-bound channels that depolarize cells when stimulated by light Channelrhodopsin-2 (ChR2) Ligand receptors Ligand-gated ion channels that depolarize cells when a ligand is added Capsaicin receptor (TRPV1) DREADD (HM3Dq) Continued 256  Guide to Research Techniques in Neuroscience Genes Used to Manipulate Neural Activity Neural activity can be manipulated through nongenetic methods, such as electrical stimulation or pharmacological inhibition. However, these methods rarely stimulate or inhibit all cells of a desired subtype, and they often affect undesired subtypes. Transgenes that allow inducible activation or inhibition of neural activity solve these problems by providing control over genetically targeted cells, with no effect on cells that do not express the transgene. These transgenes use either optically controlled or ligand-gated channels to selectively stimulate or inhibit neural activity. Chapter 8 described two optically controlled transgenes used to manipulate neural activity. Channelrhodopsin-2 (ChR2), originally derived from singlecelled algae, is a cation channel that can depolarize a neuron upon stimulation with blue light. Because ChR2 can be placed under cell-specific promoters, it is Type of transgene Utility Examples Genes for inducible inhibition of neural activity Optogenetics Membrane-bound pumps that hyperpolarize cells when stimulated by light Halorhodopsin (NpHR) Archaerhodopsin (Arch) Ligand receptors Ligand-gated ion channels that hyperpolarize cells when a ligand is added Allatostatin receptor (Alstr) DREADD (HM4Di) Shibire mutants Temperature sensitive mutants of the Drosophila shibire gene (dynamin) that prevent synaptic transmission at certain temperatures Shits Genes for neuronal ablation Toxins Kills all expressing neurons Ataxin Toxin receptors When the toxin is applied to the brain, only expressing cells will die Diphtheria toxin receptor TABLE 12.1  Examples of Commonly Used Transgenes—cont’d Making and Using Transgenic Organisms Chapter | 12  257 possible to stimulate specific cell types in a heterogeneous population of neurons by delivering blue light to those neurons. The genetically encoded chloride pump Halorhodopsin (NpHR) and the genetically encoded hydrogen pump Archaerhodopsin (Arch) hyperpolarize neurons upon stimulation with yellow light. Using these transgenes, it is possible to silence single action potentials or sustained trains of neuronal spiking. Another method of manipulating neural activity using genetically encoded transgenes is to express a ligand-gated ion channel in the brain of an animal that normally does not express this channel. For example, the capsaicin receptor (TrpV1) is a ligand-gated channel that depolarizes a neuron upon exposure to capsaicin. Flies and other invertebrates do not normally express this channel, so capsaicin exposure will depolarize only those neurons that express the receptor. A similar strategy can be used to inhibit activity and hyperpolarize neurons in the presence of a ligand. For example, the insect allatostatin receptor (Alstr) is not expressed in mammals. In the presence of the insect protein allatostatin, this receptor inhibits neural activity. Thus, if Alstr is expressed in the mammalian brain, addition of allatostatin will reversibly inactivate genetically targeted populations of neurons until the ligand is washed out from the system. Synthetic G-protein coupled receptors are also useful for modulating neural activity. Designer receptors exclusively activated by designer drugs (DREADDs) consist of artificial receptors for the normally inert ligand clozapine-N-oxide (CNO). In the presence of CNO, hM3Dq will depolarize a neuron, while hM4Di will hyperpolarize a neuron. These techniques are discussed in much more detail in Chapter 8. Genes Used to Ablate Neurons Lesion studies have historically been a staple of neuroscience research to investigate the necessity of a brain region for a specific behavior or phenotype. Using physical or pharmacological methods to ablate neurons raises concerns about the specificity and totality of the lesions. Transgenic technology allows lesions to be targeted to specific, genetically defined subsets of neurons. Genetic ablation studies can be performed using the ataxin transgene, which encodes a toxin that kills all expressing neurons. It is also possible to use a transgene that encodes a toxin’s receptor, such as the diphtheria toxin receptor. This receptor is toxic to neurons only when diphtheria toxin is present. Neurons that do not express the receptor will not be sensitive to the toxin. Using a genetically encoded toxin receptor instead of a genetically encoded toxin itself allows a scientist to ablate neurons at any time point during an animal’s lifespan. By controlling when the toxin is injected, the scientist controls the timing of ablation. Genes Used to Disrupt Endogenous Gene Function In mice, it is possible to knock out genes from the genome using gene targeting technology, as discussed in Chapter 13. It is also possible to express transgenes that disrupt endogenous gene or protein function. The overexpression of 258  Guide to Research Techniques in Neuroscience dominant negative transgenic constructs directly interferes with protein function, whereas RNA interference (RNAi) technology can be used to knock down the expression of a gene by targeting mRNA before it is translated into a functional protein. We will discuss dominant negatives, RNAi, and how they work in Chapter 13. Overexpression of Endogenous Genes For some research questions, it is useful to produce transgenic animals that overexpress an endogenous gene or express the gene in a cell-type that normally does not express it. For example, a scientist wishing to perform a gain-of-function experiment in which a gene of interest is expressed at a much higher level than normal could create a transgenic animal that highly overexpresses the gene. Scientists can also show that overexpression of a gene in a knockout background can rescue the gene’s function. Alternatively, a scientist can show that expressing a gene is sufficient to cause a particular function. For example, he or she may hypothesize that a gene encoding a cell-surface receptor is important during development to attract axons from a different brain region. A transgenic mouse that expresses the receptor in a different cell type could be used to test whether the presence of this receptor in other cells is sufficient to cause the axons to target the wrong region. Now that we have surveyed some commonly used transgenes, we examine how a scientist creates a transgenic construct and expresses foreign DNA in an animal’s genome. THE TRANSGENIC CONSTRUCT The first step in producing a transgenic organism is to use recombinant DNA technology (Chapter 10) to make a transgenic construct. This construct must contain (1) the DNA sequence encoding the transgene and (2) the necessary promoter sequences for the expression of the gene (Figure 12.1). Simply inserting a gene into the genome will not cause it to be expressed: endogenous proteins must recognize a promoter sequence to attract a transcription complex that recruits RNA polymerase to transcribe the DNA into mRNA. Promoter region Coding region Transcriptional start site 5’ 3’ FIGURE 12.1  A transgenic DNA construct.  A construct is composed of the transgene (coding region) and promoter region. The promoter region attracts endogenous cellular transcription factors and other proteins to drive expression of the transgene. Making and Using Transgenic Organisms Chapter | 12  259 Promoters control the expression of genes so that they are expressed in specific cells and tissues at specific times. For example, every cell in the human body contains the gene encoding the enzyme tyrosine hydroxylase (TH). However, this gene is only expressed in a subset of neurons in the brain because only these cells have the endogenous transcription factors and other regulatory proteins necessary to interact with the promoter for TH and activate transcription. Therefore, a transgenic construct containing all the necessary promoter elements for the TH gene will allow a transgene to be expressed specifically in the subset of cells that normally transcribe TH. A scientist can express a transgene in a different neuronal subtype by identifying a gene uniquely expressed in those cells and taking advantage of the promoter elements for that gene. Promoter sequences are often depicted as existing just upstream of the genes they regulate (Figure 12.1). However, this is not necessarily the case, especially in mammals. Promoter elements necessary to drive a gene of interest may be located thousands or even hundreds of thousands of base pairs upstream of a gene. They may be located in the introns within the coding sequences of a gene or even downstream of the coding sequence of the gene. This phenomenon restricts the promoters that can be used to drive transgene expression. In general, it is easier to use promoters to target specific cell types in invertebrate model organisms, as promoters in these species are often located just upstream of the genes they regulate. Because mammalian promoters can be difficult to isolate and place in a transgenic construct, some laboratories make transgenic mice using BAC constructs. BAC stands for bacterial artificial chromosome, a large 50-to 350-kb vector that can be used to create genetic libraries and clone mammalian DNA sequences (Chapter 10). Investigators can order a BAC that contains any gene in the genome, flanked by 50–100 kb of genomic sequences on either side of the gene. Thus, a transgenic construct could be produced by inserting a transgene at the transcriptional start site of the endogenous gene in a BAC. The transgene would then be located at the same location as the endogenous gene in relation to promoter elements. The limitation to using this technology is that recombinant DNA technology in BACs can be difficult due to the large size of the BAC compared to other expression vectors, which are usually 3–10 kb in length. Whether using small expression vectors or BACs, scientists make DNA constructs in DNA vectors that are circular, as described in Chapter 10. However, before injecting the transgene into cells to produce transgenic animals, the DNA is almost always made linear by using restriction enzymes to isolate the promoter/transgene sequence away from the vector. This helps the promoter/ transgene region integrate into the genome as a linear fragment, as well as discards unnecessary coding sequences that may be in the vector, such as antibiotic resistance genes. Transgenes typically integrate into the genome in random locations. Although there are certain regions of the genome that appear to be “hot spots” of integration activity, scientists generally cannot control where the transgene 260  Guide to Research Techniques in Neuroscience (A) (B) UAS+ Promoter A TGal4 UAS+ Promoter A Promoter B T Gal4 Gal80 (C) (D) + Promoter A Promoter B lox lox TCre + Promoter A Promoter B lox lox TCre (E) (F) + Promoter A Promoter B S lox lox STO P TCre + Promoter A Promoter B S lox lox STO P TCre-ER (G) + Promoter A Promoter B Frt Frt TFlp (H) + Promoter A Promoter B Frt Frt TFlp (I) + Promoter A TetO TtTA (J) + Promoter A TetO TrtTA FIGURE 12.2  Binary expression systems. These systems use two or more constructs to spatially or temporally restrict transgene expression. (A) The Gal4/UAS system. Promoter A drives expression of the transcription factor Gal4. Gal4 binds to UAS and activates expression of Transgene T. Therefore, T expression is indirectly controlled by Promoter A. (B) The Gal4/ Gal80/UAS system. As in (A), Promoter A drives expression of Gal4, which binds to UAS and activates expression of Transgene T. Promoter B drives expression of Gal80, which inactivates Gal4. Therefore, T will be expressed in cells in which Promoter A is active but not in cells in which Promoter B is active. (C) The Cre/lox system. Promoter A drives the expression of Cre recombinase and Promoter B drives the expression of Transgene T. Cre recognizes lox sites and catalyzes a recombination event that removes DNA within the sites. Therefore, T will be expressed in cells in which Promoter B is active but not in cells in which Promoter A is active. (D) The Cre/lox system with inverted lox sites. When Cre recognizes two lox sites facing each other, Cre rearranges the DNA between the lox sites so that it becomes “flipped” in the reverse orientation. If a transgenic construct is produced in which the transgene is oriented in the opposite direction, a Cre construct can be used to flip the transgene in the correct orientation. Thus, Transgene T will be expressed in cells in which both promoters A and B are active. (E) The Cre/ lox system used to remove a stop site. Cre is used to remove a stop codon that prevents the expression of Transgene T. Therefore, T is expressed in cells in which promoters A and B are active. (F) Promoter A expresses Cre recombinase fused to the estrogen receptor (ER). ER sequesters Cre in the cytoplasm until a scientist adds tamoxifen, which binds to ER and causes Cre to translocate to the nucleus, where it can remove a stop codon preventing expression of Transgene T. Making and Using Transgenic Organisms Chapter | 12  261 integrates unless using specific gene-targeting methods described in Chapter 13. Because of this random integration, position effects can occur that influence the transgene’s expression pattern. Depending on where the transgene lands in the genome, neighboring genes and regulatory sequences can alter the levels and pattern of expression. The number of transgene copies that integrate into the genome can also vary the levels and pattern of transgene expression. Therefore, transgenic organisms must be carefully evaluated to ensure that transgenes are expressed in the desired cell types. Just because a specific promoter is used to drive transgene expression, there is no guarantee that the transgene will be expressed in the desired cells. BINARY TRANSGENIC SYSTEMS The simplest method of making a transgenic animal is to use a single DNA construct, as just described, in which a promoter directly regulates the expression of a transgene. An alternate approach is a binary expression system that uses multiple constructs to further refine the spatial and temporal expression of a transgene (Figure 12.2 and Table 12.2). To use a binary expression system, a scientist creates two different lines of transgenic animals, each line expressing one of the required constructs, and then mates the animals to produce offspring that express both constructs. Alternatively, the scientist can express one construct in a transgenic animal and deliver the other construct using other gene delivery methods, such as viral gene delivery (Chapter 11). These flexible expression systems have numerous advantages. The main advantage is that they can provide control over the timing and location of transgene expression. Another substantial advantage is the ability to mix and match combinations of transgenes, saving time and effort in making new transgenic constructs. Therefore, T is expressed in cells in which Promoters A and B are active but only when tamoxifen is added. (G) The Flp/Frt system. Promoter A drives the expression of Flippase recombinase (Flp), and Promoter B drives the expression of Transgene T. Flp recognizes Frt sites and catalyzes a recombination event that removes DNA within the sites. Therefore, T will be expressed in cells in which Promoter B is active, but not in cells in which Promoter A is active. (H) The Flp/ Frt system with inverted Frt sites. When Flp recognizes two Frt sites facing each other, Flp rearranges the DNA between the Frt sites so that it becomes “flipped” in the reverse orientation. If a transgenic construct is produced in which the transgene is oriented in the opposite direction, a Flp construct can be used to flip the transgene in the correct orientation. Thus, Transgene T will be expressed in cells in which both promoters A and B are active. (I) The Tet-off system. Promoter A drives the expression of the tetracycline transactivator (tTA) protein. tTA binds to the tetO promoter, activating expression of Transgene T. Therefore, expression of T is indirectly controlled by Promoter A. If a scientist applies a drug called doxycycline, tTA is reversibly inhibited. (J) The Tet-on system. Promoter A drives the expression of the reverse tetracycline transactivator (rtTA). rtTA is unable to bind to the tetO promoter unless doxycycline is present. Therefore, Promoter A indirectly activates expression of Transgene T, but only when a scientist adds doxycycline. 262  Guide to Research Techniques in Neuroscience The Gal4/UAS System The Gal4/UAS system is a binary expression system primarily used in Drosophila, although it has also been applied to mice and zebrafish. In one construct, a cellspecific promoter drives the expression of the gene encoding Gal4, a transcription factor normally expressed in yeast. In a second construct, a transgene of interest is regulated by a promoter sequence called an upstream activation sequence (UAS). The Gal4 protein binds to the UAS sequence and efficiently drives expression of the transgene (Figure 12.2A). Thus, the transgene will only express in cells defined by the promoter regulating Gal4. Why not simply place the transgene under the control of the promoter regulating Gal4? There are at least two reasons. First, the UAS expresses transgenes at much higher levels than endogenous promoters, so the Gal4/UAS constructs together allow for high levels of gene expression. Second, Gal4 lines now exist for hundreds of genes in the Drosophila genome. There are also many constructs in which a UAS drives a specific transgene. Therefore, the Drosophila scientific community has created an amazingly efficient system of creating and delivering transgenes to many different cell types (Figure 12.3). Gal4/UAS can also be used in combination with another yeast protein, Gal80. The Gal80 protein binds to and inhibits the activity of Gal4. Thus, a second promoter element can drive expression of Gal80 to further restrict the cells expressing the transgene (Figure 12.2B). Furthermore, Gal80 can be made temperature sensitive: the Gal80ts protein is active at 19 °C but not 30 °C. TABLE 12.2  Promoters in Binary Transgenic Systems Binary transgenic system Cell and tissue specificity is conferred by the promoter(s) regulating: The gene of interest is directly regulated by the promoter called: Gal4/UAS Gal4 UAS Gal4/Gal80/UAS Gal4 and Gal80 UAS Cre/lox Cre Any promoter can be used…. Lox sites determine the location of DNA recombination Flp/Frt Flp Any promoter can be used…. Frt sites determine the location of DNA recombination tTA/tetO tTA tetO rtTA/tetO rtTA tetO Making and Using Transgenic Organisms Chapter | 12  263 Therefore, it is possible to express a transgene in specific cells using the Gal4/ UAS system and to regulate the timing of transgene expression by controlling the temperature, and thus the expression of Gal80ts. The Cre/Lox System The Cre/lox system is a site-specific recombination system widely used in mice. The enzyme Cre recombinase, originally derived from the P1 bacteriophage, recognizes specific 34 base-pair DNA sequences called lox sites. Lox sites are added to a transgene so that they flank a sequence of DNA. This is referred to as a “floxed” sequence, a sequence flanked by lox sites. Each lox site has an asymmetric sequence in the middle, providing the sequence with directionality. This directionality is important because Cre processes DNA depending on the orientation of the lox sites: l Cre excises any DNA sequences between two lox sites oriented in the same direction Figure 12.2(C) l Cre inverts any DNA sequences between two lox sites oriented in inverted directions Figure 12.2(D) Therefore, the Cre/lox system allows for complex rearrangement of DNA in transgenic mice. Many transgenic lines now exist in which Cre recombinase is regulated by cell-specific promoters. The ability to engineer various combinations of lox sites and specific gene sequences provides unique experimental possibilities. For example, a transgenic mouse can selectively express a transgene Gal4 lines UAS lines B X C UAS UAS Any combination of genetic coding sequence (1, 2, 3, etc.) driven by cell-specific promoter (A,B,C, etc.) UAS 1 2 3 A Gal4 Gal4 Gal4 FIGURE 12.3  The utility of the Gal4/UAS system. Scientists have created hundreds of fly stocks in which endogenous promoters drive cell-specific expression of Gal4 transcription factors. In turn, any transgene of interest can be placed under the control of a UAS sequence. Therefore, a scientist can cross any combination of Gal4 line with a UAS line to drive transgene expression in specific cell types. This system has been used extensively in the fly, as well as in Xenopus and zebrafish model organisms. 264  Guide to Research Techniques in Neuroscience with a “floxed stop” strategy: a “stop” codon is placed directly upstream from a transgene, preventing the transgene’s expression. If two lox sites flank the “stop” codon, the transgene will only be expressed in cells that express Cre to remove the “stop” (Figure 12.2E). An alternative strategy, usually referred to as a double-inverse orientation (DIO) or Flex construct, is the use of two different pairs of lox sites to cause recombination that leads to Cre-dependent expression of a transgene. Investigators can achieve temporal control of Cre/lox technology by fusing the gene encoding Cre recombinase with the gene encoding the estrogen receptor (ER). The ER is normally sequestered in the cytoplasm, but it translocates to the nucleus after binding to its ligand. The ER sequence fused to Cre is altered so it preferentially binds to the chemical tamoxifen, an estrogen antagonist, rather than endogenous estrogen present in an animal. In baseline conditions, Cre-ER fusion proteins are sequestered in the cytoplasm and unable to access lox sites in the DNA located in the nucleus. If a scientist injects tamoxifen into a mouse, it will bind to Cre-ER, causing it to translocate to the nucleus and recombine DNA (Figure 12.2F). Another method for temporal and spatial control over Cre-mediated excision is the direct introduction of Cre transgenes through gene delivery methods as described in Chapter 11. Cre transgenes can be delivered to a floxed animal through electroporation or viral transduction, or to cells cultured from a floxed animal using in vitro methods of gene delivery. The Flp/Frt System The Flp/Frt system is a site-specific recombination system analogous to the Cre/lox system just described. The enzyme Flippase recombinase (Flp), originally derived from yeast, recognizes specific 34 base-pair DNA sequences called Flippase recognition targets (Frts), excising DNA between identical Frt sites (Figure 12.2G) and inverting DNA between inversely oriented Frt sites (Figure 12.2H). This system is used in both flies and mice. The Tet-off/Tet-on System The Tet-off/Tet-on system is a tool used in the fly and mouse to temporally control transgene expression. In the Tet-off system, a promoter drives the expression of the tetracycline transactivator (tTA) protein. This protein binds to another DNA sequence called the tetO operator, which regulates the expression of a transgene. Thus, in normal conditions, the transgene expression is regulated by the promoter that regulates tTA. However, a scientist can feed or inject a derivative of the antibiotic tetracycline called doxycycline into these transgenic mice. Doxycycline binds to tTA and renders it incapable of binding to tetO. Therefore, expression of the transgene can be reversibly inhibited with the addition of doxycycline (Figure 12.2I). Making and Using Transgenic Organisms Chapter | 12  265 The Tet-on system works in the opposite way. A reversed tTA (rtTA) is only capable of binding to tetO when doxycyline is present. Therefore, the addition of doxycyline causes transcription of the transgene (Figure 12.2J). Like the Tetoff system, the effect of doxycycline is reversible. These systems have gained popularity for their reversibility and temporal control, as well as for the ability to control levels of transgene activation. Increasing the concentration of doxycycline in the Tet-off system gradually lowers and then shuts off transgene expression, while increasing concentrations of doxycycline in the Tet-on system gradually increases transgene expression. MAKING TRANSGENIC ORGANISMS After using recombinant DNA technology to produce a transgenic construct, an investigator must deliver the construct to recently fertilized eggs so that it can integrate into the genomes of the offspring. This section describes the process of creating transgenic mice, flies, worms, and other animals. Making Transgenic Mice To make a transgenic mouse, a foreign piece of DNA is injected into the nucleus of a newly fertilized egg, where it can randomly incorporate into the genome. After the egg develops into a fetus and eventually a postnatal offspring, the transgene can become stably expressed and transferred to future offspring. In many institutions, individual investigators do not inject the transgenic construct into eggs themselves but submit their constructs to a specialized transgenic animal facility that has expertise in creating transgenic mice. As opposed to flies or worms, the harvesting and injection of recently fertilized eggs is a technically challenging process that most laboratories choose to outsource. The transgenic construct is injected as a linear piece of DNA into the larger male pronucleus of a recently fertilized mouse egg (Figure 12.4). A pronucleus is the haploid nucleus of either the egg or sperm, both present in the fertilized egg prior to their fusion. This is the stage at which insertion of foreign DNA into the genome can occur. Because the embryo is at the one-cell stage, the incorporated gene is replicated and can end up in nearly all of the animal’s cells in adulthood. For the transgene to be heritable, it must be incorporated into the germline (eggs or sperm). After several embryos have been injected with the transgene, they are subsequently injected into a pseudopregnant mouse (Figure 12.5). The term pseudopregnant refers to the fact that the female is not actually pregnant but has been “tricked” into thinking she is pregnant by mating with a sterile male mouse. This is done to produce the hormones and physiological changes necessary to develop the embryo into a viable fetus. The mouse eventually carries the embryos to term, and transgenic mice are born. Each transgenic facility uses different procedures, but often 2–4 pseudopregnant females are used to produce 20–30 transgenic offspring. 266  Guide to Research Techniques in Neuroscience These transgenic offspring are called founders, as they represent the first generation of a new line of mice. However, not all offspring will have successfully incorporated the transgene, and those that do may not express the transgene in the desired cells and tissues. Therefore, once mice are born, the investigator must detect which mice have incorporated the transgene by genotyping the animals with a Southern blot and/or polymerase chain reaction (PCR; Chapter 10). Southern blots are required to fully characterize the number of transgene copies and verify FIGURE 12.4  Injection of a transgenic construct in mice.  To produce a transgenic mouse, a scientist microinjects a transgenic construct into one of two pronucei in a recently fertilized egg. Fertilized eggs injected into oviduct of pseudopregnant female mouse Some offspring might properly express the transgene, others will not FIGURE 12.5  Injection of embryos to make transgenic mice. After injecting embryos with a transgenic construct, these embryos are subsequently injected into the oviduct of pseudopregnant female mice. Each female typically produces 6–10 offspring. However, only a fraction of the offspring express the transgene. Making and Using Transgenic Organisms Chapter | 12  267 the integrity of the transgene sequence. However, PCR of genomic tail DNA is a rapid way to reveal if the transgene is present and is the standard method for genotyping future offspring.Any mice that test negative for the transgene are discarded. Mice that test positive for the transgene are mated with wild-type mice so their offspring can be further investigated. These mice are systematically examined for the presence of the transgene and rigorously tested to ensure that the gene is expressed in the appropriate tissues. Because copy number and integration position can affect the expression pattern of a transgene, multiple founder lines are characterized. The total time necessary to produce and analyze the validity of a transgenic mouse is usually about 6–9 months, including the time necessary for breeding, gestation, development, and analysis of genotype and phenotype in the laboratory. Making Transgenic Flies The process of making a transgenic fly takes advantage of specialized, endogenous segments of DNA called transposons (also called transposable elements) that can move from one position to another within the genome. These mobile genetic sequences have only modest target site selectivity within the genome and can therefore insert themselves into random genomic locations. In transposition, an enzyme called transposase acts on specific DNA sequences at the end of a transposon, first disconnecting it from the flanking DNA and then inserting it into a new target DNA site. The most commonly used transposons in Drosophila are called P elements. To make a transgenic fly, a DNA construct is first inserted between the two terminal sequences of the Drosophila P element (Figure 12.6). The terminal sequences enable the P element to integrate into Drosophila chromosomes when the P element transposase enzyme is also present. The investigator injects the transgenic construct into a very young fruit fly embryo along with a separate plasmid containing the gene encoding the transposase (Figure 12.7). If this technique is performed correctly, the injected gene enters the germ line as the result of a transposition event. Making Transgenic Worms Transgenic techniques in Caenorhabditis elegans are conceptually similar to transgenic techniques in flies. An important difference is that worms are hermaphrodites and capable of carrying their own fertilized eggs. A transgene is injected into the gonad of a young adult hermaphrodite at the time the embryos are dividing. The gonad runs the length of the worm on the dorsal side, and, with practice, an investigator can become skilled at such injections (Figure 12.8). The transgene can either be carried as an extrachromosomal array (although worms will not necessarily carry the gene in the next generation) or, if the worm is subjected to gamma or ultraviolet irradiation, it can be integrated into a chromosome to make transgene expression more stable and less variable. 268  Guide to Research Techniques in Neuroscience To ensure that the transgenic procedure was successful, investigators often use a co-injection marker that allows for easy identification of transgenic animals. For example, the investigator can co-inject the transgene with another construct encoding GFP. To avoid potential interference with the transgene of study, the co-injection marker can be controlled by a promoter that causes expression in a nonoverlapping, different cell type. GFP can then serve as a positive control to verify that the transgenic manipulation was successful. Transgenic techniques in C. elegans are good for studying gain-of-function (sufficiency) effects, rescuing mutant lines, expressing reporter genes, and expressing RNAi constructs. Because there are many promoter elements that allow targeting to individual neurons, C. elegans is a powerful genetic model system used throughout neuroscience to study the genetic basis of development and behavior. P element 5’ 3’ Transposase DNA ligates together P element ligates into new DNA region FIGURE 12.6  P elements.  The transposase enzyme splices a P element out of its genetic environment (the genome of a transgenic construct), inserting it into a new location in the genome. Anterior Posterior Trachea FIGURE 12.7  Injection of a transgenic construct in flies.  To make transgenic flies, a scientist microinjects a transgenic construct into the posterior end of a fresh Drosophila embryo. Making and Using Transgenic Organisms Chapter | 12  269 Making Other Transgenic Organisms Although mice, flies, and worms are the most frequently used genetic model organisms, in theory, a scientist can insert a transgene into any species by correctly injecting DNA into a recently fertilized egg. For example, some laboratories have created transgenic rats to study behavioral tasks that are too sophisticated for mice. Different species of fish have stably expressed transgenes (including “GloFish,” the first transgenic organism to be sold as pets). Cows, pigs, goats, and sheep have also expressed transgenes. A transgenic cow has been created that can produce human protein–enriched milk, a more nutritionally fortified product than regular bovine milk that can be provided to babies or the elderly with special nutritional needs. Other transgenic cows produce milk with less lactose or lower cholesterol. Products such as insulin, growth hormone, blood anticlotting factors, and so forth can also be produced in transgenic animals and collected through animal milk. Thus, transgenic technology is useful not only for studying the biology of organisms, but also for producing nutritional resources and biological reagents that can benefit the health of humans. CONCLUSION This chapter has surveyed the amazing variety of transgenes useful for studying neurons and neural circuits, as well as describing how scientists can stably insert these genes into the genomes of a variety of model organisms. Laboratories continually clone and create new transgenes each year that add important tools to the genetic toolkit. For example, the DREADD andArcLight transgenes have only been created and reported within the past few years, and now multiple laboratories are expressing these genes in specific cell types to answer important questions about the nervous system. The next chapter describes methods that complement transgenic technologies by targeting genes to specific locations in the genome. These methods can be used to create transgenic animals, as well as to replace functional genes with noncoding genes to produce genetic “knockouts.” These technologies further expand the neuroscientist’s toolkit of genetic techniques and allow for genetic access to specific cell types in the nervous system. Anterior Posterior Dorsal Ventral Embyros FIGURE 12.8  Injection of a transgenic construct in worms.  To make transgenic worms, a scientist microinjects a transgenic construct into a worm’s dorsal surface at the time embryos are dividing. 270  Guide to Research Techniques in Neuroscience SUGGESTED READING AND REFERENCES Books Behringer, R., Gertsenstein, M., Nagy, K., & Nagy, A. (2013). Manipulating the mouse embryo: A laboratory manual (4th ed.). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Greenspan, R. J. (2004). Fly pushing: The theory and practice of Drosophila genetics (2nd ed.). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Hofker, M. H., & van Deursen, J. (Eds.). (2010). Transgenic mouse methods and protocols (Methods in molecular biology). New York, NY: Humana Press. Pinkert, C. A. (2014). Transgenic animal technology: A laboratory handbook (3rd ed.). San Diego: Academic Press. Review Articles Barth, A. L. (2007). Visualizing circuits and systems using transgenic reporters of neural activity. Current Opinion in Neurobiology, 17, 567–571. Duffy, J. B. (2002). GAL4 system in Drosophila: a fly geneticist’s swiss army knife. Genesis, 34, 1–15. Luo, L., Callaway, E. M., & Svoboda, K. (2008). Genetic dissection of neural circuits. Neuron, 57, 634–660. Scanziani, M., & Hausser, M. (2009). Electrophysiology in the age of light. Nature, 461, 930–939. Zhang, F., Aravanis, A. M., Adamantidis, A., de Lecea, L., & Deisseroth, K. (2007). Circuit-breakers: Optical technologies for probing neural signals and systems. Nature Reviews Neuroscience, 8, 577–581. Primary Research Articles—Interesting Examples from the Literature Armbruster, B. N., Li, X., Pausch, M. H., Herlitze, S., & Roth, B. L. (2007). Evolving the lock to fit the key to create a family of G protein-coupled receptors potently activated by an inert ligand. Proceedings of the National Academy of Sciences of the United States of America, 104, 5163–5168. Boyden, E. S., Zhang, F., Bamberg, E., Nagel, G., & Deisseroth, K. (2005). Millisecond-timescale, genetically targeted optical control of neural activity. Nature Neuroscience, 8, 1263–1268. Conti, B., Sanchez-Alavez, M., Winsky-Sommerer, R., Morale, M. C., Lucero, J., Brownell, S., et al. (2006). Transgenic mice with a reduced core body temperature have an increased life span. Science, 314, 825–828. Garner, A. R., Rowland, D. C., Hwang, S. Y., Baumgaertel, K., Roth, B. L., Kentros, C., et al. (2012). Generation of a synthetic memory trace. Science, 335, 1513–1516. Gong, S., Zheng, C., Doughty, M. L., Losos, K., Didkovsky, N., Schambra, U. B., et al. (2003). A gene expression atlas of the central nervous system based on bacterial artificial chromosomes. Nature, 425, 917–925. Liu, X., Ramirez, S., Pang, P. T., Puryear, C. B., Govindarajan, A., Deisseroth, K., et al. (2012). Optogenetic stimulation of a hippocampal engram activates fear memory recall. Nature, 484, 381–385. Livet, J., Weissman, T. A., Kang, H., Draft, R. W., Lu, J., Bennis, R. A., et al. (2007). Transgenic strategies for combinatorial expression of fluorescent proteins in the nervous system. Nature, 450, 56–62. Marella, S., Fischler, W., Kong, P., Asgarian, S., Rueckert, E., & Scott, K. (2006). Imaging taste responses in the fly brain reveals a functional map of taste category and behavior. Neuron, 49, 285–295. Making and Using Transgenic Organisms Chapter | 12  271 Protocols Berkowitz, L. A., Knight, A. L., Caldwell, G. A., & Caldwell, K. A. (2008). Generation of stable transgenic Caenorhabditis elegans using microinjection. JoVE, 18. http://dx.doi.org/10.3791/833. http://www.jove.com/index/Details.stp?ID5833. Wu, J. S., & Luo, L. (2007).A protocol for mosaic analysis with a repressible cell marker (MARCM) in Drosophila. Nature Protocols, 1, 2583–2589. Yang, X. W., & Gong, S. (2005). An overview on the generation of BAC transgenic mice for neuroscience research. Current Protocols in Neuroscience, Chapter 5, Unit 5.20. http://dx.doi.org/ 10.1002/0471142301.ns0520s31. Websites Gensat Project: http://www.gensat.org/index.html. Human Genome ed. site: http://genome.wellcome.ac.uk/doc_wtd021044.html. The Jackson Laboratory Cre Repository: http://cre.jax.org/. Transgenic Fly Virtual Lab: http://www.hhmi.org/biointeractive/clocks/vlab.html. 273 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00013-7 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 13 Manipulating Endogenous Genes After Reading This Chapter, You Should be Able to: l Discuss methods of manipulating genomes in specific locations l Compare classical and nuclease-based gene targeting technologies l Compare methods that disrupt endogenous gene products Techniques covered: l Classical gene targeting: knockouts, knockins, conditional knockouts (Cre/lox, Flp/Frt) l Nuclease-based gene targeting: meganucleases, zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), CRISPR/Cas9 system l Disrupting gene products: RNA interference (RNAi), morpholinos, dominant negatives In Chapter 12, we described the process of creating a transgenic animal, an organism that integrates a piece of foreign DNA into its genome that can pass the DNA to future offspring. Using traditional transgenic technologies, the foreign DNA randomly integrates somewhere in the genome on any of the animal’s chromosomes. It is often desirable to target a DNA construct to a specific location in the genome, either to insert genetic material in a specific locus or to disrupt the function of an endogenous gene. Gene targeting is the process of incorporating a DNA construct into a specific locus in an animal’s genome. In a genetic knockout animal, gene targeting is used to replace a functional gene with a nonfunctional sequence. In a knockin animal, gene targeting is used to insert a DNA construct into the genome at a specific location. More complicated gene targeting strategies allow genes to be knocked out or knocked in to specific cell types at specific times. The purpose of this chapter is to describe methods used to deliberately target specific regions of the genome and/or interfere with functional gene products. These methods are useful for reverse genetics experiments in which a scientist tests the hypothesis that a specific gene is necessary for a defined phenotype. Gene targeting techniques have existed for about three decades, and newer cutting-edge methods are allowing for gene targeting in a wider variety of model organisms. In addition to using gene targeting technologies to manipulate the genome, a scientist can deliberately disrupt the expression of a gene at different stages of the transcription/translation process (Figure 13.1). For example, 274  Guide to Research Techniques in Neuroscience using RNA interference (RNAi) technology, it is possible to degrade mRNA transcripts, preventing the translation of the gene into protein and knocking down gene expression. Using dominant negative methods, gene function can be blocked by interfering with the protein product. Each of these methods provides scientists with powerful tools for perturbing the expression of endogenous genes to determine the effect on an animal’s anatomy, physiology, and behavior. CLASSICAL GENE TARGETING METHODS The technology necessary to target DNA constructs to specific regions of the genome was developed in the early 1980s. The scientists who developed these methods, Mario Capecchi, Sir Martin Evans, and Oliver Smithies, shared the 2007 Nobel Prize in Physiology or Medicine for discovering and pioneering the procedure for introducing specific gene modifications in mice. These techniques continue to be widely used to create knockout and knockin animals; however, these techniques are almost exclusively used in mice. Most loss-offunction mutations in flies and worms result from forward genetic screens (Chapter 9). Gene targeting technology is not widely used in other mammalian species because of the need to culture and inject embryonic stem cells during the targeting procedure, something that turns out to be surprisingly difficult for rats and other nonmurine species. Therefore, the classical gene targeting techniques described below are almost always used to target genes in mice. New methods, such as CRISPR/Cas9, allow for genome editing in theoretically any species, including humans. These techniques are described later in the chapter. DNA mRNA Protein Translation Transcription Gene targeting (knockouts, knockins, conditional knockouts) RNA interference, morpholinos Dominant negatives Stage of information flow Methods of manipulation FIGURE 13.1  Methods of genetically perturbing endogenous genes and proteins.  A scientist can manipulate gene and protein function at each stage of information processing. Gene-targeting techniques can be used to alter the genome; RNA interference and morpholino techniques can be used to functionally block mRNA transcripts; and other techniques, such as expression of dominant negative constructs, can block protein function. Manipulating Endogenous Genes Chapter | 13  275 Knockout Mouse In a knockout mouse, gene targeting is used to replace a functional gene with a nonfunctional sequence, effectively deleting the gene from the genome. There are five major steps in the creation of a knockout mouse: (1) using recombinant DNA technology to create a targeting construct; (2) injecting this construct into embryonic stem (ES) cells and selecting cells that have properly incorporated the construct into the genome; (3) inserting successfully targeted ES cells into a blastocyst; (4) inserting the blastocyst into a female mouse; and (5) breeding the offspring of this mouse until a successful knockout is achieved. This lengthy process can require 1–2 years of work and breeding time before any mice are ready to use for experiments. Creating a Targeting Construct The first step in making a knockout mouse is to use recombinant DNA technology to produce an appropriate targeting construct that will replace the gene of interest. Replacing the target gene with the genetically engineered construct relies on the natural phenomenon of homologous recombination, in which regions with strong sequence similarity can exchange genetic material (Figure 13.2A). In nature, homologous recombination produces new combinations of DNA sequences during chromosomal crossover in meiosis, resulting in enhanced genetic variation in a population. In the lab, homologous recombination is used to replace endogenous genomic DNA sequences with targeting constructs containing homologous sequences of DNA (Figure 13.2B). To use homologous recombination, a scientist designs a targeting construct with the replacement gene flanked by long segments of genomic DNA that are homologous to sequences in the endogenous mouse genome. These homology arms provide targeting specificity: the region between the two homologous segments replaces the target gene in the organism. Homologous end sequences are at least 1 kb (kilobase) in length, usually longer. A scientist can easily identify (A) (B) FIGURE 13.2  Homologous recombination.  (A) Homologous recombination is a natural phenomenon in which genetic sequences with high sequence similarity can exchange genetic material, as occurs with homologous chromosomes during meiosis. (B) In the lab, a scientist can exploit homologous recombination to target a transgene to a specific locus in the genome. If the transgene is flanked by regions homologous with the locus, homologous recombination can allow the transgene to recombine into the genome. 276  Guide to Research Techniques in Neuroscience and choose homologous sequences surrounding a gene of interest by looking up their sequences in a mouse genome database. The coding region in the middle of the construct contains a mutated version of the endogenous gene that no longer produces a functional protein product. This is often achieved by deleting the coding region within an exon. The targeting construct also contains a selection cassette, typically with two additional genes: a positive selection gene placed between the homology arms and a negative selection gene placed outside the homology arms (Figure 13.3). The positive selection gene makes cells resistant to a drug, such as neomycin, puromycin, or hygromycin. Only cells that have incorporated the targeting construct into their genome will survive if a scientist exposes the cells to the drug. The negative selection gene, often thymidine kinase or diphtheria toxin A, can be used to kill cells that randomly integrated the targeted construct somewhere in the genome other than the desired locus. These genes allow for quality control while selecting properly targeted cells if and when homologous recombination takes place. Injecting and Selecting ES Cells After the scientist makes the targeting construct, the DNA is microinjected or electroporated into ES cells (Chapter 11). Because culturing and delivering DNA to ES cells is technically challenging, most scientists outsource this step to a specialized genetic engineering facility. Technicians inject dozens or even hundreds of cells to enhance the chances of finding a cell that has properly incorporated the targeting construct. When the targeting construct is delivered to ES cells, three potential outcomes can occur: (1) it correctly recombines and integrates into the cell’s genome; (2) it integrates with the cell’s genome but in the wrong location; or (3) it fails to integrate into the genome at all. To select ES cells that have successfully recombined the targeting construct at the correct location, the technician utilizes the positive and negative selection genes in the targeting construct (Figure 13.4). For example, a selection cassette encoding the neomycin resistance gene can be used to ensure that the construct has integrated into the genome. After introducing the targeting construct, ES cells are exposed to neomycin, which kills all cells that do not express the resistance gene. Therefore, only ES cells that have incorporated the construct survive. Including the thymidine kinase (TK) gene in the selection cassette ensures that the targeting construct has inserted itself in the correct location. ES cells that have incorrectly incorporated the targeting construct into the genome will express the tk gene and die in the presence of a chemical called gancyclovir. Gancyclovir is harmless to normal cells, but it becomes highly toxic when phosphorylated by the tk protein product. Because Homology HomologyReplacement Neo TK FIGURE 13.3  A generic gene targeting construct. (A) The replacement DNA sequence is flanked by two regions of DNA that are homologous to the genomic locus where the DNA is to be inserted. The neomycin resistance cassette (Neo) and thymidine kinase gene (TK) are also placed in the construct to select embryos that have properly integrated the construct into the genome. Manipulating Endogenous Genes Chapter | 13  277 the tk gene is outside the region of homology, if homologous recombination occurs, the gene will be lost. Therefore, neomycin and gancyclovir offer positive and negative selection measures, respectively, to ensure that the targeting construct has successfully inserted itself into the genome at a precise location. Statistically, homologous recombination is an extremely rare event, and only about 1% of the ES cells will survive these positive and negative selection measures. Injecting ES Cells into a Blastocyst After several positive ES cells are identified, they are injected into a blastocyst. The blastocyst is an early embryonic structure that follows the fertilization of an egg (Figure 13.5). There are two cell types within a blastocyst: the trophoblast, which forms the outer shell of cells around the blastocyst and later FIGURE 13.4  Selecting embryonic stem cells. The gene targeting construct is delivered into mouse ES cells by microinjection or electroporation. These cells are then exposed to the compounds neomycin and gancyclovir. Neomycin will kill cells that do not express the neomycin resistance gene, cells that did not incorporate the targeting construct into the genome. Gancyclovir is harmless to cells unless the tk gene is expressed, which can occur only if the targeting construct integrates at the wrong genomic location. ES cells that survive treatment with both compounds are further tested for expression of the construct and then injected into blastocysts. Microinjection or electroporation Add neomycin and gancyclovir Targeting construct Embryonic stem cells Surviving embryonic stem cells Blastocyst TrophoblastInner cell mass FIGURE 13.5  Injecting ES cells into a blastocyst. A blastocyst is an early embryonic stage resembling a sphere of different cell types. The trophoblast forms the outer sphere of cells and gives rise to the placenta. The inner cell mass contains embryonic stem cells, which give rise to all cells and tissues in the adult organism. A scientist injects ES cells containing the targeting construct into the inner cell mass. 278  Guide to Research Techniques in Neuroscience develops into the placenta, and the inner cell mass, a collection of embryonic stem cells that eventually give rise to all cells and tissues in the adult organism. ES stem cells that have survived the positive and negative selection process are carefully injected into the inner cell mass. The future tissues that develop from this blastocyst will be chimeric, with some tissues derived from the inner cell mass of the host blastocyst and some tissues derived from the injected ES cells. Inserting a Blastocyst into a Female Mouse The next step is to insert the blastocyst into the uterus of a pseudopregnant female mouse (Figure 13.6A). The pseudopregnant female is not actually pregnant but has been psychologically tricked into believing she is pregnant by mating with a sterile male mouse. This is done so that the female’s body produces the hormones and physiological changes necessary to carry the embryo through to a viable fetus. After injection, the mouse carries the embryos to term and gives birth to offspring. Breeding Offspring Mice Once mice are born, it is necessary to determine which mice have incorporated the knockout construct. Often, the blastocyst is obtained from a white strain of mice, while the ES cells are derived from a brown strain (or vice versa). This allows the scientist to visually determine the success of the ES cell injection into the blastocyst: if the offspring of the host mother are entirely white, then it is likely that no injected cells successfully combined with the blastocyst. If the offspring have both brown and white patches of fur, the knockout ES cells were successfully incorporated into the blastocyst (Figure 13.6B). Offspring mice will never be entirely brown because the blastocyst is composed of ES cells in the inner cell mass prior to injection of the knockout ES cells, and both cell types continuously divide to produce the animal’s tissues. These mice are referred to as chimeric mice because they have a combination of tissues from the blastocyst-derived strain and the ES cell-derived strain. Some investigators use these chimeras to investigate the cell-autonomous role of the gene of interest by comparing the phenotypes of mutant cells to nonmutant cells. However, more steps must be taken to produce a full knockout. The ultimate hope of the scientist is that the chimeric mice carry the knockout sequence in their germ cells, the cells that undergo meiosis to produce either sperm in males or eggs in females. Only genetic material in the germ cells (as opposed to somatic cells) will be passed on to future offspring. Therefore, some chimeras will be able to pass along the knockout sequence in the germline; those chimeras that do not carry the knockout sequence in germ cells will never pass along the knockout sequence. After visually identifying chimeric mice, the scientist breeds them with wild-type mice to produce nonchimeric heterozygous mice with one wild-type copy of the gene and one mutant copy (Figure 13.6C). Offspring will either be all white (no germline transmission in the chimera) or all brown. The final step is to breed heterozygous mice with each other to produce full homozygous knockout animals (Figure 13.6D). This breeding scheme should Manipulating Endogenous Genes Chapter | 13  279 result in Mendelian inheritance, with one-quarter homozygous wild-type offspring, one-quarter homozygous knockout offspring, and one-half heterozygous offspring. However, knocking out certain genes can lead to an embryonic lethal phenotype, in which a homozygous knockout mouse fails to survive gestation (A) (B) (C) (D) X X +/+/- +/- 25% of progeny: +/+ 50% of progeny: +/- 25% of progeny: -/FIGURE 13.6  Generating a genetically targeted mouse.  (A) A scientist injects blastocysts into a pseudopregnant female mouse. (B) The mouse produces about 8–10 offspring, with tissues that developed from the endogenous ES cells residing within the blastocyst and the injected ES cells. The injected ES cells are typically removed from an animal with a brown coat color, allowing visualization when injected into mice with a white coat color. (C) Chimeric mice are bred with wild-type mice. Any offspring mice with a brown coat color stably express the targeting construct and are heterozygous for the endogenous gene. (D) To produce a full knockout, heterozygous mice are bred with each other. 280  Guide to Research Techniques in Neuroscience due to improper development. In these cases, scientists may still be able to examine the gene’s role in early development. If the scientist is interested in studying the gene’s role in postnatal animals, there are some alternative approaches. First, the scientist could compare homozygous wild-type animals with heterozygous animals to determine if having a single mutant version of the gene exerts any phenotypic effect. Alternatively, the scientist could employ a conditional knockout strategy in which the gene of interest is knocked out only in a particular cell type or at a particular time. Conditional knockouts are described later in the chapter, as are other alternatives to genetic knockouts, including RNAi or dominant negative strategies to disrupt gene function in a postnatal animal. In short, it takes four generations of animals to produce the first knockout mice of a colony: the pseudopregnant foster mothers, the chimeric offspring, the first heterozygous offspring of the chimeras, and the final homozygous knockout animals. Therefore, producing a knockout can require a full year or more because of the multiple generations of mice that must be bred and maintained. Validating a Knockout Mouse There are many methods that can be used to validate an animal’s genotype, and a thorough investigation uses multiple techniques (Figure 13.7). A Southern blot (Chapter 10) is a technique used to detect the presence of specific sequences of DNA from the genome. Samples from wild-type, heterozygous, and knockout mice can be compared to identify the presence or absence and the length of specific DNA sequences at the gene locus. Alternatively, a scientist may use a polymerase chain reaction (PCR) (Chapter 10) genotyping strategy and design primers around the gene’s locus. Well-designed PCR primers will produce different-sized PCR products for wild-type, heterozygote, and knockout genomic DNA. Usually, a scientist performs a Southern blot to validate a knockout and then uses PCR for routine genotyping, as PCR is a simpler and quicker method. PCR and Southern blotting techniques are the absolute gold standard in assessing the genotype of an animal. A scientist may also perform complementary tests, such as in situ hybridization and/or immunohistochemistry (Chapter 6) to confirm that there is no gene or protein expression in the tissues where they are naturally expressed. A western blot (Chapter 15) can also demonstrate the absence of a protein product. Using Knockout Mice in Experiments A knockout mouse is a powerful tool for investigating the necessity of a gene for biologicalprocesses,suchasdevelopmentorbehavior.However,scientistsmustbe careful to perform the appropriate control experiments to make valid conclusions when using this tool. Regardless of whether the scientist studies an anatomical, behavioral, or electrophysiological phenotype, knockout mice should be directly compared with their wild-type littermates. This strategy reduces potentially Manipulating Endogenous Genes Chapter | 13  281 confounding differences between mice of different genetic backgrounds. Some studies compare wild type, heterozygous, and knockout littermates to demonstrate a dose-dependent role for the relative amount of a gene product, with a heterozygote phenotype intermediate between wild-type and knockout. If no phenotype is immediately discernable, scientists must consider the possibility of a genetic compensatory mechanism. A knockout animal may compensate for the loss of a gene by altering the expression levels or pattern of another gene or adapting in some other way so that the resulting phenotype is still normal. Often, the phenotypic consequence of a genetic knockout is subtle, and more refined assays may be required to identify a difference between knockout and wild-type animals. Finally, if a knockout mouse does exhibit a phenotype, the scientist must be careful about interpreting the gene’s role. For example, if knocking out a specific (B) (A) Endogenous locus Targeting vector Integration of vector in genome Exon 1 neo thymidine kinase BglII BglII BglII BglIIBglII 4.8 kb 2.8 kb 1 2 3 4 5 4.8 kb 2.8 kb +/- -/- +/- -/-+/+ FIGURE 13.7  Analyzing genotype after gene targeting.  (A) In the literature, scientists typically depict a gene targeting strategy by showing the endogenous locus, the targeting vector, and the new genomic sequence. In this example, scientists replaced the first exon of a gene with a new genetic sequence. Note the presence of the BglII restriction enzyme sites. In the endogenous locus, digestion of DNA with BglII results in a 2.8 kb fragment. In a mouse in which gene targeting occurred, the fragment will be 4.8 kb in length. The small black box depicted just beneath the endogenous locus represents a probe that can be used to analyze the size of the fragment. (B) In a Southern blot experiment, a scientist digests the DNA with a restriction enzyme, in this case BglII, and uses a probe to detect a specific genetic sequence. Using this method, it is possible to discriminate between fragments that are 2.8 and 4.8 kb in length, allowing the scientist to determine the genotype of individual mice. 282  Guide to Research Techniques in Neuroscience gene causes an animal to lose its ability to respond to painful stimuli, it is incorrect to immediately conclude that the gene’s function is to regulate pain. The only conclusion that can be reached is that the gene is necessary for the animal to report a painful stimulus. More rigorous tests must be performed to determine whether the gene is important for tactile sensation in general, for normal development of spinal cord neurons, for correct axonal targeting to the periphery, etc. A full understanding of a gene’s function comes from phenotypic analysis combined with investigations into the specific mechanisms of the gene’s contributions to the phenotype. Knockin Mouse A knockin mouse is a genetically engineered mouse in which a genetic sequence is added at a specific locus in the genome. This differs from a standard transgenic mouse in that the DNA sequence is targeted to a particular location in the genome, rather than being randomly integrated. The procedure to make a knockin mouse uses the same homologous recombination strategy as the procedure just described for making a knockout mouse. However, a knockin mouse serves a very different purpose: to add functional genomic sequences, whether they are functional genes or promoter sequences. There are various reasons why a scientist may choose to make a knockin mouse: l To replace a functional gene with a tagged version of the same gene. The resulting reporter fusion protein can then be used to visualize gene expression patterns in fixed or living tissue. l To replace the gene of interest with a mutated version to mimic a human disease. For example, if a scientist hypothesizes that the cause of a human disorder is a specific mutation in a gene, the scientist could produce a mouse model of the disease that carries this mutation in the homologous mouse gene. l To replace a regulatory region of DNA to investigate the role of a specific promoter or repressor element in the regulation of a gene product. As with knockout mice, scientists should compare knockin animals with their wild-type littermates during experiments. Conditional Knockout Mouse It is often desirable to disrupt a gene only in a particular cell-type or at a particular time. For example, knocking out a gene in the entire organism could be lethal at the embryonic stage, but if the ablation is restricted to a specific tissue or cell-type or if the ablation occurs after the animal reaches adulthood, investigators may be able to avoid the lethality. In these scenarios, a scientist may choose to make a conditional knockout mouse in which a gene is removed from the genome in a precise spatial or temporal manner. One of the most common methods of creating a conditional knockout is by using the Cre/lox recombination system (Figure 13.8; also described in Manipulating Endogenous Genes Chapter | 13  283 Chapter 12). Homologous recombination is used to replace an endogenous target gene with a fully functional version of the gene flanked by a pair of short DNA sequences called lox sites (referred to as a floxed sequence). These lox sites do not disrupt the gene they flank, so the phenotype of floxed mice remains normal. Lox sites are specific 34 base pair sequences recognized by a protein called Cre recombinase (or simply Cre) that is not normally expressed in animals. Cre recognizes and recombines lox sites, excising any floxed genetic material. To produce conditional knockout mice, mice expressing a floxed gene are mated with transgenic mice expressing Cre under the control of a specific promoter. Cre catalyzes recombination between the lox sequences, excising the target gene from the genome. Many Cre lines now exist for distinct tissue and neural cell types. An alternative to the Cre/lox system is the Flp/Frt recombination system. This system is similar to Cre/lox, with the enzyme flippase recombinase (Flp) excising Flp recombinase target (Frt) sequences that flank the genomic region of interest. To exert temporal control over a genetic knockout, such as knocking out a gene in adulthood, scientists may use inducible knockout systems. A popular method is to combine the Cre/lox recombination system with the ER/tamoxifen system. The estrogen receptor (ER) is a protein that remains in the cytoplasm unless it binds to its chemical ligand. Cre fused to the ER remains in the cytoplasm, where it cannot act to recombine a floxed gene. The drug tamoxifen binds to the ER receptor, causing the fused Cre-ER protein to translocate into X lox lox Tissue specific promoter lox Cre Target Gene FIGURE 13.8  Producing conditional knockout mice with the Cre/lox system.  A scientist produces (or obtains) a mouse in which a gene of interest is flanked by two lox sites. If this mouse is mated with another mouse expressing Cre recombinase, Cre can recombine the genetic material between the two lox sites and excise the gene from the genome. 284  Guide to Research Techniques in Neuroscience the nucleus to exert its effects and excise the floxed gene. After mating an inducible Cre line of mice with the floxed line of mice, investigators control the timing of the knockout through the timing of tamoxifen delivery. NUCLEASE-BASED GENE TARGETING METHODS As mentioned above, one limitation to using classical gene targeting technologies is that they are only effective in mice. Attempts to use similar techniques (injecting constructs with homology arms into embryonic stem cells) in other model organisms have proven unsuccessful or inefficient, at least partially due to the difficulty of culturing ES cells in non-murine species. Even using mice, the process of creating a genetic knockout or knockin can require multiple years of laborious work. However, within the past few years, new approaches have been developed that show outstanding promise for gene editing in a wide variety of model organisms. Modern gene editing tools attempt to create a double-strand break at a precise location in the genome using a protein called a nuclease (Figure 13.9), an enzyme that cleaves nucleic acids. The nuclease is directed to a specific target in the genome by interacting with a guide strand of DNA or RNA. Nucleaseinduced double-strand breaks can be repaired by two different processes that each take advantage of endogenous cellular machinery. In nonhomologous Genomic DNA Nuclease-mediated double-strand break Homology-directed repair (HDR) Nonhomologous end joining (NHEJ) Targeting construct Inserted or modified DNA sequence Insertion or deletion mutations FIGURE 13.9  Conceptual overview of nuclease-mediated genome targeting. A nuclease causes a double-strand break at a specific site of the genome. Following the break, a scientist can employ either nonhomologous end joining (NHEJ) pathways to induce a deletion or insertion or homology-directed repair (HDR) pathways to introduce foreign DNA. Manipulating Endogenous Genes Chapter | 13  285 end-joining (NHEJ), naturally occurring DNA break repair processes rejoin the broken ends, often including an insertion or deletion mutation. These mutations can lead to a frameshift mutation of a gene, introducing a premature stop codon and effectively silencing expression. Alternatively, in a process called homology-directed repair (HDR), a scientist introduces a DNA construct with homology arms that have the same sequence as the broken DNA strands. Then the naturally occurring process of homologous recombination swaps endogenous DNA for the engineered DNA, similar to the process described above to create a knockout or knockin mouse. Many nuclease systems have been developed to target specific locations in the genome: l Meganucleases. Meganucleases are restriction enzymes (Chapter 10) that recognize longer DNA sequences (typically 20–40 bp) than typical restriction enzymes (typically 6–8 bp). Artificially engineered versions of naturally occurring meganucleases have been developed as a tool for genome engineering. Because specific 20–22 bp sequences typically occur only once in the genome, a well-engineered meganuclease should target specific genomic loci. However, despite several important advances, the engineering of meganucleases specific to desired DNA sequences has proven challenging for most research laboratories. l Zinc finger nucleases (ZFNs). ZFNs are artificially engineered restriction enzymes consisting of a zinc finger DNA-binding domain fused to a DNAcleavage domain. Zing finger domains can be engineered to target specific DNA sequences, therefore allowing for site-specific cleavage of the genome. Unfortunately, ZFNs have proven difficult to use for many research laboratories due to off-target effects in which ZFNs cut DNA at undesired sites, as well as challenges with engineering the proteins for new gene targets. l Transcription activator-like effector nucleases (TALENs). Like ZFNs, TALENs are artificially engineered restriction enzymes composed of a DNAbinding domain and a DNA-cleavage domain. The DNA-binding domain of TALENs (called a transcription activator-like effector, or TALE) is derived from Xanthomonas bacteria. TALEs can be engineered to target specific DNA sequences, but often the DNA that encodes TALEs can be quite long. Although simpler to design than ZFNs, it is still difficult to create TALEN constructs using standard recombinant DNA technology techniques (Chapter 10) or to deliver TALEN-coding sequences to cells using common gene delivery strategies (Chapter 11), such as viruses. Taken together, meganucleases, ZFNs, and TALENs have proven useful for many research groups and are still actively being optimized and improved. However, each approach continues to manifest its own set of disadvantages that has prevented widespread use for gene targeting among many research laboratories. Much recent attention has focused on a nuclease system commonly referred to as CRISPR/Cas9. CRISPR stands for “clustered regularly interspaced short 286  Guide to Research Techniques in Neuroscience palindromic repeat” and refers to clusters of short, repeated DNA sequences normally used by some strains of bacteria as a method of protection against foreign nucleic acids. A “Cas” is a CRISPR associated protein, and Cas9 is a nuclease originally derived from the bacterium Streptococcus pyogenes. In this species, the Cas9 nuclease uses templates provided by the CRISPR sequence to cleave foreign DNA. In the lab, the CRISPR/Cas9 system has been adapted for inducing site-specific double strand breaks in the genomes of many model organisms. The most common procedure utilizing CRISPR/Cas9 technology introduces two DNA sequences into cells (Figure 13.10). One DNA sequence encodes the Cas9 protein. The other DNA sequence codes for a guide RNA (gRNA) that will serve as a template for genome targeting. Twenty nucleotides at the 5′ end of the gRNA direct Cas9 to a specific genomic site using standard DNA– RNA complementary base pairing. Importantly, in the genome to be targeted, the 20 nucleotides must lie immediately 5′ of a sequence known as a protospacer adjacent motif (PAM), a three-nucleotide sequence of NGG (N can be any of the four DNA bases followed by two guanine bases). Therefore, Genomic DNA Cas9-mediated double-strand break Guide RNA (gRNA) Cas9+ FIGURE 13.10  The CRISPR/Cas9 nuclease system.  A scientist delivers DNA to a cell encoding the Cas9 protein and a guide RNA (gRNA). The gRNA contains a 20 bp sequence that allows Cas9 to recognize a specific region of the genome and cause a double-strand break. Manipulating Endogenous Genes Chapter | 13  287 a subtle limitation to using CRISPR/Cas9 gene targeting is that the break in the genome must occur next to this PAM sequence; however, as the three-nucleotide NGG sequence is incredibly common, this is almost never a problem. Thus, a scientist can design and deliver a gRNA to induce Cas9-mediated genome breaks for any genomic sequence of 20 base pairs plus the PAM. Once the Cas9 induces a double-strand break, a scientist can use NHEJ or HDR to cause a mutation in the genomic location or introduce foreign DNA. This process can take place in a fertilized egg, allowing for heritable transmission to future offspring. Because transgenic technologies are well established for many species (Chapter 12), the CRISPR/Cas9 allows for gene targeting in virtually any traditional and nontraditional model organism. The CRISPR/Cas9 system has grown quickly in popularity as the nucleasebased gene editing tool of choice and is even being explored as a potential way to treat human genetic diseases. One reason this system has overtaken other nuclease-based systems is its affordability and ease of use. To target specific regions of the genome using ZFNs or TALENs, the DNA-binding domain of the nuclease itself must be engineered, and a functional protein created. To target specific regions of the genome using CRISPR/Cas9, a scientist only needs to create a DNA sequence coding for a gRNA. Now that we have described gene targeting and methods of manipulating function at the genomic level, we survey methods of disrupting a gene by interfering with its products—either the mRNA transcript or the translated protein. DISRUPTING GENE PRODUCTS In addition to disrupting gene expression by directly targeting the genome itself, scientists can also perturb endogenous gene function by interfering with a gene’s mRNA transcript or its translated protein product (Figure 13.1). These techniques effectively inhibit gene function and are tractable in a wide variety of vertebrate and nonvertebrate species. RNA Interference (RNAi) RNA interference (RNAi) is the process of silencing gene expression through endogenous cellular mechanisms that degrade mRNA. Cells naturally use RNAi to regulate normal gene expression, as well as to degrade foreign mRNA transcripts from viruses. Scientists now use this natural process to selectively knock down gene expression and degrade mRNA for a gene of interest. Though knockdown efficiencies do not reach 100%, in good knockdown experiments, only a minimal amount of mRNA remains. The process of RNAi begins when a piece of double-stranded RNA (dsRNA) comes in contact with an enzyme called Dicer (Figure 13.11). Dicer cleaves dsRNA into small duplex fragments, about 23 base pairs long, called small interfering RNA (siRNA). The siRNA-Dicer complex recruits additional cellular 288  Guide to Research Techniques in Neuroscience proteins to form an RNA-induced silencing complex (RISC). RISC interacts with and degrades RNA strands that have a complementary nucleotide sequence to one of the siRNA strands, referred to as the guide strand. Thus, the RNAi process degrades mRNA before it is translated by a ribosome into a protein product. In theory, there are three ways to make use of the RNAi pathway to disrupt a gene: (1) deliver a dsRNA molecule that is homologous to the mRNA transcript to be silenced; (2) deliver synthetic siRNAs (19–23 bp) that will attract the RISC complex to bind to the homologous mRNA transcript; and (3) deliver short hairpin RNA (shRNA), a strand of RNA that contains complementary dsRNA comes into contact with Dicer Dicer cleaves dsRNA into siRNA Dicer/siRNA attract other proteins to form RISC complex RISC targets mRNA with a complementary sequence to siRNA RISC degrades target mRNA FIGURE 13.11  Conceptual overview of RNAi technology. When double-stranded RNA (dsRNA) comes into contact with the Dicer enzyme, Dicer cleaves the RNA into small fragments, creating small interfering RNA (siRNA) molecules. The siRNA-Dicer complex recruits other proteins to form a RISC complex. RISC targets mRNA molecules with a complementary sequence to one of the two siRNA strands, degrading the mRNA and knocking down gene expression. Manipulating Endogenous Genes Chapter | 13  289 sequences at either end that cause the molecule to fold onto itself, producing a hairpin shape (Figure 13.12). Endogenous cellular RNAi machinery can then process shRNA into small dsRNA. The choice of which RNAi molecule to use is partially determined by the model organism, as different animals respond with varied efficacy to these distinct methods. Though dsRNA works well in Drosophila and C. elegans, delivering dsRNA into mammalian cells generally causes a nonspecific interferon stress response, making experiments difficult to interpret. siRNA can knock down a gene of interest, but this is a transient effect, which may be suboptimal for certain experiments. For long-term, stable gene knockdown, scientists deliver shRNA into cells in a way that allows the shRNA to be stably integrated into the genome. This integration can be achieved by making a transgenic organism in which the shRNA sequence is placed under the control of a strong promoter. There are two critical issues to consider in the design of a good RNAi experiment: the effectiveness of the RNA probe and the use of appropriate controls. An effective probe has the ability to knock down the expression of a specific mRNA strand without off-target effects. Usually, a scientist designs multiple RNAi probes against the gene of interest and compares their knockdown efficiency in cell culture before choosing one for an actual experiment. The scientist designs these probes to have sequences unique to the gene of interest and no other genes (this can be done using the BLAST program described in Chapter 9). Other, additional control probes are also necessary for a good RNAi experiment. Typical controls include a probe with a scrambled sequence, as well as a probe against a separate, nonessential gene to show the specificity of the RNAi effect. The most conclusive experiments are not only repeated multiple times but demonstrate that multiple probes against different regions of the same gene result in the same knockdown phenotype. The gold standard of RNAi controls is a phenotypic rescue experiment, in which a version of the knocked down gene that is not targeted by the RNAi probe is delivered together with the RNAi probe to show that it can restore a normal phenotype. The rescuing gene is often an orthologue from another species with the same function but a slightly different genetic sequence. Morpholinos An alternative to RNAi is the use of morpholinos to block proper mRNA translation. Morpholinos are stable, synthetic 22–25 bp antisense oligonucleotide FIGURE 13.12  Short hairpin RNA (shRNA). An shRNA molecule is composed of two complementary sequences separated by a small spacer. The molecule naturally folds into itself, resembling a hairpin shape. Linear mRNA strand shRNA 290  Guide to Research Techniques in Neuroscience analogues designed to complement an RNA sequence. A morpholino and its complementary mRNA will hybridize to each other, preventing a ribosome from accessing the mRNA and translating it into a protein (Figure 13.13). Unlike RNAi, morpholinos do not degrade their target mRNA molecules. This is especially useful in organisms such as Xenopus frogs that do not seem to respond to an RNAi approach. Traditionally, morpholinos have been used in zebrafish, Xenopus, and chicks by injecting them into eggs and producing morphant embryos. Controls used in morpholino experiments are similar to those used for RNAi experiments. Dominant Negatives Dominant negative constructs encode protein variants that interfere with the normal function of the wild-type protein. Most dominant negatives inhibit normal protein function by competing for a substrate or binding partner without performing the protein’s regular function. There are a variety of methods used to design dominant negative molecules. One common method is to engineer a mutation that incapacitates one functional domain of a protein while leaving other regions intact to interact as normal. Therefore, dominant negative molecules will compete with the functional wild-type protein for endogenous binding partners. This approach is often used for transcription factors; the DNA-binding domain is left intact, but the (A) mRNA Ribosome (B) Morpholino FIGURE 13.13  Morpholinos.  (A) During translation, a ribosome binds to an mRNA molecule and translates specific nucleotide sequences into chains of amino acids. (B) A morpholino functionally blocks the mRNA, preventing contact with a ribosome. Manipulating Endogenous Genes Chapter | 13  291 transcriptional regulatory domain is removed. Similarly, dominant-negative receptors may have intact extracellular ligand binding domains but may be missing the cytoplasmic intracellular signaling domain; they compete for the ligand but will not pass on the signal. Another common example is creating a dominant negative subunit of an oligomeric protein (Figure 13.14). The dominant negative subunit can bind the other subunits but prevents the proper function of the protein. CONCLUSION This chapter has surveyed common methods of gene targeting and disrupting the function of specific genes in the genome. Although manipulating genes was once a set of classical techniques only possible in mice, newer technologies like the CRISPR/Cas system and RNAi allow for gene targeting and loss-offunction genetic experiments in a wide variety of model organisms. The development of newer, efficient gene editing tools is an exciting area of research that is sure to allow for even more experimental possibilities in the future. SUGGESTED READING AND REFERENCES Books Appasani, K. (Ed.). (2011). RNA interference technology: From basic science to drug development. Cambridge: Cambridge University Press. Behringer, R., Gertsenstein, M., Nagy, K., & Nagy, A. (2013). Manipulating the mouse embryo: A laboratory manual (4th ed.). New York: Cold Spring Harbor Laboratory Press. Schepers, U. (2005). RNA interference in practice: Principles, basics, and methods for gene silencing in C. elegans, Drosophila, and Mammals. Wenheim, Germany: Wiley-VCH. FIGURE 13.14  Conceptual example of a dominant negative disruption of protein function. (A) In this example, a functional protein is composed of four identical subunits. (B) A transgenic construct produces a slightly mutated subunit that retains the ability to form an oligomeric protein but blocks the protein’s function. (A) Protein with four functional subunits (B) Expression of one dominant negative subunit causes dysfunction of entire oligomeric protein 292  Guide to Research Techniques in Neuroscience Review Articles Herskowitz, I. (1987). Functional inactivation of genes by dominant negative mutations. Nature, 329, 219–222. Hsu, P. D., Lander, E. D., & Zhang, F. (2014). Development and applications of CRISPR-Cas9 for genome engineering. Cell, 157, 1262–1278. Lehner, B., Fraser, A. G., & Sanderson, C. M. (2004). Technique review: how to use RNA interference. Briefings in Functional Genomics and Proteomic, 3, 68–83. Miller, V. M., Paulson, H. L., & Gonzalez-Alegre, P. (2005). RNA interference in neuroscience: progress and challenges. Cellular and Molecular Neurobiology, 25, 1195–1207. Sander, J. D., & Joung, J. K. (2014). CRISPR-Cas systems for editing, regulating and targeting genomes. Nature Biotechnology, 32, 347–355. Sandy, P., Ventura, A., & Jacks, T. (2005). Mammalian RNAi: a practical guide. Biotechniques, 39, 215–224. Primary Research Articles—Interesting Examples from the Literature Cong, L., Ran, F. A., Cox, D., Lin, S., Barretto, R., Habib, N., et al. (2013). Multiplex genome engineering using CRISPR/Cas systems. Science, 339, 819–823. Doetschman, T., Gregg, R. G., Maeda, N., Hooper, M. L., Melton, D. W., Thompson, S., et al. (1987). Targeted correction of a mutant HPRT gene in mouse embryonic stem cells. Nature, 330, 576–578. Fire, A., Xu, S., Montgomery, M. K., Kostas, S. A., Driver, S. E., & Mello, C. C. (1998). Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature, 391, 806–811. Jessberger, S., Aigner, S., Clemenson, G. D., Toni, N., Lie, D. C., Karalay, O., et al. (2008). Cdk5 regulates accurate maturation of newborn granule cells in the adult hippocampus. PLoS Biol, 6, e272. Ma, Y., Shen, B., Zhang, X., Lu, Y., Chen, W., Ma, J., et al. (2014). Heritable multiplex genetic engineering in rats using CRISPR/Cas9. PLoS One, 9, e89413. Mezghrani, A., Monteil, A., Watschinger, K., Sinnegger-Brauns, M. J., Barrëre, C., Bourinet, E., et al. (2008). A destructive interaction mechanism accounts for dominant-negative effects of misfolded mutants of voltage-gated calcium channels. Journal of Neuroscience, 28, 4501–4511. Ralph, G. S., Radcliffe, P. A., Day, D. M., Carthy, J. M., Leroux, M. A., Lee, D. C., et al. (2005). Silencing mutant SOD1 using RNAi protects against neurodegeneration and extends survival in an ALS model. Nature Medicine, 11, 429–433. Thomas, K. R., & Capecchi, M. R. (1987). Site-directed mutagenesis by gene targeting in mouse embryo-derived stem cells. Cell, 51, 503–512. Protocols Birmingham, A., Anderson, E., Sullivan, K., Reynolds, A., Boese, Q., Leake, D., et al. (2007). A protocol for designing siRNAs with high functionality and specificity. Nature Protocols, 2, 2068–2078. Dann, C. T., & Garbers, D. L. (2008). Production of knockdown rats by lentiviral transduction of embryos with short hairpin RNA transgenes. Methods in Molecular Biology, 450, 193–209. Kappel, S., Matthess,Y., Kaufmann, M., & Strebhardt, K. (2007). Silencing of mammalian genes by tetracycline-inducible shRNA expression. Nature Protocols, 2, 3257–3269. Morozov,A. (2008). Conditional gene expression and targeting in neuroscience research. Current Protocols in Neuroscience (Chapter 4), Unit 4.31. http://dx.doi.org/10.1002/0471142301.ns0431s44. Manipulating Endogenous Genes Chapter | 13  293 Mortensen, R. (2007). Overview of gene targeting by homologous recombination. Current Protocols in Neuroscience (Chapter 4), Unit 4.29. http://dx.doi.org/10.1002/0471142301.ns0429s40. Ran, F. A., Hsu, P. D., Wright, J., Agarwala, V., Scott, D. A., & Zhang, F. (2013). Genome engineering using the CRISPR-Cas9 system. Nature Protocols, 8, 2281–2308. Yang, H., Wang, H., & Jaenisch, R. (2014). Generating genetically modified mice using CRISPR/ Cas-mediated genome engineering. Nature Protocols, 9, 1956–1968. 295 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00014-9 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 14 Cell Culture Techniques After Reading This Chapter, You Should be Able to: l Compare the advantages and disadvantages of using in vitro culture techniques with in vivo techniques l Describe types of cells used to examine nervous system function in vitro l Describe techniques for manipulating cells in cell culture conditions Techniques covered: l Tools and reagents used in culture: equipment, media l Types of cultured cells: immortalized cell lines, primary cell and tissue culture, stem cells l Manipulating cells in culture: transfection, co-culture, pharmacology, antibody interference Cell culture is the process by which a scientist grows and maintains cells under carefully controlled conditions outside of a living animal. There are many reasons why this approach is desirable. Examining the nervous system in vitro (within glass) allows scientists to simplify the cellular environment, providing greater control over experimental manipulations and reducing potentially confounding interactions with other biological systems. In vitro tools and techniques make experiments possible that would otherwise be difficult or impossible to perform (or interpret) in intact organisms, such as performing multiple assays in parallel with the exact same number of cells. In vitro experiments also tend to be faster, less expensive, and require fewer animals than experiments performed in vivo. However, cell culture experiments also raise interesting questions: Does a cell in a culture flask truly behave like it would inside the brain or body? What does it mean to culture neurons, cells notorious for their intercellular communication and neural networks, in relatively isolated conditions in which they do not form synaptic connections? Are immortalized cell lines, cells that continue to divide indefinitely, actually good models for cells that degenerate and perish in the brain? Because the goal of most in vitro experiments is to generate a hypothesis and conclusion about what occurs in vivo, it is important to design experiments carefully and extrapolate results cautiously. This chapter surveys approaches to studying the nervous system using cell culture techniques. We start with a quick summary of the equipment and 296  Guide to Research Techniques in Neuroscience reagents necessary to maintain cells outside their endogenous environments. Then we describe three categories of cells used in culture experiments: immortalized cell lines, primary cell cultures, and a special class of cell that has received much attention over the past several years—stem cells. Finally, we describe some of the methods scientists can use to perturb cells in cultured environments. CELL CULTURE EQUIPMENT AND REAGENTS Specialized equipment and reagents are necessary to provide cultured cells with an environment that can support their continued growth and health outside a living organism. Most of these supplies are used to artificially mimic the endogenous, in vivo, cellular environment. Taking a cell out of the brain and expecting it to survive in a cell culture dish without providing basic, life-sustaining factors would be like catching a fish and expecting it to stay alive in a cage: unless a scientist correctly provides the elements necessary to sustain life, the cell will quickly die. Thus, most equipment and reagents in cell culture labs are used to continually supply cells with oxygen, nutrients, growth factors, and other elements necessary to keep cells alive. Other tools and reagents prevent contamination. Cells in vivo have the benefit of an active immune system to prevent contamination from bacteria, fungi, and other microorganisms. Cells in the brain have the added benefit of the blood–brain barrier to further prevent contamination. In culture conditions, cells are incredibly vulnerable, and scientists must take great care in avoiding contamination, especially because microorganisms are ubiquitous in the environment and can easily penetrate cell culture plates. Equipment Although individual laboratories study different cell lines and ask different scientific questions, cell culture rooms tend to contain the same fundamental pieces of equipment. This equipment includes the following: l Biosafety hood. A biosafety hood or laminar flow cabinet is used to prevent contamination by microorganisms (Figure 14.1).When not in use, these chambers are often illuminated with UV light that helps sterilize exposed surfaces. Just before use and throughout experiments, scientists spray all surfaces with 70% ethanol to provide further decontamination. During experiments, scientists always keep reagents and cell culture flasks/plates covered until they are ready for use; when exposed, a scientist should never pass their gloved hands or other equipment over open bottles or, even worse, the cells themselves. l Cell incubator. Cell incubators house and store culture flasks and plates. These incubators maintain an appropriate temperature, humidity, and gas concentration to mimic endogenous conditions. They are usually set at 37 °C with 5% CO2 levels.Although 5% CO2 is not the level experienced in the body in vivo, the carbon dioxide maintains the pH of buffers in the growth media at proper physiological levels. Cell Culture Techniques Chapter | 14  297 l Specially prepared cell culture flasks/plates. Tissue culture flasks and plates come in many varieties depending on the needs of the investigator. Flasks typically range in sizes of 12.5 cm2 to large, multitiered chambers with surface areas of 1800 cm2. Multiwelled plates are often used for tissue culture experiments, as these wells can hold an exact number of cells, and it is easy to keep track of various experimental conditions across wells. Plates are typically sold with 6, 12, 24, 48, 96, or 384 wells. Many cells, including neurons, need to adhere to a substrate to grow, so tissue culture dishes often need to be specially prepared to provide a substrate for the cells. Tissue culture dishes or glass coverslips are often coated with attractive amino acids, such as lysine or ornithine or extracellular matrix components like collagen and laminin. l Refrigerator. A refrigerator (properly referred to as a 4 °C incubator) maintains cell culture media and other reagents when not in use. The growth factors and antibiotics in culture media degrade over time, but can last for weeks if stored at 4 °C. l Water bath. A water bath is often set at 37 °C and is used to quickly warm cell culture media and other reagents stored at 4 °C just prior to being added to cell culture flasks/plates. If a scientist does not warm media before adding it to cells, the cells can be shocked by the abrupt change to cold temperatures and die. l Microscope. Microscopes are used in most tissue culture rooms for routine observation of cell culture flasks/plates to inspect the health and confluence of cells. Confluence refers to the percentage of the surface of the bottom of the plate covered by cells. Most cell lines should never become 100% confluent, as cells in dense populations tend to inhibit each other’s growth, as well as quickly drain nutrients from the culture media. Biosafety hood CO2 Incubator FIGURE 14.1  Typical cell culture room with biosafety hood and CO2 incubator. 298  Guide to Research Techniques in Neuroscience Culture Media Growth media is critical to cell culture experiments, supplying nutrients (amino acids and vitamins) and a source of energy (glucose) for cells. Growth media can vary in pH, nutrient concentration, and the presence of growth factors or other biologically relevant components. To survive, cells must be bathed in an isotonic fluid that has the same concentration of solute molecules as inside the cell. The media is buffered to maintain a compatible pH (usually 7.4, though there are some cell-specific variations). In the 5% CO2 environment of culture incubators, a bicarbonate buffer maintains the physiological pH as well as providing nutritional benefits to the cells. HEPES (4-(2-hydroxyethyl)-1piperazineethanesulfonic acid) is added to the culture media for extra buffering capacity when the cells in a culture experiment require extended periods of manipulation outside a CO2 incubator. Serum is often added to culture media for its ability to promote survival through undefined mixtures of growth factors, hormones, and proteins, like platelet-derived growth factor, insulin, and transferrin. For stricter control over the cellular environment, investigators use serum-free, chemically defined supplements, such as N2 or B27, that contain known formulations of survival factors. A typical growth medium in neuronal cultures is NeuroBasal, which provides optimized amino acids and nutrients to cultured neurons, supplemented with N2 or B27. Adding or removing specific ingredients to or from the media can influence cellular behavior. For example, to maintain a neural progenitor pool that continues to divide in culture, scientists add growth factors such as epidermal growth factor (EGF) and fibroblast growth factor (FGF) to prevent differentiation. Specific recipes for culture media are tailored to provide optimal conditions for promoting the health and proper physiological function of the cultured cells and for maintaining cellular behavior. Now that we have described some of the equipment and reagents necessary for cell culture, we survey the different categories of cells commonly used in cell culture experiments. IMMORTALIZED CELL LINES Immortalized cell lines are either tumorous cells that do not stop dividing or cells that have been artificially manipulated to proliferate indefinitely and can, thus, be cultured over several generations (Table 14.1). Because immortalized cells continuously divide, they eventually fill up the dish or flask in which they grow. By passaging (also known as splitting), scientists transfer a fraction of the multiplying cells into new dishes to provide space for continuing proliferation. There are many advantages to using immortalized cell lines. Because these are standard lines used by many different labs, immortalized cells are fairly well characterized. They are, at least theoretically, homogeneous, genetically Cell Culture Techniques Chapter | 14  299 TABLE 14.1  Commonly Used Immortalized Cell Lines Origin and cell type Comments 3T3 Mouse embryonic fibroblast Robust and easy to handle; contact inhibited; stops growing at very high densities HeLa Human epithelial cell From cervical cancer in a human patient named Henrietta Lacks; may contaminate other cultured cell lines; able to grow in suspension (i.e., grow without adhering to bottom of plates) COS (Cv-1 in Origin, carrying Sv40 genetic material) Monkey kidney Efficiently transfected; commonly used as an expression system for high-level, short-term expression of proteins 293/293T/HEK-293T Human embryonic kidney Easy to transfect and manipulate; commonly used as an expression system to study signaling and recombinant proteins MDCK (Madin-Darby canine kidney) Dog kidney epithelial cell Polarized with distinct apical and basal sides, used in studying trafficking CHO Chinese hamster ovary Useful for stable gene expression and high protein yields for biochemical assays; commonly used as an expression system for studying cell signaling and recombinant proteins S2 Drosophila macrophage-like cells Well-characterized Drosophila cell line; highly susceptible to RNAi treatment PC12 Rat pheochromocytoma chromaffin cell Neuron-like, derived from a neuroendocrine adrenal tumor; can differentiate into a neuron-like cell in the presence of nerve growth factor Continued 300  Guide to Research Techniques in Neuroscience identical populations, which aid in providing consistent and reproducible results. Immortalized cells tend to be easier to culture than cells used in primary cultures in that they grow more robustly and do not require extraction from a living animal. Also, because they grow quickly and continuously, it is possible to extract large amounts of proteins for biochemical assays (Chapter 15). It is also possible to create cell lines that continuously express a gene of interest, such as a fluorescently tagged or mutant version of a protein. The major disadvantage to using immortalized cells is that these cells cannot be considered normal, in that they divide indefinitely and sometimes express unique gene patterns not found in any cell type in vivo. Therefore, they might not have the relevant attributes or functions of normal cells. Also, after several passages, cell characteristics can change and become even more different from those of a normal cell. Thus, it is important to periodically validate the characteristics of cultured cells and not use cells that have been passaged too many times. Immortalized cell lines of neuronal origin can be used to study properties unique to neurons. Scientists have used neuronal cell lines to investigate processes that occur during differentiation in neurons, such as axon selection, guidance, and growth. However, most neuronal immortalized cell culture models are derived from tumors and are sometimes genomically abnormal. One popular neuronal cell line, called PC12, is a rat pheochromocytoma cell line derived from an adrenal gland tumor. The addition of nerve growth factor (NGF) causes PC12 cells to reversibly differentiate into a neuronal phenotype (Figure 14.2). These cells can synthesize dopamine, norepinephrine, and acetylcholine. PC12 cells have been used to study molecular phenomena associated with neuronal differentiation and have even been used Origin and cell type Comments Neuro-2a/N2a Mouse neuroblastoma Model system for studying pathways involved in neuronal differentiation; can be driven to differentiate by cannabinoid and serotonin receptor stimulation SH-SY5Y Human neuroblastoma, cloned from bone marrow Dopamine beta hydroxylase activity, acetylcholinergic, glutamatergic, adenosinergic; grow as clusters of neuroblast-type cells with short, fine neurites TABLE 14.1  Commonly Used Immortalized Cell Lines—cont’d Cell Culture Techniques Chapter | 14  301 in experiments to replace dopaminergic neurons in an animal model of Parkinson’s disease. Neuroblastoma cell lines, like mouse Neuro2A, also express neurotransmitters and have been used in electrophysiology and neurodevelopment studies. As useful as immortalized cell lines of neural origin can be for certain experiments, they show obviously abnormal traits, such as the unusual combination of neurotransmitters they produce (no normal neuron produces dopamine, norepinephrine, and acetylcholine in the same cell!). Therefore, it is advantageous, when possible, to use primary cultured cells—cells extracted from living animals. PRIMARY CELL AND TISSUE CULTURE Primary cell and tissue culture is derived from tissue removed directly from a living animal rather than immortalized cells that divide indefinitely (Figure 14.3). Primary tissue culture allows scientists to directly investigate cells of interest in a carefully controlled in vitro environment. These experiments can be especially +NGF FIGURE 14.2  Neuronal differentiation. Addition of growth factor to PC12 cells causes them to differentiate into neuron-like cells that grow neurites and synthesize neurotransmitters. Extract brain Section brain Dissect brain region Culture brain slice Enzymatically dissociate Slice cultureExplant cultureDissociated culture FIGURE 14.3  Primary tissue cultures. After a brain is removed from an animal, it can be cut into thin sections (usually using a vibratome) and kept alive in slice culture. Alternatively, either a brain slice or a brain region can be finely dissected to produce an explant culture. An explant culture can be enzymatically digested into single cells to produce a dissociated cell culture. 302  Guide to Research Techniques in Neuroscience advantageous for a comparison between cells derived from a wild-type animal and cells derived from a genetically modified animal. However, there are some disadvantages to using primary cultures (Table 14.2). Unlike immortalized cell lines, primary cultured cells have a limited lifetime. The age of the animal source influences the health and robustness of the cell culture: tissue from younger, embryonic or early postnatal animals survives better and tends to be healthier than tissue from older animals. Also, a population of primary cells will always be more heterogeneous than a culture of immortalized cells, no matter how careful the scientist was in extracting and purifying the cells of interest. There are three main categories of primary tissue culture: slice cultures, explant cultures, and dissociated cultures. Slice Cultures Slice cultures maintain the structure and organization of the brain in a relatively thick (250–400 μm) section of brain tissue cut using a vibratome (Chapter 6). Slices provide greater access to and visibility of deep subcortical structures, like thehippocampusandthalamus,whicharedifficulttoaccessin vivo(Figure14.3). These slices can either be acute cultures that are used immediately (over a period of hours) or organotypic slice cultures that are maintained over multiple days. Acute slices are typically used for short-term electrophysiology experiments, while organotypic cultures are used to observe structural and morphological changes such as neuronal migration, axon outgrowth, or synapse formation. Slices require an air/liquid interface to properly regulate gas exchange throughout the slice. Different protocols for culturing slices have varied success in preserving the structure and development of the tissue to parallel that seen in vivo. TABLE 14.2  Immortalized Cell Lines versus Primary Cell Culture Advantages Disadvantages Immortalized cell lines l Easier to use (grow, transfect, etc.) l Homogeneous l Fairly well characterized l Can create stable cell lines expressing gene of interest l May not have the same properties as neurons or primary cell type of interest Primary cell culture l Relevant cell type, physiology, and circuitry l Heterogeneous populations with high variability Cell Culture Techniques Chapter | 14  303 Explant Cultures Explant cultures consist of intact fragments of tissue (Figure 14.3). Although they do not necessarily preserve the precise organization and orientation of the endogenous nervous system, explants contain the same mixture of cell types. Unlike slices, which require an air interface for proper oxygenation, explant cultures can be submerged in the bath media. Explants are often used in coculture assays and for studies on neurite outgrowth and neuronal migration. Dissociated Cell Cultures In dissociated cultures, neural tissue is separated into individual cells that are then grown on two-dimensional coated glass coverslips or within threedimensional substrates (Figure 14.3). After extracting an animal’s brain, specific regions can be micro-dissected to isolate a specific neuronal subtype (e.g., cortical, hippocampal). This region is then mechanically or enzymatically digested within a liquid suspension to separate individual cells from each other. The neurons are then removed from the suspension and plated onto a substrate on which the cells can attach and grow. Neurons dissociated from different regions of the brain retain their initial identities. The morphological, molecular, and physiological properties of cell populations present in culture correspond closely to the characteristics of the cell population present in the region of origin in a living organism. With the proper growth factors and care, it is possible to maintain a dissociated culture for weeks, during which time cells acquire properties of mature neurons: they develop characteristic axons and dendrites, form synapses with one another, and express receptors and ion channels specific to their cell types, even producing spontaneous electrical activity. Dissociated neuronal cell cultures have been used to study neurite outgrowth, synapse formation, and electrophysiological parameters. Theabilitytoprobeindividualneurons,however,comesattheexpenseoflosing the organization and connectivity critical to in vivo functions. Another limitation of dissociated cultures is the small quantity of cells relative to immortalized cell lines,whichcanmakeitdifficulttoperformbiochemicalexperiments(Chapter15) that require a high volume of starting material (e.g., millions of cells). Furthermore, most primary cell cultures are not homogeneous. Neuronal cultures are often mixtures of both glia and neurons that respond to different neurotransmitters, so identifying an individual population of cells can be difficult. To minimize heterogeneity, investigators usually attempt to dissect regions as precisely as possible to maximize the presence of the desired cell type in culture. Various methods have been developed to purify specific cell populations. In a technique called immunopanning, a scientist coats the bottom of a plate with antibodies that recognize cell-surface markers on the outside of specific cell types. When heterogeneous populations of cells are added to the plate, the scientist can purify the cells of interest by allowing the cells to bind to the bottom 304  Guide to Research Techniques in Neuroscience of the plate and then wash off the undesired, unbound cells. This technique has been used to culture oligodendrocyte precursors, retinal ganglion cells, and corticospinal motor neurons. STEM CELL CULTURES A stem cell is a pluripotent cell with the capacity to generate any cell type (e.g., neuron, muscle, blood) and an unlimited ability to renew itself. These cells are similar to immortalized cell lines in that they are theoretically able to propagate in culture continuously; however, they are also primary cells derived directly from a living organism. Stem cells are classified by their source (e.g., embryonic stem (ES) cells, adult stem cells, induced pluripotent stem (iPS) cells), as well as the tissue they typically generate (e.g., neural stem cells (NSCs), hematopoietic stem cells, skin stem cells). Stem cells defined by the tissue they generate are multipotent— able to give rise to all types of cells found in the tissue and able to continuously self-renew. For example, NSCs can give rise to all three neural lineages—neurons, astrocytes, and oligodendrocytes—as well as additional NSCs. Stem cells are distinct from progenitor cells, which have a more limited capacity for selfrenewal and may be unipotent, giving rise to a single cell type. Stem cell culture is a form of dissociated primary culture. Therefore, stem cells must be cultured in the presence of growth factors (EGF and basic fibroblast growth factor (bFGF)) in specially formulated media that preserves their multipotency and ability to self-renew. Because scientists do not yet know all of the environmental components required to recreate in vivo conditions for preserving stem cell function, in vitro studies may not completely capture the effect of endogenous environmental influences. Stem cell culture has a variety of uses in neuroscience. ES cells are examined for their ability to differentiate into specific neuronal subtypes. NSCs can be cultured to study the basic biology of development and aging. iPS cells can be used to generate clones of cells from patients with neurological disorders to characterize cellular and molecular changes in diseased neurons. While an extensive history and theory of stem cells already exists, basic stem cell biology is still a rapidly evolving field. Here we focus on specific applications of stem cell culture in neuroscience and describe common in vitro techniques used to identify stem cells. Embryonic Stem Cells Pluripotent embryonic stem (ES) cells can give rise to all tissues in an organism. Derived from the inner cell mass of a blastocyst embryo, culturing ES cells in vitro essentially traps the cells in a pluripotent state by growing them in the presence of factors that prevent the cells from differentiating. Investigators can reconstruct the environment of ES cells in a dish so that culture conditions Cell Culture Techniques Chapter | 14  305 contain the specific molecules and mitogens that specify the formation of a neuron in the developing embryo. Using special culture conditions, ES cells are first induced to become general neural progenitors, precursors that are committed to a neural fate. Once they have become neural progenitors, the specific molecules known to act during normal neuronal development are added to the culture to direct differentiation of specific neuron subtypes. For example, to make motor neurons, cells are exposed to high concentrations of sonic hedgehog (SHH) and retinoic acid, morphogens that pattern spinal cord motor neurons during development. To make midbrain dopaminergic neurons, cells are exposed to SHH and fibroblast growth factor 8 (FGF8), morphogens that specify dopaminergic fate during development. Investigators monitor the progression of differentiation by examining the culture for activation of transcription factors relevant to development and the appearance of markers known to promote differentiation of specific neuronal subtypes. For example, when trying to induce midbrain dopaminergic neurons, scientists look for midbrain-specific transcription factors (Pitx3, En1, Lmx1b, Nurr1) and later for markers of mature dopamine neurons (TH, DAT, Girk2). This process—from the initial establishment of the ES cells to their differentiation into specific neuronal subtypes—can take months. Also, despite attempts to create pure populations of a specific neuronal subtype, differentiated ES cell populations are still heterogeneous, with many other cells present in the culture. However, the ability to differentiate ES cells into specific neuronal cell types is useful for investigating therapies for a number of neurodegenerative disorders such as Parkinson’s and Alzheimer’s disease. Neural Stem Cells Multipotent NSCs exhibit self-renewal properties and the ability to differentiate into all neural subtypes. They can be extracted from regions of embryonic or adult brains, where they normally divide and give rise to neurons or glia. The embryonic brain contains many NSCs, while adult brains have far fewer NSCs. Furthermore, the ability of adult NSCs to produce neurons decreases with age. Investigators typically culture adult NSCs from the subgranular zone of the dentate gyrus in the hippocampus and the subventricular zone in the wall of the lateral ventricles. After dissection, the region containing NSCs is dissociated and can be moderately purified through centrifugation. NSCs can also be grown and passaged in special media that promote NSC survival. By removing growth factors from the media, NSCs can be differentiated into neurons, astrocytes, and oligodendrocytes. It is difficult to distinguish neural progenitors from NSCs, and scientists often confuse the terms. Both types of cells proliferate and express common sets of molecular markers (e.g., nestin, Pax6). Therefore, functional assays must also be used to determine the identity of cultured cells. Two defining features of 306  Guide to Research Techniques in Neuroscience NSC cultures that can be tested to confirm cellular identity are (1) the ability to give rise to all three types of the neural lineage (multipotency) and (2) the ability to propagate more NSCs (self-renewal). Self-renewal is often examined by performing primary neurosphere and secondary neurosphere assays. When NSCs are cultured in nonadherent conditions in the presence of the growth factors EGF and bFGF, they give rise to neurospheres, balls of dividing cells (Figure 14.4). However, neurospheres form as a general result of dividing neural precursors, including progenitor cells that do not continuously self-renew. Therefore, while this assay is frequently used, it does not definitively distinguish between stem and progenitor cells. The secondary neurosphere assay involves culturing cells from established neurospheres to see if the cells generated by the first neurosphere are able to continue proliferating, an indicator of self-renewal. These assays can be difficult to quantify, as cells and spheres can each fuse, so a sphere may not necessarily form from a single NSC. Also, neurospheres are species-specific; rat and human NSCs do not form neurospheres as often as mouse NSCs. Induced Pluripotent Stem Cells iPS cells are differentiated cells that scientists have reverted back to a stem cell state. A variety of strategies have been employed to carry out this feat. One common strategy is to use viruses to introduce specific transcription factors into differentiated somatic cells, usually fibroblasts. The presence of these transcription factors seems to allow these cells to exhibit stem cell behaviors. Other strategies employ a combination of viral delivery and chemical manipulations to coax somatic cells into a pluripotent state. Once iPS cells are generated, they are examined for characteristics of “stemness” by comparing them to ES cells in various ways. Scientists FIGURE 14.4  Neurospheres. Dividing neural progenitor and stem cells form spheres in culture. Courtesy of Dr Victoria Rafalski. Cell Culture Techniques Chapter | 14  307 typically examine transcription factor profiles, measuring the presence of markers such as Oct4, Nanog, Sox2, AP, SSEA4, and TRA-1-80 using immunohistochemistry (Chapter 6) and reverse transcription polymerase chain reaction techniques (Chapter 10). The potential for these iPS cells to differentiate is measured by driving cells into specific lineages using established protocols. The gold standard of determining whether iPS cells are like stem cells is determining whether these cells retain the ability to generate chimeric mice (Chapter 13). In neuroscience, the ability to reprogram fully differentiated cells into an ESlike state opens up the possibility of modeling human disease in vitro, directly from patient-derived cells. These cells could reveal the cellular and molecular pathogenesis of diseases such as Parkinson’s and autism spectrum disorders. iPS cells may also be used for therapeutic purposes: for example, generating new neural cell types for transplantation from a patient’s own somatic cells. Work has also progressed in bypassing the iPS cell stage to directly reprogram somatic cells into functional neurons. MANIPULATING CELLS IN CULTURE One of the major advantages of studying the nervous system in vitro is the ability to control the cellular environment. In addition to other methods of neuromodulation (Chapter 8), scientists can manipulate neurons in culture using several methods. Transfection and Infection Transfection and infection are the processes of delivering DNA to cells using nonviral and viral methods, respectively. One of the major benefits of cell culture systems, especially dissociated cultures, is that they are very amenable to both DNA delivery methods. Therefore, it is usually easy to introduce recombinant DNA molecules into cultured neurons to change gene/protein expression and alter cellular physiology. Transfection and infection procedures are discussed in much greater detail in Chapter 11. Co-Culture Systems Co-culturing allows a variety of cell types to be cultured together to examine the effect of one culture system on another (Figure 14.5). This procedure is useful when examining the effect of one type of tissue on another, one region of the brain on another, or how a particular secreted molecule leads to changes in neural development or physiology. For example, co-cultures of different regions of spinal cord explants initially revealed differing effects on the ability to attract or repel neurite outgrowth. Biochemical purification from explants in co-culture experiments led to the identification of specific molecules that could then be introduced 308  Guide to Research Techniques in Neuroscience into immortalized cell lines to express and secrete these molecular guidance cues. Transfected cell lines could then be co-cultured with spinal cord explant cultures to examine the neurite outgrowth response of the spinal cord neurons to specific guidance cues. Such approaches have deciphered a large variety of chemoattractants and chemorepellants that have since been validated in vivo. Pharmacology Because of the unparalleled access to the medium in cell culture conditions, it is relatively easy to add pharmacological agents to culture media to affect proteins, ion channels, and receptors of interest. See Chapter 8 for a detailed description of pharmacological techniques. Antibody Interference Antibodies are molecules that bind with high affinity to antigens, usually parts of a protein. In nature, these antibodies function as part of the immune system. In the laboratory, antibodies can be used in many different assays to study protein expression and binding partners (Chapter 15). If the antigen of an antibody is accessible in living cells, it is also sometimes possible to block protein function by applying antibodies against the protein of interest. These antibodies bind the protein at a site critical for its normal function, thus inhibiting function in vitro or in vivo. CONCLUSION The ability to grow cells and tissues outside of a living organism allows scientists much greater access and control over the cellular environment compared with in vivo conditions. Therefore, scientists can investigate the cellular and molecular mechanisms of nervous system function with great detail. The tools and techniques of in vitro culture techniques can provide many informative results, but ultimately this knowledge must be tested in the intact nervous system to fully understand how a biological process works in a living organism. FIGURE 14.5  The use of co-cultures to study guidance cues. In this example, midbrain explants are placed in culture medium. COS cells (top) transfected with a control plasmid (A) do not influence the direction of axon outgrowth from the explants. However, COS cells transfected with netrin-1 (B) sufficiently attract axons. Courtesy of Dr Jie Li and Dr Mary Hynes. Control Netrin Cell Culture Techniques Chapter | 14  309 SUGGESTED READING AND REFERENCES Books Barres, B. A., & Stevens, B. (Eds.). (2013). Purifying and culturing neural cells: A laboratory manual. New York: Cold Spring Harbor Laboratory Press. Butler, M. (2004). Animal cell culture and technology (2nd ed.). London: BIOS Scientific Publishers. Freshney, R. I. (2005). Culture of animal cells: A manual of basic techniques (5th ed.). Hoboken, NJ: Wiley-Liss. Martin, B. M. (1994). Tissue culture techniques: An introduction. Boston: Birkhäuser. Skloot, R. (2011). The immortal life of henrietta lacks. New York: Broadway Books. Review Articles Blau, H. M., Brazelton, T. R., & Weimann, J. M. (2001). The evolving concept of a stem cell: entity or function? Cell, 105, 829–841. Jaenisch, R., & Young, R. (2008). Stem cells, the molecular circuitry of pluripotency and nuclear reprogramming. Cell, 132, 567–582. Seaberg, R. M., & van der Kooy, D. (2003). Stem and progenitor cells: the premature desertion of rigorous definitions. Trends in Neurosciences, 26, 125–131. Primary Research Articles—Interesting Examples from the Literature Barberi, T., Klivenyi, P., Calingasan, N.Y., Lee, H., Kawamata, H., Loonam, K., et al. (2003). Neural subtype specification of fertilization and nuclear transfer embryonic stem cells and application in parkinsonian mice. Nature Biotechnology, 21, 1200–1207. Dotti, C. G., & Banker, G. A. (1987). Experimentally induced alteration in the polarity of developing neurons. Nature, 330, 254–256. Dugas, J. C., Mandemakers, W., Rogers, M., Ibrahim, A., Daneman, R., & Barres, B. A. (2008). A novel purification method for CNS projection neurons leads to the identification of brain vascular cells as a source of trophic support for corticospinal motor neurons. Journal of Neuroscience, 28, 8294–8305. Foo, L. C., Allen, N. J., Bushong, E. A., Ventura, P. B., Chung, W. S., Zhou, L., et al. (2011). Development of a method for the purification and culture of rodent astrocytes. Neuron, 71, 799–811. Pan, P. Y., Tian, J. H., & Sheng, Z. H. (2009). Snapin facilitates the synchronization of synaptic vesicle fusion. Neuron, 61, 412–424. Rafalski, V. A., Ho, P. P., Brett, J. O., Ucar, D., Dugas, J. C., Pollina, E. A., et al. (2013). Expansion of oligodendrocyte progenitor cells following SIRT1 inactivation in the adult brain. Nature Cell Biology, 15, 614–624. Raineteau, O., Rietschin, L., Gradwohl, G., Guillemot, F., & Gahwiler, B. H. (2004). Neurogenesis in hippocampal slice cultures. Molecular and Cell Neuroscience, 26, 241–250. Soldner, F., Hockemeyer, D., Beard, C., Gao, Q., Bell, G. W., Cook, E. G., et al. (2009). Parkinson’s disease patient-derived induced pluripotent stem cells free of viral reprogramming factors. Cell, 136, 964–977. Takahashi, K., & Yamanaka, S. (2006). Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell, 126, 663–676. Tessier-Lavigne, M., Placzek, M., Lumsden, A. G., Dodd, J., & Jessell, T. M. (1988). Chemotropic guidance of developing axons in the mammalian central nervous system. Nature, 336, 775–778. Vierbuchen, T., Ostermeier, A., Pang, Z. P., Kokubu,Y., Sudhof, T. C., & Wernig, M. (2010). Direct converstion of fibroblasts to functional neurons by defined factors. Nature, 463, 1035–1041. Watanabe, K., Kamiya, D., Nishiyama, A., Katayama, T., Nozaki, S., Kawasaki, H., et al. (2005). Directed differentiation of telencephalic precursors from embryonic stem cells. Nature Neuroscience, 8, 288–296. 310  Guide to Research Techniques in Neuroscience Protocols Current protocols in neuroscience. Chapter 3: Cellular and developmental neuroscience. (2006). Hoboken, NJ: John Wiley & Sons, Inc. Loring, J. F., Wesselschmidt, R. L., & Schwartz, P. H. (Eds.). (2007). Human stem cell manual: A laboratory guide. Amsterdam: Academic Press. Poindron, P., Piguet, P., & Förster, E. (Eds.). (2005). New methods for culturing cells from nervous tissues. Basel: Karger. Zigova, T., Sanberg, P. R., & Sanchez-Ramos, J. R. (Eds.). (2002). Neural stem cells: Methods and protocols. Totowa, NJ: Humana Press. Websites American type culture collection: http://www.atcc.org. European collection of cell cultures: http://www.hpacultures.org.uk/aboutus/ecacc.jsp. Life technologies protocols: http://www.lifetechnologies.com/us/en/home/references/protocols.html. 311 Guide to Research Techniques in Neuroscience. http://dx.doi.org/10.1016/B978-0-12-800511-8.00015-0 Copyright © 2015 Elsevier Inc. All rights reserved. Chapter 15 Biochemical Assays and Intracellular Signaling After Reading This Chapter, You Should be Able to: l Discuss why studying proteins is important to neuroscience l Explain why antibodies are used to study proteins and how they are produced l Describe common biochemical assays used to investigate intracellular signaling Techniques covered: l Antibody production: making monoclonal and polyclonal antibodies l Techniques used to purify proteins: affinity chromatography, immunoprecipitation assay l Techniques used to measure the amount and location of proteins within cells and tissue: western blot (immunoblot), enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), immunohistochemistry (IHC), immunoelectron microscopy, reporter proteins l Techniques used to investigate protein–protein interactions: co-immunoprecipitation, protein affinity chromatography, yeast two-hybrid assay l Techniques used to investigate posttranslational modifications: posttranslational modification assays (kinase assay), posttranslational modification-specific antibodies l Techniques used to investigate protein–DNA interactions: electrophoretic shift assay (EMSA), chromatin immunoprecipitation assay (ChIP), ChIP on chip assay, luciferase assay Proteins are molecular machines responsible for virtually all of the structural and functional properties of cells. Structural proteins maintain a cell’s shape, size, and durability. Transmembrane proteins serve as receptors for hormones and neurotransmitters, allowing the cell to receive information from the extracellular environment. Other proteins act as downstream messengers from cell surface receptors and mediate the response to an extracellular cue. Some proteins serve as enzymes that facilitate the metabolic and physiological needs of the cell. Finally, some proteins regulate the transcription and expression of genes that code for other proteins to further affect a cell’s physiology. Because of the important role that proteins play in neural structure and function, anyone interested in how the brain works should appreciate the techniques used by scientists for their investigation. 312  Guide to Research Techniques in Neuroscience Proteins interact with each other and other biomolecules much like neurons interact in functional circuits. Instead of acting in isolation, they communicate with upstream signaling molecules and subsequently affect downstream proteins and biochemical substrates. Intracellular signaling (or signal transduction) is a term used to describe the biochemical pathways that affect a cell’s physiology and metabolism. Often, these pathways begin with the activation of a cell surface receptor, which leads to a series of biochemical events that culminate in a change in gene expression in the nucleus. Just as an organism senses a change in the environment and uses neural circuits to produce an appropriate response, an individual neuron senses a change in its environment and uses intracellular signaling pathways to change its transcriptional, physiological, and metabolic properties (Figure 15.1). The purpose of this chapter is to survey the methods used to study the important roles that proteins play within neurons. We will begin with a general description of intracellular signaling events and why they are useful for understanding the properties of neurons. Next, we will describe two fundamental methods used to study proteins: producing and using antibodies, and purifying proteins of interest. The bulk of the chapter then surveys a variety of methods that answer useful questions about intracellular signaling events: (1) whether a protein is expressed in a particular population of cells and how much is present; (2) whether a protein interacts with another protein; (3) what types of posttranslational modifications can occur for a protein and which proteins cause these modifications; and (4) whether a protein is capable of interacting with DNA and regulating gene expression. Environment Response Environment Response FIGURE 15.1  A metaphor for intracellular signaling. Just as the brain senses changes in the environment and uses neural circuitry to execute an appropriate behavioral response, a neuron senses changes in its extracellular environment and uses intracellular signaling to produce changes in metabolism and gene expression. Biochemical Assays and Intracellular Signaling Chapter | 15  313 INTRODUCTION TO SIGNAL TRANSDUCTION AND INTRACELLULAR SIGNALING The study of signal transduction pathways is fundamental to several subfields of neuroscience including clinical neuroscience, learning and memory, regulation of homeostasis, and neural development. For example, many researchers have investigated the signaling pathways responsible for axon guidance and synapse formation. Extracellular guidance molecules bind to membrane receptors, signaling which direction the neuron should grow. Intracellular signaling cascades then provide the metabolic and structural changes necessary for the axon to extend in the correct direction. When an axon reaches its postsynaptic target, other extracellular cues shut off this signaling cascade and new cascades develop to strengthen synaptic connections. Clearly, an understanding of the biochemical events within a neuron is necessary to truly understand the nature of axon guidance, as well as other vital neuronal functions. An example of a well-known cell signaling pathway is the JAK–STAT path- way(Figure15.2).Variousextracellularproteinssuchascytokinesandgrowth factors bind to specific transmembrane receptors. Some of these receptors bind to proteins called Janus Kinases (JAKs). Protein kinases are proteins that, when activated, have the ability to add a phosphate group to other proteins. When a cytokine binds to its receptor, JAKs transfer a phosphate group to the receptor. This phosphate group causes a conformational change in the three-dimensional structure of the receptor that allows other proteins to bind called Signal Transducers and Activators of Transcription (STATs). When STAT proteins bind to the receptor, JAKs add an additional phosphate group to the STAT proteins, causing them to dissociate from the receptor and bind to each other. These new STAT dimers translocate to the nucleus and activate specific target genes. The JAK–STAT pathway is one of several pathways that play a prominent role in the development and functional physiology of the nervous system. Several elements of this pathway are consistent with other signaling pathways: an extracellular cue binds to a membrane protein, causing an intracellular signaling cascade that eventually affects the expression of genes. These genes code for other proteins or genetic regulatory sequences that allow the cell to adapt to its environment. FUNDAMENTAL TOOLS USED TO STUDY PROTEINS This chapter describes many techniques that can be used to answer specific questions about intracellular signaling. Many of these techniques routinely make use of two fundamental biochemical tools that are essential for studying proteins: antibodies and protein purification techniques. Therefore, before discussing any specific assays, we first examine these fundamental tools. 314  Guide to Research Techniques in Neuroscience Making and Using Antibodies Antibodies are proteins used by an organism’s immune system to detect foreign proteins, called antigens. An antibody recognizes a specific region of an antigen, called the epitope. Antibodies are extremely useful to scientists because they bind specific proteins with high affinity. For example, an antibody can be tagged with a marker to bind and indicate the presence of a protein in a sample. Alternatively, an antibody can be attached to a column or plate in order to collect and purify a protein of interest. Antibodies for research are classified as either monoclonal or polyclonal antibodies. Monoclonal antibodies are collections of antibodies that recognize the same epitope (Figure 15.3). They are derived from single cell lines, usually cells obtained from mice. Once produced, these cell lines provide a renewable source of antibodies and all batches of antibody will be identical. Because they are directed against the same epitope, these antibodies may not detect Ligand Intracellular Extracellular Ligand binds to its receptor on the cell membrane P P P P P P PP Nucleus Target genes Promoter Binding of ligand to receptor causes phosphorylation and dimerization of STAT proteins Dimerized STAT translocates to the nucleus and binds to the promoter of target genes to activate transcription STAT proteins are eventually dephosphorylated and exported to the cytoplasm STAT proteins JAK proteins FIGURE 15.2  The JAK–STAT signaling pathway. Some receptors for ligands (e.g., cytokines and growth factors) interact with Janus Kinase (JAK) proteins. When a ligand binds to the receptor, JAKs add a phosphate group to the intracellular domain of the receptor. These phosphate groups attract Signal Transducers and Activators of Transcription (STAT) proteins to the receptor. JAKs add a phosphate group to STAT proteins, causing them to dissociate from the receptor and bind to each other. These STAT dimers then translocate to the nucleus, where they can affect gene transcription. Biochemical Assays and Intracellular Signaling Chapter | 15  315 homologous epitopes across species. In contrast, polyclonal antibodies are a collection of antibodies that may not be completely homogeneous, recognizing different epitopes of the same antigen (Figure 15.3). Therefore, these antibodies may detect homologous epitopes across different species. The physical accessibility of an epitope of a protein varies; therefore, polyclonal antibodies may be better able to bind to a target protein because a heterogeneous collection of antibodies increases the likelihood of one or more epitopes being accessible. However, polyclonal antibodies are prone to batch-to-batch variability, so the quality may not be consistent over time. These antibodies are commonly produced in rats, rabbits, chickens, goats, sheep, and donkeys. The procedures for producing monoclonal and polyclonal antibodies are described in Figures 15.4 and 15.5, respectively. A great variety of antibodies are commercially available and range in cost from $50 to as much as $1000 ($200–300 is common). If an antibody for a protein is not commercially available, individual labs may produce their own antibodies, but usually this work is contracted to an antibody-production service that produces custom antibodies on demand. If an antibody is difficult to produce, scientists can use recombinant DNA techniques (Chapter 10) to produce proteins tagged with a short amino acid sequence that can be easily detected by commercially available antibodies (Box 15.1). Several biochemical techniques depend on the use of antibodies (Table 15.1).A specific antibody may be useful for one, several, or (rarely) all of these techniques, depending on the particular binding affinity of the antibody–protein complex. For example, an antibody that works well for immunohistochemistry (IHC) may work very poorly for chromatin immunoprecipitation (ChIP). Even within IHC, the antibody may work much better for labeling proteins in cell culture than labeling proteins in tissue sections. Therefore, the only way to know if an antibody is useful for a particular technique is to try it. Often, commercial manufacturers indicate which biochemical techniques an antibody is best suited for based on internal testing. While antibody-based techniques are overwhelmingly used in neuroscience research to identify and report the presence of proteins, antibodies cannot always be generated to recognize a target molecule of interest. Recently, Monoclonal antibody Polyclonal antibody FIGURE 15.3  Monoclonal versus polyclonal antibodies. A monoclonal antibody recognizes a single epitope of an antigen. A polyclonal antibody is a mixture of several antibodies that recognize different epitopes of an antigen. 316  Guide to Research Techniques in Neuroscience scientists have developed an alternative to antibodies called aptamers.Aptamers are nucleic acid or peptide-based molecules that are engineered to bind to a specific target molecule (a chemical compound, protein, nucleic acid sequence, or even cell types or entire organisms). To create aptamers, a lab (or a scientific supply company) exposes the target molecule to a library of random DNA, RNA, and/or protein molecules to determine which molecules have the highest affinity. Aptamers have proven useful for identifying and labeling substances within the brain, such as amyloid plaques in murine and human brains. Because Inject antigen into mice Fuse cells to produce hybridomas Culture myeloma tumor cells Extract immunogenic B cells from spleen Culture hybridomas and select for cells that express antibodies Clone and expand colony of antibody-producing cells to mass-produce antibody FIGURE 15.4  Producing a monoclonal antibody. A scientist injects an antigen into an animal, usually a mouse. The mouse’s immune system naturally produces antibodies as a response. After several days, the scientist extracts immunogenic B cells that produce antibodies from the spleen. These cells are fused with tumor cells that divide indefinitely. After culturing these cells and selecting for hybrids that continuously produce an antibody of interest, a scientist can mass produce a monoclonal antibody. Biochemical Assays and Intracellular Signaling Chapter | 15  317 Inject animal with antigen and allow time for B cells to produce antibodies Remove blood Centrifuge to collect blood serum Purify antibodies on a column containing the antigen FIGURE 15.5  Producing a polyclonal antibody. A scientist injects an antigen into an animal, often multiple times over the course of weeks. The animal’s immune system produces many antibodies against the antigen as a response. The scientist then collects blood, separates the serum from the red blood cells, and purifies the serum using affinity chromatography. BOX 15.1  Creating and Using Tagged Proteins Most of the techniques used to detect specific proteins rely on antibodies. However, antibodies do not exist for all native proteins. Furthermore, producing good antibodies can sometimes be difficult and expensive, and the antibodies are not guaranteed to be effective for all techniques. Therefore, many scientists use standard recombinant DNA techniques (Chapter 10) to add short amino acid tags to a protein of interest. The genetic sequence for the protein of interest can be inserted into a plasmid vector containing the sequence for one of these tags to create a fusion protein. The scientist can then use antibodies against the tag to examine or purify the tagged protein. Common tags include c-Myc, His (histidine), FLAG, GST (glutathione-S-transferase), and HA (hemagglutinin). FLAG and GST tags are often used for western blot and IP assays, while c-Myc and HA tags are often used in immunohistochemistry. These peptide tags have a variety of origins: the polyhistidine (His)-tag was developed commercially for protein purification, while HA is derived from human influenza hemagglutinin, a viral gene, which may explain the ability to easily produce antibodies to this peptide sequence. 318  Guide to Research Techniques in Neuroscience aptamers are able to isolate molecules in vivo, they also have therapeutic potential in the treatment of neurodegenerative diseases and cancer. Purifying Proteins The ability to purify a protein of interest is fundamental to biochemical research. Purifying proteins allows an investigator to generate antibodies, determine a protein’s sequence, and determine a protein’s binding partners. Two common TABLE 15.1  Techniques that Depend on Protein-Specific Antibodies Technique Role of antibody in technique Western blot Antibody binds to a protein that has been run on a polyacrylamide gel to mark the presence, approximate size, and relative quantity of the protein in a sample. Enyzme-linked immunosorbent assay (ELISA) Antibody binds to a protein to measure the quantity of the protein in a sample. Radioimmunoassay Antibody binds to radioactive proteins of known concentration. A sample containing nonradioactive proteins is added to compete with the radioactive proteins, allowing a scientist to determine the protein concentration in a sample. Immunohistochemistry Antibody binds to a protein to show spatial expression in tissue or cells. Immunoprecipitation Antibody is bound to small beads. A sample is mixed with the beads, so the antibody binds and purifies a protein. The protein sample is then eluted off the beads. Co-immunoprecipitation Antibody is bound to small beads. A sample is mixed with the beads, so the antibody binds and purifies a protein. After the protein is eluted, the investigator uses a separate antibody in conjunction with a western blot or ELISA experiment to detect the presence of any secondary proteins that may interact with the immunoprecipitated protein. Chromatin immunoprecipitation Antibody is bound to small beads in a column. A sample is run through the column, so the antibody binds and purifies a protein. After the protein is eluted, the investigator uses PCR to determine the presence of any bound DNA sequences. Antibody interference Antibody binds to a protein, preventing the proper function of the protein. Biochemical Assays and Intracellular Signaling Chapter | 15  319 methods of purifying proteins include chromatography and immunoprecipitation (IP). Chromatography Chromatography allows a mixture of proteins (and other biomolecules) in a solution to be separated on a column. In this procedure, a scientist runs the solution through a column containing a porous matrix (Figure 15.6). The proteins pass through the column differently depending on how they interact with the matrix, and therefore, proteins can be differentially collected as they flow out the bottom of the column. Within the matrix, proteins can be separated on the basis of many different properties. For example, in gel-filtration chromatography, proteins descend through a column based on their size. In ion-exchange chromatography, proteins are separated based on their charge. A more specific way to purify a protein, affinity chromatography, uses small molecules fixed to the porous matrix to bind to proteins that may have a high affinity, such as a substrate binding to an enzyme. After the solution is run through the column and the protein of interest is bound to the matrix, the proteins are eluted from the column by changing the salt concentration or pH and the purified sample is acquired in a small collection tube. Immunoprecipitation Immunoprecipitation (IP) uses antibodies to purify a protein of interest out of solution (Figure 15.7). Specific antibodies are first bound to a mixture of beads. When a solution is mixed with these antibody-conjugated beads, the antibody FIGURE 15.6  Chromatography. A biological sample is run through a column composed of a porous matrix. Different columns have different properties and separate proteins based on properties such as their size, charge, and binding partners. Unpurified sample Purification column Collection tube 320  Guide to Research Techniques in Neuroscience will bind to its epitope, precipitating the protein out of solution. After the initial solution is washed away, a scientist can elute the protein off the beads by removing the antibodies, usually by changing the salt concentration or pH of the solution. This elution product can be further purified, although this can also reduce the yield of the final product. Taken together, the ability to purify proteins based on their binding partners, charge, size, and other properties, as well as the ability to use antibodies to purify proteins out of solutions, allow scientists to perform many cell signaling assays to investigate proteins in the nervous system. INVESTIGATING PROTEIN EXPRESSION A fundamental question in any cell signaling experiment is whether a particular protein is expressed in a population of cells and, if so, how much of the protein is present. There are a variety of techniques a scientist may use to answer these questions, some more quantitative and others with better spatial resolution. A quick comparison of the strengths and limitations of each technique is presented in Table 15.2. Western Blot The western blot (WB) (also known as an immunoblot) is the most commonly used method to measure the expression of a protein in a tissue or cell sample (Figure 15.8). Before performing a WB, scientists extract proteins from freshly dissected brain tissue or cultured cells using a simple protocol that lyses the cells and removes DNA contaminants. The sample is then processed using what is known as sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (PAGE), or simply SDS-PAGE. SDS is a negatively charged detergent that (A) (B) (C) (D) (E) FIGURE 15.7  Immunoprecipitation. (A) Before an immunoprecipitation procedure begins, a protein of interest exists in a heterogeneous sample with many other proteins. (B) A scientist adds antibodies that recognize the protein to small beads and then adds the beads to the sample. (C) The antibodies bind to the proteins, causing them to adhere to the beads and precipitate out of solution. (D) The heterogeneous sample is washed away from the beads, and (E) the proteins can be removed from the beads by changing the salinity or pH of the solution. Biochemical Assays and Intracellular Signaling Chapter | 15  321 binds to hydrophobic regions of the protein, allowing it to become soluble. The detergent also causes the protein to take on a net negative charge, regardless of its amino acid composition. Therefore, the protein will migrate toward a positive electrode when voltage is applied during electrophoresis. The scientist loads the protein sample onto a polyacrylamide gel (Figure 15.8A). In the presence of an electric field, the gel acts as a physical filter so that larger proteins travel relatively slowly and smaller proteins travel relatively quickly. This allows the investigator to separate a heterogeneous mixture of proteins into a spectrum of proteins with different molecular weights. At this stage of the WB process, the investigator can visualize the protein bands on the polyacrylamide gel using a stain such as Coomassie blue. This dye will stain all of the proteins in each lane of the gel. The result can appear like a smear, with some prominent bands representing highly expressed proteins in the sample. A scientist might perform this step to demonstrate that the total amount of protein in each lane is the same. After the protein sample is run on a gel, the final step in a WB procedure is to identify the presence of a specific protein. Because the gel is relatively thick and extremely delicate, a scientist first transfers the proteins onto a thin, nitrocellulose membrane (Figure 15.8B). This procedure, referred to as the blotting process, uses current to transfer the proteins from the gel onto the membrane. Once the proteins have transferred to the membrane, the investigator can detect the presence of a particular protein by incubating the membrane in a solution TABLE 15.2 Methods of Measuring Protein Expression Technique Description Western blot Allows for the detection of a protein in a sample, with measurements of relative concentration between samples. Enzyme-linked immunosorbent assay Allows for measurement of protein concentration in a sample. More sensitive than a western blot. Can quantify specific amount of protein, rather than determine the relative amount of protein (as in a western blot). Radioimmunoassay Precise measurement of protein concentration in samples in which the protein may be very dilute. Immunohistochemistry Characterizes the spatial distribution of the expression of a protein in cells and tissues. Not very quantitative. Immunoelectron microscopy Characterizes the spatial distribution of the expression of a protein within cells at a very high resolution. Reporter proteins Characterizes the spatial distribution of the expression of a protein without the need to use immunohistochemistry. Can be performed in live cells or tissues. 322  Guide to Research Techniques in Neuroscience with an antibody specific to the protein (Figure 15.8C). Then the membrane is incubated in a solution with a secondary antibody that recognizes the primary antibody; this secondary antibody is coupled to a radioactive isotope, an easily detectable enzyme, or a fluorescent dye. The labeled antibody marks the presence of the protein as a discrete band within a lane of the gel. The scientist can compare the location of the protein on the blot to a standardized protein ladder that highlights specific molecular weights in kilodaltons. Therefore, the final result of a WB experiment is a depiction of the band of protein, along with its molecular weight and necessary controls (Figure 15.9). The most important control in a WB is a loading control, a measurement of a separate protein in the sample to ensure that each lane contains the same quantity of total protein. For example, an antibody to β-actin or GAPDH (glyceraldehyde 3-phosphate dehydrogenase), proteins expressed in all cells, is often used to ensure that each lane contains the same total amount of protein, allowing a scientist to compare the intensity of bands in different samples Sample wells Cathode (-) Anode (+) Polyacrylamide gel Protein samples added to gel Proteins migrate down gel and are separated by size Cathode (-) Anode (+) Gel Nitrocellulose Absorbent paper Proteins transferred from gel to nitrocellulose Nitrocellulose Secondary antibody Primary antibody Bound protein Marker Protein of interest detected and visualized using antibodies (C) (B) (A) FIGURE 15.8  The western blot. (A) Protein samples are run on an SDS-PAGE gel, which separates proteins on the basis of size. (B) The proteins within the gel are transferred to a thin nitrocellulose membrane. The gel and membrane are stacked and an electric field causes the proteins to migrate from the gel to the membrane. (C) To detect a specific protein of interest, a scientist incubates the nitrocellulose membrane in a primary antibody, and then a secondary antibody that recognizes the primary antibody. The secondary antibody contains a marker that can be used to visualize the protein on the membrane. Biochemical Assays and Intracellular Signaling Chapter | 15  323 to determine relative protein amounts. When characterizing a new antibody, a scientist should perform a negative control with a sample in which the specific protein is absent, as well as a positive control with a sample in which the specific protein is definitely present. These controls allow the scientist to verify that the antibody is specific and capable of reliably indicating the presence of the protein of interest. An investigator may wish to know not only which cells express a protein of interest, but also where, within those cells, that protein is localized. In this case, the investigator can use a process called cell fractionation to divide a sample into its cellular components before a WB experiment begins (Figure 15.10). Individual organelles within a cell vary in their weight and can be pulled out of solution through a series of centrifugation steps. Tissue is first dissociated into a liquid homogenate and then centrifuged through a series of spins, each increasing in time and force. After each step, the pellet containing specific cellular components is collected. Finally, the scientist can run each sample in a separate lane of a WB experiment to determine which fraction contains a particular protein. Enzyme-Linked Immunosorbent Assay An enzyme-linked immunosorbent assay (ELISA) is an alternative to a WB, used to measure the expression of a protein, especially a protein that may be expressed in extremely low quantities. A traditional WB requires milligrams of protein, whereas an ELISA is able to quantify nanograms. Antibody to protein of interest Antibody to 2nd protein of interest Loading control Protein samples FIGURE 15.9  How data are often presented in a western blot experiment. Western blot data are often presented as blots on gels. These dark blots represent signal detected from an antibody. The darker the band, the more protein was present in the sample. Each row usually represents detection from a specific antibody.A loading control is included to show that each well contains the same amount of starting protein. Usually, this is an antibody against β-actin or glyceraldehyde-3-phosphate dehydrogenase (GAPDH). A reasonable conclusion from this mock figure would be that the first protein of interest is detected in samples 1, 3, and 4, but not sample 2. Furthermore, a second protein of interest is detected in samples 1 and 2, but not 3 or 4. 324  Guide to Research Techniques in Neuroscience A scientist binds an antibody to the bottom surface of a plate (Figure 15.11). A sample is added to the plate, and any protein present that can bind to the antibody becomes adhered to the surface. The sample is washed away, but the bound protein remains.A secondary detecting antibody that is linked to an enzyme is then applied to the plate, such that the protein of interest is sandwiched between both antibodies. A chemical substrate is added to the sample that is converted by the enzyme into a chromogenic or fluorescent signal. Finally, the scientist uses an optical device to measure the signal and determine the quantity of protein present. Radioimmunoassay A radioimmunoassay (RIA) is a highly sensitive method of measuring very low concentrations of proteins. Samples are usually derived from blood, extracellular fluid, or cerebrospinal fluid. To perform an RIA reaction (Figure 15.12A), a scientist first labels a known quantity of protein with a radioisotope, usually an isotope of iodine (125I or 131I), as iodine readily attaches to the amino acid tyrosine. The radiolabeled protein is then mixed with an antibody specific for that protein. This antibody–protein mixture is divided into several samples. To the first sample, the scientist adds a known amount of unlabeled, nonradioactive protein. This protein competes for antibody binding sites with the radioactive form of the protein. In the subsequent samples, the scientist continually increases the concentration of nonradioactive protein, increasing the amount of displaced radioactive protein. Finally, the scientist removes all proteins bound to antibodies from the samples using a secondary antibody so that only unbound proteins remain. The radioactivity of each sample is measured. From these data, a scientist can create a standard binding curve reflecting the ratio of bound to unbound protein versus the concentration of nonradioactive protein (Figure 15.12B). To measure the concentration of a protein from a biological sample, the scientist runs this unknown quantity of protein in parallel with the known preceding quantities. Once the scientist identifies the ratio of bound to unbound Biological sample 800 g 10 min 15 min 60 min 2 h 20,000 g 100,000 g 150,000 g Pellet contains nuclei, cell debris Pellet contains mitochondria Pellet contains membrane and vesicles Pellet contains ribosomes FIGURE 15.10  Cell fractionation. A biological sample can be centrifuged at successively higher speeds to collect specific organelles within cells. Biochemical Assays and Intracellular Signaling Chapter | 15  325 protein, it is possible to use the standard binding curve to deduce the concentration of the unknown sample (Figure 15.12C). A RIA is sensitive and can measure extremely small protein concentrations, such as the concentration of neuropeptides in the extracellular fluid of brain tissue. However, this technique requires specialized equipment and is relatively expensive. It also requires safety precautions due to the use of radioactivity. Immunohistochemistry IHC was previously described in Chapter 6 as a method to visualize the expression of proteins in cells or brain sections. This technique can be used to determine the spatial expression pattern of a protein, a relative advantage over the WB, ELISA, and RIA techniques described previously. IHC is not a very good technique for quantifying the amount of protein present in a sample, though it FIGURE 15.11  Enyzme-linked immunosorbent assay. A scientist coats the bottom of a plate with antibodies specific to a protein of interest and then adds a biological sample in solution. The proteins bind to the antibodies, and the sample is washed off the plate. Other antibodies conjugated to a marker are added to the plate, allowing a scientist to quantify the amount of protein in the sample. Add solution with heterogeneous mixture of proteins Wash solution off plate Add additional layer of antibodies with marker Add antibodies to bottom of plate 326  Guide to Research Techniques in Neuroscience is possible to compare the relative signal from two different samples to show a relative difference in expression. Immunoelectron Microscopy Immunoelectron microscopy (IEM) is an extension of IHC used in combination with electron microscopy to visualize protein expression within subcellular structures. Like a cell fractionation procedure followed by a WB, IEM can be (A) (B) (C) Standard binding curve %ofbound radioactiveprotein Concentration of nonradioactive protein Determination of concentration of unknown value %ofbound radioactiveprotein Concentration of nonradioactive protein Solutions with mixtures of radioactive and nonradioactive proteins Radioactive protein Nonradioactive protein FIGURE 15.12  A radioimmunoassay. (A) A scientist mixes different concentrations of nonradioactive and radioactive versions of a protein in a series of reaction tubes with an antibody against the protein. The nonradioactive and radioactive proteins compete to bind with the antibody. The scientist then collects the antibodies and bound proteins using a secondary antibody so that only unbound proteins remain. (B) By measuring the radioactivity of each sample, the scientist can produce a standard binding curve, comparing the amount of radioactive protein bound to an antibody with the concentration of nonradioactive protein added to the sample. (C) This binding curve can be used to determine the unknown concentration of a protein from an experimental sample. Biochemical Assays and Intracellular Signaling Chapter | 15  327 used to show that a protein is localized in a particular subcellular compartment. A cell fractionation experiment is more quantitative, whereas IEM is more spatially precise. Therefore, these techniques can be used in combination to provide strong evidence about protein distribution within a cell. Reporter Proteins To visualize proteins without the need for IHC, scientists can use recombinant DNA technology to attach a reporter protein, such as green fluorescent protein, to proteins of interest (Chapter 6). Tagging a protein of interest with a reporter protein is more time-consuming than IHC, as various recombinant DNA constructs, gene delivery strategies, and/or transgenic strategies must be employed before an experiment can take place. However, this method overcomes some of the limitations of IHC. For example, reporter genes allow an investigator to determine the expression and location of a protein in living cells. INVESTIGATING PROTEIN–PROTEIN INTERACTIONS Almost all proteins function within the cell by interacting and/or forming a complex with other proteins. As the term intracellular signaling implies, proteins communicate with other proteins within the cell in a chain of events that serve as the basis of normal cellular physiology and metabolism. Therefore, a scientist might ask whether a specific protein interacts with another protein of interest. Alternatively, a scientist might want to know the identity of all proteins that interact with a protein of interest. The following techniques address these questions. Co-Immunoprecipitation One of the simplest methods of determining if a protein physically interacts with another protein is by performing a co-immunoprecipitation (Co-IP) assay. This technique is an extension of the IP assay described previously (Figure 15.7). A scientist lyses cells to extract proteins and then mixes the solution with beads linked to antibodies. An antibody directed toward one of the proteins in the complex immunoprecipitates the protein out of solution. Any proteins that are tightly associated with this protein will also precipitate out of solution and be released in the elution process (Figure 15.13). The eluted proteins can be separated and identified using a WB protocol with antibodies for the immunoprecipitated protein and potential binding partner (Figure 15.14). If the identities of the potential binding partners are unknown, they may be determined using a technique called mass spectrometry (Box 15.2). Note that Co-IP does not show if a protein directly interacts with another protein; the two proteins may exist in a complex, indirectly bound with intermediary proteins. 328  Guide to Research Techniques in Neuroscience Protein Affinity Chromatography A scientist can use protein affinity chromatography to isolate and identify proteins that physically interact with each other. To capture the interacting proteins, a target molecule is attached to polymer beads within a column. These target molecules have specific interactions with the protein of interest that will capture IP: Protein A WB: Protein A IP: Protein A WB: Protein B Protein B 1 2 3 4 Used to ensure that the starting sample had an equal amount of Protein B This is a western blot against Protein B, after an IP experiment with an antibody against Protein A This is a western blot against Protein A after an IP experiment with an antibody against Protein A Protein samples FIGURE 15.14  How data are often presented in a co-immunoprecipitation experiment. Co-immunoprecipitation experiments are often followed by western blots. Just as in a western blot, the dark blots represent signal detected from an antibody. The darker the band, the more protein was present in the sample. A co-IP figure must report the antibody used to immunoprecipitate a protein as well as the antibody used to detect the protein in the western blot. A reasonable conclusion from this mock figure would be that Protein B interacts with Protein A in samples 1 and 3, but not in samples 2 and 4. This result is not because Protein B is absent from samples 2 and 4, as the bottom gel shows presence of this protein in all four initial samples (before the IP took place). (A) (B) (C) (D) (E) FIGURE 15.13  Co-immunoprecipitation. (A) Before an experiment begins, a protein of interest exists in a heterogeneous sample bound to other proteins. (B) A scientist adds antibodies that recognize the protein to small beads and then adds the beads to the sample. (C) The antibodies bind to the proteins, causing them to adhere to the beads and precipitate out of the solution. Any proteins that interact with the protein will also adhere to the beads. (D) The heterogeneous sample is washed away from the beads, and (E) the proteins can be removed from the beads by changing the salinity or pH of the solution. At this point, the scientist can use other methods, such as a western blot, to identify the interacting proteins. Biochemical Assays and Intracellular Signaling Chapter | 15  329 BOX 15.2  Mass Spectrometry Mass spectrometry (MS) is a technique used to identify and sequence proteins by breaking them apart and determining the mass to charge ratio of their fragments. Automated programs compare the calculated mass of protein and peptide fragments from a sample with the masses computed from protein databases. Thus, MS can identify components of an unknown mixture. In de novo sequencing, MS can determine the amino acid sequence of an unknown peptide. By assembling many overlapping peptides, the whole sequence of a protein may be determined, though this is rarely done in practice. MS is often used to identify posttranslational modifications (PTMs), such as phosphorylation or acetylation, and the residues on which they reside. This technique is extremely sensitive and requires little material (subpicomole amounts). Determining the masses of proteins and peptides requires the sample to be in the gas phase, and two techniques have been developed to do this: matrix-assisted laser desorption ionization (MALDI) and electrospray ionization (ESI). In MALDI, peptides are mixed with a UV-absorbing organic acid and dried on a metal slide. The sample is then blasted with a UV laser; the matrix absorbs the laser light, causing it to go into the gas phase carrying the sample (often peptides), which become charged. In ESI, the sample is in an acidic solution that charges the sample. The charged sample travels through a capillary into a heated area where solvent molecules are lost due to the heat and the increasing charge on the droplets: as the charge density becomes higher and higher, the droplets explode, becoming finer and finer until only single molecules remain. Once the molecules enter a gas phase, their mass-to-charge ratio can be measured. Many different types of detectors exist for this purpose; commonly coupled with MALDI is a time of flight (TOF) detector. In such an instrument, ionized peptides are accelerated in an electrical field and fly toward a detector. The time it takes each peptide to reach the detector is determined by its mass and charge, with large peptides moving slowly and highly charged peptides moving relatively quickly. Other detectors use electrical or magnetic fields to selectively filter or detect different mass to charge ratios. Many molecules can have the same mass-to-charge ratio despite having very different chemical structures. For example, the peptides PEPTIDE and TIDEPEP are indistinguishable by their mass alone. This situation is common and occurs with proteins, nucleic acids, and organic molecules. To distinguish these types of molecules, they may be fragmented and the masses of their constituent components measured; with peptides, this usually involves smashing them with gas molecules such as argon or nitrogen, to cleave the peptide bonds. The result of this collision induced dissociation (CID) is overlapping fragments that reveal the original sequence of the peptide. In the preceding example, observing EPEP or PEPT would distinguish between the two possibilities. Ideally, collisions create a virtual ladder of fragments, each differing by only one amino acid. In practice, a perfect ladder is rarely observed, but enough detail is present to match a peptide to known protein sequences. MS is a core tool in the field of proteomics, the systematic characterization of the complete set of proteins in a system at a particular time, allowing for insight Continued 330  Guide to Research Techniques in Neuroscience the protein as it passes through the column. They can be ligands for a receptor, substrates for an enzyme, or antibodies for an antigen. Cellular proteins are washed through the column, and the proteins that interact with the target attach to the matrix, while all other proteins pass through. The proteins that interact with the target are eluted off the beads, and their identity can be determined using a WB. Just as in the co-IP assay, if the identities of the interacting proteins are unknown, they can be determined using mass spectrometry (Box 15.2). Also similar to a co-IP assay, this technique does not show if the two proteins directly interact or exist in a larger protein complex. Yeast Two-Hybrid Assay A yeast two-hybrid assay investigates protein–protein interactions by exploiting a transcription system normally used by yeast cells. In a normal yeast cell, a transcription factor called Gal4 binds to a promoter region called an upstream activating sequence (UAS). Gal4 is composed of a binding domain (BD), which binds to the UAS, as well as an activation domain (AD), which initiates transcription of a target gene. In a yeast two-hybrid assay, a scientist uses recombinant DNA technology to divide the Gal4 protein into these two separate domains (Figure 15.15A). The Gal4 binding domain is fused to a protein of interest thought to interact with another protein. This protein can be considered the bait in that its role in the experiment is to attract other binding partners. The scientist fuses the Gal4 activation domain to a potential binding partner. This binding partner can be considered prey in that it could potentially bind with the bait. If the two proteins do not interact, the BD and AD fragments will be physically separated, and no transcription will occur. However, if the two proteins are able to bind, the Gal4 BD and AD fragments will become physically close enough to cause transcription of the downstream coding sequence (Figure 15.15B; Box 15.3). If the gene adjacent to the UAS is a reporter gene, such as lacZ, the scientist will be able to identify the interaction between the two proteins. A yeast two-hybrid assay is useful to test hypotheses that two proteins interact, but this assay can also be used as a screen to identify potential binding partners for a protein of interest (Figure 15.16). For example, a scientist can identify unknown “prey” for a protein of interest by attaching the activation domain to a large mixture of DNA fragments from a cDNA library (Chapter 10). into signaling networks and protein interactions. Proteomics involves the analysis of many proteins at the same time—for example, identifying variations in protein expression in the brains of patients with Alzheimer’s disease compared to healthy brains. MS (along with bioinformatics tools to manage the large amounts of data) allows for the profiling of these thousands of different proteins. BOX 15.2  Mass Spectrometry—cont’d Biochemical Assays and Intracellular Signaling Chapter | 15  331 Individual ligation products are then introduced into yeast cells containing the target protein. If one of the members of the library encodes a protein that interacts with the target protein, the two Gal4 fragments will interact and activate transcription of the reporter. The scientist can then identify the yeast colony expressing the reporter, purify and sequence the DNA attached to the activation domain, and determine the identity of the binding protein. While this assay shows that the proteins can directly interact, it does not show that they interact in vivo or even that they can interact in mammalian cells. INVESTIGATING POSTTRANSLATIONAL MODIFICATIONS A posttranslational modification (PTM) is a biochemical modification that occurs to one or more amino acids on a protein after the protein has been translated by a ribosome. One of the most commonly studied PTMs is the addition Promoter Reporter gene (A) (B) Expression of reporter gene DNA-binding domain Transcriptional activation domain Target protein Binding partner FIGURE 15.15  Yeast two-hybrid assay. (A) A scientist uses recombinant DNA technology to fuse the DNA binding domain of Gal4 to a protein of interest and the transcriptional activation domain of Gal4 to a potential binding partner. These constructs are introduced into yeast cells. If the two proteins do not interact, the DNA binding domain and transcriptional activation domain of Gal4 do not interact. However, (B) if the two proteins do interact, then the Gal4 complex is complete and can activate transcription of a reporter gene. 332  Guide to Research Techniques in Neuroscience Recombined cDNA library clones tagged to transcriptional activation domain of Gal4 Recombined protein of interest tagged to DNA-binding domain of Gal4 Randomly transform yeast cells Yeast cells on plate Yeast cells on plate Wait 1-3 days for transcription to occur Examine cells for presence of reporter protein FIGURE 15.16  Using a yeast two-hybrid assay as a screen. Recombinant DNA technology is used to fuse thousands of genetic sequences from a cDNA library to the Gal4 transcriptional activator. These constructs, as well as a construct in which a protein of interest is fused to the Gal4 DNA binding domain, are randomly introduced into yeast cells. In some yeast cells, a target protein will interact with the protein of interest, and the Gal4 will activate expression of a reporter gene. A scientist can visualize the expression of the reporter by examining the color of cell colonies grown on a plate. To determine the identity of the protein binding partner, the scientist can extract and sequence the DNA from these colonies. Biochemical Assays and Intracellular Signaling Chapter | 15  333 BOX 15.3  Walkthrough of an Intracellular Signaling Experiment Let us say that you work in a laboratory that studies a specific aspect of neural development: how a neuron’s axon correctly extends to its postsynaptic target. The tip of an axon is referred to as the axonal growth cone, and it typically reaches its destination by responding to extracellular guidance cues produced by other cells. There are three general steps that occur to guide the growth cone: (1) an extracellular guidance cue is secreted to guide the axon’s position; (2) a receptor on the axon’s surface recognizes that cue and transduces it into an intracellular signaling cascade that affects other proteins within the developing neuron; and (3) various downstream organelles receive and respond to the signal, including the cytoskeleton that must be reorganized to move the growth cone toward the site where the initial cue was detected. If you know the identity of the receptor and the extracellular cue, you might be interested in figuring out how the receptor translates the cue into a response. What is the intracellular signaling pathway that leads to the cytoskeletal changes necessary for the growth cone to turn toward the cue? Let us focus on identifying the next step of signaling that occurs after the receptor is activated: the activation of a second protein that binds to the intracellular domain of the receptor. There are multiple approaches for finding proteins that interact with the receptor. For example, you could immunoprecipitate the receptor and then perform mass spectrometry to identify any other proteins that co-precipitated. However, this approach would not tell you if the two proteins interacted directly or were simply part of a larger complex of proteins. Another approach would be to perform a yeast two-hybrid assay. This technique is used to identify protein-binding partners that directly interact with the receptor. To perform this assay, you would begin by genetically engineering a construct that attaches the receptor’s intracellular domain to the DNA-binding domain of a transcriptional activator, such as Gal4, to act as the bait. Then you would create or use a cDNA library to attach different cDNA molecules to the activation domain of Gal4 to act as the prey. Finally, you would introduce the bait and prey constructs into yeast cells that express a reporter gene, often lacZ, under the control of a promoter that is recognized by the transcriptional activator (UAS for the Gal4 activator). Thus, the bait will bind to the promoter but cannot initiate transcription of the reporter until a specific prey with the activation domain interacts with the bait to create the full activator. Then, the reporter gene will be transcribed and can be detected. To detect yeast cells that express lacZ, you could grow thousands of cells on an agar plate that also contains X-gal, the substrate that causes a visible blue byproduct to form in those cells. The final step would be to pick blue yeast colonies off the plate, extract and sequence the DNA of the prey, and identify the protein that interacts with the receptor based on the cDNA sequence. This screen allows you to conclude that the prey you found can interact with the intracellular domain of the receptor. Most likely, you will have discovered a number of positive hits to sequence and should confirm the interaction’s specificity to determine which would be a good candidate for a follow-up. This yeast two-hybrid experiment is a good start, but you could learn even more about the receptor–protein interaction by attaching different functional Continued 334  Guide to Research Techniques in Neuroscience or removal of a phosphate group, PO4 (Figure 15.17). Other common PTMs are listed in Table 15.3. These modifications often change the ultrastructural and functional properties of a protein, causing substantial downstream signaling effects. For example, the phosphorylation state of a protein often indicates whether it is functional or inactive. There are multiple questions a scientist may ask about a protein of interest regarding PTMs. For example, a scientist might want to know if a protein of interest receives a PTM.This question can be answered using mass spectrometry (Box 15.2) regions of the receptor’s intracellular end to use as bait. This would tell you with finer resolution exactly what part of the receptor is interacting with the prey or allow you to capture only prey that will interact with a specific region—say, the kinase domain—or a particular region you know to be functionally important. To convince yourself and other scientists of the interaction between the receptor and intracellular protein, you could perform additional, complementary experiments. For example, co-immunoprecipitation, as just described, is often used to verify yeast two-hybrid results. You could also use protein affinity chromatography. In cell signaling studies, there is no such thing as too much evidence; figures often contain several different methods to validate the same result. Now that you have a candidate (or several candidates, based on the results of the yeast two-hybrid experiments), you should follow up with other experiments to make sure that your candidate protein makes sense as a protein that mediates the intracellular response to a guidance cue. One question you might ask right away is whether the candidate protein is expressed in the right place (the axon growth cone) and at the right time (the embryonic or postnatal period during which the axon grows toward its target). If the interacting protein is not present in axonal growth cones or at a time when axons are growing out, the interaction may not be relevant to the growth cone turning behavior you’re interested in, or the interaction may not actually occur. To examine protein expression, you could use immunohistochemistry to stain for the expression of the protein in histological preparations of the growth cone at different developmental time points. You could also collect tissue samples containing the growth cones from different time points and perform western blots to examine the expression of the protein at each time point. If you have determined that a protein binds to the receptor, is expressed in the growth cone, and is expressed during the developmental period in which the growth cone migrates toward its target, you have laid the foundation for many future experiments. For example, you could ask whether the interaction between the receptor and the protein causes the protein to be phosphorylated. You could use techniques described in other chapters, such as RNA interference (RNAi— Chapter 13), to investigate the consequence of knocking down the expression of the intracellular protein. There are many future experiments that can and should be performed, and the more complementary lines of evidence you collect, the more you will know, and the better your study will be received. BOX 15.3  Walkthrough of an Intracellular Signaling Experiment—cont’d Biochemical Assays and Intracellular Signaling Chapter | 15  335 to analyze the molecular weights of protein fragments, identifying fragments with a PTM. Other questions include whether a specific enzyme is responsible for modifying a protein of interest or how the PTM changes over time following a stimulus. These questions can be answered using the following techniques. ATP P Protein kinase ADP FIGURE 15.17  Phosphorylation. A protein kinase adds a phosphate group from ATP to a substrate protein. TABLE 15.3  Common Posttranslational Modifications (most of these posttranslational modifications can be reversed by opposing enzymes) Classification Chemical modification Type of enzyme mediating the modification Phosphorylation Addition of a phosphate group (-PO4) to a residue of serine, tyrosine, threonine, or histidine Protein kinase Acetylation Addition of an acetyl group (-COCH3) at the N-terminus of a protein or a lysine residue Acetyltransferase Methylation Addition of a methyl group (-CH3) at a residue of lysine or arginine Methyltransferase Glycosylation The addition of a glycosyl group to a residue of asparagine, hydroxylysine, serine, or threonine, resulting in a glycoprotein (Multiple types of enzymes mediate this reaction) Sulfation Addition of a sulfate group (SO4) to a tyrosine residue Tyrosylprotein sulfotransferase Farnesylation Addition of a farnesyl group (C15H26O) to the C-terminus of a protein. Usually causes the protein to anchor to the cell membrane Farnesyltransferase Ubiquitination Addition of a ubiquitin protein to a lysine residue. Usually targets proteins for proteasomal degradation. Ubiquitin ligase 336  Guide to Research Techniques in Neuroscience PTM-Specific Assays PTM-specific assays are designed to test the hypothesis that a particular protein mediates the PTM of another protein. Each PTM has its own biochemical assay. Here, we describe a common type of assay: a kinase assay (Figure 15.18). A kinase assay is used to determine whether one protein is capable of phosphorylating another protein. As mentioned previously, protein kinases are enzymes that catalyze the covalent transfer of phosphate from adenosine triphosphate (ATP) to a substrate protein. A scientist can incorporate a radioactive label, such as [32P] orthophosphate, into molecules of ATP. This radioactive ATP acts as the phosphate donor for the protein kinase substrate. To perform this assay, the scientist combines the hypothetical protein kinase, the substrate protein, and radioactive ATP into a reaction tube. Next, the scientist isolates the substrate protein using immunoprecipitation and visualizes the presence of 32P using autoradiography (Figure 15.19). FIGURE 15.18  A kinase assay. To determine if a protein kinase is able to phosphorylate a protein, a scientist mixes the kinase, the protein, and radioactive ATP into a reaction tube. Fifteen minutes later, the scientist immunoprecipitates the protein and runs samples on a gel to detect for the presence of radioactivity. Substrate protein Protein kinase 32P] ATP + + 32P WB with antibody against substrate protein 1 2 3 4 Protein samples FIGURE 15.19  How data are often presented in a kinase assay. The 32P will only be present if the protein kinase is able to phosphorylate the substrate protein. To ensure that the amount of substrate protein was equal in each sample, a scientist typically performs a western blot using an antibody against the substrate protein. Alternatively, a scientist could stain the gel with Coomassie blue. A reasonable conclusion from this mock figure would be that the protein kinase under investigation was able to phosphorylate a protein in sample 2 but not samples 1, 3, or 4. Biochemical Assays and Intracellular Signaling Chapter | 15  337 Kinase assays are easily the most common PTM assay represented in the literature, as at least 2% of the genes in the mammalian genome code for protein kinases. However, there are many different PTM-specific assays: for example, methyltransferase or acetyltransferase assays are used to determine whether a protein is capable of adding a methyl or acetyl group, respectively, to a target protein. PTM-Specific Antibodies It is often possible to produce an antibody to a protein of interest that is directed against a site on the protein that receives a PTM (Figure 15.20). Such an antibody binds to its substrate protein only when the PTM is present. PTM-specific antibodies can be used in any of the previously described antibody-based techniques to assay the extent of modifications, or with spatially sensitive techniques such as IHC or IEM to determine the localization of modified proteins. Such analysis is especially informative when dealing with dynamic changes in protein localization and activation. By exposing a given sample to various molecular signals, such as growth factors, a scientist can assay any changes in localization as well as changes in activity corresponding to the modifications. INVESTIGATING PROTEIN–DNA INTERACTIONS The final step in many intracellular signaling pathways is the activation of a protein that directly affects gene transcription. Scientists often ask whether a specific protein of interest is able to bind to DNA and affect the expression of a specific target gene. Three techniques are commonly used to examine interactions between proteins and DNA: the electrophoretic mobility shift assay, the ChIP assay, and the luciferase assay. Because these techniques answer slightly different questions about protein–DNA interactions, scientists often use them in concert to strengthen the evidence that a protein interacts with DNA and affects the expression of target genes. P FIGURE 15.20  A posttranslational modification-specific antibody. The antibody binds to an epitope on a protein that is phosphorylated at a specific amino acid residue, but not when the phosphate group is not present. 338  Guide to Research Techniques in Neuroscience Electrophoretic Mobility Shift Assay An electrophoretic mobility shift assay (EMSA, also known as a gel shift assay) is used to determine if a protein is able to directly interact with a short, specific sequence of DNA. Before the experiment begins, the investigator hybridizes two complementary DNA strands (about 30–40 bp in length) and labels the strands with a radioactive probe. The investigator also purifies the protein that is hypothesized to interact with the strand of DNA. During the experiment, the radiolabeled DNA is run in several lanes of a polyacrylamide gel (Figure 15.21A). The speed at which the DNA molecules Cathode (-) Anode (-) DNAmigration 1 2 6543 1 2 6543 (A) (B) FIGURE 15.21  Electrophoretic mobility shift assay. (A) To determine if a protein is capable of interacting with a short sequence of double-stranded DNA, a scientist radioactively labels short DNA sequences, mixes these sequences with purified proteins, and runs the mixture on a gel. Any DNA sequences that interact with a protein will migrate much slower on the gel. Lane 1 represents DNA alone. Lane 2 represents DNA with an interacting protein. Lane 3 represents a scrambled control, in which the nucleotide sequence is jumbled. Lane 4 represents a reaction in which a sufficient amount of nonradiolabeled DNA is added to the mixture, taking a high percentage of the interactions with the binding protein, allowing the radiolabelled DNA band to be shifted back to normal. Lane 5 represents a reaction that includes an antibody to the protein. The antibody–protein interaction increases the macromolecular complex, “supershifting” the DNA so it moves more slowly down the gel. Lane 6 represents a reaction in which a protein is included that is known not to bind to DNA, such as a membrane-bound protein. (B) How data are often presented in an EMSA reaction, corresponding to the treatments described above. The dark gradient at the bottom represents the unbound, radioactive DNA strands that have migrated to the bottom of the gel. Biochemical Assays and Intracellular Signaling Chapter | 15  339 move from one side of the gel to the other during electrophoresis is determined by their size. In one lane, the DNA is run by itself. In another lane, the DNA is run with the purified protein. If the protein interacts with the DNA strand, the size of the DNA–protein complex will be greater than the DNA strand alone, and therefore, the band of DNA will be shifted upward on the gel (Figure 15.21B). Several controls are necessary for a proper EMSA experiment: (1) in addition to the DNA sequence of interest, the investigator should run a slightly scrambled DNA sequence to ensure that the DNA–protein interaction is specific to the exact sequence. (2) The investigator should perform a competitive binding assay in which nonradiolabeled DNA is added to the reactions with the radiolabeled DNA. If there is sufficient nonradiolabeled DNA, the amount of protein bound to the radiolabeled DNA will decrease, and the size of the radiolabeled DNA band will be shifted back to normal. (3) An antibody to the purified protein should be added to a lane with the protein and labeled DNA. This will further increase the size of the DNA–protein complex and create a “supershifted” band, demonstrating that it is indeed the protein of interest that is causing the shift in the first place. (4) The investigator could run a nonspecific protein known not to interact with DNA, such as a membrane-bound protein, to ensure that there is nothing “sticky” about the DNA strand in the EMSA reaction. One of the strengths of an EMSA assay is that it demonstrates that a protein can directly bind to a sequence of DNA. One of the limitations is that all reactions take place in a test tube. Therefore, the only conclusions that can be reached using this assay is that a protein is capable of binding DNA, but not that it actually does so in a living cell. For this conclusion, it is necessary to perform a ChIP assay. Chromatin Immunoprecipitation ChIP is used to determine if a protein interacts with a specific region of DNA. The advantage of ChIP over an EMSA assay is that the starting material comes from living cells, most often cultured cells, though fresh tissue is sometimes used. However, unlike an EMSA, it is impossible to determine if a protein interacts with a sequence of DNA directly or if it is part of a greater macromolecular complex. The fundamental concept behind a ChIP assay is that before the experiment begins, a protein of interest is physically bound to DNA or is part of a DNA–protein complex in living cells (Figure 15.22). The scientist begins the experiment by quickly fixing the cells in a diluted formaldehyde solution so that DNA–protein complexes are cross-linked and remain intact. Next, the DNA is sonicated and sheared into approximately 600–1000 bp fragments. If the crosslinking is adequate, the proteins will still be bound to sonicated DNA. At this stage, the scientist immunoprecipitates the DNA–protein complexes using an antibody against the protein of interest. 340  Guide to Research Techniques in Neuroscience Aftertheimmunoprecipitationstep,thepurifiedproteinsarereversecross-linked so that the DNA fragments can be purified. If the scientist has a hypothesis about the identity of the bound DNA sequence, he or she can then determine the identity of the DNA fragments by performing a polymerase chain reaction (PCR) reaction (Chapter 10) with primers for a specific region of DNA. The final result of a ChIP assay is usually a photograph of an agarose gel that reveals the end product of PCR reactions (Figure 15.23). Alternatively, quantitative real time PCR (qRT-PCR) can also be performed to better quantify the amount of PCR product in each lane. What if a scientist does not have a hypothesis as to which DNA sequences are bound to a protein of interest? It is possible to screen all DNA sequences in the genomes of commonly used organisms by performing a ChIP on chip Genetic sequence A IgG Protein X Input Genetic sequence B FIGURE 15.23  How data are often presented in a ChIP assay. The bands represent PCR products, i.e., DNA run on an agarose gel. The labels at top are the immunoprecipitating antibodies. IgG is a random immunoglobulin and should not show any signal. The input lane represents the solution before running the immunoprecipitation, so all DNA should be present. A reasonable conclusion from this mock figure would be that Protein X interacts with DNA sequence A but not sequence B. 1. Crosslink protein to DNA 2. Sonicate to shear DNA 3. Immunoprecipitate with antibody to protein 4. Reverse crosslinks and purify DNA 5. Run PCR using primers for DNA sequence of interest FIGURE 15.22  Chromatin immunoprecipitation. Biochemical Assays and Intracellular Signaling Chapter | 15  341 assay. In this assay, the DNA sequences obtained at the end of a ChIP experiment are run on a DNA microarray (Chapter 9) to determine all of the DNA sequences bound to the immunoprecipitated protein. Several positive and negative controls are necessary for a proper ChIP experiment.An important positive control is an input lane containing PCR products from DNA obtained after sonication but prior to immunoprecipitation. This reaction should result in a strong signal and ensures that the DNA of interest was properly obtained in the first place. Another important positive control is an immunoprecipitation reaction with an antibody for a protein known to bind to any region of DNA, such as a histone protein. ChIP results using a histone antibody should, therefore, produce a strong DNA signal. Negative controls include performing an immunoprecipitation with nonspecific antibodies (random IgG is usually used) and using primers that amplify nonspecific DNA regions. Luciferase Assay A luciferase assay is used to determine if a protein can activate or repress the expression of a target gene. Unlike the ChIP or EMSA assays, which only assess the ability of a protein to interact with a region of DNA, a luciferase assay is able to establish a functional connection between the presence of the protein and the amount of gene product that is produced. This assay is unable to determine whether the protein directly interacts with DNA itself; the protein could indirectly affect transcription by activating or repressing a separate protein or protein complex that affects transcription. Luciferase is an enzyme used for bioluminescence by various organisms in nature, most famously the firefly. The scientist produces a construct in which the regulatory region of a target gene is fused with the DNA coding sequence for luciferase (Figure 15.24). A separate DNA construct encodes the protein hypothesized Promoter region Luciferase reporter 5’ 3’ ? Hypothetical transcriptional activator FIGURE 15.24  A luciferase assay. To determine if a protein is able to activate (or suppress) transcription of a gene of interest, a scientist uses recombinant DNA technology to produce a construct in which the gene’s promoter is placed adjacent to a luciferase reporter gene. This construct, as well as a construct coding for the protein, is introduced into cultured cells. If the protein is able to activate transcription, the cell will produce the luciferase reporter. The amount of luciferase produced can be quantified using a luminometer. 342  Guide to Research Techniques in Neuroscience to affect transcription. The scientist transfects a cell culture system, such as HEK 293T cells, with both constructs. If the protein is able to upregulate transcription of the target gene, the cells will express luciferase. If the protein downregulates transcription, the cells will express less luciferase than normal. The scientist examines the expression of luciferase about 2–3 days after the initial cell transfection. The cells are lysed, and the cell contents are placed in a reaction tube. If the proper substrate is added, luciferase will catalyze a reaction that produces light. This light is detected with a luminometer, a device that precisely quantifies how much light is produced in each reaction tube. The amount of light produced provides a quantitative measure of the effect of the protein on expression of the target gene. Due to variability in conditions within each well in a cell culture plate, a proper luciferase assay experiment should repeat each condition at least in triplicate. The scientist should compare luciferase expression between conditions in which (1) no protein is present, (2) protein is present, and (3) the regulatory sequence regulating luciferase expression is mutated so that the protein is no longer able to bind. If a scientist knows enough about the structure of a protein such that it is possible to mutate the protein’s DNA-binding domain or enhance the ability of the protein to affect transcription, such conditions would also strengthen the conclusions of a luciferase assay experiment. CONCLUSION The purpose of this chapter was to survey common methods used to investigate protein expression and signaling. Although these methods are numerous, they essentially fall into four categories: (1) determining whether a protein is expressed in a particular population of cells and how much; (2) determining whether a protein interacts with other proteins; (3) determining the types of PTMs a protein can receive and which proteins cause these modifications; and (4) determining whether a protein is capable of interacting with DNA and/or regulating gene expression. Anyone outside the fields of biochemistry and cell signaling should not be intimidated by the huge variety of techniques. Focusing on the experimental questions rather than the methods can clarify the results and conclusions of an intracellular signaling study. SUGGESTED READING AND REFERENCES Books Berg, J. M., Tymoczko, J. L., & Stryer, L. (2010). Biochemistry (7th ed.). NewYork: W. H. Freeman. Fields, R. D. (2008). Beyond the synapse: Cell-cell signaling in synaptic plasticity. Cambridge: Cambridge University Press. Moss, T., & Leblanc, B. (2009). Protein interactions: Principles and protocols (3rd ed.). New York: Humana Press. Wilson, K., & Walker, J. M. (2010). Principles and techniques of biochemistry and molecular biology (7th ed.). New York: Cambridge University Press. Biochemical Assays and Intracellular Signaling Chapter | 15  343 Review Articles Araujo, S. J., & Tear, G. (2003). Axon guidance mechanisms and molecules: lessons from invertebrates. Nature Review Neuroscience, 4, 910–922. Grant, S. G. N., & Blackstock, W. P. (2001). Proteomics in neuroscience: from protein to network. Journal of Neuroscience, 21(21), 8315–8318. Greer, P. L., & Greenberg, M. E. (2008). From synapse to nucleus: calcium-dependent gene transcription in the control of synapse development and function. Neuron, 59, 846–860. Guan, K. L., & Rao, Y. (2003). Signalling mechanisms mediating neuronal responses to guidance cues. Nature Review Neuroscience, 4, 941–956. Tronson, N. C., & Taylor, J. R. (2007). Molecular mechanisms of memory reconsolidation. Nature Review Neuroscience, 8, 262–275. Primary Research Articles—Interesting Examples from the Literature Cowan, C. W., Shao, Y. R., Sahin, M., Shamah, S. M., Lin, M. Z., Greer, P. L., et al. (2005). Vav family GEFs link activated Ephs to endocytosis and axon guidance. Neuron, 46, 205–217. Flavell, S. W., Cowan, C. W., Kim, T. K., Greer, P. L., Lin, Y., Paradis, S., et al. (2006). Activitydependent regulation of MEF2 transcription factors suppresses excitatory synapse number. Science, 311, 1008–1012. Kim, M. S., Pak, Y. K., Jang, P. G., Namkoong, C., Choi, Y. S., Won, J. C., et al. (2006). Role of hypothalamic Foxo1 in the regulation of food intake and energy homeostasis. Nature Neuroscience, 9, 901–906. Kondo, M., Takei, Y., & Hirokawa, N. (2012). Motor protein KIF1A is essential for hippocampal synaptogenesis and learning enhancement in an enriched environment. Neuron, 73, 743–757. Koyano, F., Okatsu, K., Kosako, H., Tamura, Y., Go, E., et al. (2014). Ubiquitin is phosphorylated by PINK1 to activate parkin. Nature, 510, 162–166. Soden, M. E., Jones, G. L., Sanford, C. A., Chung, A. S., Guler, A. D., Chavkin, C., et al. (2013). Disruption of dopamine neuron activity pattern regulation through selective expression of a human KCNN3 mutation. Neuron, 80, 997–1009. Visel, A., Blow, M. J., Li, Z., Zhang, T., Akiyama, J. A., Holt, A., et al. (2009). ChIP-seq accurately predicts tissue-specific activity of enhancers. Nature, 457, 854–858. Protocols Aparicio, O., Geisberg, J. V., Sekinger, E., Yang, A., Moqtaderi, Z., & Struhl, K. (2005). Chromatin immunoprecipitation for determining the association of proteins with specific genomic sequences in vivo. Current Protocols in Molecular Biology (Chapter 21), Unit 21.3, http://dx.doi.org/10.1002/0471142727.mb2103s69. Golemis, E. A., Serebriiskii, I., Gyuris, J., & Brent, R. (2011). Interaction trap/two-hybrid system to identify interacting proteins. Current Protocols in Neuroscience (Chapter 4), Unit 4.4, http://dx.doi.org/10.1002/0471142301.ns0404s55. Harlow, E., & Lane, D. (1999). Using antibodies: A laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Pandey, A., Andersen, J. S., & Mann, M. (2000). Use of mass spectrometry to study signaling pathways. Science’s STKE, 2000, PL1. Walker, J. M. (2002). The protein protocols Handbook (2nd ed.). Totowa, NJ: Humana Press. Websites Resource for information on protein methods – expression, purification, antibody production: http://www.piercenet.com/method/protein-methods-library 345 Glossary Ablation  Permanent destruction of a cell or tissue, creation of a lesion. Absolute recovery  In microdialysis, the total amount of substance collected in the dialysate over a set amount of time. Action potential  A rapid, all-or-none, transient depolarization of the cell membrane of a neuron. Acute culture  A primary cell or tissue culture used for short-term experiments, usually lasting less than a day. For example, a slice culture for patch clamping experiments. Adeno-associated virus (AAV)  A type of virus used as a DNA delivery tool that infects dividing and postmitotic cells. The DNA integrates into the nucleus, permitting long-term gene expression. Adenovirus  A type of virus used as a DNA delivery tool that infects dividing and postmitotic cells. The DNA does not integrate into the genome, so expression is often transient, lasting weeks or months. Aequorin  A chemiluminescent protein derived from the jellyfish Aequorea victoria that emits blue light when bound to calcium. This protein has been useful for imaging calcium activity in living cells. Affinity chromatography  A form of chromatography that uses small molecules fixed to a porous column to bind to proteins with a high affinity, such as a substrate binding to an enzyme. Agonist  A compound that can bind and activate an endogenous receptor, thus mimicking an endogenous ligand. Allatostatin receptor  A ligand-gated receptor found in insects. In the presence of the insect hormone allatostatin, the receptor causes hyperpolarization of cells. Thus, this receptor can be used as a genetically encoded tool to inhibit neural activity. Amino acid  A category of molecule consisting of amine and carboxyl functional groups, as well as a variable side chain. Used as the monomeric building blocks of proteins. Amperometry  A form of voltammetry that uses an electrode to detect neurochemicals in living tissue. Unlike in other voltammetry techniques, the electrode is held at a specific, constant voltage. Amplifier  A device used in electrophysiology recordings to amplify the relatively weak electrical signals derived from a recording electrode. Anhedonia  The loss of desire to pursue an activity that is normally enjoyable. 346 Glossary Anisotropic  Describes diffusion of substances that is not equal in all directions but instead tends to move randomly along a single axis. The opposite of isotropic. Antagonist  A compound that can bind to a receptor but does not cause activation. Therefore, it prevents the biological activity of an endogenous ligand. Anterograde tracer  A chemical probe that labels efferent axon tracts from the cell body to the presynaptic terminals. Antibody  Specialized protein used by an organism’s immune system to detect and bind to foreign proteins. Used by neuroscientists to identify or purify proteins of interest. Antigen  A foreign substance recognized by an antibody. Aptamer  Engineered nucleic acid or peptide-based molecule that binds to a specific target molecule. Used as an alternative to antibodies. Archaerhodopsin (Arch)  A genetically encoded transmembrane pump that hyperpolarizes neurons upon stimulation with yellow light. ArcLight  A genetically encoded voltage indicator. Array tomography  A technique that combines immunohistochemistry techniques with ultrathin, serial sections. This technique improves spatial resolution of traditional immunohistochemistry experiments and allows scientists to create highly detailed three-dimensional images of protein expression. Artificial chromosome  A large DNA molecule that contains the telomeric, centromeric, and replication origin sequences necessary for replication and preservation in yeast, bacteria, or phage. Useful to neuroscientists as a vector for cloning relatively large DNA fragments (50–350 kb). Ataxia  A neurological condition exhibited as a gross lack of coordination of muscle movements. Ataxin  A genetically encoded toxin that can be used to ablate specific neurons of interest. Autoclave  A large device used to sterilize tools, glassware, and reagents. Autofluorescence  The natural ability of some structures to fluoresce, even without the addition of exogenous fluorescent tags or dyes. Autoradiography  A technique that uses the pattern of decay emissions produced from a radioactive substance to form an image on an X-ray film. Used in neuroscience to detect radioactive probes. Avidin  A small molecule with an extremely high affinity for another molecule called Biotin. Axial  An imaginary plane that divides the brain and body into superior and inferior sections (top to bottom). BAC  See Bacterial artificial chromosome. BAC construct/transgenic  A recombinant DNA or recombineering construct that uses a bacterial artificial chromosome as the vector. Useful for creating transgenic animals with large, endogenous sequences of DNA. Glossary  347 Bacterial artificial chromosome (BAC)  A large (50–350 kb) DNA vector capable of replicating in bacteria. Useful for manipulating large sequences of DNA, such as in the generation of transgenic or knockout constructs. Bacteriophage  Any one of a number of viruses that infect bacteria. Sometimes simply referred to as phage. Barnes maze  A maze used to test spatial learning and memory in rodents. The animal must learn the location of a drop box on a large circular table using spatial cues placed around the maze. Base pair  Describes a single nucleotide–nucleotide complementary match between the two strands of DNA. These base pairs are always hydrogen bonds between adenine–thymine or cytosine–guanine bases. Basic Local Alignment Search Tool (BLAST)  A free, public access database used to identify and compare DNA and protein sequences. Basophilic stain  A histological stain used to visualize cell bodies. Between-subjects study  A study that compares two different groups of human subjects or animals. For instance, in functional imaging experiments, a between-subjects study may test the differences of old versus young, or male versus female participants. Bimolecular fluorescence complementation (BiFC)  A technique used to visualize the interaction between two proteins. Each protein is fused to half of a fluorescent protein. When the proteins interact, the two halves of the fluorescent protein can fold together to emit light and report the interaction. Binary expression system  An approach that uses two or more constructs to express a transgene in cells. Used to refine the spatial or temporal properties of transgene expression. Biolistics  A method of DNA delivery that uses a gene gun to physically shoot DNAcoated bullets into cells. Biotin  A small molecule with an extremely high affinity for another molecule called avidin. Used to tag molecular probes so that future histochemical processing with avidin-bound probes can enhance a signal. BLAST  See Basic Local Alignment Search Tool. Blastocyst  An early embryonic structure that follows the fertilization of an egg. Bleed-through  A phenomenon in fluorescent microscopy that occurs when the signal of one fluorophore appears when a scientist uses a filter designed to detect the signal of a separate fluorophore. Blocked design  A strategy for presenting stimuli to human subjects in functional imaging experiments. Stimuli of the same category are presented together in blocks. An alternative to an event-related design. Blood oxygen level-dependent (BOLD) effect  The phenomenon in which the brain microvasculature increases the flow of oxygen-rich blood to neurons following an increase of neural activity. Blunt ends  Ends of fragments of DNA cut with restriction enzymes that result in no 5’ or 3’ overhangs; both strands of DNA terminate at the same complementary base pair. Compare with sticky ends. 348 Glossary BOLD effect  See Blood Oxygen Level-Dependent effect. Brain atlas  A tool that depicts brain anatomy in three-dimensional coordinates. BrdU  Bromodeoxyuridine, a synthetic analog of the DNA base thymidine, commonly used to indicate cell proliferation and neurogenesis. Bregma  A landmark on the surface of the skull, defined as the intersection between the sagittal and coronal sutures. Brightfield microscopy  A type of light microscopy in which white light passes directly through or is reflected off a specimen. Brodmann’s areas  Divisions of the cortex defined by the cytoarchitecture of the cortical layers. Developed by Korbinian Brodmann in the early twentieth century and still used today. Calcium phosphate transfection  A method of delivering DNA into cells by using chemical reactions that form calcium phosphate precipitates. Canine Adenovirus (CAV)  A viral vector with retrograde transport properties. Cannula  A narrow, cylindrical tube that can be implanted on the surface of the skull for long-term access to deep brain structures in a living animal. Capsaicin receptor  A genetically encoded, ligand-gated ion channel that depolarizes neurons in response to the capsaicin molecule. Can be used by neuroscientists to stimulate genetically defined populations of neurons. Cas9  See CRISPR/Cas9. Case study  A natural event that demonstrates something important about the nervous system. cDNA  See Complementary DNA. cDNA library  A collection of hundreds or thousands of DNA plasmids, each containing a unique cDNA molecule. Used to collect and store DNA sequences that were actively expressed in a sample prior to exposure with the reverse transcriptase enzyme. cFos  See Fos. Cell-attached recording  A patch-clamp technique in which a glass micropipette forms a gigaseal with the membrane of a cell, but the membrane is not punctured. Cell culture  An in vitro approach in which a scientist grows and maintains cells under carefully controlled conditions outside of a living animal. Cell fractionation  A process in which a scientist centrifuges a sample multiple times at various speeds to divide the sample into different cellular components and organelles. Cell signaling  See Intracellular signaling. Central dogma of molecular biology  The overarching model of information flow in cells, briefly stated as “DNA codes for molecules of RNA, and molecules of RNA code for proteins.” Cerebral angiogram  A structural brain imaging technique that uses X-rays and radio-opaque dyes to produce images of the brain’s vasculature. Glossary  349 Channelrhodopsin-2 (ChR2)  A genetically encoded cation channel capable of depolarizing a neuron with millisecond-timescale resolution in response to stimulation with blue light. Chemical gene delivery  The process by which a scientist uses a chemical reaction to deliver DNA into cells. Chemical mutagenesis  The use of chemical agents, such as EMS or ENU, to mutagenize hundreds or thousands of eggs/larvae for the purpose of performing a forward genetic screen. Chemosensory jump assay  A behavioral test in flies that exploits the tendency of flies to exhibit a startle response and jump when encountering a novel odor. Chimeric mice  Mice composed of patches of tissues from different genetic sources. An intermediate step in the procedure to create knockout or knockin mice. ChIP on chip assay  An assay in which chromatin immunoprecipitation is followed up with a microarray to determine the identity of all genetic sequences bound to a particular protein in vitro or in vivo. ChR2  See Channelrhodopsin-2. Chromatin immunoprecipitation (ChIP)  A technique used to determine whether a sequence of DNA physically interacts with a specific protein, usually a transcription factor or histone protein. Chromatography  A category of biochemistry techniques used to separate proteins on a column based on their size, charge, substrates, or other properties. Chromogenic label  A label on a probe that allows a scientist to perform a simple biochemical reaction that produces a colored byproduct. This byproduct can be visualized using simple light microscopy. Chronic mild stress  A series of mildly stressful events (periods of constant light, tilting an animal’s cage, wetting the animal’s bedding, etc.) that is used to create rodent models of depression and anxiety. Circadian rhythms  The regular, roughly 24-h cycle of stereotyped biochemical, physiological, and behavioral processes. CLARITY  A method of making extracted tissue (such as a brain or other organ) transparent to visualize structures in an intact system. Classical conditioning  A paradigm of animal learning in which a scientist couples an initially neutral stimulus, such as a tone or light, with a salient stimulus, such as food or an electric shock. An animal eventually associates the neutral stimulus with the salient stimulus. Cloning vector  A DNA vector capable of storing DNA sequences for recombinant DNA techniques but lacking the necessary noncoding sequences that allow the DNA to be expressed in a host cell. Co-culture system  An in vitro culture system in which two different cultured cell types or explants are cultured together in the same chamber, allowing a scientist to examine the effect of one type of cell or tissue on another. Co-immunoprecipitation (Co-Ip)  A method that determines whether a protein interacts with another protein. 350 Glossary Codon  A series of three nucleotides that code for a specific amino acid during translation. Cognitive neuroscience  The field of neuroscience dedicated to elucidating the neural basis of thought and perception. Cohesive ends  See Sticky ends. Colorimetric label  See Chromogenic label. Competent cell  A bacterial cell transiently capable of receiving DNA sequences through its cell wall. Complementary DNA (cDNA)  A fragment of DNA produced from an RNA template following a reaction with the enzyme reverse transcriptase. Complementary strands  Refers to two strands of DNA that have complementary base pairs, allowing the two strands to hybridize into a double-stranded molecule. Complementation test  A test used to determine if a mutant animal with an abnormal phenotype has a mutation in the same gene as a different mutant animal with the same abnormal phenotype. Compound microscope  A microscope in which two or more lenses are used in concert to enhance the magnification power. Computerized tomography (CT)—also called computerized axial tomography (CAT)  A structural imaging technique in which several X-rays are taken from multiple orientations around a single plane to produce an image of the brain. Condenser  The part of a microscope that focuses light from a light source onto a specimen. Conditional knockout  An animal in which a gene is knocked out of the genome using gene targeting measures, but only in a specific cell type, and/or at a specific time. Conditioned place preference/aversion  Rodent behavioral assay demonstrating an association between a specific environment and a positively reinforcing stimulus (preference) or negatively reinforcing stimulus (aversion). Conditioned stimulus (CS)  In classical conditioning, the neutral stimulus that eventually becomes associated with a salient, unconditioned stimulus (US). Confluence  The relative amount of space that a growing population of cultured cells takes up on the surface of a cell culture flask/plate. Confocal microscope  A specialized fluorescent microscope that can produce clear images of structures within specimens by selectively exciting and collecting emitted light from single points on the specimen. Construct  A recombinant DNA sequence capable of coding for a functional protein when properly expressed in host cells. Construct validity  Refers to a state in which an animal model and human model have the same underlying genetic or cellular mechanism that may result in a certain behavior, such as a transgenic mouse engineered to have the same mutation that causes Alzheimer’s disease in humans. Contextual fear conditioning  A form of classical conditioning in which an animal associates a neutral stimulus with an aversive stimulus, such as a foot shock, and the animal subsequently displays a fearful behavior, such as freezing, when placed in the training chamber, even in the absence of the aversive stimulus. Glossary  351 Continuous reinforcement  In operant conditioning, a paradigm in which an animal performs a single response to receive a rewarding stimulus. Compare with fixed ratio reinforcement or progressive ratio reinforcement. Coomassie blue  A blue dye used to stain proteins in SDS-PAGE gels. Coronal plane  A neuroanatomical plane that divides the brain into anterior and posterior sections (from ear to ear). Cosmid  A large (30–50 kb) DNA plasmid capable of replicating in bacteria. Useful for engineering very large DNA constructs. Countercurrent apparatus  A device used to fractionate a population of flies based on their behavioral responses to a sensory stimulus. Cranial window  An implant or thinning of a rodent’s skull that allows for direct observation of fluorescent neurons in vivo. Craniotomy  A procedure used to remove a small section of the skull during a surgical procedure. Cre/lox system  A binary expression system used to control the spatial expression of a transgene. Cre recombinase  A protein derived from bacteriophage that recognizes lox sites. Cre excises any DNA sequences between two lox sites oriented in the same direction and inverts any DNA sequences between two lox sites oriented in inverted directions. CRISPR/Cas9  A nuclease system used to introduce double-strand breaks at precise genomic locations. Cross-linking fixative  A chemical fixative that creates covalent chemical bonds between proteins in tissue and cells. Cross-talk  See Bleed-through. Cryostat  A device used to freeze and cut tissue specimens into 5–50 μm sections. Cued fear conditioning  A form of classical conditioning in which an animal associates a neutral stimulus with an aversive stimulus, such as a foot shock, and then displays freezing behaviors in the presence of the formerly neutral stimulus, even in the absence of foot shock. Current (I)  The rate of flow of electrical charge over time. Current clamp  An electrophysiological technique in which a scientist injects current into a cell and measures the change of voltage over time. Cyclic voltammogram  The end result of a fast-scan cyclic voltammetry experiment that plots the amount of measured current versus the applied voltage. These data are useful for detecting the presence of various neurochemicals in neural tissue. Cyclotron  A large machine that produces positron-emitting isotopes, such as the kind used in positron emission tomography (PET) experiments. Cytokine  Extracellular signaling molecule used in intercellular communication. DAPI  A fluorescent stain used to label nuclei acids (and therefore also cell nuclei). 352 Glossary Darkfield microscopy  A form of light microscopy that uses an oblique light source to illuminate a specimen from the side so that only scattered light enters the objective lens. This causes some organelles to stand out from a dark background. Defensive marble burying assay  A behavioral paradigm used to assay anxiety in a rodent. Marbles are placed in an animal’s cage, and a scientist measures how many marbles an animal buries in the cage bedding over a fixed amount of time. Animals that are more anxious typically bury more marbles than nonanxious animals. Dehydrating fixative  A histological fixative that disrupts lipids and reduces the solubility of protein molecules, precipitating them out of the cytoplasmic and extracellular solutions. Dependent variable  The variable in an experiment that is measured and dependent upon the value of an independent variable. Depolarization  A decrease in the absolute value of a cell’s membrane potential. Descriptive science  The act of observing properties of the nervous system without experimental manipulation. Designer receptor exclusively activated by a designer drug (DREADD)  An artificially engineered membrane-bound receptor capable of stimulating or inhibiting neural activity upon binding to a normally inert ligand. Dicer  An enzyme that recognizes double-stranded RNA and cleaves it into fragments about 23 bp long (siRNA). This siRNA-dicer complex then recruits additional cellular proteins to form an RNA-induced silencing complex (RISC), which mediates the phenomenon of RNA interference (RNAi). Differential interference contrast (DIC) microscopy  A type of brightfield microscopy that uses optical modifications within the microscope to exaggerate changes in the light-scattering properties of cellular structures, producing a three-dimensional, textured appearance. Diffuse optical imaging (DOI)  A type of noninvasive optical imaging method in which light is reflected off neural structures through the scalp. Diffusion  The passive transport of molecules from a region of higher concentration to a region of lower concentration caused by random molecular interaction. Diffusion MRI  A technique used to image fiber tracts in the brain by measuring the anisotropic diffusion of water molecules along fiber bundles. Diffusion tensor imaging (DTI)  A diffusion MRI method that measures the anisotropic diffusion of water molecules along fiber tracts in the brain to image connections between different brain regions. Digoxigenin  A steroid molecule frequently used to tag molecular probes in histology experiments. Diphtheria toxin receptor  A genetically encoded receptor that will kill an expressing cell in the presence of the diphtheria toxin, which is normally harmless to rodent neurons. Dipstick assay  See Olfactory avoidance assay. Glossary  353 Direct IHC  A category of immunohistochemistry experiments in which the primary antibody is directly conjugated to a fluorescent or chromogenic label. This typically produces a weaker signal than indirect IHC. Dissecting microscope  See Stereomicroscope. Dissociated cell culture  A primary cell culture method in which tissue removed directly from an animal is separated into individual cells. DNA ladder  A sample of various DNA fragments of known sizes, run on an agarose gel next to experimental DNA samples to determine the size of fragments in the samples. DNA library  A comprehensive collection of hundreds or thousands of cloned DNA fragments from a biological sample, stored in expression vectors or artificial chro- mosomes. DNA vector  A nucleic acid storage system that can hold an isolated DNA fragment of interest. Dominant negative  The twisted ladder shape of a double-stranded DNA molecule. Double helix  A mutated protein that interferes with the function of the wild-type protein. Double inverse orientation (DIO) construct  A recombinant DNA construct that allows Cre-dependent expression of a transgene due to the presence of two pairs of lox sites. Doxycycline  A drug that binds to the tetracycline transactivator protein (tTA) or reversed transactivator protein (rtTA) in Tet-off/Tet-on transgenic animals. This drug allows scientists to gain inducible control of the Tet-off/Tet-on system, allowing temporal control of transgene activity. DREADD  See Designer receptor exclusively activated by a designer drug. Dynamic clamp  Electrophysiological method that uses computer simulation to model artificial membrane or synaptic conductances in live cells. Electrode  A metal or glass probe used to measure neural activity from biological tissue. EEG  See Electroencephalography. Electroencephalography (EEG)  A noninvasive technique used to measure the gross electrical activity of the brain. Electrolytic lesion  A physical lesion in the nervous system caused by injection of current. Electromyogram  A technique used to record the electrical potential generated by muscle cells and, therefore, to record motor activity. Electron microscope (EM)  A form of microscopy that uses a beam of electrons rather than a beam of photons to image a specimen, greatly enhancing the resolving power. Electron microscope tomography  See Electron tomography. Electron tomography (ET)  A procedure used to collect and combine several transmission electron microscope images from the same specimen, allowing scientists to image the three-dimensional ultrastructure of organelles and macromolecules. Electrophoretic mobility shift assay (EMSA)  A technique used to determine if a protein is capable of directly binding to a short sequence of DNA. 354 Glossary Electrophysiology  The branch of neuroscience that explores the electrical activity of living neurons and investigates the molecular and cellular processes that govern their signaling. Electroporation  The process of using an electric field to deliver DNA constructs into cells. Electrospray ionization (ESI)  A technique used to cause a sample to enter a gaseous phase for mass spectrometry experiments. Elevated plus maze  A behavioral assay used to assay anxiety in rodents. Embedding  The process of surrounding a brain or tissue section with a substance that infiltrates and forms a hard shell around the tissue. Embryonic lethal  A phenotype caused by a transgenic or knockout construct that causes the death of an animal during gestation. Embryonic stem (ES) cell  A pluripotent cell capable of giving rise to any tissue in an organism. Emission filter  A filter on a fluorescent microscope that blocks extraneous wavelengths of light, including the light used to illuminate the specimen, but allows emitted light wavelengths to pass through to a detector. Ensembl  A public genomic database, useful for looking up genomic sequences and genetic information for a variety of organisms. Enzyme-linked immunosorbent assay (ELISA)  A technique used to quantify the amount of protein in a sample. Enzymatic histochemistry  A histochemical procedure that utilizes an endogenous protein’s activity to generate a visible signal. Epifluorescent microscopy  Standard fluorescent microscopy. Specimens labeled with fluorescent probes are illuminated by light of the excitation wavelength. The specimen is then viewed using a second filter that is opaque to the excitation wavelength but transmits the longer wavelength of the emitted light. Epitope  The specific region of an antigen to which an antibody binds. ER/tamoxifen system  A system used in genetic engineering experiments to allow inducible gene activity/suppression at specific time points. The estrogen receptor (ER) is sequestered in the cytoplasm. In the presence of tamoxifen, ER translocates to the nucleus. In genetic engineering experiments, ER can be fused to other useful proteins, such as Cre recombinase, to allow for inducible activation of Cre activity when a scientist adds tamoxifen. Estrogen receptor (ER)  An endogenous receptor that resides in the cytoplasm until binding with its ligand, at which point it translocates to the nucleus. Ethology  The study of natural animal behaviors. Event-related design  A strategy for presenting stimuli to human subjects in functional imaging experiments in which stimuli are presented as isolated, individual events of short duration. An alternative to a blocked design. Event-related field  A distinct, stereotyped waveform in magnetoencephalography (MEG) that corresponds to a specific sensory, cognitive, or motor event. Glossary  355 Event-related potential  A distinct, stereotyped waveform in electroencephalography (EEG) that corresponds to a specific sensory, cognitive, or motor event. Excitation filter  A filter on a fluorescent microscope that allows only light in a specific range of wavelengths to pass through to an objective to illuminate a specimen. Excitatory postsynaptic potential (EPSP)  A localized potential change in a small region of a neuron that causes the membrane potential to depolarize, bringing the membrane potential closer to the threshold to generate an action potential. Exon  A sequence of an RNA molecule that remains in the final mRNA sequence after splicing occurs. Explant culture  A primary cell culture method in which intact chunks of tissue are removed from a living organism and cultured in an in vitro environment. Expression vector  A DNA vector containing all of the necessary noncoding sequences necessary for a host cell to express a coding DNA sequence as a protein. Extracellular recording  An electrophysiological recording in which the recording electrode is placed outside a neuron in the extracellular environment. Eye coil  A wire loop implanted around the outer circumference of a monkey’s eye. Used to ensure that the monkey correctly fixes its gaze during a visual task. Face validity  A concept that establishes the validity of an animal model by demonstrating that the animal’s behavior is similar to the analogous human behavior. Fast-scan cyclic voltammetry  A technique used to measure neurochemicals in neural tissue. Feeding acceptance assay  A Drosophila assay that examines taste preferences. Slightly starved flies are offered a choice of appetitive, aversive, or neutral stimuli, each dyed a different color. The amount of each stimulus can be scored by examining the color of the fly’s stomach. Fiber stain  A histological stain used to mark the presence of fiber tracts in the nervous system. Fixation  The process of using chemical methods to preserve, stabilize, and strengthen a biological specimen for subsequent histological procedures and microscopic analysis. Fixed ratio reinforcement  In operant conditioning, a paradigm in which an animal must perform a set number of responses to receive a rewarding stimulus. Compare with continuous reinforcement or progressive ratio reinforcement. Five choice serial reaction time task  A rodent behavioral assay used to study attention and impulsivity. Flat skull position  The placement of an animal on a stereotaxic frame such that the top of the skull is perfectly level, at the same height along the rostral–­caudal axis. Flex construct  A recombinant DNA construct that allows Cre-dependent expression of a transgene due to the presence of two pairs of lox sites. Flight simulators  Specialized chambers that suspend a fly from a thin pin and project visual stimuli on a screen, depicting the external, visual environment. Used to determine the effect of visual stimuli on flying behavior. 356 Glossary Flippase recognition target (Frt)  A 34 bp sequence recognized by the flippase recombinase (Flp) enzyme. Flippase recombinase (Flp)  A protein that recognizes flippase recognition target (Frt) sites. Flp excises any DNA sequences between Frt sites oriented in the same direction, and inverts any DNA sequences between two Frt sites oriented in inverted directions. Floxed  Term used to describe any DNA sequence flanked by two lox sites. Flp/Frt system  A binary expression system used to control the spatial expression of a transgene. Fluorescence recovery after photobleaching (FRAP)  A protein visualization technique in which a scientist uses strong laser illumination to bleach fluorescently tagged proteins in a particular region of a cell and then measures the time course of fluorescence that returns to the bleached region. Fluorescence resonance energy transfer (FRET)  A technique used to visualize if and when two proteins interact. Fluorescent in situ hybridization (FISH)  A technique used to examine the location of specific DNA sequences within chromosomes, commonly used to detect chromosomal abnormalities. Fluorodeoxyglucose (FDG)  A radioactive form of glucose used to image functional brain activity in positron emission tomography (PET) experiments. Fluorophore  A molecule that has the ability to absorb light at a specific wavelength and then emit light at a different, typically longer wavelength. FM dyes  Lipophilic dyes that fluoresce when bound to a membrane. They are particularly useful for reporting the exocytosis and endocytosis events that occur during synaptic vesicle recycling and can stain nerve terminals in an activitydependent manner. Footprint pattern assay  A behavioral assay used to examine a rodent’s motor coordination and balance by observing its footprint patterns. Forced swim test  A behavioral experiment used to assay depression by placing an animal in a narrow chamber of water and measuring how long it takes for the animal to stop trying to escape. Formalin assay  An assay used to examine an animal’s sensitivity to noxious chemical stimuli by observing its response to a small injection of the noxious chemical formalin. Förster resonance energy transfer (FRET)  See Fluorescence resonance energy transfer. Forward genetics  A method used to identify genes that contribute to a phenotype. Scientists generate random mutations in eggs/larvae and then examine the phenotypes of the offspring. Contrast with Reverse genetics. Fos  A transcription factor used as an immediate early gene, i.e., an indirect marker of neural activity. Founder  Individuals in the first generation of a transgenic or genetically targeted animal. Fractional intensity change  The change in fluorescent signal during calcium or voltage imaging compared to baseline fluorescence. Denoted as ΔF/F or ΔI/I. Glossary  357 Freezing microtome  A microtome with a cooling unit that allows a tissue sample to be frozen during sectioning. Functional brain imaging  Techniques used to image neural activity in the central nervous system. Functional connectivity  In whole-brain imaging, brain regions that demonstrate temporally correlated fluctuations in BOLD activity. Functional magnetic resonance imaging (fMRI)  A technology that uses the principles of magnetic resonance imaging (MRI) to indirectly measure neural activity in the nervous system over time. Gain of function experiment  An experiment in which an independent variable (e.g. neural activity or gene expression) is increased to determine its effect on a dependent variable. Gal4  A transcription factor derived from yeast that binds to a specialized upstream activation sequence (UAS) to activate the transcription of a gene. Gal4/UAS system  A binary expression system used to control the spatial expression of a transgene. Widely used with invertebrate and zebrafish model organisms. Gal80  A protein derived from yeast that inhibits Gal4 transcription factors. GCaMP  A genetically encoded calcium indicator. Gel electrophoresis  The process of using an electrical field to force DNA through an agarose gel. Used to isolate, identify, and characterize the properties of DNA fragments. Gel-filtration chromatography  A chromatography technique that separates proteins on the basis of their size. Geller–Seifter conflict test  A behavioral paradigm that assays anxiety in rodents. A rodent is trained to press a lever to receive a food reward. In subsequent experiments, the lever is paired with an aversive foot shock. Anxious animals tend to press the lever significantly less to receive food than nonanxious animals. Gel shift assay  See Electrophoretic mobility shift assay. Gene  The fundamental molecular unit of heredity, a segment of DNA that codes for a functional protein. Gene gun  A device used in biolistic experiments to physically shoot DNA-coated bullets onto tissue specimens. Gene targeting  The process of incorporating a DNA construct into a specific locus in an animal’s genome. Genetically encoded calcium indicator (GECI)  A genetically encoded transgene used to indirectly measure neural activity by imaging calcium dynamics. Genetically encoded voltage indicator (GEVI)  A genetically encoded transgene used to visually measure membrane potential. Genome  The complete set of genes or genetic material present in a cell or organism. Genomic DNA library  A collection of thousands of plasmids that each contain fragments of an animal’s genome. Genotype  The genetic constitution of an animal. 358 Glossary Germ cells  Cells that pass on genetic information to an organism’s offspring (sperm in males and eggs in females). GFP  See Green fluorescent protein. Go/no-go task  A form of operant conditioning in which a rodent is trained to produce a response in the presence of a “go” cue yet withhold a response in the presence of a separate “no-go” cue. Used to test attention and impulsivity. Golgi stain  A classic histological stain used to label a small subset of neurons in their entirety. Graded potential  See Localized potential. Green fluorescent protein (GFP)  A protein derived from jellyfish, capable of absorbing wavelengths of blue light and emitting wavelengths of green light. Ubiquitously used in the biosciences as a reporter protein. Growth factors  Proteins produced by certain cells as cell communication molecules that stimulate the growth and differentiation of other cells. Guide RNA  An RNA molecule that serves as a template for genomic targeting in the CRISPR/Cas9 system. Gyri  Ridges on the cerebral cortex (plural for “gyrus”). Halorhodopsin (NpHR)  A genetically encoded transmembrane pump that hyperpolarizes neurons upon stimulation with yellow light. Hanging wire assay  A behavioral assay used to measure neuromuscular deficits in rodents. Hargreaves assay  A behavioral assay used to measure pain in rodents by aiming a high-intensity beam of light at the hind paw and measuring the time it takes for the rodent to withdraw its paw. Hemoglobin  The protein that transports oxygen throughout the bloodstream. Herpes simplex virus  A neurotropic virus that can be used to deliver DNA to some cell types and also serve as a multisynaptic tracer. Heterologous expression systems  A dissociated cell culture system that can easily be transfected with DNA. HM3Dq  A DREADD transgene capable of stimulating neurons upon binding the normally inert ligand clozapine-n-oxide (CNO). HM4Di  A DREADD transgene capable of inhibiting neurons upon binding the normally inert ligand clozapine-n-oxide (CNO). Holding potential  The set voltage that is held constant in a voltage clamp experiment. Homologous recombination  A natural phenomenon in which DNA regions with strong sequence similarity can exchange genetic material. Homology arms  Regions of homology that flank a gene targeting construct to exploit the phenomenon of homologous recombination in the production of a knockout or knockin mouse. Homology-directed repair (HDR)  A method of inserting DNA into the genome following nuclease-mediated double-strand break repair. A scientist introduces a DNA construct with homology arms that have the same sequence as the broken DNA strands. Glossary  359 Horizontal plane  A neuroanatomical plane that divides the brain into superior and inferior sections such that the brain is complete from the anterior-posterior and leftto-right planes. Hot plate assay  A behavioral paradigm used to assay nociception in rodent models. A rodent is enclosed on a heated plate and observed for licking and paw withdrawal. Hyperpolarization  Any change in the membrane potential of a neuron that causes a net increase in the absolute value of the membrane potential. IACUC  See Institutional Animal Care and Use Committee. I/V curve  A plot of the voltage across a cell membrane compared with the current that flows through the ion channels of the membrane. Ibotenic Acid  A pharmacological agent used to irreversibly ablate neurons when directly injected into the nervous system. Immediate early gene (IEG)  A gene that tends to be actively transcribed during periods of high neural activity. Immersion  The process of fixing small brains or even entire animals by immersing them in fixative solutions. Immortalized cell line  A cell line manipulated so that it will continually divide and multiply. Immunoblot  See Western blot. Immunocytochemistry (ICC)  A histological technique that uses antibodies to stain proteins in cells. Immunoelectron microscopy  A combination of immunohistochemistry and electron microscopy used to visualize protein expression within subcellular compartments. Immunofluorescence (IF)  A histological technique that uses antibodies to stain proteins with fluorescent reagents. Immunohistochemistry (IHC)  A histological technique that uses antibodies to stain proteins in tissues. Immunopanning  A technique used to purify certain cell types. A scientist coats the bottom of a plate with antibodies that recognize cell surface markers on the outside of specific cells. When heterogeneous populations of cells are added to the plate, the scientist can purify the cells of interest by allowing the cells to bind to the bottom of the plate and then wash off the undesired, unbound cells. Immunoprecipitation (IP)  A protein purification method that uses antibodies to purify specific proteins out of solution. In silico screen  A method of identifying genes or proteins of interest by searching public genome and bioinformatics databases. In situ hybridization (ISH)  A histological method used to label the location of mRNA transcripts. In vitro  Any process that takes place in a controlled environment outside a living organism, such as a cell culture dish or test tube. 360 Glossary In vivo  Any process that takes place in a whole, living organism. Independent variable  The experimental variable that is intentionally manipulated by the researcher and is hypothesized to cause a change in the dependent variable. Index of refraction  A measure of the relative speed of light in a specific medium, such as air, water, or oils. Indirect IHC  A method of performing an immunohistochemistry experiment in which a primary antibody binds to an antigen and then a secondary antibody binds to the primary antibody. The secondary antibody is conjugated to a fluorescent or chromogenic tag that allows for subsequent visualization. Induced pluripotent stem (iPS) cell  A somatic stem cell that has been manipulated into reverting back to a pluripotent stem cell state, capable of giving rise to multiple cell types. Infection  The process of delivering DNA into cells using viruses. Inhibitory postsynaptic potential (IPSP)  A localized potential change in a small region of a neuron that causes the membrane potential to hyperpolarize, bringing the membrane potential farther from the threshold to generate an action potential. Inner cell mass  A group of cells within a blastocyst that give rise to all cells and tissues in an organism. Insertional mutagenesis  The process of causing genetic mutations by insertion of mobile genetic elements called transposons into the genomes of offspring. These transposons insert at random locations in the genome, occasionally disrupting an endogenous gene. Inside-out recording  A patch clamp recording method in which a tiny patch of the membrane is isolated and oriented such that the intracellular surface is exposed to the medium and the extracellular surface resides on the inside of the glass pipette. Institutional Animal Care and Use Committee (IACUC)  An academic institution’s internal review committee that approves and inspects all procedures using vertebrate animals. Institutional Review Board (IRB)  An academic institution’s internal review committee that approves and inspects all procedures that use human subjects. Intracellular recording  An electrophysiology recording technique in which an electrode is gently placed inside a cell. Intracellular signaling  The biochemical pathways that take place within a cell that affect the cell’s physiology and metabolism. Intracerebroventricular (i.c.v.) injection  The injection of a solution directly into the ventricular system of the brain. Intraperitoneal (i.p.) injection  The injection of a solution into the peritoneum (body cavity) of an animal. Intron  A sequence of an RNA molecule that is cut out of a final mRNA sequence during RNA splicing. Inverted microscope  A microscope in which the objective lens is placed beneath the sample and the light source and condenser are placed above the sample. Glossary  361 Ion-exchange chromatography  A chromatography technique that separates proteins based on their charge. Irradiation mutagenesis  The use of high-intensity UV light to mutagenize hundreds or thousands of eggs/larvae for the purpose of performing a forward genetic screen. Isofluorane  A gaseous anesthetic agent used in mammals. Isotropic  The diffusion of a substance in all directions within a medium. The opposite of anisotropic. Ketamine  A chemical anesthetic agent that acts by inhibiting NMDA and HCN1 ion channels. Often used in concert with Xylazine. Kinase assay  An assay used to determine if a protein kinase is capable of phosphorylating a substrate protein or protein fragment. Knockdown  The disruption of mRNA transcript expression, therefore causing a decrease in functional protein products. Knockin  The use of gene targeting procedures to add a functional gene or gene sequence to a specific location in a genome. Knockout  The use of gene targeting procedures to remove a functional DNA sequence from the genome, thus knocking out a gene of interest. LacZ  A bacterial gene that encodes the enzyme β-galactosidase. This gene is often used as a reporter gene, as β-galactosidase can react with the substrate X-gal to produce a dark blue byproduct. Lambda  Landmark on the surface of the skull, defined as the intersection between the lines of best fit through the sagittal and lambdoid sutures. Learned helplessness  A paradigm used to create rodent models of depression. A scientist exposes an animal to aversive stimuli at random intervals. Theoretically, this treatment creates a condition in which the animal experiences a lack of control, causing it to show symptoms of behavioral despair. Lectins  Proteins that exhibit extremely high binding affinities for sugars, many of which can be used as transsynaptic neural tracers. Lentivirus  A retrovirus that can be used to deliver transgenes to both dividing and postmitotic cells and integrate these sequences in the genome of the host cell. Lesion  A physical cut or destruction of biological tissue. Linkage analysis  A strategy to map a gene’s location in the genome by identifying the position of the novel gene in relation to the location of known genes. Lipofection  A chemical transfection method that uses a lipid complex to deliver DNA to cells. Liposomes  Tiny vesicular structures composed of a sphere of phospholipids that have the same composition as a plasma membrane. Local field potential  An electrophysiological measurement of the sum of all dendritic synaptic activity within a volume of tissue. 362 Glossary Localized potential  A local change in membrane potential caused by the activity of individual ion channels. These potentials include excitatory postsynaptic potentials (EPSPs) and inhibitory postsynaptic potentials (IPSPs). Loss of function experiment  An experiment in which an independent variable (e.g. neural activity or gene expression) is decreased or lost to determine its effect on a dependent variable. Lox site  A 34 bp DNA sequence recognized by the Cre recombinase enzyme. Luciferase  A genetically encoded bioluminescent protein cloned from various species of fireflies. Luciferase assay  A technique used to determine whether a protein is able to interact with a genetic regulatory sequence and activate the transcription of a gene. Magnetic resonance imaging (MRI)  A non-invasive structural imaging technique that utilizes the magnetic properties of different tissues to produce highly detailed structural images of the brain and body. Magnetoencephalography (MEG)  A functional imaging technique that measures changes in magnetic fields on the surface of the scalp produced by changes in underlying patterns of neural electrical activity. Magnification  A critical parameter in microscopy that measures how much larger the sample appears compared to its actual size. Mass spectrometry  A technique used to identify proteins and/or peptide sequences by determining the mass to charge ratio of proteins, peptides, and their fragments. Match to sample task  A rodent behavioral assay that uses operant conditioning to test working memory. Matrix-assisted laser desorption ionization (MALDI)  A mass spectrometry method of vaporizing proteins into the gas phase by blasting them with a UV laser so that their mass to charge ratio can be measured. Meganuclease  A restriction enzyme that recognizes longer DNA sequences (typically 20–40 bp) than typical restriction enzymes (typically 6–8 bp). Membrane potential  The voltage across the cell membrane generated by differential ionic charges on the intracellular versus extracellular sides of the membrane. Microarray  A tool used to measure the expression of thousands of genes between one or more tissue samples. It consists of a grid (array) of known nucleic acid sequences spotted on a slide so that cDNA from a biological sample can be hybridized with the sequences to detect whether any of the known sequences are present in the sample. Microdialysis  A method to sample neurochemicals from brain extracellular fluid based on the diffusion of chemicals from the extracellular fluid into a microdialysis probe’s semipermeable membrane. Microelectrode  A small (micrometer scale) metal or glass probe used to measure electrical signals from cells or tissue. Microinjection  The process of injecting small volumes of a solution (usually containing DNA) into cells with a thin needle. Glossary  363 Microiontophoresis  The process of using a small electrical current to infuse substances out of a glass electrode into the brain. Microscope  An instrument that manipulates the trajectory of light rays so that small objects appear larger. Microstimulation  The process of sending electrical current through an electrode to induce changes in membrane potential. Microtome  A piece of equipment with a sharp knife to cut medium-thickness (25–100 μm) sections of tissue. Midsagittal  A view or section of the brain that perfectly divides the left and right hemispheres. Mixed design  A whole brain imaging task paradigm that uses elements of both blocked and event-related designs. MNI template  A standardized three-dimensional coordinate map of the human brain produced by the Montreal Neurological Institute. Based on the average of hundreds of individual MRI brain scans and matched to landmarks in Talairach space. Molecular cloning  The process of identifying, isolating, and making copies of a particular DNA fragment through recombinant DNA technology. Monoclonal antibody  An antibody made by a single clone of cells that only recognizes a single epitope, or specific region of a protein. Contrast with polyclonal antibody. Morpholinos  Stable, synthetic 22–25 bp antisense oligonucleotide analogues that complement an RNA sequence to block proper mRNA translation as an alternative to RNAi. Morris water maze  A spatial learning and memory task for rodents that involves a hidden platform submerged in opaque water whose location can be recognized through distant visual cues. MRI  See Magnetic resonance imaging. Multielectrode array (MEA)  A group of individual electrodes or tetrodes arranged into one electrical recording unit to record extracellular activity from multiple cells at the same time. Multiple cloning site  A region in a DNA vector with a variety of recognition sequences for different restriction enzymes that make it simple to insert a foreign DNA sequence into the vector. Multipotent  The ability of some cells to give rise to all types of cells found in a particular tissue—for example, the ability of a neural stem cell to give rise to neurons, astrocytes, and oligodendrocytes. Muscimol  A GABA receptor agonist used to reversibly inhibit neural activity. Myelin  The fatty insulating material surrounding axons produced by glial cells that provide insulation to neurons. Near-infrared spectroscopy (NIRS)  A noninvasive, optical brain imaging method that records neural activity-induced changes in the reflectance properties of nearinfrared light as it is reflected off the scalp. Necessity experiment  See Loss of function experiment. 364 Glossary Neural progenitor  A dividing cell with a limited ability to give rise to another dividing cell but that has the ability to generate either a neuron, astrocyte, or oligodendrocyte. Neural stem cell (NSC)  A multipotent, self-renewing cell that divides and gives rise to neurons, astrocytes, or oligodendrocytes, as well as other neural stem cells. Neurosphere  A ball of cells generated from a dividing neural progenitor or neural stem cell that indicates an ability to proliferate. Neurosphere assay  See Primary neurosphere assay. Neuroethology  The study of the neural basis of an animal’s natural behaviors. Neuron doctrine  The theory, advocated by Ramón y Cajal, that discrete neurons are the basic structural and functional units of the nervous system. Next-generation sequencing (NGS)  Methods of identifying DNA base pair sequences quickly, with use of fewer reagents, and less expensively than the standard Sanger sequencing method. Enables large-scale genome exploration. Nissl stain  A basophilic stain that highlights RNA (the Nissl substance) in cells. Cresyl violet is a commonly used Nissl stain. Nociception  The detection of a noxious stimulus, typically perceived as pain. Nomarski microscopy  See Differential interference contrast (DIC) microscopy. Nonhomologous end-joining  A naturally occurring DNA break repair process that rejoins double-strand breaks, occasionally by introducing an insertion or deletion mutation. Nonmatch to sample task  A rodent assay that uses operant conditioning to test working memory. Nonratiometric dye  An organic chemical whose fluorescence intensity varies directly with its ability to bind a substance such as calcium. Contrast with ratiometric dyes. Northern blot  A nucleic acid hybridization technique used to detect the presence and relative concentration of mRNA in a sample. Novel object recognition  A learning and memory task in which a rodent will spend more time exploring an object that has not previously been presented, demonstrating memory for the previously presented object. NpHR  See Halorhodopsin. Nuclease  An enzyme that cleaves nucleic acids. Nucleotide  A molecule composed of a phosphate group, a nitrogenous base (adenine, thymine, cytosine, guanine, or uracil), and a sugar molecule (2′-deoxyribose for DNA, ribose for RNA) that combines with other nucleotides to form a DNA or RNA polymer. Numerical aperture (NA)  A measure of the light-collecting ability of a microscope objective lens. The NA depends on the angle by which light enters the objective and the medium light must pass through. Higher NA objectives collect more light and consequently have better resolving power. Objective lens  The lens on a microscope that gathers and focuses light from the specimen in a microscope. This lens is typically placed adjacent to the specimen. Glossary  365 Ocular lens  The lens, typically 10× magnification, located in the eyepiece of a compound microscope. Ohm’s law  The relationship between voltage (V), current (I), and resistance (R) in an electrical circuit, expressed as V = I × R. Olfactory avoidance assay  A Drosophila chemosensory assay in which a scientist presents an odor on a stick in a fly chamber and measures the distance a fly maintains from the odor. Also called a dipstick assay. Olfactory jump response  See Chemosensory jump assay. Open field test  A rodent assay used to measure exploratory behavior in a large open chamber. More anxious animals stay near the walls of the chamber and do not explore the center area. Operant conditioning  The use of consequences (a positive reward or negative punishment) in response to a specific behavior in an attempt to influence that behavior. For example, a rodent learns to press a lever to receive a pleasurable food reward. Optical imaging  A brain imaging technique that measures changes in light reflectance from the surface of the brain due to changes in blood flow and metabolism caused by neural activity. Optodes  See Optrodes. Optogenetics  An approach to stimulating and inhibiting neurons by expressing genetically encoded neuronal actuators capable of depolarizing or hyperpolarizing neurons in response to light. Optrodes  Either (1) A recording device that records changes in light reflectance in optical brain imaging techniques, or (2) the combination of an optic fiber and microelectrode in an optogenetic experiment. Organotypic slice culture  A slice of brain tissue kept in culture conditions for multiple days to weeks. Such cultures require an air/liquid interface to properly regulate gas exchange throughout the slice. Oscilloscope  An instrument that displays the membrane potential over time. Outside-out recording  A patch-clamp technique that exposes the extracellular side of an excised membrane patch to the bath solution and the intracellular side to the inside of the recording pipette. P element  A transposable genetic element used in Drosophila that is able to change locations in the genome when the P element transposase enzyme is also present. Packaging cells  An immortalized cell line that is transfected with the necessary recombinant DNA sequences to make viral particles. The endogenous cell machinery produces virus, which is eventually harvested in the extracellular media. Partial agonist  A compound that can bind and activate an endogenous receptor, but not to the maximal extent of a full agonist or endogenous ligand. Particle-mediated gene transfer  The process of introducing DNA into cells by physically shooting DNA-coated pellets through the cell membrane. Also called biolistics. Passaging  The process of transferring a fraction of cells into a new container to provide space for the cells to continue to divide. 366 Glossary Patch-clamp techniques  A set of electrophysiological methods in which a glass electrode forms a tight seal with a patch of membrane, allowing the study of single or multiple ion channels. Pavlovian conditioning  See Classical conditioning. PCR  See Polymerase chain reaction. Perforated patch  A category of patch-clamp technique in which a chemical is added to the recording pipette to cause small holes to form in the membrane. This is useful for making the contents of the glass pipette continuous with the cell but without the disadvantage of cytoplasmic contents leaking into the pipette. Perfusion  A method used to circulate a chemical fixative through the cardiovascular system so that the brain is thoroughly preserved. Peri-stimulus time histogram (PSTH)  A graph of the number of action potentials recorded with respect to the time a stimulus was presented. PET  See Positron emission tomography. Phage  See Bacteriophage. Pharmacogenetics  A method of manipulating neural activity by expressing genetically encoded ion channels and receptors not usually found in the species of interest. Phase-contrast microscopy  An optical method of enhancing contrast in a light microscope to amplify small differences in the index of refraction of different cellular structures. Phenotype  An observable trait or set of traits in an animal. Photoactivation  In the context of photoactivatable fluorescent proteins, a change in the fluorescence properties of a fluorophore, usually causing it to fluoresce more or less. Photobleaching  The phenomenon in which fluorescence intensity emitted from a fluorophore decreases over time as it is continuously exposed to light. Photoconversion  The process of delivering a pulse of light to a fluorophore whose emission spectra change from one color to another. Phototaxis  The tendency of an organism to move toward light and away from dark. Phototoxicity  The phenomenon in which illumination leads to the death of cells expressing a fluorophore, typically because illumination of the fluorophore generates free radicals. Physical gene delivery  A method of introducing foreign DNA into cells by physically disrupting the cell’s membrane. Includes microinjection, electroporation, and particlemediated gene transfer (biolistics). Plasmids  Independently replicating accessory chromosomes in bacteria that exist as circles of DNA. Commonly used as vectors for molecular cloning. Pluripotent  The ability of a cell to generate any type of cell, including more pluripotent cells. Polyclonal antibody  A collection of antibodies that each recognizes a distinct epitope, or region, of the same antigen. Contrast with monoclonal antibody. Glossary  367 Polymerase chain reaction (PCR)  A biochemical reaction that uses controlled heating and cooling in the presence of DNA synthesizing enzymes to exponentially amplify a small DNA fragment. Porsolt test  See Forced swim test. Position effects  The phenomenon of transgenes having different expression patterns depending on where the transgene randomly integrates in the genome. Positron  An antimatter counterpart of an electron. When a positron contacts an electron, an annihilation event occurs that generates gamma photons. Positron emission tomography (PET)  A brain imaging technique in which positronemitting isotopes generate annihilation events that can be detected by a gamma detector. Depending on the positron-emitting isotope, PET can detect neural activity or the location of specific receptors in the brain. Positron-emitting isotope  An unstable version of a molecule that emits positrons as it decays, allowing it to be detected in positron emission tomography (PET). Posttranslational modification (PTM)  A modification to a protein, usually the addition or subtraction of a functional chemical group, after it has been translated by a ribosome. For example, the addition or removal of a phosphate group. Potential  See Membrane potential. Precess  The circular movement of the axis of a spinning body around another axis due to a torque. Describes the circular movement of a proton. Predictive validity  An aspect of an animal model’s validity that is fulfilled if treatments used in human patients have the same effect on the animal model. For example, rodent models of depression have predictive validity if human antidepressants relieve behavioral measures of depression in rodents. Prepulse inhibition (PPI)  The attenuation of a startle response caused by the presentation of a weaker sensory stimulus before the presentation of the startlecausing stimulus. Primary antibody  The antibody that recognizes and binds to a specific antigen. Primary cell culture  The culture of cells extracted directly from a tissue of interest. Can be in the form of dissociated cells, explants, or slices. Primary neurosphere assay  An assay of cultured cells for the ability to form neurospheres to examine whether they are proliferative. Cannot distinguish between progenitor cells and stem cells. Compare to secondary neurosphere assay. Primer  A short, single-stranded oligonucleotide used in PCR to hybridize to a template strand in the region flanking the region to be amplified. Proboscis extension response (PER)  A reflex in Drosophila that can be used to test the responses of specific gustatory receptors. A scientist applies taste ligands to gustatory receptor neurons on the leg or proboscis and examines proboscis extension. Progenitor  A cell that divides, but with a limited capacity for self-renewal. 368 Glossary Progressive ratio reinforcement  In operant conditioning, a paradigm in which an animal must perform an action more and more (such as pressing a lever an increasing number of times) to receive a rewarding stimulus. Compare with continuous reinforcement or fixed ratio reinforcement. Promoter  A regulatory region of DNA that controls the spatial and temporal expression of a gene under its control. Pronucleus  One of the nuclei present in a recently fertilized egg—either the male sperm nucleus or the female egg nucleus. Protein  A biological macromolecule composed of one or more sequences of amino acids. Proteins serve as the structural and enzymatic molecules that regulate virtually every process within cells. Protein kinase  A protein that adds a phosphate group to another protein. Proteomics  The systematic, large-scale study of the complete set of proteins in a system (such as a cell or tissue) at a particular time, making use of mass spectrometry and bioinformatics to identify proteins and potential functional interactions. Protospacer adjacent motif (PAM)  A three-nucleotide sequence of NGG that must lie immediately 3′ of the genomic sequence to be targeted in CRISPR/ Cas9-mediated double-strand breaks. Pseudopregnant mouse  A female mouse that has been mated with a sterile male mouse so she produces the proper hormones to act as a surrogate mother. Pulse chase labeling  A process in which a labeled probe (the pulse) is briefly injected into animals or added to cultured cells, then washed away and replaced by unlabeled molecules (the chase). By following the changes in localization of the labeled probe, different protein trafficking pathways can be observed. qRT-PCR  See Quantitative real time PCR. Quantitative real time PCR (qRT-PCR)  A method to both amplify and quantify a specific DNA fragment (especially a cDNA fragment produced from mRNA). Quantum dots  Semiconductor nanocrystals that can be used as bright fluorophores. Radial arm maze  A behavioral test used to assay spatial learning and memory in rodents. It utilizes a maze consisting of an array of arms radiating out from a central starting point, with only some arms containing food. A rodent is trained to traverse down only one arm and not others. Radiofrequency (RF) pulse  A pulse of electromagnetic energy used in MRI to excite hydrogen protons in a human (or sometimes, an animal) subject. After the RF pulse is switched off, the protons relax, generating the MR signal used to form images of the brain. Radioimmunoassay (RIA)  A sensitive method for measuring extremely low concentrations of proteins. Raster plot  A graph that displays each action potential (or membrane potential over a threshold) as a dot or tick mark in relation to a certain time period. Often, the Y-axis represents categories of stimuli, while the X-axis represents time. Glossary  369 Ratiometric dye  An organic molecule that is either excited at or emits at a slightly different wavelength when bound to a substance, such as calcium. Therefore, the ratio of fluorescence intensity at either the excitation or emission wavelengths reveals changes in the concentration of the substance. Recognition site  See Restriction site. Recombineering  A method of creating recombinant DNA molecules using homologous recombination to avoid restriction enzyme digests and ligation reactions. Reduction  One of the guidelines for animal welfare stating that scientists must use the minimum number of animals required to obtain statistically significant data. Refinement  One of the guidelines for animal welfare stating that scientists must minimize distress and pain and enhance animal well-being. Refractive index  See Index of refraction. Region-of-interest (ROI) analysis  Method of brain imaging data analysis that divides the brain into discrete regions that can be compared for significant differences in signal intensity. Relative recovery  In microdialysis, the relative concentration of a substance in the collected dialysate from the probed brain region compared to concentration in the perfusion solution. Replacement  One of the guidelines for animal welfare stating that scientists must do as much research without using animals as possible, including using computational methods and cell culture methods. Also applies to choice of a less-sentient animal species, such as choosing to study invertebrates over vertebrates or mice rather than monkeys. Reporter gene/protein  A visible, nonendogenous gene or protein that is controlled under the same promoter as a gene of interest to report the expression pattern of that gene. Alternatively, a reporter gene is fused to a gene of interest to report the subcellular localization of a reporter fusion protein. Rescue experiment  An experiment in which an aspect of the nervous system that is lost is deliberately reintroduced to determine if it restores a normal phenotype. Resident intruder assay  A social behavioral assay in rodents that measures territorial behavior in males by adding an intruder animal into the cage of a resident animal and measuring specific aggressive behaviors. Resistance (R)  An electrical property that restricts the movement of charge, for example, across a membrane. Resolution  In microscopy, the minimum distance by which two points can be separated and still distinguished as two separate points. Resolving power  See Resolution. Resting potential  The voltage across the cell membrane of a cell at rest, generated by differential buildup of ionic charges on the intracellular vs. extracellular sides of the membrane. Neurons typically have a resting potential of about −70 mV. Restriction digest  The process of using restriction enzymes to cut DNA into fragments. Restriction endonuclease  See Restriction enzyme. 370 Glossary Restriction enzyme  An enzyme that recognizes and cuts specific sequences of DNA. Restriction site  An approximately four to eight unique base pair sequence that a particular restriction enzyme recognizes and cuts. Retrograde tracer  A chemical probe that labels an axon path by being transported from the synaptic terminal back up to the cell body, revealing what areas of the brain project to the location where the tracer was deposited. Retrovirus  A type of virus that uses RNA rather than DNA as its genetic material. Reversal potential  The membrane voltage at which there is no overall flow of ions across the membrane. Determining this potential can be used to identify the ion species that passes through a particular channel. Reverse genetics  The directed mutagenesis of a specific gene so that a scientist can determine if it is necessary for a certain phenotype to occur. Contrast with forward genetics. Reverse microdialysis  The use of a microdialysis probe to introduce chemical substances into a brain region. Reverse transcriptase  The enzyme that uses RNA as a template for creating strands of DNA, called complementary DNA (cDNA). Reverse transcription  The process of creating complementary DNA (cDNA) from RNA templates using reverse transcriptase. Reverse transcription PCR (RT-PCR)  A PCR reaction that amplifies a cDNA fragment generated from reverse transcription of an RNA fragment. Ribosome  Macromolecular complexes of RNA and protein that translate mRNA into a protein by assembling amino acids into polypeptide chains. RNA-induced silencing complex (RISC)  A complex of proteins that uses siRNA as a template for finding a complementary mRNA to degrade. RNA interference (RNAi)  The process of silencing gene expression through endogenous cellular mechanisms that degrade mRNA. RNA polymerase  The enzyme that binds to and transcribes RNA molecules from DNA templates. RNA splicing  The removal of introns and joining of exons by a spliceosome of a newly transcribed mRNA before it leaves the nucleus for translation. Rotarod  A rodent behavioral assay of muscle coordination and balance. Consists of a rotating rod on which rodents must walk and maintain their balance. This device can also be used to examine motor learning, as animals will improve after multiple trials of walking on the rotarod. RT-PCR  See Reverse transcription PCR. Run  In a brain imaging experiment, one complete scan of the brain, during which a volume of information is collected. Sagittal plane  The plane dividing the brain into left and right portions such that the brain is complete in the superior to inferior and anterior to posterior planes. Glossary  371 Sanger dideoxy chain termination method  A technique for determining the sequence of a DNA molecule. Scanning electron microscope (SEM)  A form of microscopy that uses electron interactions at the surface of a specimen to generate an image of the specimen’s surface that appears three dimensional. Screen  A method used to determine which nuclei, neurons, genes, or proteins may be important for a particular biological process. SDS-PAGE  See Sodium dodecylsulfate polyacrylamide gel electrophoresis. Secondary neurosphere assay  An assay that measures whether cells taken from a primary neurosphere are able to form new neurospheres, indicating the cell is still proliferative. Secondary antibody  An antibody that binds to a primary antibody. Sectioning  The process of physically or optically creating thin sections of a specimen to gain greater access and/or visibility of internal structures. Selection cassette  A sequence within a genetic targeting construct that encodes genes that can be used to identify cells that have properly incorporated the targeting construct. Typically, this cassette includes a positive selection gene, such as neomycin resistance gene, that can be used to kill cells that have not incorporated the construct. It also typically includes a negative selection gene, such as thymidine kinase, that can be used to kill cells that have improperly incorporated the targeting construct. Session  In a brain imaging experiment, the scheduled time when an actual experiment is conducted. Short hairpin RNA (shRNA)  A strand of RNA containing complementary sequences at either end that cause it to fold onto itself, producing a hairpin shape. ShRNA  See Short hairpin RNA. Signal transduction  The conversion of an extracellular cue, such as a ligand binding to a receptor, into a signal that begins an intracellular signaling cascade. Single-proton emission computerized tomography (SPECT)  A functional brain imaging technique in which radioactive probes injected into the blood generate high-energy photons that can be detected using a gamma camera placed around a subject’s head. SiRNA  See Small interfering RNA. Slice  In brain imaging experiments, two-dimensional representations of a plane of the brain that is one voxel thick. Slice culture  A primary tissue culture technique in which sections of brain tissue are sliced using a vibratome and then placed in a culture medium. These slices can be used acutely for short-term experiments, or organotypically, for multiple days. Small interfering RNA (siRNA)  Short (19–23 bp) double-stranded RNA fragments used as a template in the RNAi pathway to target mRNA molecules for degradation. Sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE)  A technique used to separate proteins based on their size and folding properties. Often used while performing a western blot. 372 Glossary Somatic cell  A differentiated cell that is not part of the germline (eggs or sperm), such as skin or blood cells. Southern blot  A nucleic acid hybridization technique that can identify the presence of a specific sequence of DNA. Spatial resolution  The minimum size that can be resolved to distinguish individual components of an organ or tissue. Compare to temporal resolution. Spike  Another name for an action potential. Spike sorting  Differentiating action potentials recorded in an extracellular electrophysiology experiment and assigning them to different neurons based on their pattern, shape of spiking activity, and signal size. Spliceosome  The macromolecular complex that catalyzes RNA splicing. Splitting  See Passaging. Startle response assay  A behavioral assay that measures the ability of an animal to exhibit a startle response, such as an eye blink or sudden muscle contraction, in response to the presentation of an unexpected sensory stimulus. This assay can be used to indicate whether sensory processing is intact. Stem cell  A pluripotent cell that has an unlimited ability to renew itself. Stereomicroscope  A microscope that magnifies small objects and provides a threedimensional perspective for examining the surface of brains or large neural structures. Often used for dissections, surgeries, or examining electrodes and implants. Stereotaxic instrument  A piece of equipment that stabilizes an animal’s head and allows precise positioning for injections or implants. Stereotaxic surgery  A survivable procedure in which a scientist physically penetrates the brain of an animal under anesthesia for the injection of substances or implantation of devices. Sterile  Aseptic and free from microorganisms. Important condition for performing surgeries on live animals or culturing cells and tissue outside of an organism. Sterile field  A dedicated aseptic surface where sterile instruments can be placed when not in use. Sterilizer  A small device filled with heated glass beads used to sterilize tools and fine instruments when autoclave sterilization is inconvenient, such as between surgeries. Sticky end  Single-stranded DNA overhang generated by some restriction enzyme digests. Sticky ends can be matched to a complementary sticky end produced by the same restriction enzyme on a different sample of DNA. Compare to blunt end. Structural brain imaging  Techniques that are used to resolve brain anatomy in a living subject without physically penetrating the skull. Examples include cerebral angiography, CT scans, and magnetic resonance imaging (MRI). Subcloning  The process of copying or moving a fragment of DNA from one recombinant DNA construct into a new vector to create a new recombinant DNA construct. Sucrose preference test  A behavioral measure of anhedonia in rodents. A scientist places a bottle of plain tap water and a bottle containing sweet water in a rodent’s cage. Normal rodents prefer sweet water, but rodent models of depression do not. Glossary  373 Sulci  The indentations created by the folds in the brain (plural for “sulcus”). Suture  Either (1) a seam between different parts of the skull or (2) a thread used after a surgery to close a wound. Synapto-pHlorin  A pH-sensitive fluorescent protein variant of GFP that is used to examine synaptic vesicle recycling and release. It does not fluoresce when present inside synaptic vesicles, but does fluoresce when the vesicle fuses during exocytosis. Systems neuroscience  The branch of neuroscience that examines the coordinated activity of neural ensembles or the firing properties of individual neurons that generate behavior and cognition. T1  In MRI and fMRI, the time (in milliseconds) required for a certain percentage of protons that had been excited by radiofrequency pulses to realign in the longitudinal direction. T1-weighted image  An MRI image formed using T1 signal intensity. T2  In MRI and fMRI, the time (in milliseconds) required for a certain percentage of protons that had been excited by radiofrequency pulses to relax in the transversal direction. T2-weighted image  An MRI image formed using T2 signal intensity. T-maze  A behavioral apparatus with two arms oriented in opposite directions, giving it a T shape. A Drosophila T-maze contains an additional loading arm to move flies into the choice point between the T. This maze can be used to test sensory preferences as well as learning and memory. Tail flick assay  A rodent assay for nociception in which painful heat or cold is applied to the tail, causing the rodent to rapidly flick its tail to the side. Tail suspension assay  An assay for depression in which a rodent is held by the tail and the amount of time the animal struggles is measured. Rodent models of depression stop struggling sooner than normal rodents but struggle for a normal amount of time if given antidepressant medication. Talairach space  A coordinate system based on the stereotaxic measurements of a single postmortem brain used to normalize MRI data so that anatomical comparisons can be made among different brains. Tamoxifen  A molecule that is able to bind to the estrogen receptor, causing its translocation from the cytoplasm to the nucleus. Useful to exert temporal control over the Cre/lox system when Cre recombinase is fused to the estrogen receptor. Task paradigm  A strategy for presenting stimuli to subjects during an experiment. In human brain imaging experiments, these paradigms are typically classified as a blocked design, event-related design, or mixed design. Temporal delay  In functional brain imaging techniques, the time delay between the presentation of a stimulus and measurement of changes in neural activity. Temporal resolution  The minimum distance in time necessary to distinguish between discrete neural events. Tesla  Units of measurement for magnetic field strength. Conventional MRI scanners create external magnetic fields at 1.5–3 T. 374 Glossary Tet-off/Tet-on system  An inducible promoter system that allows transgene expression to be either turned off (Tet-off) or turned on (Tet-on) through the use of an antibiotic called doxycycline. Tetrode  A bundle of four microelectrodes used to perform extracellular electrophysiological recordings, often in vivo recordings in awake, behaving rodents. Tetrodotoxin (TTX)  A toxin produced by specific species of fish, such as the pufferfish, that binds pores of voltage-gated sodium channels to block action potentials. The purified chemical is often used as a pharmacological agent to block action potentials or sodium channel current. Thermocycler  A machine that controls the heating and cooling of samples, most often used for PCR. Time of flight (TOF) detector  In mass spectrometry, an instrument that measures the mass to charge ratio of molecules in the gas phase. Total internal reflection fluorescence (TIRF) microscopy  A form of microscopy used to image molecules and/or events that occur at the cell surface. A TIRF microscope uses a rapidly decaying evanescent wave of fluorescence excitation to restrict imaging to a thin (∼100 nm) region of contact between a specimen and the surface it is on (usually a glass coverslip). Tracer  A chemical that can be injected and transported along neural processes to highlight axonal paths and connections. Transcranial magnetic stimulation (TMS)  A noninvasive method used to manipulate neural activity in the brain. A coil is positioned adjacent to a subject’s head. This coil generates magnetic field pulses that induce electrical activity in superficial brain structures. Depending on the specific sequence and strength of pulses, TMS can reversibly activate or inactivate brain regions. Transcription  The process of synthesizing RNA from a DNA template, performed by the enzyme RNA polymerase. Transcription activator-like effector nucleases (TALENs)  Artificially engineered restriction enzymes composed of a DNA-binding domain and a DNA-cleavage domain. Transcriptional start site  The location within a gene where RNA polymerase begins transcription. Transcription factor  A DNA-binding protein that recognizes a specific genetic regulatory element and regulates transcription by recruiting or blocking RNA polymerase. Transduction  The process of using a nonreplicating viral vector to deliver foreign DNA into a cell. Transfection  The process of delivering DNA to cells using nonviral methods. Transformation  The process of introducing a DNA vector into competent cells. Transgene  A foreign gene expressed in an organism that does not normally express the gene. Transgenic organism  An organism that carries foreign DNA that it does not normally express. Translation  The process by which a ribosome synthesizes a chain of amino acids based on an mRNA template. Glossary  375 Transmission electron microscopy (TEM)  A form of microscopy that transmits an electron beam through a thin specimen to generate an image based on differences in electron density of the material. Produces extremely high-resolution images of cellular ultrastructure. Transposable element  See Transposon. Transposase  An enzyme that acts on specific DNA sequences at the end of a transposon to disconnect the sequence from flanking DNA and insert it into a new target DNA site. Transposition  The natural process of a transposon moving to different positions within the genome. Transposon  A genetic sequence capable of translocating to different positions within the genome. Transsynaptic tracer  A tracer capable of passing through synapses useful for labeling multisynaptic connections. Trophoblast  Cells in a blastocyst that form the outer spherical shell and eventually develop into the placenta. Two-photon microscopy  A specialized form of fluorescent microscopy that sends two relatively low energy pulses of laser light that can add up to excite fluorophores in a restricted section of a specimen. Regions beyond the restricted section do not receive enough energy for excitation, greatly reducing out-of-focus fluorescence. UAS  See Upstream activation sequence. Ultrasonic neuromodulation  A noninvasive neuromodulation technique in which focused sound waves slightly heat an area of the brain, causing transient effects. Unconditioned stimulus (US)  In classical conditioning, the salient stimulus that eventually becomes associated with a previously neutral stimulus. Unipotent  A cell that is capable of dividing but only producing a single cell type. Upright microscope  A light microscope in which the objective lens is above the specimen and the light source and condenser are beneath the specimen. Upstream activation sequence (UAS)  A strong promoter derived from yeast that is activated by the Gal4 transcription factor. Validity  A subjective measure of how relevant an animal model is toward understanding humans. Vector  See DNA vector. Vertical pole test  A behavioral paradigm used to assay motor coordination and balance in rodents. Vibratome  A device that uses a vibrating blade to section a fresh, unfrozen specimen into tissue slices that can be kept alive. Viral gene delivery  The delivery of DNA constructs into cells in vivo or in vitro using one of many available viral vectors. Visual cliff assay  An assay used to examine normal visual function in rodents. Voltage (V)  An electromotive force that exists across a membrane. 376 Glossary Voltage clamp  An electrophysiology technique that allows a scientist to hold the membrane potential at a constant voltage to measure currents generated by ions moving across the membrane. Voltage-sensitive dye imaging (VSDI)  A cell or tissue visualization technique that uses dyes that shift their absorption or emission fluorescence based on the membrane potential, thus allowing scientists to visualize changes in membrane potential over time. Voltammetry  A method used to measure the presence and relative amounts of neurochemicals in living tissue. Volume  In brain imaging experiments, a complete scan of the brain composed of multiple, individual slices. Von Frey assay  A behavioral test used to assay noxious mechanical and pinch stimuli in rodents. Voxel  A three-dimensional unit of space, often used to describe a unit of brain space during whole-brain imaging experiments. Voxelwise analysis  A strategy for analyzing data in a brain imaging experiment, in which each voxel of data is examined for significant changes in signal intensity. Western blot  A method used to detect the expression of a protein in a biological sample. Whole-cell recording  A patch clamp recording technique in which a scientist places a glass pipette adjacent to a cell membrane, forms a tight gigaseal with the membrane, and provides enough suction so that the membrane patch is removed, allowing the interior of the pipette to become continuous with the cytoplasm of the cell. Whole-mount preparations  A histological preparation in which a chunk of tissue or an entire small animal brain is mounted onto a slide for analysis. Wide-field fluorescent microscopy  See Epifluorescent microscopy. Within-subjects study  A study that compares the effects of two different stimuli within the same subject pool. Worm tracker  Computer software that can automatically track locomotor behavior in Caenorhabditis elegans. X-ray  A structural visualization technique used to image parts of the body with natural contrast, such as a bone in tissue. This technique is inadequate by itself to visualize individual structures within the brain, but serves as the basis of cerebral angiogram and computerized tomography technology. Xylazine  A chemical sedative and analgesic, often used in combination with ketamine to anesthetize small rodents. Yeast artificial chromosome (YAC)  A large (100–2000 kb) DNA vector capable of replicating in yeast. Useful for manipulating large sequences of DNA, such as in the generation of transgenic or knockout constructs. Yeast two-hybrid assay  A method that takes advantage of yeast transcriptional machinery to determine whether two proteins directly interact with each other. Zinc finger nucleases  Artificially engineered restriction enzymes consisting of a zinc finger DNA-binding domain fused to a DNA-cleavage domain. 377 Index Note: Page numbers followed by “f”, “t”, and “b” indicate figures, tables and boxes, respectively. A Ablation, 187–188 Absolute recovery, 83 Action potential, 91 ionic basis of, 92f Activity sensors, 172t Acute cultures, 302 Adeno-associated virus, 250 Adenovirus, 248 Adobe Photoshop®, 136 Aequorin, 175 Affinity chromatography, 319 Agonist, 189–190, 190f Allatostatin receptor (Alstr), 253, 257 Amino acids, 208–209 Amperometry, 84–85 Amplifier, 92–93, 94b–96b Anatomical landmarks, 74–75, 74f Animal behavior, 39–40 behavioral assay. See Behavioral assay Animal use, ethical consideration, 42b Anisotropic diffusion, 13 Antagonist, 189–190, 190f, 198t–199t Anterograde tracers, 162–164, 162f, 163t Antibodies interference, 308 making and using, 314–318 protein-specific antibodies, in technique, 318t Antigens, 314 Anxiety, 58–60 defensive marble burying, 59 elevated plus maze, 59, 59f Geller-Seifter conflict test, 59–60 open field test, 58, 58f Aptamers, 315–318 Archaerhodopsin (Arch), 256–257 Archaerhodopsin-3 (Arch), 194 ArcLight, 254 Array tomography, 159–160 Assaying cellular function, 169–170 cell proliferation with thymidine analogs, 169–170, 170f protein trafficking with pulse-chase labeling, 170 Ataxia, 46 Ataxin, 191, 257 Attention and impulsivity, 56 five choice serial reaction time task, 56 go/no-go task, 56 Autofluorescence, 128–129 Autoradiography, 154b–156b B Bacterial artificial chromosome (BAC) constructs, 259 Bacteriophage, 227–228 Base pair, 206 Basic fibroblast growth factor (bFGF), 304 Basic Local Alignment Search Tool (BLAST), 214 Basophilic stain, 150 Behavioral assay behavioral paradigm selection, 41–42 model organism selection, 40–41 using animal behavior as a model for, 43–44 Between subjects, 26–27 Bimolecular fluorescence complementation (BiFC), 180 Binary expression systems, 260f–261f, 261 Binary transgenic systems, 260f–261f, 261–265 Cre/Lox system, 263–264 Flp/Frt system, 264 Gal4/UAS system, 262–263 promoters in, 262t Tet-off/Tet-on system, 264–265 378 Index Biochemical assays. See individual entries Biolistics, 243 Biological ballistics. See Biolistics Blastocyst, 277–278 inserting into female mouse, 278 Bleed-through, 139 Blocked designs, 27–28 Blood oxygen level-dependent (BOLD) effect, 14–17, 15f Blunt ends, 220–221 Brain atlases, 74–75, 74f Brain composition, 3f X-ray technology, 4f Bregma, 74–75 Brightfield microscopy, 126–127, 140t–142t Brodmann’s areas, 32–33 C Caenorhabditis elegans (C. elegans), 211b, 267–268, 289 advantages and disadvantages, 212t behavioral paradigms, 65–67 locomotor behavior, 66 sensory behavior, 66–67 Calcium indicator dyes, 174 genetically encoded, 173–175, 176f Canine adenovirus (CAV), 248 Cannulae, 80–81 Capsaicin receptor (TrpV1), 257 Carbon fiber microelectrode, 85 cDNA microarray screen, 215–216, 215f Cell body stains, 150–152 Cell Cell-attached mode, 105 culture. See Cell culture flow of information in, 205f fractionation, 323, 324f Cell culture, 295 equipment, 296–297 immortalized cell lines, 298–301 manipulating cells in, 307–308 media, 298 primary culture. See Primary cell/tissue culture stem cell cultures. See Stem cell cultures techniques, 295–310 Central dogma of molecular biology, 204 DNA, 204–206 transcription, 206–208 translation, 208–209 Cerebral angiogram, 3 Cerebral angiography, 3, 5f Cerebrospinal fluid (CSF), 2 Channelrhodopsin-2 (ChR2), 194, 253, 256–257 Chemical gene delivery, 244–247 calcium phosphate transfection, 244–246, 245f lipid transfection, 246–247 Chemical mutagenesis, 210 Chemosensory jump assay, 64 Chimeras, 278 Chromatin immunoprecipitation (ChIP), 338–341, 340f Chromatography, 319, 319f Chromogenic labels, 154b–156b Classical conditioning, 55b Clear lipid-exchanged acrylamide-hybridized rigid imaging/immunostaining/in situ hybridization-compatible tissue hydrogel (CLARITY), 149–150 Clone, 227b Cloning DNA, 227–231 ligation, 229–230 purifying DNA from host cells, 231 transformation, 231 vector. See Vectors Cloning vectors, 228 Clozapine-N-oxide (CNO), 257 Co-culture systems, 307–308 Codon, 208–209 Cognitive neuroscience, 24 Cohesive ends, 220–221 Co-immunoprecipitation (Co-IP), 327, 328f Colorimetric labels, 154b–156b Competent cells, 231 Complementary DNA (cDNA), 215–216 library, 223b–224b Complementary strands, 206 Complementation test, 213 Compound microscope, 121–125, 123f–124f Computer, 92–93, 94b–96b Computerized tomography (CT), 4–6, 5f Condenser, 122 Conditional knockout mice, 282–284 with the Cre/lox system, 281, 283f Conditioned stimulus (CS), 55b Confluence, 297 Confocal microscopy, 129–132, 131f advantages and disadvantages, 140t–142t Construct validity, 43–44 Contextual fear conditioning, 55b Coomassie blue, 321 Coronal plane, 149 Countercurrent apparatus, 63, 63f Index  379 Cranial windows, 81 Cre recombinase, 263, 282–283 and lox recombination system, 281 CRISPR (clustered regularly interspaced short palindromic repeat), 285–286 Cas9 nuclease system, 274, 285–287, 286f Cross talk. See Bleed-through Cross-linking fixatives, 146 Cryostat, 148, 148t Cued fear conditioning, 55b Current (I), 90 clamp mode, 96–99 I/V curves, 98b–99b Cyclic voltammogram, 84, 84b Cytokines, 313 D Darkfield microscopy, 127f, 140t–142t Data analysis current over time, 100b I/V curves, 98b–99b spikes, 101b–103b Defensive marble burying, 59 Dehydrating fixatives, 146 Dependent variable, 26 Depolarization, 90–91 Depression, 60–61 forced swim test, 60, 61f sucrose preference test, 60–61 tail suspension assay, 60 Designer receptors exclusively activated by designer drugs (DREADDs), 192–193, 257 Dicer, 287–288 Differential interference contrast (DIC) microscopy, 127, 140t–142t Diffuse optical imaging (DOI), 22 Diffusion, 13 Diffusion MRI, 12–13 Diffusion tensor imaging (DTI), 13, 13f Digoxigenin, 154b–156b Diphtheria toxin receptor, 191, 257 Dipstick assay. See Olfactory avoidance assay Direct IHC, 158–159 Dissecting microscope, 125 Dissociated cell cultures, 303–304 DNA, 204–206 cloning. See Cloning DNA identifying. See Identifying DNA ladder, 226 libraries, 223b–224b microinjection, 240–241 molecular structure of, 205f purifying from host cells, 231 sequencing, 233–234, 234f synthesis, 224 vector, 227–229 Dominant negatives, 290–291, 291f Double helix, 206, 207f Double-inverse orientation (DIO), 263–264 Doxycycline, 264 Drosophila melanogastor, 211b advantages and disadvantages, 212t behavioral paradigms, 61–65, 289 flight, 62–63 learning and memory, 65. See also Learning and memory locomotor behavior, 62 sensory function, 63–65. See also Sensory function social behaviors, 65 E Electroencephalography (EEG), 20, 21f Electrolytic lesion, 189 Electron microscope tomography, 135, 135f Electron microscopy, 133–135, 134f Electron tomography (ET), 135, 140t–142t Electrophoretic mobility shift assay (EMSA), 338–339, 338f Electrophysiology tissue preparations, 108–113 in vitro recordings, 108–112. See also In vitro recordings, in electrophysiology in vivo recordings, 112–113. See also In vivo recordings, in electrophysiology Electrophysiology, 81–82 Electroporation, 241–242, 242f in utero, 243f Electrospray ionization (ESI), 329b–330b Elevated plus maze, 59, 59f Embryonic lethal, 278–280 Embryonic stem (ES) cells, 304–305 inserting into blastocyst, 277–278 selecting, 276–277, 277f Emission filter, 128 Endogenous genes genetically perturbing, 274f manipulating, 273–294 Ensembl, 214 Enzymatic histochemistry, 160, 161f Enzyme-linked immunosorbent assay (ELISA), 323–324, 325f Epidermal growth factor (EGF), 298, 304 380 Index Epifluorescent microscopy, 129, 131f, 140t–142t Epitope, 314 Equipment, for cell culture, 296–297 biosafety hood, 296 cell incubator, 296 microscope, 297 refrigerator, 297 specially prepared cell culture flasks/plates, 297 water bath, 297 Estrogen receptor (ER), 283–284 and tamoxifen system, 283–284 Ethnology, 40 Event-related designs, 27–28 Event-related potentials (ERPs), 20 Excitation filter, 128 Excitatory postsynaptic potential (EPSP), 90–91 Exon, 208 Explant cultures, 303 Expression vector, 228–229 Extracellular recording, 89–90, 99–104, 101f F Face validity, 43–44 Faraday cage, 94b–96b Fast-scan cyclic voltammetry, 84 Fiber stains, 152 Fibroblast growth factor (FGF), 298 Flat skull position, 78 Flex construct. See Double-inverse orientation (DIO) Flight simulators, 62–63, 62f Flippase recognition targets (Frts), 264 Flippase recombinase (Flp), 264 Frt recombination system, 282–283 Floxed sequence, 263. See also Lox sites Fluorescence microscopy, 128–133 Fluorescence photoactivation localization microscopy (FPALM), 122b Fluorescence recovery, after photobleaching, 180 Fluorescence resonance energy transfer (FRET), 177–180, 179f Fluorescent activity indicators, 82 Fluorescent labels, 154b–156b Fluorodeoxyglucose (FDG), 17 Fluorophores, 128, 154b–156b Footprint pattern assay, 46, 48f Forced swim test, 60, 61f Formalin assay, 51 Forward genetic screen, 209–213 Founders, 266–267 Fractional intensity change, 171–173 Freezing microtome, 147–148 Functional brain imaging electroencephalography, 20, 21f experimental design and analysis, 24–35. See also Functional brain imaging experiments figure preparation, 33–35, 34f fMRI. See Functional magnetic resonance imaging (fMRI) magnetoencephalography, 20–22, 21f optical imaging, 22–24, 23f positron emission tomography, 17–19, 19f single-proton emission computerized tomography, 19–20, 21f techniques, comparison of, 23t Functional brain imaging experiments, 35b–36b conducting, 29–31 design and analysis, 24–35 human experiment. See Human experiments pilot experiments, 28 planning, 25–29 practical consideration, 25 structure of, 26–27 task paradigm, 27–28 Functional magnetic resonance imaging (fMRI), 14–17 animal experiment, 18b BOLD effect, 15f example, experiment, 16f G Gain-of-function experiment, 186–187 Gal4, 262 UAS system, 262–263, 263f Gal80, 263 Gaussian distribution curve, 101b–103b GCaMPs, 254 Gel electrophoresis, 229f isolation and characterization of DNA fragments using, 226–227 Gel shift assay. See Electrophoretic mobility shift assay (EMSA) Gel-filtration chromatography, 319 Geller-Seifter conflict test, 59–60 Gene delivery strategies, 239–252 chemical gene delivery. See Chemical gene delivery choosing, 251t Index  381 physical gene delivery. See Physical gene delivery viral gene delivery. See Viral gene delivery Gene gun, 243, 244f Gene products, disrupting, 287–291 dominant negatives, 290–291, 291f morpholinos, 289–290, 290f RNA interference, 287–289 Gene targeting, 273, 279f analyzing genotype after, 281f classical methods, 274–284 nuclease-based methods, 284–287, 284f Genes, 203–204 review of, 204–209 screening. See Genetic screens; In silico screens; Molecular screens Genetic model organisms, 211b relative advantages and disadvantages of different organisms, 212t Genetic screens, 209–213 forward genetic screen, 210–213 reverse genetic screen, 213 Genetically encoded calcium indicators (GECIs), 173–175, 176f, 254 imaging synaptic transmission, 176–177 Genetically encoded voltage indicators (GEVIs), 171, 173, 254 calcium indicator dyes, 174. See also Calcium indicator dyes imaging calcium dynamics, 173–175, 175f Genomic DNA library, 223b–224b Genotype, 203 Germ cells, 278 Gold labels, 154b–156b Golgi stain, 152, 153f, 153b Graded potential, 90–91 Gray matter, 2 Green fluorescent protein (GFP), 253 Ground electrode, 94b–96b Ground-state depletion (GSD), 122b Growth factors, 313 Guide RNA (gRNA), 285–286 Gyri, 33 H Halorhodopsin (NpHR), 194, 256–257 Hanging wire assay, 46–47 Hargreaves assay, 50 Headstage, 94b–96b Health Insurance Portability and Accountability Act’s (HIPAA) Privacy Rule, 28–29 Hemoglobin, 14 Herpes simplex virus (HSV), 249–250 Heterologous expression systems, 109 Histochemical stains, 151t. See also individual entries hM3Dq, 257 hM4Di, 257 Hodgkin, Alan, 105 Holding potential, 96–99 Homologous recombination, 275, 275f Homology arms, 275–276 Homology-directed repair (HDR), 284–285 Hooke, Robert, 117 Horizontal plane, 149 Hot plate assay, 50–51, 51f Human experiments data acquisition, 30–31 figure preparation, 33–35, 34f manipulating neural activity, 31–32 planning, 28–29 postexperimental data analysis, 32–35 working with humans, 29–30 Huxley, Andrew, 105 Hybridization, 234 Hyperpolarization, 90–91 Hypothesis generating, 209–210 Hypothesis testing, 209–210 I I/V curves. See under Current (I) Ibotenic acid, 190 Identifying DNA, 231–236 northern blot, 236 nucleic acid hybridization techniques, 234–236 sequencing, 233–234, 234f Southern blot, 235–236 Imaging voltage, 171–173 voltage-sensitive dyes, 171–173, 173f Immediate early genes (IEGs), 168 Immortalized cell lines, 298–301, 299t–300t versus primary cell culture, 302t Immunoblot. See Western blot (WB) Immunoelectron microscopy (IEM), 326–327 Immunofluorescence (IF), 158 Immunohistochemistry (IHC), 158–160, 159f, 280, 325–326 Immunopanning, 303–304 382 Index Immunoprecipitation, 319–320, 320f Implants, for long-term access, 80–85 amperometry, 84–85 cannulae, 80–81 cranial windows, 81 electrophysiology, 81–82 fluorescent activity indicators, 82 microdialysis, 82–84 neural activity in vivo, 81, 84–85 neural chemistry in vivo, 82 sealable chambers, 80 voltammetry, 84–85 In silico screens, 214 BLAST, 214 Ensembl, 214 In situ hybridization (ISH), 156–158, 157f, 280 In vitro recordings, in electrophysiology, 108–112 experiment, 110b–111b heterologous expression systems, 109 primary cultures, 109 slice cultures, 111–112 In vivo recordings, in electrophysiology, 112–113 acute versus chronic, 112 anesthetized versus awake animals, 112–113 In vivo techniques manipulating neural activity, 86 measuring neural activity, 81 measuring neural chemistry, 82 microdialysis, 77–78, 82–84 Independent variable, 26 Index of refraction, 118–120 Indirect IHC, 158–159 Induced pluripotent stem (iPS) cells, 304, 306–307 Infection, 239, 307 Inhibitory postsynaptic potential (IPSP), 90–91 Insertional mutagenesis, 210 Inside-out recording, 107 Institutional Animal Care and Use Committee (IACUC), 42b, 75 Institutional Review Board (IRB), 28–29 Intracellular recording, 89–90, 101f, 105 Intracellular signaling, 312–313, 312f experiment, 333b–334b Intracerebroventricular (i.c.v.) injection, 190 Intraperitoneal (i.p.) injection, 190 Introns, 208, 259 Inverted microscope, 124–125, 125f Ion-exchange chromatography, 319 Irradiation mutagenesis, 210 Isoflurane, 77 Isolating DNA fragments, 220–227 isolation and characterization of, 226–227 polymerase chain reaction, 221–226. See also Polymerase chain reaction (PCR) restriction enzymes, 220–221, 221f Isotropic diffusion, 13 J JAK-STAT pathway, 313, 314f Janus Kinases (JAKs), 313 K Ketamine, 77 Kinase assay, 336, 336f Knock down gene expression, 287 Knockin animal, 273 Knockin mouse, 282 Knockout animal, 273 Knockout mouse, 275–282 breeding offspring mice, 278–280 conditional, 282–284 creating targeting construct, 275–276, 276f injecting and selecting ES cells, 276–277 inserting blastocyst into female mouse, 278 inserting ES cells into blastocyst, 277–278 using in experiment, 280–282 validating, 280 L Label specific neural cell types, 158t lacZ, 253 Lambda, 74–75 Learning and memory, 65 nonspatial, 54–56, 55b spatial, 52–54 Lectins, 164 Leica (microscope manufacturer), 139 Lentivirus, 250 Lesion. See Ablation Light emitting diodes (LEDs), 195 Light microscopy, 126–127, 127f Linkage analysis, 213 Lipid transfection, 246–247 Lipofection, 246, 246f. See also Lipid transfection Liposomes, 246 Local field potentials, 101–103 Localized potential, 90–91 Locomotor activity, 44–45 exploratory, 46f home cage activity, 44–45, 45f Index  383 open field locomotion test, 45 running wheel, 44 Loss-of-function experiment, 186–187 Loudspeaker system, 92–93, 94b–96b Lox sites, 282–283 Luciferase, 253 assay, 341–342, 341f M Magnetic resonance imaging (MRI), 6–12 diffusion. See Diffusion MRI electromagnetic basis of, 6–7, 8f features, 11 functional. See Functional magnetic resonance imaging (fMRI) image generation, 8–11 slice selection, 11–12 subject in, 7f Magnetoencephalography (MEG), 20–22, 21f Magnification, 118 Magnifying glass, 123f Making transgenic organisms, 265–269 other organisms, 269 transgenic flies, 267, 268f transgenic mice, 265–267, 266f transgenic worms, 267–268, 269f Manipulating neural activity, 185–186 electrical manipulation, 189 experimental logic of, 186–187 genetic manipulation, 191–192 methods of, 186f, 198t optogenetic manipulation, 193–196. See also Optogenetic manipulation; neural activity pharmacogenetic manipulation, 192–193 pharmacological manipulation, 189–191 physical manipulation, 187–189. See also Physical manipulation; neural activity Mass spectrometry, 327, 329b–330b Matrix-assisted laser desorption ionization (MALDI), 329b–330b Meganucleases, 285 Microarray, 215 Microdialysis, 82–84, 82f Microdrive, 94b–96b Microelectrode, 92–93, 94b–96b Microinjection, 240–241 DNA, 241f Microiontophoresis, 190–191 Micromanipulator, 94b–96b Microscope, 94b–96b Microscopy, 117–118. See also individual entries comparison of different forms of, 130f, 140t–142t data preparation. See Microscopy data preparation fundamental parameters in, 118–121 principles, 118–126 resolving power, 118, 119f Microscopy data preparation, 136–139 image interpretation, 137–139, 138f image processing, 136–137 merging images, 137f Microstimulation, 189 Microtome, 147–148, 148t Midsagittal plane, 149 Mixed designs, 27–28 MNI (Montreal Neurological Institute) template, 32–33 Molecular cloning, 219–220, 227 definition, 219 isolating DNA fragments, 220–227 Molecular screens, 214–216 cDNA microarray screen, 215–216, 215f RNAi screen, 216 Monoclonal antibodies, 314–315 versus polyclonal antibodies, 315f producing, 316f Morpholinos, 289–290, 290f Motor coordination and balance, 45–48 footprint pattern assay, 46 hanging wire assay, 46–47 rotarod, 46, 47f vertical pole test, 47–48 Multielectrode array (MEA), 81, 103, 104f Multipotent, 304 Muscimol, 190 Myelin, 152 N Near-infrared spectroscopy (NIRS), 22 Neher, Erwin, 105 Neomycin resistance gene (neoR), 276–277 Nervous system visualizing, tissue preparation, 146–150 embedding, 147 fixation, 146–147 intracellular labeling, 153, 154b–156b juxtacellular labeling, 153, 154b–156b neuroanatomical planes, 149f sectioning, 147–150, 148f 384 Index Neural activity in vivo, 81 Neural chemistry in vivo, 82 Neural progenitors, 305 Neural stem cells (NSCs), 304–306 Neuroethnology, 41b Neuromodulation techniques, used in humans, 197 Neuronal differentiation, 301f Neurons, electrical properties of, 90–92 Neurospheres, 306, 306f NIH ImageJ®, 136 Nikon (microscope manufacturer), 139 Nissl stains, 150 Nomarski microscopy, 127, 140t–142t. See also Differential interference contrast (DIC) microscopy Nonhomologous end-joining (NHEJ), 284–285 Nonhuman primate behavioral paradigms, 67–69 Nonratiometric dyes, 174 Nonspatial learning and memory, 54–56 delayed/nonmatch to sample, 55–56 novel object recognition, 54–55 Northern (RNA) blot, 225–226, 235f, 236 Nuclease, 284–285 Nuclease-based gene targeting methods, 284–287, 284f Nucleic acid hybridization techniques, 225–226 Numerical aperture (NA), 118–120, 120f O Objective lens, 122 Ocular lens, 122 Ohm’s law, 90 Olfaction, 50 nociception, 50–51 Olfactory avoidance assay, 64 Olfactory jump response. See Chemosensory jump assay Open field locomotion test, 45 Open field test, 58, 58f Operant conditioning, 55b Optical imaging, 22–24, 23f Optogenetic manipulation, neural activity, 193–196 genetically encoded actuators, 193f methods of delivering light, to nervous system, 195, 195f optogenetic transgenes. See Optogenetic transgenes Optogenetic transgenes, 193–194 optogenetic inhibition, 194 optogenetic stimulation, 194 Optogenetics, 193 stimulation of projections using, 196f strategies for studying neural circuits using, 196 Organotypic slice culture, 302 Oscilloscope, 92–93, 94b–96b Outside-out recording, 107 P P elements, 267, 268f Packaging cells, 247–248 Particle-mediated gene transfer, 243 Passaging, 298 Patch clamp techniques, 89–90, 101f, 105–108 relative advantages and disadvantages of, 107t Pavlovian conditioning. See Classical conditioning Perforated patch, 107–108 Perfusion, 146–147 Peri-stimulus time histogram (PSTH), 101b–103b Phage. See Bacteriophage Pharmacogenetics, 192 Pharmacology, 308 Phase-contrast microscopy, 127, 140t–142t Phenotype, 203 Phosphorylation, 335f Photoactivated localization microscopy (PALM), 122b Photoactivation, 180 Photobleaching, 128–129, 180, 181f Photoconversion, 154b–156b, 180 Phototaxis, 63 Phototoxicity, 128–129 Physical gene delivery, 239–243 biolistics, 243 categories, 240t electroporation, 241–242, 242f microinjection, 240–241 Physical manipulation, neural activity, 187–189 ablation, 187–188 thermal cooling, 188–189 Plasmids, 227–228 Pluripotent, 304 Polyclonal antibodies, 314–315 versus monoclonal antibodies, 315f producing, 317f Index  385 Polymerase chain reaction (PCR), 221–226, 221f, 280 modifying DNA sequences using, 225 PCR cycle, 222–224 quantitative real-time, 225–226 reverse transcription, 225 standard PCR, 221–224 Porsolt test. See Forced swim test Position effects, 259–261 Positron emission tomography (PET), 17–19, 19f Positron-emitting isotope, 17 Poststimulus time histogram, 101b–103b Posttranslational modifications (PTMs), 208–209 investigating, 331–337, 335t specific antibodies, 337, 337f specific assays, 336–337 Prepulse inhibition (PPI), 50b Primary antibody, 158–159 Primary cell/tissue culture, 301–304, 301f dissociated cell cultures, 303–304 explant cultures, 303 versus immortalized cell lines, 302t slice culture, 302 Primate behavioral task, 68f Primers, 221 Proboscis extension response (PER), 64–65 Progenitor cells, 304 Promoters, 206, 258 Protein expression, investigating, 320–327 ELISA. See Enzyme-linked immunosorbent assay (ELISA) immunohistochemistry, 325–326 methods of measuring, 321t radioimmunoassay, 324–325 western blot. See Western blot (WB) Protein kinases, 313 Protein–DNA interactions, 337–342 chromatin immunoprecipitation, 339–341, 340f electrophoretic mobility shift assay, 338–339, 338f luciferase assay, 341–342, 341f Protein–protein interaction, 327–331 co-immunoprecipitation, 327, 328f protein affinity chromatography, 328–330 yeast two-hybrid assay, 330–331, 331f–332f Proteins, 203–204, 311 definition, 203 and DNA interaction. See Protein–DNA interactions expression. See Protein expression, investigating fundamental tools used, 313–320 and protein interaction. See Protein–protein interaction purifying, 318–320 review of, 204–209 tagged, 317b Proteomics, 329b–330b Protons alignment in magnetic field, 6, 8f process around axis, 7, 9f Protospacer adjacent motif (PAM), 286–287 Pseudopregnant, 265 Pulse-chase labeling, 170 Q Quantitative real-time PCR (qRT-PCR), 225–226, 225f Quantum dots, 154b–156b R Radioactive amino acids, 164 Radioactive labels, 154b–156b Radiofrequency (RF) pulses, 8–9, 9f Radioimmunoassay (RIA), 324–325, 326f Raster plot, 101b–103b Ratiometric dyes, 174 Recognition sites, 220 Recombinant DNA plasmid, 230f Recombinant DNA technology, 219–238 Recombineering, 229–230, 231b Refractive index, 118–120 Region-of-interest (ROI) analysis, 33 Relative recovery, 83 Reporter genes, 160, 161f, 177, 254 Reporter proteins, 327 Resistance (R), 90 Resolution, 118 diffraction limit of, 122b Resolving power, 118, 119f. See also Resolution Resting potential, 90 ionic basis of, 91f Restriction digest, 220–221 Restriction endonucleases. See Restriction enzymes Restriction enzymes, 220–221, 221f Restriction sites, 220 Retrograde tracers, 162–164, 162f, 163t Reverse genetic screen, 209–210, 213 Reverse genetics, 273–274 Reverse microdialysis, 83 386 Index Reverse transcription PCR (RT-PCR), 225 Reward-related behaviors, 56–57 conditioned place preference, 57 self-administration, 56–57 Ribosomes, 208–209 RNA RNA-induced silencing complex (RISC), 287–288 polymerase, 206 splicing, 208f RNA interference (RNAi) screen, 216, 287–289, 288f Rodent behavioral paradigms, 44–61 anxiety, 58–60. See also Anxiety attention and impulsivity, 56 depression, 60–61. See also Depression locomotor activity, 44–45. See also Locomotor activity motor coordination and balance, 45–48. See also Motor coordination and balance nonspatial learning and memory. See Nonspatial learning and memory reward-related behaviors, 56–57 sensory function, 48–50. See also Sensory function social behaviors, 57–58 spatial learning and memory. See Spatial learning and memory S Sagittal plane, 149 Sakmann, Bert, 105 Sanger dideoxy chain-termination method, 233–234 Saturated structured-illumination microscopy (SSIM), 122b Scanning electron microscopy (SEM), 133–135, 140t–142t Sealable chambers, 80 Secondary antibody, 158–159 Sensory behavior, 66–67 chemosensation, 67 mechanosensation, 66–67 thermosensation, 67 Sensory function, 48–50, 63–65 formalin assay, 51 Hargreaves assay, 50 hot plate assay, 50–51, 51f olfaction, 64. See also Olfaction prepulse inhibition (PPI), 50b startle response assay, 49 tail flick assay, 50 taste, 49, 64–65 T-maze, 64f vision, 63 visual cliff assay, 48, 49f Von Frey assay, 51 Short hairpin RNA (shRNA), 288–289, 289f Signal Transducers and Activators of Transcription (STATs), 313 Signal transduction, 313 Single electrodes, 81 Single-proton emission computerized tomography (SPECT), 19–20, 21f Slice culture, 302 Slices, 12f, 28–29 definition, 11 selection, 11–12 Small interfering RNA (siRNA), 287–288 Social behaviors, 57–58, 65 aggression, 65 courtship, 65 resident-intruder assay, 57 social approach/avoidance, 58 Sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (PAGE), 320–321 Southern (DNA) blot, 220–2270, 235f, 280 Spatial learning and memory, 52–54 Barnes maze, 53, 53f Morris water maze, 52–53, 52f radial arm maze, 53–54, 54f Spatial resolution, 11 Spike, 91. See also Action potential Spike sorting, 103–104 Spliceosome, 208 Splitting, 298 Startle response assay, 49 Static markers, nervous system’s function, 168–170 assaying cellular function. See Assaying cellular function assaying neural activity, 168–169, 169f Stem cell cultures, 304–307 embryonic stem cells, 304–305 induced pluripotent stem cells, 306–307 neural stem cells, 305–306 Stereomicroscope, 125–126, 126f Stereotaxic instrument, 74–75, 75f Stereotaxic surgeries and in vivo techniques, 73–88 in nonhuman primates, 79–80 in rodents. See Stereotaxic surgeries; in rodents Stereotaxic surgeries, in rodents, 73–80 accessing brain, 78 anesthesia, 77–78 Index  387 finishing surgery, 79 penetrating brain, 78–79 positioning in instrument, 78 sterile environment, 76 Sterile environment, 76 equipment, 77f Sticky ends, 220–221 Stimulated emission depletion (STED), 122b Stochastic optical reconstruction microscopy (STORM), 122b Structural brain imaging, 2–13 cerebral angiography, 3, 5f computerized tomography, 4–6, 5f magnetic resonance imaging. See Magnetic resonance imaging (MRI) Subcloning, 227, 232b–233b Sulci, 33, 34f Synaptic transmission, 176–177 dynamics, 177f FM dyes, 176 pH-sensitive fluorescent proteins, 176–177 T T1-weighted image, 10–11, 10f T2-weighted image, 10–11, 10f Tagged proteins, 317b Tail flick assay, 50 Talairach space, 32–33 TaqMan assay, 225f Task paradigms, 27–28 relative advantages and disadvantages of different paradigms, 28t Temporal delay, 15–17 Temporal resolution, 15–17 Tetracycline transactivator (tTA) protein, 264 Tetraethylammonium (TEA), 189–190 Tetrodes, 103, 104f Tetrodotoxin (TTX), 189–190 The Principles of Humane Experimental Research (Russell and Burch), 42b Thermal cooling, 188–189 Thermocyclers, 224 Thymidine analogs, 169–170 Time of flight (TOF) detector, 329b–330b Total internal reflection fluorescence (TIRF) microscopy, 132–133, 133f, 140t–142t Transcranial magnetic stimulation (TMS), 31–32, 31f, 197 Transcription, 206–208 of DNA into RNA, 207f Transcription activator-like effector nucleases (TALENs), 285 Transcription factors, 206, 259 Transcriptional start site, 259 Transfection, 239, 307 calcium phosphate transfection, 244–246, 245f lipid transfection, 246–247 Transgenes, 253–258 to ablate neurons, 257 commonly used, 255t–256t to disrupt endogenous gene function, 257–258 to manipulate neural activity, 256–257 to measure neural activity, 254 overexpression of endogenous genes, 258 reporter genes, 254 Transgenic organisms, 253–272 binary systems. See Binary transgenic systems definition, 253–254 making. See Making transgenic organisms transgenes. See Transgenes transgenic construct, 258–261, 258f Translation, 208–209 of mRNA to protein, 209f Transmission electron microscopy (TEM), 133–134, 140t–142t Transposable elements, 267 Transposase, 267 Transposition, 267 Transposons, 210, 267 Transsynaptic labeling, 164–165 plant lectins, 164 radioactive amino acids, 164 viruses, 165 Trophoblast, 277–278 Two-photon laser scanning microscopy (TPLSM), 132 Two-photon microscopy, 132, 140t–142t U Ultrasonic neuromodulation (USNM), 32, 197 Unconditioned stimulus (US), 55b Unipotent, 304 Untranslated regions (UTRs), 208 Upright microscope, 124–125, 125f Upstream activation sequence (UAS), 262 V Vectors, 227–229 features, 228 types of, 228–229 viral, in neuroscience, 249t 388 Index Vertical pole test, 47–48 Vibration isolation system, 94b–96b Vibratome, 148, 148t Viral gene delivery, 247–250, 247f adeno-associated virus, 250 adenovirus, 248 canine adenovirus, 248 herpes simplex virus, 249–250 lentivirus, 250 in neuroscience, 249t Visible spectrum of light, 121f Visual cliff assay, 48, 49f Visualizing circuitry, 162–165 anterograde tracers, 162–164, 162f, 163t retrograde tracers, 162–164, 162f, 163t transsynaptic labeling. See Transsynaptic labeling Visualizing gene and protein expression, 156–160 enzymatic histochemistry, 160, 161f immunohistochemistry, 158–160, 159f in situ hybridization, 156–158, 157f reporter genes, 160, 161f Visualizing nervous system, function, 167–168 neural activity. See Visualizing neural activity static markers of activity. See Static markers, nervous system’s function Visualizing nervous system, structures, 145–146 morphology, 150–153 tissue preparation. See Nervous system visualizing, tissue preparation Visualizing neural activity, 171–177 imaging voltage, 171–173. See also Imaging voltage Visualizing protein function, 177–180 bimolecular fluorescence complementation, 180 comparison of techniques, 178t fluorescence recovery after photobleaching, 180 fluorescence resonance energy transfer, 177–180 photoactivation, 180 photoconversion, 180 time-lapse imaging with reporter genes, 177, 179f Voltage clamp, 109 Voltage-sensitive dye imaging (VSDI), 171 Voltammetry, 84–85 analyzing and depicting data, 85f Von Frey assay, 51 von Leeuwenhoek, Anton, 117 Voxel, 11, 28–29 Voxelwise analysis, 33 W Western blot (WB), 280, 320–323, 322f–323f White matter, 2 Whole brain imaging, 1–2 functional brain imaging, 14–24. See also Electroencephalography (EEG); Functional magnetic resonance imaging (fMRI); Magnetoencephalography (MEG); Optical imaging; Positron emission tomography (PET); Single-proton emission computerized tomography (SPECT) structural brain imaging, 2–13. See also Cerebral angiography; Computerized tomography (CT); Magnetic resonance imaging (MRI) Whole-cell recording, 106–107 Whole-mount preparations, 149 Wide-field fluorescent microscopy, 129 Within subjects, 26–27 Worm Tracker (software), 66, 66f www.brain-map.org (online resource), 157 X Xanthomonas bacteria, 285 Xenopus laevis oocytes, 263f, 289–290 X-gal, 333b–334b X-ray technology, 2–3, 4f Xylazine, 77 Y Yeast protein. See Gal80 Yeast two-hybrid assay, 330–331, 331f–332f Z Zebra fish (Danio rerio), 129, 211b, 212t, 263f, 289–290 Zeiss (microscope manufacturer), 139 Zinc finger nucleases (ZFNs), 285