REVIEWS
Functional genomics
The study of gene functions and interactions in relationship to RNA transcripts and protein products using genome-wide data, and often involving high-throughput methods.
RNA interference
(RNAi). A technique used to knock down the expression of a specific gene by introducing a double-stranded RNA molecule that complements the gene of interest and triggers the degradation of the target mRNA.
'Broad Institute of Massachusetts Institute of Technology and Harvard, Cambridge, Massachusetts 02142, USA.
2McCovern Institute for Brain Research, Massachusetts Institute of Technology. 3Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology. "Department of Biological Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA.
Correspondence to F.Z. zhang@broadinstitute. org
doi:10.1058/nm.2015.2 Published online 10 Dec 201 5
Applications of CRISPR-Cas systems in neuroscience
Matthias Heidenreich'~4 and Feng Zhang'~4
Abstract | Genome-editing tools, and in particular those based on CRISPR-Cas (clustered regularly interspaced short palindromic repeat (CRISPR)-CRISPR-associated protein) systems, are accelerating the pace of biological research and enabling targeted genetic interrogation in almost any organism and cell type. These tools have opened the door to the development of new model systems for studying the complexity of the nervous system, including animal models and stem cell-derived in vitro models. Precise and efficient gene editing using CRISPR-Cas systems has the potential to advance both basic and translational neuroscience research.
Our understanding of brain function at the cellular and circuit level has been greatly advanced by functional genomics and the availability of various genetic tools to decipher neuronal diversity and function and model human brain disorders in non-mammalian and mammalian organisms. Just as the development of chemical DNA mutagens1 and RNA interference (RNAi)2 led to huge leaps in the fields of genetics and developmental biology — mainly as a result of research in non-mammalian organisms such as flies, worms and fish3"5 — precise genetic modifications introduced by homologous recombination (HR) in embryonic stem cells (ESCs)6 paved the way for studying the mammalian brain and modelling human diseases in mice and rats. For example, many neurological disorders, such as Alzheimer disease, are associated with genetic risk factors that can be introduced and studied in animal models7. In addition, novel approaches based on human ESCs and induced pluripotent stem cells (iPSCs) are changing the way that we model cellular processes under normal and pathological conditions in vitro. For example, human stem cells can be differentiated into neurons or glia to genetically dissect the molecular mechanisms of complex brain disorders in vitros~u. Genome-editing technologies are allowing researchers to take full advantage of both animal and cellular models and to work more easily with non-traditional model organisms for neuroscience research.
Genome-editing tools based on site-specific DNA nucleases, including zinc-finger nucleases (ZFNs)13"15, transcription activator-like effector nucleases (TALENs)16"19 and the CRISPR-associated (Cas) effector proteins of clustered regularly interspaced short palindromic repeat (CRISPR) systems, such as Cas9 (REFS 20-25) and Cpfl (REFS 26,27), have been developed to facilitate site-specific genomic modifications. In addition, ZFs28, TALEs29 and enzymatically inactive versions of Cas9 (known as dead
Cas9 (dCas9))30 can be coupled to functionally different enzymatic domains30"35 or fluorescent proteins36 to achieve targeted transcriptional control, epigenetic modification and DNA labelling (FIG. 1).
ZFNs and TALENs recognize specific DNA sequences through protein-DNA interactions, whereas the DNA-specificity of Cas proteins is RNA-guided. To target Cas proteins to specific genomic loci, dual-guide RNAs or single-guide RNAs (sgRNAs)24,25,27,37,38 can be designed and generated quickly. Another key advantage of Cas proteins is that multiple sgRNAs can be used simultaneously to edit multiple genes, which can be useful for studying genetic interactions and modelling multigenic disorders, something that previously required multiple cloning and complex protein engineering steps to achieve using ZFNs and TALENs.
The benefits of using CRISPR-Cas systems to study the nervous system are highlighted by several successful applications in different animal species and cell types to study synaptic and circuit function39"41, neuronal development42"45 and diseases41'46. Here, we describe how genome-editing tools, and in particular those based on CRISPR-Cas enzymes, are opening new avenues for neuroscientific and biomedical research through the generation of new model systems, both in vivo and in vitro, and discuss the challenges and possible future applications of this technology for understanding the brain.
Overview of genome-editing strategies
Site-specific nucleases, including ZFNs, TALENs and Cas proteins, enable precise genetic modifications by inducing double-strand DNA breaks (DSBs) at target locations in the genome. Two highly conserved DNA-repair machinery pathways typically repair DSBs that would otherwise result in cell death: non-homologous end joining (NHEJ) and homology-directed repair
36 I JANUARY 2016 I VOLUME 17
© 2016 Macmillan Publishers Limited. All rights reserved
www.nature.com/nrn
REVIEWS
a Precise gene editing
iDNA-binding domains lofZFN or TALEN
^ ^^-Nuclease
DSB
Cas9-	\    ▼ \		
			
J 1		ÄTTTI7	III)
		IsgRNA	
DSB 5' — 3' —
NHEJ
Indel mutation.
Premature stop codon
HDR
I HR DNA template
Knock-in or gene correction
b Chromosomal rearrangement
Homologous recombination
(HR). The exchange of homologous DNA strands between similar DNA molecules, an event that occurs naturally during meiosis to generate genetic variation. HR is used to direct error-free repair of DNA double-strand breaks induced by DNA nucleases, such as zinc-finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs) and clustered regularly interspaced short palindromic repeat (CRISPR)-associated (Cas) proteins.
Embryonic stem cells
(ESCs). Totipotent cells derived from embryos that can be genetically manipulated in vitro to generate transgenic, knock-in and knockout mice. ESCs can also be directed to differentiate into various cell types in vitro, including neurons and glial cells.
Induced pluripotent stem cells
(iPSCs). Pluripotent cells derived from reprogrammed differentiated adult cells; iPSCs have properties similar to those of embryonic stem cells and therefore can, in principle, be differentiated into all cell types of the body.
c Large chromosomal deletion
Transcriptional control
Transcriptional activator or repressor,
dCas9
Epigenetic modulation
Epigenetic modifie
DNA labelling
Figure 11 Genome-editing applications of CRISPR-Cas9. a | Non-homologous end-joining (NHEJ) and homology-directed repair (HDR) after a DNA double-strand break (DSB) is induced by zinc-finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs) or clustered regularly interspaced short palindromic repeat (CRISPR)-associated protein 9 (Cas9). ZFNs and TALENs recognize their DNA-binding site via protein domains that can be modularly assembled for each DNA target sequence. Cas9 recognizes its DNA-binding site via RNA-DNA interactions mediated by the short single-guide RNA (sgRNA), which can be easily designed and cloned. The error-prone NHEJ repair pathway53 can result in the introduction of insertion or deletion (indel) mutations that can lead to a frame shift, the introduction of a premature stop codon and, consequently, gene knockout. The alternative repair pathway, HDR14,47-53, can be used to introduce precise genetic modifications if a homologous DNA template is present, b | Two different sgRNAs guide Cas9 to induce DNA cleavage at two different genes, resulting in chromosomal rearrangements116117, c | Two proximate sgRNAs guide Cas9 to induce DNA cleavage at two different loci of the same gene, introducing large deletions118119, d | The nuclease-inactivated version of Cas9 (dead Cas9 (dCas9)) can be fused to different functional enzymatic domains to mediate transcriptional control, epigenetic modulation or fluorescent DNA labelling of specific genetic loci30-36. HR, homologous recombination; M, methyl group.
NATURE REVIEWS I NEUROSCIENCE VOLUME 17 I JANUARY 2016 I 37
© 2016 Macmillan Publishers Limited. All rights reserved
REVIEWS
(HDR)14,47"55 (FIG. la). The highly error-prone NHEJ pathway induces insertions and deletions (indels) of various lengths that can result in frameshift mutations and, consequently, gene knockout. By contrast, the HDR pathway directs a precise recombination event between a homologous DNA donor template and the damaged DNA site, resulting in accurate correction of the DSB. Therefore, HDR can be used to introduce specific mutations or transgenes into the genome. Because ZFNs and TALENs achieve specific DNA binding via protein domains, individual nucleases have to be synthesized for each target site. By contrast, Cas9 is guided by a specificity-determining guide-RNA sequence (CRISPR RNA (crRNA)) that is associated with a trans-activating crRNA (tracrRNA) and forms Watson-Crick base pairs with the complementary DNA target sequence, resulting in a site-specific DSB22,23,37,56. A simple two-component system (consisting of Cas9 from the bacterial species Streptococcus pyogenes2*25 or Staphylococcus aureus57 and a fusion of the tracrRNA-crRNA duplex to a sgRNA)37 has been engineered for expression in eukaryotic cells and can achieve DNA cleavage at any genomic locus of interest. More recently, Cpfl, a single-RNA-guided nuclease that does not use tracrRNA, has also been adapted for genome editing27. Hence, different Cas proteins can be targeted to specific DNA sequences simply by changing the short specificity-determining part of the guide RNA, which can be easily achieved in one cloning step.
Gene editing across species
Non-human animal models provide an experimental platform to dissect the complexity of the brain and study the cellular and molecular underpinnings of brain disorders. Neuroscience in particular benefits from exploiting a wide range of species, including worms, flies, fish and mammals, as well as non-traditional model systems, such as birds and amphibians58. Disrupting gene expression is a common approach to study gene function and understand loss-of-function disease mutations. For many years, RNAi was the 'gold standard' for gene silencing and studying gene function in vitro and in vivo59,60; however, genome editing based on engineered designer nucleases offers several advantages over RNAi (TABLE 1). For example, genome-editing tools can be modified to allow for more refined control of gene expression beyond simple gene knockdown, adding to their versatility (FIG. Id).
Multiplying the power of simple model organisms. At the molecular level, non-mammalian model systems can provide important information about fundamental features of the nervous system as a result of their well-characterized genetic and cellular organization and amenability to a range of genetic tools. For example, many evolutionarily conserved genes involved in human neurological disorders such as Alzheimer disease and Parkinson disease have been extensively studied
Table 11 Comparison of approaches for gene knockdown or knockout
Approach
	RNAi	ASO	ZFN	TALEN	CRISPR-Cas
Molecular target	RNA	RNA	DNA	DNA	DNA
Result of targeting	Reversible knockdown	Reversible knockdown	Irreversible knockout	Irreversible knockout	Irreversible knockout
Ease of generating target specificity	Easy: simple oligo synthesis and cloning steps and limited chemical modifications to enhance RNA degradation	Easy: simple oligo synthesis and cloning steps; often chemical modification is required to enhance RNA binding and ASO stability	Difficult: substantial cloning and protein engineering required	Moderate: substantial cloning steps required	Easy; simple oligo synthesis and cloning steps
Off-target activity	High	High	Moderate	Low	Low
Ease of multi-plexing	High	High	Low	Moderate	High
Transcriptional and epigenetic control	Direct control not possible	Direct control not possible; TSOs can interfere with protein translation	DNA-binding ZF domains can be fused to new functional domains	DNA binding domains can be fused to new functional domains	Enzymatically inactive dCas9 can be fused to new functional domains
Ease of delivery into the mammalian CNS	High: delivered by nanoparticles, bioconjugates, cell-penetrating peptides or viral vectors	High: delivered by nanoparticles, bioconjugates, cell-penetrating peptides orviral vectors	Moderate: delivered by viral vectors	Moderate: delivered by viral vectors but large size makes packaging intoviral vectors challenging	Moderate: delivered by electroporation, PEI-mediated transfection, nanoparticles and viral vectors
Ease of generating large-scale libraries	High: simple oligo synthesis and cloning required	High: simple oligo synthesis and cloning required	Low: complex protein engineering required for each gene	Moderate: technically challenging cloning steps	High: simple oligo synthesis and cloning required
Costs	Low	Low	High	Moderate	Low
ASO, DNA antisense oligonucleotide; Cas, CRISPR-associated protein; CRISPR, clustered regularly interspaced short palindromic repeat; dCas9, dead Cas9; PEI, polyethylenimine; RNAi, RNA interference; TALEN, transcription activator-Like effector nuclease; TSOs, translation-suppressing oligonucleotides; ZF, zinc-finger; ZFN, ZF nuclease.
38 I JANUARY 2016 | VOLUME 17 www.nature.com/nrn
© 2016 Macmillan Publishers Limited. All rights reserved
REVIEWS
Epigenetic mechanisms
Multilayered cellular processes that modulate gene expression and function in response to interoceptive and environmental stimuli during development, adult life and ageing, including DNA methylation, post-translational histone modifications, ATP-dependent nucleosome and higher-order chromatin remodelling, non-coding RNA deployment and nuclear reorganization.
using flies, worms and fish61-63. For years, studies using these simple model organisms relied mainly on genetic screens using chemical mutagenesis and RNAi3"5 or imprecise methods for transposon excision and retroviral insertion64"66. More-precise genetic modifications have been achieved using ZFNs67"69, TALENs70"73 and Cas proteins (reviewed in REF. 74). In the case of Cas proteins, large numbers of RNA guides can be easily synthesized to study gene function on a large scale. By contrast, generating large libraries based on ZFNs and TALENs is challenging owing to difficulties in designing and synthesizing these proteins with varying DNA binding specificities (TABLE 1). In a proof-of-concept study, approximately 50 genes were screened with Cas9 and novel loci involved in electrical synapse formation in zebrafish were identified43. Such in vivo screening approaches in small model organisms offer an accessible platform to identify the genes involved in various aspects of nervous system function and dysfunction.
Rapid generation of mammalian models. The development of methods facilitating HR in ESCs6 enabled neuroscientists to study the effects of gene knockouts, mainly in mice. This approach has been significantly enhanced by genome-editing technologies (FIG. 2a,b). Genome editing in single-cell embryos has been used to generate mouse75, rat76 and primate models77'78 that can be used to study the role of specific proteins in nervous system function. Mouse and rat models provide a bridge between our understanding of the molecular underpinnings of the nervous system gleaned from studies in non-mammalian systems and the complex pheno-types observed in human brain disorders. However, in some cases, a comprehensive understanding of the human brain will require primate models, which have brains that are more similar to the human brain in terms of neuroanatomical, physiological, perceptual and behavioural characteristics.
Transgenic approaches in primates are generally very inefficient. However, successful insertion of transgenic alleles in primates, including macaques79'80 and the common marmoset81, has been achieved using retroviral and lentiviral approaches in early embryos. For example, the viral insertion of a disease-related version of the human gene huntingtin (HTT) into the macaque genome recapitulated clinical features of Huntington disease80, representing an important step forward for genetic-disease modelling in non-human primates. TALENs have also been successfully used in monkeys to model mutations in methyl-CpG-binding protein 2 (MECP2), an X-linked Rett syndrome gene77, and genome-engineered primates have been generated by precise disruption of single and multiple genes with Cas9 (REF. 78). The simplicity of the use of Cas proteins relative to that of ZFNs and TALENs, and the ability to modify multiple genes simultaneously, is a breakthrough that is already catalysing molecular interrogation of neurological and psychiatric dysfunctions in disease-relevant brain circuits using primate models78'82. The ability to examine brain function in genetically modified non-human primates has the potential to contribute significantly to
our understanding of higher cognitive functions and to the development of new therapeutic strategies for diseases that cannot be adequately modelled in rodents. However, such research raises important bioethical questions and requires careful consideration of the costs and benefits before moving forward.
In vivo gene editing in the brain
In vivo gene editing allows the systematic genetic dissection of neuronal circuits and the ability to model pathological conditions while bypassing the need to engineer germline-modified mutant strains. This experimental approach is fast, independent of genetic background, animal species and availability of ESCs, and can be applied to existing disease models and transgenic strains, as well as to aged animals to study age-related neurological changes (FIG. 2c). In vivo methods based on RNAi have been commonly used to reduce the expression of genes in the brain83. In addition, alternative methods based on DNA antisense oligonucleotides (ASOs) can be used for gene silencing and have been shown to be promising therapeutic molecules for suppressing pathogenic protein aggregates in the brain84,85. However, neither strategy allows the generation of stable gene knockouts or site-specific epigenetic modifications (TABLE 1). In the mouse brain, histone modifications and transcriptional control have been achieved using ZFs86 and TALEs32, and Cas9 has been used to induce indel mutations in neurons to achieve stable gene knockouts in living animals39'41. This demonstrates the capacity for spatial and temporal control of gene expression in fully developed circuits and also opens the door to probing epigenetic dynam-ics30-33,35 in the brain. Epigenetic control is of particular interest, as there is increasing evidence that epigenetic mechanisms, such as histone modifications and DNA methylation, play a part in learning, memory formation and the pathology of neuropsychiatric disorders87. Using Cas proteins, functional domains of DNA-methylation or -demethylation enzymes or histone modifiers can be easily targeted to specific DNA sequences to edit the epigenome with high spatial and temporal specificity in vivo (FIG. 1 d).
Delivery to the brain. Viral vectors are a promising mode for delivery of Cas proteins to the brain. Viral vectors have defined, tissue-specific or cell type-specific tro-pism and can be admitted either locally to the brain or through the bloodstream to achieve more-systematic tissue penetration88. The most-attractive gene-delivery vectors are adeno-associated viruses (AAVs), which afford long-term expression without genomic integration, are relatively safe and are non-pathogenic89'90. However, AAV vectors have limited transgene capacity, and the large size of the commonly used S. pyogenes Cas9 variant poses a significant challenge for AAV-mediated delivery41'91. AAV-mediated delivery may become even more challenging when Cas9 is enlarged by the fusion of additional functional domains. Smaller Cas9 ortho-logues, such as those derived from S. aureus, are easier to pack57, making them an attractive option for in vivo genome editing in the brain.
NATURE REVIEWS I NEUROSCIENCE VOLUME 17 I JANUARY 2016 I 39
© 2016 Macmillan Publishers Limited. All rights reserved
REVIEWS
Vector design Ar-and cloning
ESC transduction with HR template
I
[     Clonal selection
6-8 weeks
Determination of HR (Southern blotting and sequencing)
(    Clonal expansion "")
_*
(   Blastocyte injection ")
C
Chin
1 year
Verification of germline transmission and breeding of founders
1-2 years
Genetic backcrossing and/or cross-breeding
iCas9
gRNA design and ClpRNA :loning (single or ^~tr
ling (singl. multiple guides^
sgRNA validation (surveyor nuclease and sequencing)
i
Cas9-sgRNA injection into zygotes with desired genetic background
Breeding of founders (single or multiplexed)
sgRNA design and cloning (single or multiple guides)
Cell type-specific-'
promoter (e.g. SYN,CFAP)
j^^j)iiirm -jimnm^jjjiP
i
(    sgRNA validation ) 1
(     AAV production )
I"
Stereotactic injection of AAV
^ uniím
Cas9 mouse
Wild-type animal
• Aged animal or disease model
• Reporter line
Non-human primate
Validation of Cas9-sgRNA efficiency and specificity
[ Readout of phenotype J
Figure 2 | Using Cas9 to generate genetically modified rodents and for in vivo genome editing, a,b | Comparison of the timelines of traditional gene targeting using classic homologous recombination (HR) in embryonic stem cells (ESCs; part a) and clustered regularly interspaced short palindromic repeat (CRISPR)-associated protein 9 (Cas9) gene targeting in one-cell embryos (part b). There are two main time- and cost-intensive phases of the HR approach. First, the design and cloning of the targeting vector, ESC transduction and selection, and generation of chimaeras. Second, the backcrossing of mice to a desired background and/or cross-breeding to generate multiple genetically modified animals. By contrast, cloning of short single-guide RNA (sgRNA) into a targeting vector, verification of sgRNA on-target efficiency (through the surveyor nuclease assay or sequencing), Cas9-sgRNA microinjection and founder identification are relatively easy and fast120. Because embryos can be obtained from any mouse strain and multiple genes can be targeted simultaneously, genetic backcrossing and cross-breeding are not required, c | Cas9 nucleases also enable precise in vivo genome editing of specific cell types in the mammalian brain on a relatively short timescale. Cas9 is cloned under the control of cell type-specific promoters, and sgRNA efficiency is validated in vitro before being packaged into viral vectors, such as adeno-associated viruses (AAVs). sgRNA can then be stereotactically delivered into the brains of mice that have endogenous Cas9 expression (Cas9 mice)91, or the sgRNA can be delivered together with Cas9 into wild-type mice41 or rats, aged animals, disease models or reporter lines. In vivo genome editing in the brain is not limited to rodents and can theoretically be applied to other mammalian systems, including non-human primates. CFAR glial fibrillary acidic protein; Neo, neomycin anitibiotic selection marker; SYN, human synapsin promoter.
Other techniques have been also used to deliver Cas9 and RNA guides to the brain, including in utero electroporation39 and polyethylenimine (PEI)-mediated transfection46. In rodents, electroporation and PEI-mediated transfection are easy to use, fast and efficient at delivering large plasmid DNA into a high
number of neurons. However, two drawbacks of these techniques are their low spatial accuracy of transgene expression and the necessity of prenatal intervention, which often results in low viability and targeting of mitotic neuronal precursors instead of postmitotic, differentiated neurons.
40 I JANUARY 2016 I VOLUME 17 www.nature.com/nrn
© 2016 Macmillan Publishers Limited. All rights reserved
REVIEWS
Liposomes
Lipid vesicles artificially formed by sonicating lipids in an aqueous solution. Liposomes can be packed with negatively charged molecules to deliver them into cells and are therefore promising vehicles for therapeutic applications.
Cre-foxP recombination
A site-specific recombination system derived from Escherichia coii bacteriophage PI. Two short DNA sequences [loxP sites) are engineered to flank the target DNA. Activation of the Cre recombinase enzyme catalyses recombination between the loxP sites, leading to excision of the intervening sequence.
Alternatively, Cas9 protein itself, rather than the DNA or RNA that encodes it, could be delivered, an approach that is particularly interesting for protein-based therapeutics. The anionic nature of sgRNA allows the integration of Cas9-sgRNA complexes into cationic liposomes, a commonly used DNA-, RNA- and protein-delivery tool. Liposome Cas9-sgRNA complexes have already been successfully used to achieve genome editing in the mouse inner ear92. Therefore, lipid-mediated delivery of Cas9 may also serve as powerful tool for genome editing in the brain in the future.
Cell type-specific genome editing. In the mammalian brain, there are probably several hundred neuronal subtypes, each with distinct morphological, biophysical, biochemical and computational functions. Thus, cell type-specific tools are required to dissect this heterogeneous tissue. Research has shown that the malfunction of specific cell types in different brain regions contributes to diverse symptoms usually connected with neuro-psychiatric symptoms, such as hallucinations, depression and repetitive motor behaviour93. This highlights the need to pinpoint causal relationships between cell types within the context of relevant neuronal networks, genetics and behavioural dysfunction, which will require precise genome editing in specific cellular subtypes. Site-specific Cre-loxP recombination elements that enable the control of the spatiotemporal expression of Cas9 have been introduced in fish94 and mouse embryos91, and
Box 11 Validating Cas nuclease efficiency and specificity in the brain
Validating clustered regularly interspaced short palindromic repeat (CRISPR)-associated protein (Cas) nuclease efficiency and specificity is particularly challenging in the mammalian brain because of its complex architecture and cellular diversity. To precisely validate nuclease efficiency and specificity, targeted cells first have to be identified and sorted out from the heterogeneous cell population in the brain. Recently, an easy and efficient method for this was developed in which fluorescent-activated cell sorting (FACS) is used to isolate fluorophore-tagged nuclei of targeted cells to purify and analyse genomic DNA and nuclear RNA with high resolution and sensitivity41.
Cas efficiency
Cas nuclease efficiency can be validated using enzymatic DNA cleavage assays (such as surveyor nuclease technology110) or DNA sequencing4146'91. DNA-sequencing analysis provides a complete picture of insertion and deletion (indel) frequency, types of frame-shift and in-frame mutations, length and exact sequence of indels, and information about mono- and bi-allelic modifications when applied to single cells41. In addition, RNA levels of the targeted gene can be determined using quantitative PCR (qPCR) or RNA-sequencing methods. Depending on the targeted exon (that is, whether it is an early or late exon), truncated transcripts might be expressed from the target gene and should also be considered when qPCR probes are designed. Ideally, effective protein knockdown should also be measured using histological, biochemical and/or functional (for example, electrophysiology and enzymatic activity assays) readouts.
Cas specificity
Similarly to zinc-finger nucleases and transcription activator-like effector nucleases, Cas proteins can cleave off-target sites in the genome. Many software tools predict potential off-target effects and help to choose optimal target sequences to reduce off-target activity (see Further information for a non-comprehensive list of online tools). On-target specificity can be further improved by using double-nicking111112 or truncated single-guide RNA approaches113. In addition, sensitive readout methods for identifying genome-wide Cas9 off-target activity have been developed that provide useful tools for evaluating specificity and safety of Cas9 in basic and clinical research (see Further information)57114115.
similar approaches could achieve precise gene editing in defined cell types in vivo. The vast number of established Cre-driver mouse lines95 and inducible Cas9 systems96"98 can, when combined with conditional gene-targeting strategies, provide enormous combinatorial power to decipher the logic of complex neuronal networks and their role in neurological disorders in vivo.
In vivo efficiency and specificity. In postmitotic neurons, Cas9 has been successfully used to introduce single39-414691 and multiple DSBs4146 resulting in NHEJ and efficient formation of indel mutations. For example, AAV delivery of Cas9 and sgRNA targeting Mecp2 in the adult mouse brain resulted in the local loss of more than 70% of MECP2, which was sufficient to recapitulate phe-notypes observed in classic Mecp2-mutant mouse models and patients with Rett syndrome41. In another study, Cas9-mediated deletion of common tumour-suppressor genes, such as patched homologue 1 (Ptchl), Trp53 (also known as Tp53), phosphatase and tensin homologue (Pten) and neurofibromin 1 (Nfl), in the cerebellum or forebrain efficiently induced the formation of medullo-blastoma or glioblastoma tumours, respectively46. Despite this success, the validation of Cas-mediated gene editing in the brain is still challenging, and sensitive methods are required for analysing Cas efficiency and specificity in targeted brain regions (BOX 1).
Although NHEJ in postmitotic neurons has been demonstrated to be active, it remains unclear how efficient HDR is in postmitotic cells. It is commonly believed that HDR predominantly occurs in the S and G2 phases of the cell cycle99 100, and HDR is therefore thought to be rare in non-dividing cells, such as neurons. Introduction or correction of precise genetic mutations via HDR in the brain would validate disease mutations in vivo and open the door to therapeutic applications of genome editing in brain disorders. Thus, future work should focus on identifying and activating signalling pathways required for triggering HDR in differentiated cells. However, it should also be noted that gene insertion has been achieved through NHEJ pathways, which may allow us to insert DNA into neurons and glia101.
In contrast to precise gene knockout and insertion, genome editing aimed at transcriptional regulation and epigenetic modulation may be less challenging in the brain, as these approaches are independent of DNA-repair pathways. Achieving epigenetic and transcriptional control in neurons can aid in the study of the molecular mechanisms of natural gene silencing in the nervous system and can help us to better understand neurological disorders associated with gene imprinting, such as Angelman syndrome102.
Gene editing in human iPSCs
Combinatorial approaches based on iPSC technology and genome editing offer another approach to model human neurological disorders in vitro. A key advantage of this approach is that genetic modifications can be studied in different human genetic backgrounds, because iPSCs retain all of the individual donor s genetic information. This is particularly important for complex
NATURE REVIEWS I NEUROSCIENCE VOLUME 17 I JANUARY 2016 I 41
© 2016 Macmillan Publishers Limited. All rights reserved
REVIEWS
-(CWAS)-
Patient with a neurological disorder
iPSC culture
Cas9,
HRDNA template^
sgRNAB
Defective neuron sgRNA candidate gene A
Healthy neuron sgRNA candidate gene B
t
Healthy individual
(Gene interrogation)
(Gene interrogation)
iPSC culture
Healthy neuron Defective neuron sgRNA candidate sgRNA candidate gene A geneB
Healthy neuron sgRNA candidate gene A
Neuron with mild phenotype sgRNA candidate gene B
Defective neuron sgRNA candidate gene A + B
Figure 3 | In vitro applications of Cas9 in human iPSCs. a | Evaluation of disease candidate genes from large-population genome-wide association studies (CWASs). Human primary cells, such as neurons, are not easily available and are difficult to expand in culture. By contrast, induced pluripotent stem cells (iPSCs) derived from somatic cells (such as fibroblasts) of healthy individuals or patients with neurological disorders can be differentiated into neurons and cultured in vitro8-12. Disease candidate genes can be examined in two ways. Site-specific homologous recombination (HR) of the candidate gene using clustered regularly interspaced short palindromic repeat (CRISPR)-associated protein (Cas) nucleases can be applied in disease-affected cells (left). If this rescues disease phenotypes (as for candidate gene B in the example shown), the validity of the candidate gene is confirmed. Alternatively, candidate genes can be mutated in healthy cells (right). Where this recapitulates disease pathogenesis in vitro (as in the case of candidate gene B), the validity of the candidate gene is confirmed, b | The contribution of specific genetic loci to multigenic disorders, such as Alzheimer or Parkinson diseases, can also be systematically evaluated using Cas-mediated single and multiplex genome editing. This may enable dissection of possible synergistic effects (as shown for candidate genes A and B) and screening for functional correlations between disease phenotypes and distinct gene mutations. sgRNA, single-guide RNA.
neurological disorders, because genetic variants associated with such diseases act in concert with many other alleles. Another advantage is that the genetically modified cells can be differentiated into almost any cell type, including those that are not easily accessible in patients, such as neurons and glia.
iPSC-based disease models have been generated for several neurological disorders, including Parkinson1011, Alzheimer9 and Huntington8 diseases, and they have been proven to closely mimic cellular and molecular features of human diseases. Genome-editing tools applied to these models can be used to examine the genetic link between risk variants and cellular pathways involved in multigenic neurological disorders in a high-throughput manner (FIG. 3). Furthermore, specific signalling pathways involved in the pathogenesis of the disease can be precisely dissected to gain insight into the molecular mechanisms of the disease and to identify new drug targets10. Gene editing may be performed either in iPSCs or induced later in differentiated cells96'98, allowing for
the investigation of phenotypes that arise during cell differentiation, which may be relevant when studying neurodevelopmental aspects of a disease such as Rett syndrome103"105. In addition, inducing or rescuing a phenotype in differentiated cells will be useful for validating potential therapeutic applications.
Future perspectives
Genome-editing technologies allow for the introduction of genetic modifications into almost any cell type and organism. For example, Cas9 has already been used to alter genes in species such as killifish106 and salamander107, which are commonly used to study ageing and tissue regeneration, respectively. It may also open up the possibility of developing models in other species of interest to neuroscience research, such as social insects or songbirds58, which have thus far been intractable to genetic modification. In addition to the generation of new model systems, including iPSC-derived in vitro models, genome editing in combination with single-cell
42 I JANUARY 2016 I VOLUME 17 www.nature.com/nrn
© 2016 Macmillan Publishers Limited. All rights reserved
REVIEWS
transcriptomics108 provides a route to understanding cell type-specific gene function within a heterogeneous tissue, allowing for precise dissection of genetic networks in the brain. Furthermore, together with genome-wide association studies, in vivo genome editing holds potential for personalized therapeutic applications for brain disorders109. However, to realize these advances, several open challenges have to be addressed. First, existing methods for delivering Cas proteins and RNA guides to the brain must be optimized and new methods must
be developed to achieve sufficient levels of specificity and efficiency. Second, new methods for stimulating efficient gene insertion and correction in postmitotic cells have to be established. Third, safety and ethical concerns have to be carefully addressed. Nevertheless, we believe that novel genome-editing technologies based on CRISPR-Cas systems, together with powerful readout methods, will help us better understand the logic of neuronal circuits and unravel some of the mysteries of complex neurological disorders in the near future.
1. Lewis, E. B. & Bacher, F. Methods for feeding ethyl methane sulfonate (EMS) to Drosophila males. Drosoph. Inf. Serv. 43, 193-194 (1968).
2. Fire, A. et ai. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391, 806-811 (1 998).
3. St Johnston, D. The art and design of genetic screens: Drosophila melanogaster. Nat. Rev. Genet. 3, 176-188 (2002).
4. Jorgensen, E. M. & Mango, S. E. The art and design of genetic screens: Caenorhabditis elegans. Nat. Rev. Genet. 3, 356-369 (2002).
5. Patton, E. E. & Zon, L. I. The art and design of genetic screens: zebrafish. Nat. Rev. Genet. 2, 956-966 (2001).
6. Thomas, K. R. SiCapecchi, M. R. Site-directed mutagenesis by gene targeting in mouse embryo-derived stem cells. Ce//51, 503-512 (1987).
7. Oddo, S. etal. Triple-transgenic model of Alzheimer's disease with plaques and tangles: intracellular Aß and synaptic dysfunction. Neuron 39, 409-421 (2003).
8. Jeon, I. et al. Neuronal properties, in vivo effects, and pathology of a Huntington's disease patient-derived induced pluripotent stem cells. Stem Cells 30, 2054-2062 (2012).
9. Yagi, T. etal. Modeling familial Alzheimer's disease with induced pluripotent stem cells. Hum. Mol. Genet. 20,4530-4539 (2011).
10. Ryan, S. D. et al. Isogenic human iPSC Parkinson's model shows nitrosative stress-induced dysfunction in MEF2-PGC1 a transcription. Cell 1 55, 1 351 -1 364 (2013).
11. Soldner, F. et al. Generation of isogenic pluripotent stem cells differing exclusively at two early onset Parkinson point mutations. Cell 146, 318-331 (2011). References 10 and 11 combine ZFN-mediated genome-editing and human-stem-cell technologies for studying neurological disorders in vitro.
1 2.  Pak, C. et al. Human neuropsychiatric disease modeling using conditional deletion reveals synaptic transmission defects caused by heterozygous mutations in NRXN1. Cell Stem Cell 1 7, 31 6-328 (201 5). 1 3.  Kim, Y. G., Cha, J. & Chandrasegaran, S. Hybrid restriction enzymes: zinc finger fusions to Fok I cleavage domain. Proc. Natl Acad. Sei. USA 93, 1156-1160 (1996). 1 4.  Bibikova, M. et al. Stimulation of homologous
recombination through targeted cleavage by chimeric nucleases. Mol. Cell. Biol. 21, 289-297 (2001). This early study shows the use of ZFNs in Xenopus iaevis for stimulating HR. 1 5.  Urnov, F. D. et al. Highly efficient endogenous human gene correction using designed zinc-finger nucleases. Nature^S, 646-651 (2005). Boch, J. et ai. Breaking the code of DNA binding specificity of TAL-type III effectors. Science 326, 1509-1512(2009).
Moscou, M. J. & Bogdanove, A. J. A simple cipher governs DNA recognition by TAL effectors. Science 326, 1501 (2009). 1 8.  Christian, M. et ai. Targeting DNA double-strand breaks with TAL effector nucleases. Genetics 1 86, 757-761 (2010). 1 9.  Miller, J. C. etal. A TALE nuclease architecture for efficient genome editing. Nat. Biotechnol. 29, 143-148 (2011).
This paper uses an improved TALEN architecture to introduce gene knockouts in human cells.
20.  Bolotin, A., Quinquis, B., Sorokin, A. & Ehrlich, S. D. Clustered regularly interspaced short palindrome repeats (CRISPRs) have spacers of extrachromosomal origin. Microbiology 1 51, 2551-2561 (2005).
16.
17
Barrangou, R. et al. CRISPR provides acquired resistance against viruses in prokaryotes. Science 315, 1709-1712 (2007). This work provides the first experimental demonstration of the adaptive immune function of the CRISPR-Cas9 system in bacteria. Deltcheva, E. et ai. CRISPR RNA maturation by trans-encoded small RNA and host factor RNase III. Nature 471,602-607 (2011).
Garneau, J. E. et ai. The CRISPR/Cas bacterial immune
system cleaves bacteriophage and plasmid DNA.
Nature^S, 67-71 (2010).
This paper demonstrates that Cas9 facilitates
RNA-guided DNA cleavage in bacteria.
Cong, L. et al. Multiplex genome engineering using
CRISPR/Cas systems. Science 339, 819-823
(2013) .
Mali, P. et al. RNA-guided human genome engineering via Cas9. Science 339, 823-826 (2013). References 24 and 25 describe the successful harnessing of the CRISPR-Cas9 system for editing the mammalian genome in cell lines. Makarova, K. S. & Koonin, E. V. Annotation and classification of CRISPR-Cas systems. Methods Mol. Biol. 1311,47-75(2015). Zetsche, B. et al. Cpf 1 is a single RNA-guided endonuclease of a class 2 CRISPR-Cas system. Cell 163, 759-771 (2015).
Beerli, R. R., Segal, D. J., Dreier, B. & Barbas, C. F. Toward controlling gene expression at will: specific regulation of the erbB-2IHER-2 promoter by using polydactyl zinc finger proteins constructed from modular building blocks. Proc. Natl Acad. Sei. USA 95, 14628-14633 (1998). Zhang, F. et al. Efficient construction of sequence-specific TAL effectors for modulating mammalian transcription. Nat. Biotechnol. 29, 149-153 (2011). Gilbert, L. A. et ai. CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes. Cell 154, 442-451 (2013).
Qi, L. S. et al. Repurposing CRISPR as an RNA-guided platform for sequence-specific control of gene expression. Ceil 152, 1173-1183 (2013). Konermann, S. et al. Optical control of mammalian endogenous transcription and epigenetic states. Nature 500, 472-476 (2013). Kearns, N. A. et al. Functional annotation of native enhancers with a Cas9-histone demethylase fusion. Nat. Methods 12,401-403 (2015). Konermann, S. et al. Genome-scale transcriptional activation by an engineered CRISPR-Cas9 complex. Nature 51 7, 583-588 (201 5). Hilton, I. B. et ai. Epigenome editing by a CRISPR-Cas9-based acetyltransferase activates genes from promoters and enhancers. Nat. Biotechnol. 33, 510-517 (2015).
Chen, B. et al. Dynamic imaging of genomic loci in living human cells by an optimized CRISPR/Cas system. Ceil 1 55, 1479-1491 (2013). Jinek, M. et ai. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337, 816-821 (2012). Hsu, P. D. et ai. DNA targeting specificity of RNA-guided Cas9 nucleases. Nat. Biotechnol. 31, 827-832 (2013).
Straub, C, Granger, A. J., Saulnier, J. L. & Sabatini, B. L. CRISPR/Cas9-mediated gene knockdown in post-mitotic neurons. PLoS ONE 9, el 05584
(2014) .
Incontro, S., Asensio, C. S., Edwards, R. H. & Nicoll, R. A. Efficient, complete deletion of synaptic proteins using CRISPR. Neuron 83, 1051 -1057 (2014).
This paper reports the delivery of Cas9 and guide RNAs in organotypic brain-slice cultures and the disruption NMDA receptor and AM PA receptor su bun its.
Swiech, L. et ai. In vivo interrogation of gene function in the mammalian brain using CRISPR-Cas9. Nat. Biotechnol. 33, 102-106(2015). This paper demonstrates the delivery of Cas9 and guide RNAs into the mouse brain using AAV, and single and multiplex gene editing in vivo. It also shows a purification method of genetically tagged Cas9-targeted cell nuclei for DNA and RNA sequencing.
Shen, Z. et ai. Conditional knockouts generated by engineered CRISPR-Cas9 endonuclease reveal the roles of coronin in C. elegans neural development. Dev. Cell 30, 625-636(2014). This paper describes a conditional-knockout strategy using Cas9 in Caenorhabditis elegans for studying gene function in neural development. Shah, A. N., Davey, C. F., Whitebirch, A. C, Miller, A. C. & Moens, C. B. Rapid reverse genetic screening using CRISPR in zebrafish. Nat. Methods 12, 535-540 (2015).
This paper represents a useful application of Cas9 for studying neurodevelopmental processes on a large scale in zebrafish.
Auer, T. O., Duroure, K., De Cian, A., Concordet, J. P. & Del Bene, F. Highly efficient CRISPR/Cas9-mediated knock-in in zebrafish by homology-independent DNA repair. Genome Res. 24, 142-153 (2014). Jao, L. E., Wente, S. R. & Chen, W. Efficient multiplex biallelic zebrafish genome editing using a CRISPR nuclease system. Proc. Natl Acad. Sci. USA 110, 13904-13909 (2013). Zuckermann, M. et al. Somatic CRISPR/ Cas9-mediated tumour suppressor disruption enables versatile brain tumour modelling. Nat. Commun. 6, 7391 (2015).
This paper describes methods for delivering Cas9 and guide RNA into the brains of newborn mice and embryos. By targeting multiple tumour-suppressor genes, the development of the medulla and glioblastoma was induced. Plessis, A., Perrin, A., Haber, J. E. & Dujon, B. Site-specific recombination determined by l-Scel, a mitochondrial group I intron-encoded endonuclease expressed in the yeast nucleus. Genetics 1 30, 451-460 (1992).
Rudin, N., Sugarman, E. & Haber, J. E. Genetic and physical analysis of double-strand break repair and recombination in Saccharomyces cerevisiae. Genetics 122, 519-534 (1989). Fishman-Lobell, J. & Haber, J. E. Removal of nonhomologous DNA ends in double-strand break recombination: the role of the yeast ultraviolet repair gene RAD 1. Science 258, 480-484 (1992). Fishman-Lobell, J., Rudin, N. & Haber, J. E. Two alternative pathways of double-strand break repair that are kinetically separable and independently modulated. Moi. Ceil. Biol. 12, 1292-1303 (1992).
Liang, F, Han, M., Romanienko, P. J. & Jasin, M. Homology-directed repair is a major double-strand break repair pathway in mammalian cells. Proc. Natl Acad. Sci. USA 95, 5172-5177 (1998). Johnson, R. D., Liu, N. & Jasin, M. Mammalian XRCC2 promotes the repair of DNA double-strand breaks by homologous recombination. Nature 401, 397-399 (1999).
Bibikova, M., Beumer, K., Trautman, J. K. & Carroll, D. Enhancing gene targeting with designed zinc finger nucleases. Science 300, 764 (2003).
NATURE REVIEWS | NEUROSCIENCE VOLUME 17 | JANUARY 2016 | 43
© 2016 Macmillan Publishers Limited. All rights reserved
REVIEWS
54. Rouet, P., Smih, F. & Jasin, M. Introduction of double-strand breaks into the genome of mouse cells by expression of a rare-cutting endonuclease. Mol. Ceil. Biol. 14,8096-8106(1994).
55. Rouet, P., Smih, F. & Jasin, M. Expression of a site-specific endonuclease stimulates homologous recombination in mammalian cells. Proc. Natl Acad. Sci. USA 91, 6064-6068(1994).
56. Gasiunas, G., Barrangou, R., Horvath, P. & Siksnys, V. Cas9-crRNA ribonucleoprotein complex mediates specific DNA cleavage for adaptive immunity in bacteria. Proc. Natl Acad. Sci. USA 109, E2579-E2586 (2012).
This paper, along with reference 37, characterizes Cas9-mediated DNA cleavage in vitro.
57. Ran, F. A. et al. In vivo genome editing using Staphylococcus aureus Cas9. Nature 520, 186-191 (2015).
58. Brenowitz, E. A. & Zakon, H. H. Emerging from the bottleneck: benefits of the comparative approach to modern neuroscience. Trends Neurosci. 38, 273-278 (2015).
59. Elbashir, S. M. et al. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411, 494-498 (2001).
60. Hommel, J. D., Sears, R. M., Georgescu, D., Simmons, D. L. & DiLeone, R. J. Local gene knockdown in the brain using viral-mediated RNA interference. Nat. Med. 9, 1539-1544 (2003).
61. Wittenburg, N. et al. Presenilin is required for proper morphology and function of neurons in C. elegans. Nature 406, 306-309 (2000).
62. Geling, A., Steiner, H., Willem, M., Bally-Cuif, L. & Haass, C. Ay-secretase inhibitor blocks Notch signaling in vivo and causes a severe neurogenic phenotype in zebrafish. EMBO Rep. 3, 688-694 (2002).
63. Clark, I. E. et al. Drosophila pink 1 is required for mitochondrial function and interacts genetically with parkin. Nature 441, 1162-1166 (2006).
64. Cooley, L, Kelley, R. & Spradling, A. Insertional mutagenesis of the Drosophila genome with single P elements. Science 239, 1121-11 28 (1988).
65. Gaiano, N. et al. Insertional mutagenesis and rapid cloning of essential genes in zebrafish. Nature 383, 829-832 (1996).
66. Bessereau, J. L. etal. Mobilization of a Drosophila transposon in the Caenorhabditis elegans germ line. Nature W5, 70-74 (2001).
67. Beumer, K. J. etal. Efficient gene targeting in Drosophila by direct embryo injection with zinc-finger nucleases. Proc. Natl Acad. Sci. USA 105, 19821-19826 (2008).
68. Morton, J., Davis, M. W., Jorgensen, E. M. & Carroll, D. Induction and repair of zinc-finger nuclease-targeted double-strand breaks in Caenorhabditis elegans somatic cells. Proc. Natl Acad. Sci. USA 103, 16370-16375 (2006).
69. Doyon, Y. etal. Heritable targeted gene disruption in zebrafish using designed zinc-finger nucleases. Nat. Biotechnol. 26, 702-708 (2008).
70. Bedell, V. M. etal. In vivo genome editing using a high-efficiency TALEN system. Nature 491, 114-118 (2012).
71. Wood, A. J. etal. Targeted genome editing across species using ZFNs and TALENs. Science 333, 307 (2011).
72. Sander, J. D. etal. Targeted gene disruption in somatic zebrafish cells using engineered TALENs. Nat. Biotechnol. 29, 697-698 (2011).
73. Katsuyama, T. et al. An efficient strategy for TALEN-mediated genome engineering in Drosophila. Nucleic Acids Res. 41, el 63 (2013).
74. Sander, J. D. & Joung, J. K. CRISPR-Cas systems for editing, regulating and targeting genomes. Nat. Biotechnol. 32, 347-355 (2014).
75. Carbery, I. D. etal. Targeted genome modification in mice using zinc-finger nucleases. Genetics 1 86, 451-459 (2010).
76. Geurts, A. M.etai. Knockout rats via embryo microinjection of zinc-finger nucleases. Science 325, 433 (2009).
77. Liu, H. et al. TALEN-mediated gene mutagenesis in rhesus and cynomolgus monkeys. Cell Stem Cell 1 4, 323-328 (2014).
78. Niu, Y. et al. Generation of gene-modified cynomolgus monkey via Cas9/RNA-mediated gene targeting in one-cell embryos. Ceil 156, 836-843 (2014). References 77 and 78 describe the successful generation of genetically modified non-human primates using genome-editing technologies in early embryos.
79. Chan, A. W., Chong, K. Y., Martinovich, C, Simerly, C. & Schatten, G. Transgenic monkeys produced by retroviral gene transfer into mature oocytes. Science 291,309-312(2001).
80. Yang, S. H. et al. Towards a transgenic model of Huntington's disease in a non-human primate. Nature 453, 921-924 (2008).
81. Sasaki, E. et al. Generation of transgenic non-human primates with germline transmission. Nature 459, 523-527 (2009).
82. Belmonte, J. C. et al. Brains, genes, and primates. Neuron 86, 617-631 (2015).
83. Xia, H., Mao, Q., Paulson, H. L. & Davidson, B. L. siRNA-mediated gene silencing in vitro and in vivo. Nat. Biotechnol. 20, 1006-1010 (2002).
84. Kordasiewicz, H. B. et al. Sustained therapeutic reversal of Huntington's disease by transient repression of huntingtin synthesis. Neuron 74, 1031-1044 (2012).
85. Smith, R. A. et al. Antisense oligonucleotide therapy for neurodegenerative disease. J. Clin. Invest. 116, 2290-2296 (2006).
86. Garriga-Canut, M. etal. Synthetic zinc finger repressors reduce mutant huntingtin expression in the brain of R6/2 mice. Proc. Natl Acad. Sci. USA 109, E3136-E3145 (2012).
87. Sweatt, J. D. The emerging field of neuroepigenetics. Neuron 80, 624-632 (2013).
88. Murlidharan, G., Samulski, R. J. &Asokan, A. Biology of adeno-associated viral vectors in the central nervous system. Front. Mol. Neurosci. 7, 76 (2014).
89. Burger, C, Nash, K. & Mandel, R. J. Recombinant adeno-associated viral vectors in the nervous system. Hum. Gene Ther. 16, 781-791 (2005).
90. Taymans, J. M. et al. Comparative analysis of adeno-associated viral vector serotypes 1, 2, 5, 7, and 8 in mouse brain. Hum. Gene Ther. 18, 195-206 (2007).
91. Platt, R. J. etal. CRISPR-Cas9 knockin mice for genome editing and cancer modeling. Cell 1 59, 440-455 (2014).
This paper describes how the CRISPR-Cas9 knock-in mouse can be used for cell type-specific gene editing in the brain.
92. Zuris, J. A. et al. Cationic lipid-mediated delivery of proteins enables efficient protein-based genome editing in vitro and in vivo. Nat. Biotechnol. 33, 73-80 (2015).
93. Akil, H. et al. Medicine. The future of psychiatric research: genomes and neural circuits. Science 327, 1580-1581 (2010).
94. Yin, L. et al. Multiplex conditional mutagenesis using transgenic expression of Cas9 and sgRNAs. Genetics (2015).
95. Harris, J. A. et al. Anatomical characterization of Cre driver mice for neural circuit mapping and manipulation. Front. Neural Circuits 8, 76 (2014).
96. Polstein, L. R. & Gersbach, C. A. A light-inducible CRISPR-Cas9 system for control of endogenous gene activation. Nat. Chem. Biol. 11,198-200 (2015).
97. Dow, L. E. et al. Inducible in vivo genome editing with CRISPR-Cas9. Nat. Biotechnol. 33, 390-394 (2015).
98. Zetsche, B., Volz, S. E. & Zhang, F. A split-Cas9 architecture for inducible genome editing and transcription modulation. Nat. Biotechnol. 33, 139-142 (2015).
99. Pardo, B., Gomez-Gonzalez, B. & Aguilera, A. DNA repair in mammalian cells: DNA double-strand break repair: how to fix a broken relationship. Cell. Mol. Life Sci. 66, 1039-1056 (2009).
100. van Gent, D. C. & van der Burg, M. Non-homologous end-joining, a sticky affair. Oncogene 26, 7731-7740 (2007).
101. Maresca, M., Lin, V. G., Guo, N. & Yang, Y. Obligate ligation-gated recombination (ObLiGaRe): custom-designed nuclease-mediated targeted integration through nonhomologous end joining. Genome Res. 23,539-546 (2013).
102. Peters, J. The role of genomic imprinting in biology and disease: an expanding view. Nat. Rev. Genet. 1 5, 517-530 (2014).
103. Kim, K. Y., Hysolli, E. & Park, I. H. Neuronal maturation defect in induced pluripotent stem cells from patients with Rett syndrome. Proc. Natl Acad. Sci. USA 108, 14169-14174 (2011).
104. Cheung, A. Y. et al. Isolation of M£CP2-null Rett Syndrome patient hiPS cells and isogenic controls through X-chromosome inactivation. Hum. Mol. Genet. 20, 2103-2115 (2011).
105. Marchetto, M. C. et al. A model for neural development and treatment of Rett syndrome using human induced pluripotent stem cells. Cell 143, 527-539 (2010).
106. Harel, I. et al. A platform for rapid exploration of aging and diseases in a naturally short-lived vertebrate. Cell 160, 1013-1026 (2015).
107. Flowers, G. P., Timberlake, A. T, McLean, K. C, Monaghan, J. R. & Crews, C. M. Highly efficient targeted mutagenesis in axolotl using Cas9 RNA-guided nuclease. Development 141, 21 65-21 71 (2014).
108. Zeisel.A. etal. Brain structure. Cell types in the mouse cortex and hippocampus revealed by single-cell RNA-seq. Science 347, 1138-1142 (2015).
109. Cox, D. B., Platt, R. J. & Zhang, F. Therapeutic genome editing: prospects and challenges. Nat. Med. 21, 121-131 (2015).
110. Qiu, P. etal. Mutation detection using Surveyor nuclease. Biotechniques 36, 702-707 (2004).
111. Ran, F. A. et al. Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity. Cell 154, 1380-1389 (2013).
11 2. Mali, P. etal. CAS9 transcriptional activators for target specificity screening and paired nickasesfor cooperative genome engineering. Nat. Biotechnol. 31, 833-838 (2013).
11 3. Fu, Y., Sander, J. D., Reyon, D., Cascio, V. M. & Joung, J. K. Improving CRISPR-Cas nuclease specificity using truncated guide RNAs. Nat. Biotechnol. 32, 279-284 (2014).
11 4. Tsai, S. Q. et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nat. Biotechnol. 33, 187-197 (2015).
11 5. Kim, D. et al. Digenome-seq: genome-wide profiling of CRISPR-Cas9 off-target effects in human cells. Nat. Methods 12, 237-243 (2015).
11 6. Blasco, R. B. et al. Simple and rapid in vivo
generation of chromosomal rearrangements using CRISPR/Cas9 technology. Cell Rep. 9, 1219-1227 (2014).
11 7. Maddalo, D. et al. In vivo engineering of oncogenic
chromosomal rearrangements with the CRISPR/Cas9
system. Nature 516, 423-427 (2014). 11 8. Xiao, A. et al. Chromosomal deletions and inversions
mediated by TALENs and CRISPR/Cas in zebrafish.
Nucleic Acids Res. 41,el41 (2013). 11 9. Essletzbichler, P. etal. Megabase-scale deletion
using CRISPR/Cas9 to generate a fully haploid
human cell line. Genome Res. 24, 2059-2065
(2014).
1 20. Wang, H. et al. One-step generation of mice
carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering. Cell 1 53, 910-918 (2013).
This paper describes the generation of mice that were genetically modified using Cas9.
Acknowledgements
The authors thank R. Macrae for manuscript review and all members of the Zhang laboratory for discussions. M.H. is supported by the Human Frontiers Scientific Program. F.Z. is supported by the US National Institute of Mental Health (NIMH; grants 5DP1-MH 100706 and 1 R01-MH 110049), the US National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK; grant 5R01 DK097768-03), the Poitras Center for Affective Disorders Research, the Vallee, Simons, Paul G. Allen Family, and New York Stem Cell Foundations, D. R. Cheng, T. Harriman and B. Metcalfe. F.Z. is a New York Stem Cell Foundation Robertson Investigator.
Competing interests statement
The authors declare competing interests: see Web version for details.
FURTHER INFORMATION
Benchling: https://benchling.com/crispr
CasOT (CRISPR/Cas system (Cas9/gRNA) Off-Targeter):
http://eendb.zfgenetics.org/casot/
CHOPCHOP: https://chopchop.rc.fas.harvard.edu/
COSMID: http://omictools.com/cosmid-s9S90.html
CRISPR Design: http://crispr.mit.edu/
DESKGEN: https://www.deskgen.com/landing/#/
E-CRISP: http://www.e-crisp.org/E-CRISP/
sgRNA designer: http://www.broadinstitute.org/rnai/public/
analysis-tools/sgrna-design
ZiFiT: http://zifit.partners.org/ZiFiT/
ALL LINKS ARE ACTIVE IN THE ONLINE PDF
I JANUARY 2016 | VOLUME 17 www.nature.com/nrn
© 2016 Macmillan Publishers Limited. All rights reserved