Neuron
NeuroResource
Imaging Neural Activity
Using Thy1-GCaMP Transgenic Mice
Qian Chen,1 Joseph Cichon,3,10 Wenting Wang,1,10 Li Qiu,4 Seok-Jin R. Lee,4 Nolan R. Campbell,1 Nicholas DeStefino,6
Michael J. Goard,2 Zhanyan Fu,1,5 Ryohei Yasuda,4 Loren L. Looger,7 Benjamin R. Arenkiel,8 Wen-Biao Gan,3
and Guoping Feng1,9,*
1McGovern Institute for Brain Research, Department of Brain and Cognitive Sciences
2Picower Institute for Learning and Memory, Department of Brain and Cognitive Sciences
Massachusetts Institute of Technology, Cambridge, MA 02139, USA
3Molecular Neurobiology Program, Skirball Institute, Department of Physiology and Neuroscience, New York University School of Medicine,
New York, NY 10016, USA
4Department of Neurobiology
5Department of Psychiatry and Behavioral Science
Duke University Medical Center, Durham, NC 27710, USA
6MD-PhD Program, Harvard Medical School, Boston, MA 02115, USA
7Howard Hughes Medical Institute, Janelia Farm Research Campus, Ashburn, VA 20147, USA
8Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA
9Stanley Center for Psychiatric Research, Broad Institute, Cambridge, MA 02142, USA
10These authors contributed equally to this work
*Correspondence: fengg@mit.edu
http://dx.doi.org/10.1016/j.neuron.2012.07.011
SUMMARY
The ability to chronically monitor neuronal activity in
the living brain is essential for understanding the
organization and function of the nervous system.
The genetically encoded green fluorescent proteinbased
calcium sensor GCaMP provides a powerful
tool for detecting calcium transients in neuronal
somata, processes, and synapses that are triggered
by neuronal activities. Here we report the generation
and characterization of transgenic mice that express
improved GCaMPs in various neuronal subpopulations
under the control of the Thy1 promoter.
In vitro and in vivo studies show that calcium transients
induced by spontaneous and stimulus-evoked
neuronal activities can be readily detected at the
level of individual cells and synapses in acute brain
slices, as well as chronically in awake, behaving
animals. These GCaMP transgenic mice allow investigation
of activity patterns in defined neuronal populations
in the living brain and will greatly facilitate
dissecting complex structural and functional relationships
of neural networks.
INTRODUCTION
Monitoring neuronal activity is critical for our understanding
of both normal brain function and pathological mechanisms
of brain disorders. Because neuronal activity is tightly coupled
to intracellular calcium dynamics, calcium imaging has proven
invaluable for probing the activities of neuronal somata, processes,
and synapses both in vitro and in vivo (Andermann
et al., 2011; Chen et al., 2011; Kerr and Denk, 2008; Yasuda
et al., 2004). Compared to multielectrode recording approaches,
calcium imaging has the advantages of detecting activity in
large or disperse populations of neurons simultaneously over
extended periods of time with little or no mechanical disturbance
to brain tissues.
Synthetic calcium dyes have been widely used to monitor
intracellular calcium dynamics in cultured neurons, brain slices,
as well as in the intact brain (Chen et al., 2011; Dombeck et al.,
2007; Kerr and Denk, 2008; Marshel et al., 2011; Rothschild
et al., 2010; Yasuda et al., 2004). However, loading calcium
dyes into specific neuronal populations is technically challenging.
It is difficult, if not impossible, to image activities of
the same neuronal populations repeatedly over extended
periods of time. Genetically encoded calcium indicators (GECIs)
overcome these difficulties, permitting chronic imaging of
calcium dynamics within specific cell types. GECIs are composed
solely of translated amino acids and do not require the
addition of synthetic compounds or cofactors. They can be
targeted to specific cell types or subcellular compartments
when used in combination with cell type-specific promoters or
cellular targeting sequences. In addition, GECIs can be delivered
and expressed in brain tissues via viral vectors, in utero electroporation,
or through transgenic techniques (Hasan et al., 2004;
Mao et al., 2008; Wallace et al., 2008; Yamada et al., 2011).
Importantly, recently developed GECIs are capable of detecting
calcium dynamics at the sensitivity level close to that of synthetic
calcium dyes (Hendel et al., 2008; Pologruto et al., 2004).
At least one class of green fluorescent protein (GFP)-based
GECIs, the GCaMP family, has been effective for detecting
calcium dynamics induced by neuronal activity in multiple model
organisms (Muto et al., 2011; Reiff et al., 2005; Tian et al., 2009;
Warp et al., 2012). Recently, a new generation of GCaMPs (e.g.,
GCaMP3) has been successfully used to monitor neuronal
Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc. 297
activity in rodents using viral approaches (Borghuis et al., 2011;
Dombeck et al., 2010; Mittmann et al., 2011; Osakada et al.,
2011; Tian et al., 2009). Here we report the generation and
characterization of new transgenic mouse lines that express
the improved GCaMP2.2c and GCaMP3 indicators (Tian et al.,
2009) in subsets of excitatory neurons in the mouse brain using
the Thy1 promoter. We demonstrate long-term, stable expression
of GCaMPs in subpopulations of neurons with no apparent
toxicity. Both GCaMP2.2c and GCaMP3 show strong and
sensitive changes in fluorescence upon neuronal stimulation.
We further demonstrate the broad utility of Thy1-GCaMP2.2c
and Thy1-GCaMP3 transgenic mice in reporting neuronal activity
in vitro and in vivo.
RESULTS
Generation of Thy1-GCaMP2.2c and Thy1-GCaMP3
Transgenic Mice
To generate GCaMP transgenic mice, we utilized the previously
described GCaMP3 and a further modified GCaMP2.2b (Tian
et al., 2009). Previous studies suggested that the N-terminal
arginine located immediately after the initiator methionine of
GCaMP2.0 destabilizes the protein, and changing the serine at
118 to cysteine could improve brightness and sensor response
(Tian et al., 2009). Thus, we changed the second arginine in
GCaMP2.0 to valine to increase its stability according to the
N-terminal rule of protein degradation (Varshavsky, 2011) and
changed the serine at 118 to cysteine as in GCaMP2.2b to create
GCaMP2.2c. The domain structure and specific mutations of
GCaMP2.2c and GCaMP3 are summarized in Figure S1A,
available online.
Two important properties to consider when evaluating
GECIs are basal levels of fluorescence and stimulation-induced
changes in fluorescence (DF/F). To assess these properties for
GCaMP2.2c and GCaMP3, we coexpressed GCaMPs and the
red fluorescence protein tdTomato in the same construct using
the 2A peptide (P2A) sequence (Szymczak et al., 2004) in
HEK293 cells. To normalize for transfection efficiency, we used
the fluorescence intensity ratio of GCaMPs/tdTomato. We found
that GCaMP3 showed significantly higher basal fluorescence
compared to GCaMP2.2c and GCaMP2.0, whereas there was
no significant difference between the basal fluorescence of
GCaMP2.0 and GCaMP2.2c (Figures S1B and S1C). In addition,
we found that fluorescence intensity changes elicited by 100 mM
ATP are $1.9-fold (1.9 ± 0.1, n = 56) and $3.2-fold (3.2 ± 0.3, n =
61) higher in cells expressing GCaMP2.2c and GCaMP3 than in
cells expressing GCaMP2.0, respectively (Figure S1D).
The studies above indicate that GCaMP2.2c has a low basal
fluorescence with a modest fluorescence change in response
to stimulation, whereas GCaMP3 shows higher basal fluorescence
and a more robust change in fluorescence after stimulation.
Because GCaMP (and any GECI) binds calcium, there is
a risk of neuronal toxicity associated with calcium binding and
expression level. To increase the chances of finding lines
with both strong signal and low toxicity, we generated both
GCaMP2.2c and GCaMP3 transgenic lines.
To generate GCaMP transgenic mice, we used the wellcharacterized
Thy1 promoter to express GCaMPs in neurons.
We generated eight founder lines of GCaMP2.2c and six founder
lines of GCaMP3. Our previous studies have shown that the Thy1
promoter predominantly drives transgene expression in projection
neurons in the CNS. Due to strong transgenic position-effect
variegation, a Thy1-driven transgene is often stochastically and
differentially expressed in subsets of neurons in different transgenic
lines (Feng et al., 2000; Young et al., 2008). Consistent
with these findings, we found that all founder lines differed in
levels and patterns of expression. For further characterization,
we focused on Thy1-GCaMP2.2c line 8 and Thy1-GCaMP3 line
6 because these lines had the highest levels of transgene
expression. Both lines of mice are born at the expected
Mendelian rate and are healthy with no apparent histological or
behavioral abnormality. GCaMP2.2c and GCaMP3 expression
in these lines was widespread in the CNS including cortex,
hippocampus, thalamus, cerebellum, superior colliculus, amygdala,
brain stem, retina, and spinal cord (Figures 1A, 1B, and 2;
Figure S2). However, some notable differences in expression
Figure 1. Expression Patterns of Thy1-GCaMP3 Transgenic Mice
(A) A sagittal section of brain from a Thy1-GCaMP3 mouse. GCaMP3 is highly
expressed in the cortex, hippocampus, thalamus, superior colliculus, and
mossy fiber in the cerebellum and various nuclei in the brain stem.
(B) Confocal images showing the expression in motor cortex (M1) of Thy1GCaMP3
mouse brains. The insets (b1 and b2) show a higher-magnification
view of layer II/III neurons in the cortex (boxed area). GCaMP3 is expressed in
most layer II/III neurons. In (b1) and (b2), the same section was costained with
an antibody against the neuronal marker NeuN (red) showing the localization of
GCaMP in the cytoplasm of neurons. OB, olfactory bulb; CTX, cortex; HP,
hippocampus; STR, striatum; TH, thalamus; SC, superior colliculus; CB,
cerebellum; BS, brain stem. I, II, III, IV, V, and VI refer to cortical layers. Scale
bars represent 1 mm in (A) and 50 mm in (B) and (Bb2).
Neuron
Thy1-GCaMP Transgenic Mice
298 Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc.
between the two lines were observed. For example, although
both lines have expression in layer V neurons of the cortex,
expression in layer II/III is more widespread in the Thy1-GCaMP3
line (Figures 1B, 1Bb1, and 1Bb2) compared to the Thy1GCaMP2.2c
line (Figures S2B, S2Bb1, and S2Bb2). In addition,
Thy1-GCaMP3 mice, but not Thy1-GCaMP2.2c mice, showed
high expression in olfactory bulb (Figures 1A and 2). At the single
cell level, both GCaMP2.2c and GCaMP3 were homogeneously
distributed in the cytoplasm without nuclear localization (Figures
1B, 1Bb1, and 1Bb2; Figures S2B, S2Bb1, and S2Bb2).
We further examined the effect of long-term GCaMP expression
in both GCaMP transgenic lines. We did not find any cells with
GCaMP-filled nuclei from 2- to 12-month-old animals in either
Thy1-GCaMP2.2c or Thy1-GCaMP3 lines (Figure S3A). There
were very few cells expressing GCaMP in layer II/III before
4 months in Thy1-GCaMP2.2c lines. The expression of GCaMP
in Thy1-GCaMP3 lines was widespread in layer II/III and layer
V from 2 to 12 months (Figure S3B). The brightness of GCaMP
in both lines increased from 2 to 4 months and was stable after
4 months of age (Figure S3C).
In Vitro Characterization of Thy1-GCaMP Mice
To determine GCaMP reporter function in transgenic brain
tissues, we used laser-scanning confocal microscopy to monitor
Ca2+
responses in acute brain slices from 1-month-old Thy1GCaMP2.2c
and Thy1-GCaMP3 mice. First, we noted that the
improved properties of GCaMP2.2c and GCaMP3 allowed
for robust calcium imaging of spontaneous activity in layer V
neurons of the cortex and in neurons from the CA1 and the dentate
gyrus of hippocampus (Movies S1 and S2). To test whether
these spontaneous fluorescence changes were associated with
neuronal activities, we performed cell-attached recording of
spontaneous spike activity and imaged fluorescence changes
simultaneously in the hippocampal pyramidal cells. As expected,
Figure 2. GCaMP Fluorescence in Thy1GCaMP3
Transgenic Mice Showing Labeling
of Subsets of Neurons in Different Brain
Areas
Green, GCaMP fluorescence; blue, DAPI staining.
GCL, ganglion cell layer; IPL, inner plexiform layer;
INL, inner nuclear layer; OPL, outer plexiform layer;
ONL, outer nuclear layer. Scale bars represent
50 mm.
the fluorescence changes were well
correlated with the spontaneous spiking
activities in these neurons (Figure S4).
Next, we measured action potential
(AP)-triggered fluorescence responses of
GCaMP2.2c and GCaMP3. We made
whole-cell recordings from GCaMP-expressing
hippocampal dentate granular
cells and evoked APs by brief current
injections(3–5 nA,2 ms).SingleAPevoked
Ca2+
transients with average DF/F amplitudes
of 21.6% ± 1.4% and 25.8% ±
2.0% (n = 9 cells) in Thy1-GCaMP2.2c
and Thy1-GCaMP3 acute slices, respectively. Moreover, the
average DF/F and the number of APs were well correlated. The
average DF/F of GCaMP2.2c (n = 9 cells) was 65.0% ± 10.5%,
96.6% ± 13.0%, 126.1% ± 15.2%, 146.6% ± 16.8%, 261.4% ±
23.3%, and 308.9%± 24.2% for3, 5, 7, 9, 20, and 40 APs, respectively.
Similarly, the average DF/F of GCaMP3 (n = 9 cells) was
69.8% ± 14.5%, 119.5% ± 16.1%, 159.8% ± 19.9%, 200.2% ±
22.1%, 343.5% ± 31.2%, and 396.6% ± 28.2% for 3, 5, 7, 9, 20,
and 40 APs, respectively (Figures 3A–3D). The signal-to-noise
ratio (SNR) of GCaMP2.2c and GCaMP3 was 7.5 ± 0.4 versus
11.9 ± 0.9, 32.7 ± 6.0 versus 46.9 ± 5.5, and 110.5 ± 15.4 versus
148.1 ± 13.6 for 1, 5, and 40 APs, respectively (Figure 3E). The
rise times of fluorescence changes range from 214.1 ms to
374.1 ms for both GCaMP2.2c and GCaMP3. Decay times were
between 0.9 s and 1.9 s for GCaMP2.2c and 1.4 s and 2.6 s for
GCaMP3 (Figures 3F and 3G). Finally, we tested Thy1GCaMP2.2c
and Thy1-GCaMP3 for the ability to image calcium
transients in populations of neuronal somata. For this, we treated
acute brain slices from Thy1-GCaMP2.2c and Thy1-GCaMP3
mice with a high-potassium bath solution. We found that depolarization
with high potassium (10 mM and 30 mM KCl) induced
dramatic fluorescence changes in dentate granular neurons of
the hippocampus in both transgenic lines (Movie S3). The averaged
DF/F of GCaMP3 was higher than GCaMP2.2c (217.9% ±
9.4% versus 136.3% ± 6.1% with 10 mM KCl and 310.1% ±
11.8% versus 186.5% ± 10.2% with 30 mM KCl) (Figures S5A
and S5B). Thus, at the single cell or population levels, both
GCaMP2.2c and GCaMP3 robustly detect spontaneous and
evoked responses in vitro in acute brain slice preparations.
In Vivo Ca2+
Imaging of Apical Dendrites of Layer V
Pyramidal Neurons in the Motor Cortex
To evaluate GCaMP expression in the intact brain, we performed
transcranial two-photon imaging of the motor cortex of adult
Neuron
Thy1-GCaMP Transgenic Mice
Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc. 299
Thy1-GCaMP2.2c and Thy1-GCaMP3 mice. Under in vivo
imaging conditions in both transgenic lines, GCaMP expression
was clearly perimembrane and was never detected in the
nucleus (Figures 4A–4F and Movies S4 and S5). The baseline
fluorescence intensity of GCaMP was similar in both lines in
5-month-old animals (Figure 4G). In Thy1-GCaMP2.2c mice,
densely packed yet resolvable individual apical tuft dendrites
were clearly visible in superficial cortical layers (Figures 4A
and 4B). In comparison, the density of labeled dendrites was
substantially higher in Thy1-GCaMP3 animals, making individual
dendritic imaging difficult (Figures 4D and 4E). Consistent
with the expression data from fixed brain slices (Figure
S2B), Thy1-GCaMP2.2c mice had mainly layer V neuron
labeling with very rare layer II/III neuron labeling (Figures 4C
and 4H), whereas GCaMP3 was expressed in layer V neurons
as well as in the majority of layer II/III neurons (Figures
4F and 4H). Therefore, unlike Thy1-GCaMP3 mice, Thy1GCaMP2.2c
mice offer an opportunity to image the activity of
Figure 3. Action Potential-Evoked Response
of GCaMP2.2c and GCaMP3 in Hippocampal
Granular Cells of Thy1-GCaMP
Transgenic Mice
(A) A granular cell patched with internal solution
containing 100 mM Alexa Fluor 555 is shown in
red. The recording pipette is indicated with white
lines.
(B) Fluorescence changes of GCaMP2.2c and
GCaMP3 to different numbers of action potentials
evoked at 83 Hz (n = 9 cells).
(C and D) Representative DF/F traces to different
numbers of action potentials across cells from
Thy1-GCaMP2.2c (C) and Thy1-GCaMP3 (D)
transgenic mice. The insets show the evoked
action potentials from each cell.
(E) Signal-to-noise ratio of GCaMP2.2c and
GCaMP3 to different numbers of action potentials
evoked at 83 Hz (n = 9 cells).
(F and G) The means of rise time (F) and decay time
(G) of fluorescence responses corresponding
to the number of stimulating action potentials
(n = 9 cells).
Data are presented as mean ± SEM. *p < 0.05.
Scale bar represents 50 mm in (A).
apical dendrites and spines of layer V
pyramidal neurons in the cortex.
We next investigated whether Thy1GCaMP2.2c
and Thy1-GCaMP3 mice
could report neuronal activity responses
in the intact brain. Since individual
dendrites are clearly resolvable in Thy1GCaMP2.2c
mice compared to Thy1GCaMP3
mice, we tested whether
calcium transients could be detected in
the apical dendrites of layer V neurons
of Thy1-GCaMP2.2c mice using twophoton
microscopy in the primary motor
cortex (M1). In awake, head-fixed animals,
we observed numerous dendritic Ca2+
transients with
large amplitudes (Figures 5A and 5C). These dendritic Ca2+
transients typically lasted several hundreds of milliseconds
with a DF/F ranging from $50% up to 200% (Figure 5B). The
duration and amplitude of these dendritic calcium transients
are comparable to dendritic calcium spikes observed in vitro
(Larkum et al., 2009). In contrast, we rarely observed such
robust Ca2+
transients in dendritic branches in anesthetized
mice (Figure 5C). Furthermore, in the awake state, large
elevations of calcium influx were readily detected not only
in the entire dendritic shafts but also in their associated
dendritic spines (Figures 5A and 5D and Movie S6). In both
anesthetized and awake mice, we were able to detect transient
calcium elevations within single dendritic spines over tens of
milliseconds (Figure 5E). Thus, Thy1-GCaMP2.2c mice provide
a means for investigating calcium transients over time in
dendritic spines, as well as dendritic branches in layer V pyramidal
neurons.
Neuron
Thy1-GCaMP Transgenic Mice
300 Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc.
In Vivo Ca2+
Imaging of Somatic Activity in the Motor
Cortex
Next, we examined whether calcium transients could also be
detected in individual somata of Thy1-GCaMP mice. We first
tested whether we could detect GCaMP responses in the
soma of neurons in layer II/III of the motor cortex in anesthetized
animals. In Thy1-GCaMP2.2c mice, we observed GCaMPexpressing
neurons but could not detect activated cells within
an imaging window of 250 3 250 mm during a 10 min recording
period. In contrast, in Thy1-GCaMP3 mice, we observed 9.3 ±
0.5 cells (n = 4 areas from 3 mice) using the same imaging conditions
(Figures 6A and 6B). Differences between Thy1GCaMP2.2c
and Thy1-GCaMP3 mice are likely due to the fact
that significantly fewer layer II/III neurons are labeled in Thy1GCaMP2.2c
mice (Figure 1; Figure S2).
We also imaged neuronal activity in the motor cortex of awake
mice using a fixed-head imaging design (Dombeck et al., 2007,
2009). In awake, behaving animals, we were able to detect activated
neurons both in Thy1-GCaMP2.2c and Thy1-GCaMP3
mice (Figures 6A and 6B; Figure S6; Movie S7). In Thy1-GCaMP3
mice, we detected many more activated neurons (34.4 ± 1.7 cells
in a 250 3 250 mm imaging window, n = 5 areas from 3 mice) over
a 10 min period (Figures 6A and 6B) compared to anesthetized
animals (see above). The population activity in the primary motor
cortex of both Thy1-GCaMP mice was correlated with locomotor
activity (Figure 6C; Figure S6). Repeated imaging of the same
brain area at 15 days after the first imaging showed that most
of the same neurons were active in both views (Figure 6D).
From these observations, we conclude that both Thy1GCaMP2.2c
and Thy1-GCaMP3 mice can be used to monitor
neuronal activity over extended periods of time in the motor
cortex of living animals.
In Vivo Imaging of Sensory Stimulation-Evoked Ca2+
Transients in the Somatosensory Cortex
In recent studies, viral expression of the ratiometric GECIs
(YC3.60 and D3cpV) and GCaMP3 in pyramidal neurons of the
mouse somatosensory (barrel) cortex allowed the detection of
neuronal activity induced by whisker stimulation (Lu¨ tcke et al.,
2010; Mittmann et al., 2011; O’Connor et al., 2010; Wallace
et al., 2008). To determine whether Ca2+
transients could be detected
in the barrel cortex in response to sensory stimulation in
our Thy1-GCaMP mice, we performed a similar test. We used
Thy1-GCaMP3 mice because in vivo two-photon imaging in
the barrel cortex revealed sparse labeling of layer II/III pyramidal
neurons in Thy1-GCaMP2.2c mice and dense labeling in Thy1GCaMP3
mice (data not shown). To induce sensory stimulation,
we deflected multiple mystacial vibrissae ten times using 500 ms
air puffs with 10 s interpuff intervals. In Thy1-GCaMP3 transgenic
mice, we routinely detected calcium transients associated with
whisker stimulation in both cell somata and the adjacent layer
II/III neuropil (Figures 7A–7C and Movie S8). To further characterize
the kinetics of GCaMP3 signals in vivo, we next recorded
changes in reporter fluorescence in 64 3 64 pixel frames at
a frame rate of 40 Hz from layer II/III GCaMP3-expressing
neurons. Single air puffs induced DF/F amplitude changes of
242.9% ± 14.2% (n = 8 from 3 mice), similar to those seen with
virally transduced GCaMP3 (O’Connor et al., 2010) and higher
than the ratio changes seen from YC 3.60 and D3cpV (Lu¨ tcke
et al., 2010; Wallace et al., 2008). The rise and decay time of
the calcium transients in GCaMP3 were 477.9 ± 17.1 ms and
1,072.5 ± 29.4 ms, respectively (n = 8 from 3 mice; Figures
7D–7F). Thus, Thy1-GCaMP3 mice allow the detection of
dynamic changes in neuronal activity in vivo in response to
sensory stimulation.
In Vivo Ca2+
Imaging of Odor Responses in the Olfactory
Bulb
In Thy1-GCaMP3 transgenic mice, GCaMP is expressed in the
glomerular layer, the external plexiform layer, and the mitral
cell layer, but not within the olfactory nerve layer or the granule
cell layer (Figures S7A and S7B). Two-photon imaging showed
that GCaMP3 fluorescence was detected in the olfactory bulb
Figure 4. In Vivo Two-Photon Imaging of GCaMP2.2c- and GCaMP3Expressing
Neocortical Neurons
In vivo two-photon images of GCaMP-expressing neurons in the motor cortex
of 5-month-old Thy1-GCaMP2.2c (A–C) and Thy1-GCaMP3 (D–F) mice. The
depth below the pial surface is shown above each panel. In Thy1-GCaMP2.2c
mice, densely packed yet resolvable individual dendrites were clearly visible in
the superficial layers (A and B), whereas the density of labeled dendrites was
much higher in Thy1-GCaMP3 animals (D and E). Thy1-GCaMP2.2c mice had
few layer II/III neurons labeled (C), whereas GCaMP3 was expressed in almost
all layer II/III neurons (F). Red arrows mark individual layer II/III neurons. See
also Movies S4 and S5.
(G) Quantification of brightness of neuronal somata from Thy1-GCaMP2.2c
and Thy1-GCaMP3 transgenic mice ($150–160 mm below the pial surface,
n = 75 from 3 mice).
(H) Quantification of neuron number in 250 3 250 mm area from Thy1GCaMP2.2c
and Thy1-GCaMP3 transgenic mice ($150–160 mm below the pial
surface, n = 10 from 3 mice). A.U., arbitrary units. Data are presented as
mean ± SEM. Scale bar represents 50 mm.
Neuron
Thy1-GCaMP Transgenic Mice
Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc. 301
in vivo (Figure S7C and Movie S9). Based on the location and
soma size, GCaMP3-expressing cells appeared to be mainly
mitral cells, in addition to a small subset of periglomerular
and external tufted cells. GCaMP fluorescence can be seen
throughout the soma and the dendrites.
To characterize activity-induced GCaMP3 responses in the
olfactory bulb, we performed in vivo two-photon Ca2+
imaging
in the dorsal olfactory bulb during odor presentation. For odor
stimulation, we chose four odorants, methyl salicylate, amyl
acetate, eugenol, and 1-pentanol, because they have different
molecular structures and have previously been shown to
strongly activate distinct glomeruli in the dorsal olfactory
bulb (Lin et al., 2006; Rubin and Katz, 1999; Wachowiak and
Cohen, 2001). As shown in Figure S7D, 1% odorants trigger
strong calcium responses in the olfactory bulbs of Thy1GCaMP3
mice. Similar to previous in vivo imaging data using
Kv3.1 potassium channel promoter-driven expression of
GCaMP2.0 in the olfactory bulb (Fletcher et al., 2009), each
odor induced two types of signals within the odor maps. The
first response type was relatively weak and diffuse, whereas
the second type of response was more focused and formed
‘‘hot spots’’ that corresponded to individual glomeruli (Figure
S7D). Consistent with previous studies (Wachowiak and
Cohen, 2001; Fried et al., 2002; Bozza et al., 2004), we found
that different odorants activated discrete glomeruli in Thy1GCaMP3
mice (Figure S7D). We also found that initial odor
responses were often higher than subsequent stimuli (Figure
S7E), a phenomenon we attributed to odor habituation
(Holy et al., 2000; Verhagen et al., 2007). Notably, we found
that odorant-triggered fluorescence changes with GCaMP3
are in the range of 30%–150%, much greater than in previous
reports that used other calcium indicators (De Saint Jan et al.,
2009; Fletcher et al., 2009).
Olfactory coding is multidimensional. In addition to response
profiles being dictated by the molecular structure of odorants,
odorant concentration also influences the receptor repertoire
recruited to the stimulus and thus shapes the composite
response pattern. With increasing concentrations of odorants,
new glomeruli may be recruited into the response pattern, while
previously active glomeruli often respond more intensively
(Fletcher et al., 2009; Johnson and Leon, 2000). These effects
were observed for all odorants tested in Thy1-GCaMP3 mice;
the representative odor maps are shown in Figure S7F. Taken
together, these data show that Thy1-GCaMP3 mice can detect
changes of neuronal activity in mitral cells in response to specific
odorants in a population and concentration-dependent manner.
Furthermore, unlike previous methods for monitoring Ca2+
mediated
olfactory responses at presynaptic terminals (Bozza
et al., 2004; Fried et al., 2002), the Thy1-GCaMP3 reporter line
described here reflects postsynaptic responses.
Figure 5. In Vivo Ca2+
Imaging of Apical
Dendrites and Dendritic Spines of Layer V
Pyramidal Neurons in Thy1-GCaMP2.2c
Transgenic Mice
(A) Two-photon Ca2+
imaging of layer V apical
dendrites in the motor cortex of an awake, headfixed
Thy1-GCaMP2.2c mouse (2 months old). A
representative time-lapse sequence shows a
dendritic calcium transient in both the dendritic
shaft and spines.
(B) A representative DF/F tracing of a dendritic
calcium transient within the apical tuft in an awake,
head-fixed mouse. Three segments (red, green,
and blue boxes) of the dendritic arbor were
measured before and after the calcium transient.
Note that the calcium transient lasts for hundreds
of milliseconds.
(C) Quantification of the number of dendritic
calcium transients within 10 min in a 100 3
100 mm imaging window under anesthetized
(n = 4 from 3 mice) and awake (n = 6 from 3 mice)
state.
(D) Dendritic calcium transient caused a transient
calcium elevation in dendritic spines in motor
cortex of an awake, behaving mouse. Images of
the same apical dendritic segment before (45 s),
during (50 s), and after (55 s) its activation are
shown. The blue, green, and red circles mark the
location of three different spines along the
dendrite. See also Movie S6.
(E) A transient calcium elevation could be detected
in dendritic spines under anesthetized
and awake states. Fluorescence images were
acquired from a line scan intersecting a spine (S)
and the dendrite (D, gray trace). The increases in fluorescence indicate Ca2+
entry within the bulbous spine (black trace).
Data are presented as mean ± SEM. Scale bars represent 10 mm for (A) and (B), 5 mm for (D), and 2 mm for (E). **p < 0.005.
Neuron
Thy1-GCaMP Transgenic Mice
302 Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc.
DISCUSSION
In this study, we generated transgenic mice that stably express
improved GCaMPs, GCaMP2.2c and GCaMP3 (Tian et al.,
2009), in subsets of CNS neurons under the control of the mouse
Thy1 promoter. Our findings indicate that these GCaMP transgenic
lines provide an excellent tool for detecting neural activity
in acute brain slices as well as the intact brain. First, we show that
both spontaneous and evoked calcium transients can be
detected in acute brain slices prepared from both transgenic
lines. Notably, we were able to detect small calcium transients
in neuronal somata triggered by a single action potential.
Second, calcium transients were also readily detected in apical
dendrites and dendritic spines in the living cortex of Thy1GCaMP2.2c
transgenic mice. Third, large, robust calcium
signals can be detected in populations of layer II/III cortical
neurons in both GCaMP transgenic lines with natural motor or
sensory stimuli. Lastly, odor-evoked calcium transients can be
detected at single glomerulus resolution in Thy1-GCaMP3
mice. Together, these results indicate that GCaMP2.2c and
GCaMP3 mice provide a sensitive means to detect patterns of
neuronal activity at the level of individual neurons and synapses,
as well as populations of neurons in vitro and in vivo.
Until recently, calcium imaging with synthetic calcium dyes
has been the method of choice to monitor activity in neuronal
cultures, acute brain slices, and intact brains. However, routine
and reliable loading of Ca2+
dyes into targeted neuronal populations
in vivo has proven difficult. Bulk loading of calcium indicators
indiscriminately labels mixtures of cells, making cell
type-specific labeling nearly impossible. Single-cell labeling is
technically challenging and allows for only a few cells to be
loaded during a given imaging session. Furthermore, imaging
with calcium dyes can only last for periods of hours, making
chronic recordings of neuronal activity over extended times difficult,
if not impossible. Genetically encoded calcium indicators
overcome many of these limitations (Hasan et al., 2004; Looger
and Griesbeck, 2012; Miyawaki et al., 1997). Incremental rounds
of reporter optimization have resulted in new GCaMPs with
significantly improved fluorescence characteristics and higher
sensitivity to calcium (Muto et al., 2011; Ohkura et al., 2005;
Souslova et al., 2007; Tallini et al., 2006; Tian et al., 2009; Zhao
et al., 2011b).
A number of methods are available for transgene delivery,
including in utero electroporation, biolistic delivery, and viral
transduction. Some viral delivery methods have distinct advantages,
e.g., the retrograde transsynaptic tracing ability afforded
by rabies virus, into which GCaMP3 has recently been incorporated
(Osakada et al., 2011). However, they have a number of
drawbacks as well: limited payload capacity, inherent tropism,
local delivery, incompatibility with early developmental events,
and the requirement that each experimental animal be subjected
to a survival surgery. Only transgenic incorporation into the
genome affords stable expression of a transgene in all target
tissues, reliable animal-to-animal comparisons, and the ability
to image the embryo and other early developmental states. In
this study, we demonstrated the feasibility and functionality of
long-term expression of GCaMPs from the Thy-1 promoter for
in vitro and in vivo calcium imaging.
As any GECI buffers Ca2+
and may interfere with endogenous
signaling events, there is an inherent risk of neuronal toxicity with
long-term and/or high levels of expression. Indeed, overexpression
of GCaMP3 using in utero electroporation or viral infection
Figure 6. In Vivo Ca2+
Imaging of Neuronal Activity in the Motor Cortex with Thy1-GCaMP3 Transgenic Mice
(A and B) In vivo two-photon time-lapse images of layer II/III neurons in the motor cortex of 5-month-old Thy1-GCaMP3 mice.
(A) Top: examples of neuronal activity in the anesthetized state. Bottom: neuronal activity of the same area in the awake state. Red arrows mark activated neurons.
(B) Quantification of the number of activated neurons within 10 min in anesthetized (n = 4 areas from 3 mice) and awake (n = 5 areas from 3 mice) state. Data are
presented as mean ± SEM. **p < 0.005.
(C and D) Repeated imaging of calcium dynamics of layer II/III neurons in the motor cortex.
(C) A raw fluorescence image of layer II/III neurons in the motor cortex of Thy1-GCaMP3 mice at 7 days after surgery (top) and DF/F traces of each circled neuron
(bottom). Shaded part of the traces indicates that the mice were moving.
(D) The fluorescence image of the same field and fluorescent traces of the same neurons as in (C) at 22 days after surgery. Scale bars represent 50 mm for (A), (C),
and (D).
Neuron
Thy1-GCaMP Transgenic Mice
Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc. 303
showed that high expression levels can induce neural dysfunction
and altered subcellular localization (e.g., nuclear), particularly
near the injection site (Dombeck et al., 2010; Tian et al.,
2009). In our transgenic animals, GCaMP was widely expressed
in many neuronal subtypes throughout the CNS. Analysis of
Thy1-GCaMP2.2c and Thy1-GCaMP3 transgenic mice did not
reveal any obvious gross or cellular abnormalities. Importantly,
distribution of GCaMP was cytosolic and homogeneous, with
no signs of aggregation or compartmentalization in the nucleus
in vivo. These results suggest that our transgenic mice exhibit
stable, long-term expression of GCaMPs in neurons with normal
functions and thus allow sensitive detection of calcium transients
in vivo.
One key advantage of calcium imaging is that it allows the
simultaneous mapping of neuronal activities from numerous
cells within complex neuronal networks. Given that GCaMP3
transgenic expression targets most pyramidal neurons ($90%)
throughout the cortical layers, this mouse line could allow activity
monitoring from large populations of neurons across various
cortical layers in behaving animals. The stable expression of
GCaMP3 at nontoxic levels in our transgenic mice makes their
application ideal for long-term in vivo monitoring of somatic
activity. On the other hand, due to the low basal fluorescence
Figure 7. In Vivo Imaging of Sensory Stimulation-Evoked
Calcium Transients in the
Somatosensory Cortex of Thy1-GCaMP3
Transgenic Mice
(A) In vivo two-photon time-lapse images of layer
II/III neurons in the somatosensory cortex of
Thy1-GCaMP3 mice. Red arrows mark activated
neurons and a red arrowhead marks an activated
neuronal process.
(B and C) Calcium dynamics of layer II/III neurons
in the somatosensory cortex. A raw fluorescence
image of layer II/III neurons is shown in (B). Fluorescence
traces of the neurons (green circles) and
neuropil (red circle) are shown in (C). The blue bars
show the onset of air puffs. See also Movie S8.
(D) Three examples of individual fluorescence
traces of layer II/III neurons in the somatosensory
cortex using 40 Hz scanning speed.
(E) The average maximal fluorescence changes in
response to a single air puff.
(F) Decay time and rise time of neurons in response
to a single air puff.
Data are presented as mean ± SEM (n = 8 cells
from 3 mice for E and F). Scale bars represent
50 mm for (A) and (B).
and sparser labeling, Thy1-GCaMP2.2c
mice provide a suitable means for imaging
of Ca2+
transients in the dendrites of
layer V neurons in vivo. Large calcium
transients could readily be detected
in dendrites and dendritic spines of
layer V pyramidal neurons in Thy1GCaMP2.2c
mice and in the somata of
layer II/III neurons in Thy1-GCaMP3
transgenic mice.
In the CNS, odors are represented as patterns of neural
activity encoded by time and space. Previous mapping
approaches in the olfactory bulb have used 2-deoxyglucose
staining, intrinsic optical signal imaging, and pH-sensitive exocytosis
detection to monitor odor-induced changes in neuronal
activity. Such functional mapping strategies can provide
temporal and spatial resolution of neuronal activity but to date
have primarily reported olfactory nerve presynaptic activity,
with little (or no) contribution from postsynaptic neurons. On
the other hand, odor responses imaged by bulk-loaded
voltage-sensitive dyes comprise a mixture of both pre- and
postsynaptic components and do not show genetic specificity
(Friedrich and Korsching, 1998; Spors and Grinvald, 2002).
More recently, GCaMP2.0 transgenic mice, driven by a Kv3.1
potassium channel promoter, allowed detection of postsynaptic
odor representation within the glomerular cell layer, but
responses were relatively weak and did not span a dynamic
range of odor concentration or specificity (Fletcher et al.,
2009). In Thy1-GCaMP3 mice, GCaMP3 is expressed strongly
in the glomerular and mitral cell layers, and responses to odorants
were encoded by distinct sets of glomeruli. Concentration
coding involved both graded responses from each activated
glomerulus, as well as an increase in the total number of
Neuron
Thy1-GCaMP Transgenic Mice
304 Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc.
glomeruli that responded. Compared to GCaMP2.0 transgenic
mice, baseline expression and odor-induced changes in GCaMP
fluorescence was significantly higher in Thy1-GCaMP3 mice.
These findings suggest that the Thy1-GCaMP3 transgenic
mouse is an improved genetic tool to investigate neuronal
activity changes within the olfactory system.
Although our studies only tested the utility of the Thy1-GCaMP
mice in the motor cortex, somatosensory cortex, and the olfactory
bulb, GCaMP expression in these mice was widespread
(Figures 1 and 2; Figures S2 and S3), and the strains are likely
to be useful for monitoring neuronal activity in many brain areas.
Stable expression of GCaMP via transgenic mice will enhance
our ability to study how information is processed in both the
healthy and diseased brain. Together with the recently generated
Cre-inducible GCaMP3 mice (Zariwala et al., 2012), these tools
may provide important insights into disease processes and
activity-related pathological changes when combined with
animal models of neurological disorders. Furthermore, chronic
imaging of various subtypes of neurons with GCaMP will help
to pinpoint the important groups of neurons, brain regions, and
characteristic abnormalities involved in the onset, progression,
and end stages of neurological disorders.
EXPERIMENTAL PROCEDURES
Animal Use
All experiments were conducted according to protocols approved by the Institutional
Animal Care and Use and Institutional Biosafety Committees of MIT
and the NYU School of Medicine.
DNA Constructs
GCaMP2.0 and GCaMP3 expression constructs were previously
reported (Tian et al., 2009). GCaMP2.2c was generated by changing the
second arginine to valine and serine at 118 to cysteine of GCaMP2.0. All
in vitro expression constructs of GCaMPs were connected with the coding
sequence of tdTomato via a 2A peptide (P2A) sequence and subcloned into a
modified pBluescript plasmid, which contained the CAG promoter (a combination
of the cytomegalovirus early enhancer element and chicken beta-actin
promoter). To generate the Thy1-GCaMP transgenic mouse, we subcloned
GCaMP2.2c and GCaMP3 coding sequences into a Thy1 transgenic construct
(Arenkiel et al., 2007; Feng et al., 2000). All constructs were verified by
sequencing.
Fluorescence Measurements in HEK293 Cells
HEK293 cells were cultured in DMEM/F12 containing 10% FBS and GCaMPP2A-tdTomato
plasmid transfection was performed with Lipofectamine 2000.
Imaging experiments were performed $36–48 hr after the transfection as
described previously (Nakai et al., 2001). Imaging was performed using an
Olympus Fluoview 1000 confocal microscope equipped with multiline argon
laser (457 nm, 488 nm, and 515 nm) and HeNe (G) laser (543 nm) using the
203 water-immersion objective (NA = 0.5). Green GCaMP fluorescence was
excited at 488 nm and isolated using a band-pass filter (505–525 nm). Red
tdTomato fluorescence was excited at 543 nm and isolated using a bandpass
filter (560–660 nm). The time-series images (XYT) were acquired at frame
rates of 1 Hz at a resolution of 256 3 256 pixels. For ATP stimulation, the
solution contained 135 mM NaCl, 5.4 mM KCl, 2 mM CaCl2, 1 mM MgCl2,
10 mM glucose, 5 mM HEPES (pH 7.4), and 100 mM ATP. All experiments
were performed at room temperature.
Transgenic Mice
Thy1-GCaMP2.2c and Thy1-GCaMP3 transgenic mice were generated by
injection of gel-purified DNA into fertilized oocytes using standard techniques
(Feng et al., 2004; Zhao et al., 2011a). Embryos for injection were obtained by
mating (C57BL6/J and CBA) F1 hybrids. Transgenic founders were backcrossed
to C57BL6/J mice for analysis of expression patterns. Primers for
genotyping were 50
- TCT GAG TGG CAA AGG ACC TTA GG À30
(forward)
50
- TTA CGA CGT GAT GAG TCG ACC À30
(reverse). The mouse strains
have been deposited at The Jackson Laboratory. The JAX stock number for
Thy1-GCaMP2.2c line 8 is 017892 and the JAX stock number for Thy1GCaMP3
line 6 is 017893.
Immunohistochemistry
GCaMP mice were anesthetized by the inhalation of isoflurane and were intracardially
perfused with 20 ml 13 PBS, followed by 20 ml 4% paraformaldehyde
(PFA) in PBS. Mouse brains were then postfixed in 4% PFA/PBS overnight at
4
C. We cut 50 mm sagittal sections using a vibratome. Rabbit anti-GFP antibody
(Invitrogen, 1:1,000) was used to enhance the GCaMP fluorescence.
Briefly, sections were incubated with blocking buffer (5% normal goat serum,
2% BSA, and 0.2% Triton X-100 in 13 PBS) for 1 hr at room temperature and
then incubated with rabbit anti-GFP antibody overnight at 4
C. After incubation
with the first antibody, sections were washed with 13 PBS three times
for 20 min each, followed by incubation with Alexa 488-conjugated goat
anti-rabbit secondary antibody (Invitrogen) for 2–4 hr at room temperature
and then washed with 13 PBS. Sections were transferred onto slides,
mounted with 0.1% paraphenylinediamine in 90% glycerol/PBS, and imaged
with a microscope (BX61, Olympus).
Slice Preparation and Electrophysiology
Acute slices were prepared according to published procedures (Pec¸ a et al.,
2011). Briefly, mice were anesthetized with Avertin solution (20 mg/ml,
0.5 mg/g body weight) and perfused through the heart with 20 ml of icecold
oxygenated (95% O2, 5% CO2) cutting solution containing 105 mM
NMDG, 105 mM HCl, 2.5 mM KCl, 1.2 mM NaH2PO4, 26 mM NaHCO3,
25 mM glucose, 10 mM MgSO4, 0.5 mM CaCl2, 5 mM L-ascorbic acid,
3 mM sodium tyruvate, and 2 mM thiourea (pH was 7.4, with osmolarity of
295–305 mOsm). The brains were rapidly removed and placed in ice-cold
oxygenated cutting solution. Coronal or transverse hippocampal slices
(300 mm) were prepared using a slicer (Vibratome 1000 Plus, Leica Microsystems)
and then transferred to an incubation chamber (BSK4, Scientific System
Design) at 32
C with carbogenated cutting solution, which was gradually
replaced with artificial cerebral spinal fluid (ACSF) in 30 min through a
peristaltic pump (Dynamax Model RP-1; Rainin Instruments), allowing a
precise regulation of fluid flow rates. The slices were then kept in the ACSF
that contained 119 mM NaCl, 2.3 mM KCl, 1.0 mM NaH2PO4, 26 mM NaHCO3,
11 mM glucose, 1.3 mM MgSO4, and 2.5 mM CaCl2 (pH was adjusted to 7.4
with HCl, with osmolarity of 295–305 mOsm) at room temperature for at least
30 min.
Recordings were performed in oxygenated ACSF. Intracellular solution consisted
of 130 mM KMeSO3, 10 mM HEPES, 4 mM MgCl2, 4 mM Na2ATP,
0.4 mM NaGTP, 10 mM Na-phosphocreatine, and 3 mM Na-L-ascorbate;
pH was adjusted to 7.3 with KOH. Recordings were performed at room
temperature in ACSF. To evoke APs, we held cells in the current-clamp configuration,
and we injected 3–5 nA of current for 2 ms through the recording electrode.
Cells were selected if their GCaMP fluorescence was homogeneously
distributed in the cytoplasm.
Fluorescent signals were imaged by a confocal microscope (Fluoview FV
1000; Olympus) with a 30 mW multiline argon laser, at 5%–10% laser power.
The laser with a wavelength of 488 nm was used for excitation, and fluorescence
was recorded through a band-pass filter (505–525 nm). The images
were acquired using 403 water-immersion objectives (NA = 0.8) with 5 Hz
scanning speed. XYT image galleries were collected and average fluorescence
intensity in the soma was measured for the quantification by Fluoview data
processing software. We report time series as DF/F = [(F À FB) À (F0 À FB)]/
(F0 À FB), where F is the raw fluorescence signal, FB is the background, and
F0 is the mean fluorescence signal in a baseline period prior to the action
potential stimuli. SNR was calculated as the ratio of DF/F to SD of the basal
fluorescence, 1 s before the stimulus up to stimulus onset. Rise time was
measured as the time between onset of current injection and the maximal
response. Decay time was measured as the time between the maximal
response and the decay back to baseline.
Neuron
Thy1-GCaMP Transgenic Mice
Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc. 305
In Vivo Ca2+
Imaging of Layer V Apical Dendrites in Motor Cortex
A head holder composed of two parallel micrometal bars was attached to the
animal’s skull to reduce motion-induced artifact during imaging. First, surgical
anesthesia was achieved with an intraperitoneal injection (5–6 ml/g) of
a mixture of ketamine (20 mg/mL) and xylazine (3 mg/mL). A midline incision
of the scalp exposed the periosteum, which was manually removed with
a microsurgical blade. A small skull region ($0.2 mm in diameter) was located
over the left motor cortex based on stereotactic coordinates (0.5 mm
posterior from the bregma and 1.5 mm lateral from the midline) and marked
with a pencil. A thin layer of cyanoacrylate-based glue was first applied to
the top of the entire skull surface and to the metal bars, and the head holder
was then further fortified with dental acrylic cement (Lang Dental
Manufacturing). The dental cement was applied so that a well was formed
leaving the motor cortex with the marked skull region exposed between the
two bars. All procedures were performed under a dissection microscope.
After the dental cement was completely dry, the head holder was screwed
to two metal cubes that were attached to a solid metal base, and a cranial
window was created over the previously marked region. The procedures for
preparing a thinned skull cranial window for two-photon imaging have been
described in detail in previous publications (Yang et al., 2010). Briefly,
a high-speed drill was used to carefully reduce the skull thickness by approximately
50% under a dissecting microscope. The skull was immersed in
artificial cerebrospinal fluid during drilling. Skull thinning was completed by
carefully scraping the cranial surface with a microsurgical blade to $20 mm
in thickness. For anesthetized imaging, animals were immediately imaged
under a two-photon microscope tuned to 910 nm with a 403 objective
immersed in an artificial cerebrospinal fluid solution and a 33 digital zoom.
For awake animal imaging, the completed cranial window was covered
with silicon elastomer (World Precision Instruments) and mice were given at
least 4 hr to recover from the surgery-related anesthesia. Mice with head
mounts were habituated for a few times (10 min for each time) in the imaging
apparatus to minimize potential stress effects of head restraining and
imaging. To image dendrites in awake mice, we screwed the head holder to
two metal cubes attached to a solid metal base, and the silicon elastomer
was peeled off to expose the thinned skull region and ACSF was added to
the well. The head-restrained animal was then placed on the stage of a
two-photon microscope.
These in vivo two-photon imaging experiments were performed using
a Bio-Rad Radiance 2000 two-photon system equipped with a Tsunami
Ti:sapphire laser pumped by a 10 W MilleniaXs laser (Spectra-Physics).
The average laser power on the sample is $20–30 mW. Most experiments
were acquired at frame rates of 1 Hz at a resolution of 512 3 512 pixels
using a 403 water-immersion objective (Nikon). Image acquisition was
performed using Laser Sharp 2000 software and analyzed post hoc using
ImageJ software (NIH). DF/F was calculated identical to slice imaging
experiments. For detecting calcium signals in layer V apical tuft dendritic
spines, a line crossing the dendrite and the middle of the spine head
was drawn and fluorescence intensity along the line was measured using
ImageJ (NIH).
In Vivo Ca2+
Imaging of Neuronal Activity in the Motor Cortex,
Somatosensory Cortex, and Olfactory Bulb
Imaging experiments were performed on 4- to 5-month-old mice. The surgery
was performed as described previously (Dombeck et al., 2009). Briefly, the
mice were anesthetized with Avertin solution (20 mg/ml, 0.5 mg/g body weight)
and were placed in a stereotactic apparatus with a heating pad underneath to
maintain body temperature. A 2 3 2 mm piece of bone was removed above the
motor cortex, somatosensory cortex, or olfactory bulb as determined by
stereotactic coordinates, and the dura was kept intact and moist with saline.
To dampen heartbeat and breathing-induced motion, we filled the cranial
window with Kwik-sil (World Precision Instruments) and covered it with an
immobilized glass coverslip. A custom-designed head plate was cemented
on the cranial window with Meta-bond (Parkell) when the Kwik-sil set. For
chronic imaging, two coverslips were joined with ultraviolet curable optical
glue (NOR-138, Norland). A smaller insert fit into the craniotomy and a larger
piece was attached to the bone. Imaging was performed 7 days postsurgery
to allow the window to clear.
During imaging of neuronal activity in motor cortex, the head-fixed animals
were placed in water to induce swimming-like behavior. The animals were kept
alert by presenting a pole or by mild air puffs to the whisker field. An infrared
charge-coupled device camera (CCTV) was used for observing the animal’s
behavior during imaging sessions.
Sensory stimulations, consisting of puffs of compressed air delivered by
a Picospritzer unit (Picospritzer II; General Valve), were applied through
a 1-mm-diameter glass pipette placed 15–25 mm rostrolateral from the
whiskers. Air puffs (500 ms duration) were given ten times with 10 s intervals
to prevent adaptation of whisker-evoked responses.
Odorants were delivered using a custom-built odor delivery system
in which the saturated vapor of an odorant was diluted into a main
stream of clean air. The clean air stream was fixed at 0.6–0.8 L/min
throughout the experiment and the odor vapor stream was adjusted
to give the final concentration to the animal. A tube opening was positioned
<1 cm from the animal’s nostrils. To avoid cross-contamination,
we used separate Teflon tubing in parallel for delivery of different odorants.
Odorants were usually presented with pulse duration of 1 s and interstimulus
interval of 30 s to avoid potential sensory adaptation. A constant
suction system was positioned close to the odorant delivery system and
used to quickly remove remnant odorants. The odorants used in this study
included methyl salicylate, amyl acetate, eugenol, and 1-pentanol (Sigma-
Aldrich).
In these experiments, in vivo two-photon imaging was performed at the
McGovern Institute two-photon microscopy core facility. Imaging was
performed on a custom two-photon laser-scanning microscope (Ultima; Prairie
Technologies) coupled with a Mai Tai Deep See laser (Spectra Physics). The
laser was operated at 910 nm ($30–40 mW average power on the sample).
The emission filter set for imaging GCaMP fluorescence consisted of
a 575 nm dichroic mirror and a 525/70 nm band-pass filter. Fluorescence signal
was detected using Hamamatsu multialkali PMTs. In most experiments,
images were acquired at frame rates of 1.5–2 Hz at a resolution of 512 3 256
pixels using a 203, 1.0 NA water-immersion objective (Zeiss). For in vivo z stack
imaging, images were taken at a resolution of 512 3 512 pixels with 2 mm
intervals. Image acquisition was performed using custom Prairie View
Software. The images were analyzed post hoc using NIH ImageJ and ImagePro
Plus 5.0 software (Media Cybernetics). DF/F was calculated identical to
slice imaging experiments.
Statistical Analysis
All statistical analyses were performed using SPSS (IBM) software and graphs
were drawn in SigmaPlot 2000 (Systat Software). Values are expressed as
mean ± SEM. The data between two groups were compared using unpaired
t test. The data among three groups were compared using one-way ANOVA.
Statistical significance was defined as *p < 0.05 or **p < 0.005.
SUPPLEMENTAL INFORMATION
Supplemental Information includes seven figures and nine movies and can
be found with this article online at http://dx.doi.org/10.1016/j.neuron.2012.
07.011.
ACKNOWLEDGMENTS
We thank the members of the Feng laboratory for helpful discussions. We
would like to give special thanks to Peimin Qi, Ethan Skowronski-Lutz, Tyler
Clark Brown, Mriganka Sur, Caroline Runyan, and Holly Robertson for their
intellectual input and technical help. We also thank Charles Jennings and
Thomas J. Diefenbach in the McGovern Institute two-photon microscopy
core facility for their technical support. This work was made possible by the
support from an anonymous grant and from The Poitras Center for Affective
Disorders Research to G.F, by National Institutes of Health Grant NS047325
to W.-B.G, and by The McNair Foundation and NINDS R00NS64171 and
NIH grant R01NS078294 to B.R.A.
Accepted: July 3, 2012
Published: October 17, 2012
Neuron
Thy1-GCaMP Transgenic Mice
306 Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc.
REFERENCES
Andermann, M.L., Kerlin, A.M., Roumis, D.K., Glickfeld, L.L., and Reid, R.C.
(2011). Functional specialization of mouse higher visual cortical areas.
Neuron 72, 1025–1039.
Arenkiel, B.R., Peca, J., Davison, I.G., Feliciano, C., Deisseroth, K., Augustine,
G.J., Ehlers, M.D., and Feng, G. (2007). In vivo light-induced activation of
neural circuitry in transgenic mice expressing channelrhodopsin-2. Neuron
54, 205–218.
Borghuis, B.G., Tian, L., Xu, Y., Nikonov, S.S., Vardi, N., Zemelman, B.V., and
Looger, L.L. (2011). Imaging light responses of targeted neuron populations in
the rodent retina. J. Neurosci. 31, 2855–2867.
Bozza, T., McGann, J.P., Mombaerts, P., and Wachowiak, M. (2004). In vivo
imaging of neuronal activity by targeted expression of a genetically encoded
probe in the mouse. Neuron 42, 9–21.
Chen, X., Leischner, U., Rochefort, N.L., Nelken, I., and Konnerth, A. (2011).
Functional mapping of single spines in cortical neurons in vivo. Nature 475,
501–505.
De Saint Jan, D., Hirnet, D., Westbrook, G.L., and Charpak, S. (2009).
External tufted cells drive the output of olfactory bulb glomeruli. J. Neurosci.
29, 2043–2052.
Dombeck, D.A., Khabbaz, A.N., Collman, F., Adelman, T.L., and Tank, D.W.
(2007). Imaging large-scale neural activity with cellular resolution in awake,
mobile mice. Neuron 56, 43–57.
Dombeck, D.A., Graziano, M.S., and Tank, D.W. (2009). Functional clustering
of neurons in motor cortex determined by cellular resolution imaging in awake
behaving mice. J. Neurosci. 29, 13751–13760.
Dombeck, D.A., Harvey, C.D., Tian, L., Looger, L.L., and Tank, D.W. (2010).
Functional imaging of hippocampal place cells at cellular resolution during
virtual navigation. Nat. Neurosci. 13, 1433–1440.
Feng, G., Mellor, R.H., Bernstein, M., Keller-Peck, C., Nguyen, Q.T., Wallace,
M., Nerbonne, J.M., Lichtman, J.W., and Sanes, J.R. (2000). Imaging neuronal
subsets in transgenic mice expressing multiple spectral variants of GFP.
Neuron 28, 41–51.
Feng, G., Lu, J., and Gross, J. (2004). Generation of transgenic mice. Methods
Mol. Med. 99, 255–267.
Fletcher, M.L., Masurkar, A.V., Xing, J., Imamura, F., Xiong, W., Nagayama, S.,
Mutoh, H., Greer, C.A., Kno¨ pfel, T., and Chen, W.R. (2009). Optical imaging
of postsynaptic odor representation in the glomerular layer of the mouse
olfactory bulb. J. Neurophysiol. 102, 817–830.
Fried, H.U., Fuss, S.H., and Korsching, S.I. (2002). Selective imaging of
presynaptic activity in the mouse olfactory bulb shows concentration and
structure dependence of odor responses in identified glomeruli. Proc. Natl.
Acad. Sci. USA 99, 3222–3227.
Friedrich, R.W., and Korsching, S.I. (1998). Chemotopic, combinatorial, and
noncombinatorial odorant representations in the olfactory bulb revealed using
a voltage-sensitive axon tracer. J. Neurosci. 18, 9977–9988.
Hasan, M.T., Friedrich, R.W., Euler, T., Larkum, M.E., Giese, G., Both, M.,
Duebel, J., Waters, J., Bujard, H., Griesbeck, O., et al. (2004). Functional
fluorescent Ca2+ indicator proteins in transgenic mice under TET control.
PLoS Biol. 2, e163.
Hendel, T., Mank, M., Schnell, B., Griesbeck, O., Borst, A., and Reiff, D.F.
(2008). Fluorescence changes of genetic calcium indicators and OGB-1
correlated with neural activity and calcium in vivo and in vitro. J. Neurosci.
28, 7399–7411.
Holy, T.E., Dulac, C., and Meister, M. (2000). Responses of vomeronasal
neurons to natural stimuli. Science 289, 1569–1572.
Johnson, B.A., and Leon, M. (2000). Modular representations of odorants in
the glomerular layer of the rat olfactory bulb and the effects of stimulus
concentration. J. Comp. Neurol. 422, 496–509.
Kerr, J.N., and Denk, W. (2008). Imaging in vivo: watching the brain in action.
Nat. Rev. Neurosci. 9, 195–205.
Larkum, M.E., Nevian, T., Sandler, M., Polsky, A., and Schiller, J. (2009).
Synaptic integration in tuft dendrites of layer 5 pyramidal neurons: a new
unifying principle. Science 325, 756–760.
Lin, Y., Shea, S.D., and Katz, L.C. (2006). Representation of natural stimuli in
the rodent main olfactory bulb. Neuron 50, 937–949.
Looger, L.L., and Griesbeck, O. (2012). Genetically encoded neural activity
indicators. Curr. Opin. Neurobiol. 22, 18–23.
Lu¨ tcke, H., Murayama, M., Hahn, T., Margolis, D.J., Astori, S., Zum Alten
Borgloh, S.M., Go¨ bel, W., Yang, Y., Tang, W., Ku¨ gler, S., et al. (2010).
Optical recording of neuronal activity with a genetically-encoded calcium
indicator in anesthetized and freely moving mice. Front Neural Circuits 4, 9.
Mao, T., O’Connor, D.H., Scheuss, V., Nakai, J., and Svoboda, K. (2008).
Characterization and subcellular targeting of GCaMP-type geneticallyencoded
calcium indicators. PLoS ONE 3, e1796.
Marshel, J.H., Garrett, M.E., Nauhaus, I., and Callaway, E.M. (2011).
Functional specialization of seven mouse visual cortical areas. Neuron 72,
1040–1054.
Mittmann, W., Wallace, D.J., Czubayko, U., Herb, J.T., Schaefer, A.T., Looger,
L.L., Denk, W., and Kerr, J.N. (2011). Two-photon calcium imaging of evoked
activity from L5 somatosensory neurons in vivo. Nat. Neurosci. 14, 1089–1093.
Miyawaki, A., Llopis, J., Heim, R., McCaffery, J.M., Adams, J.A., Ikura, M., and
Tsien, R.Y. (1997). Fluorescent indicators for Ca2+ based on green fluorescent
proteins and calmodulin. Nature 388, 882–887.
Muto, A., Ohkura, M., Kotani, T., Higashijima, S., Nakai, J., and Kawakami, K.
(2011). Genetic visualization with an improved GCaMP calcium indicator
reveals spatiotemporal activation of the spinal motor neurons in zebrafish.
Proc. Natl. Acad. Sci. USA 108, 5425–5430.
Nakai, J., Ohkura, M., and Imoto, K. (2001). A high signal-to-noise Ca(2+)
probe composed of a single green fluorescent protein. Nat. Biotechnol. 19,
137–141.
O’Connor, D.H., Peron, S.P., Huber, D., and Svoboda, K. (2010). Neural activity
in barrel cortex underlying vibrissa-based object localization in mice. Neuron
67, 1048–1061.
Ohkura, M., Matsuzaki, M., Kasai, H., Imoto, K., and Nakai, J. (2005).
Genetically encoded bright Ca2+ probe applicable for dynamic Ca2+ imaging
of dendritic spines. Anal. Chem. 77, 5861–5869.
Osakada, F., Mori, T., Cetin, A.H., Marshel, J.H., Virgen, B., and Callaway, E.M.
(2011). New rabies virus variants for monitoring and manipulating activity and
gene expression in defined neural circuits. Neuron 71, 617–631.
Pec¸ a, J., Feliciano, C., Ting, J.T., Wang, W., Wells, M.F., Venkatraman, T.N.,
Lascola, C.D., Fu, Z., and Feng, G. (2011). Shank3 mutant mice display
autistic-like behaviours and striatal dysfunction. Nature 472, 437–442.
Pologruto, T.A., Yasuda, R., and Svoboda, K. (2004). Monitoring neural activity
and [Ca2+] with genetically encoded Ca2+ indicators. J. Neurosci. 24, 9572–
9579.
Reiff, D.F., Ihring, A., Guerrero, G., Isacoff, E.Y., Joesch, M., Nakai, J., and
Borst, A. (2005). In vivo performance of genetically encoded indicators of
neural activity in flies. J. Neurosci. 25, 4766–4778.
Rothschild, G., Nelken, I., and Mizrahi, A. (2010). Functional organization and
population dynamics in the mouse primary auditory cortex. Nat. Neurosci. 13,
353–360.
Rubin, B.D., and Katz, L.C. (1999). Optical imaging of odorant representations
in the mammalian olfactory bulb. Neuron 23, 499–511.
Souslova, E.A., Belousov, V.V., Lock, J.G., Stro¨ mblad, S., Kasparov, S.,
Bolshakov, A.P., Pinelis, V.G., Labas, Y.A., Lukyanov, S., Mayr, L.M., and
Chudakov, D.M. (2007). Single fluorescent protein-based Ca2+ sensors with
increased dynamic range. BMC Biotechnol. 7, 37.
Spors, H., and Grinvald, A. (2002). Spatio-temporal dynamics of odor representations
in the mammalian olfactory bulb. Neuron 34, 301–315.
Szymczak, A.L., Workman, C.J., Wang, Y., Vignali, K.M., Dilioglou, S., Vanin,
E.F., and Vignali, D.A. (2004). Correction of multi-gene deficiency in vivo using
Neuron
Thy1-GCaMP Transgenic Mice
Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc. 307
a single ‘self-cleaving’ 2A peptide-based retroviral vector. Nat. Biotechnol. 22,
589–594.
Tallini, Y.N., Ohkura, M., Choi, B.R., Ji, G., Imoto, K., Doran, R., Lee, J., Plan,
P., Wilson, J., Xin, H.B., et al. (2006). Imaging cellular signals in the heart in vivo:
Cardiac expression of the high-signal Ca2+ indicator GCaMP2. Proc. Natl.
Acad. Sci. USA 103, 4753–4758.
Tian, L., Hires, S.A., Mao, T., Huber, D., Chiappe, M.E., Chalasani, S.H.,
Petreanu, L., Akerboom, J., McKinney, S.A., Schreiter, E.R., et al. (2009).
Imaging neural activity in worms, flies and mice with improved GCaMP calcium
indicators. Nat. Methods 6, 875–881.
Varshavsky, A. (2011). The N-end rule pathway and regulation by proteolysis.
Protein Sci. Published online June 1, 2011. http://dx.doi.org/10.1002/pro.666.
Verhagen, J.V., Wesson, D.W., Netoff, T.I., White, J.A., and Wachowiak, M.
(2007). Sniffing controls an adaptive filter of sensory input to the olfactory
bulb. Nat. Neurosci. 10, 631–639.
Wachowiak, M., and Cohen, L.B. (2001). Representation of odorants by
receptor neuron input to the mouse olfactory bulb. Neuron 32, 723–735.
Wallace, D.J., Meyer zum Alten Borgloh, S., Astori, S., Yang, Y., Bausen, M.,
Ku¨ gler, S., Palmer, A.E., Tsien, R.Y., Sprengel, R., Kerr, J.N., et al. (2008).
Single-spike detection in vitro and in vivo with a genetic Ca2+ sensor. Nat.
Methods 5, 797–804.
Warp, E., Agarwal, G., Wyart, C., Friedmann, D., Oldfield, C.S., Conner, A., Del
Bene, F., Arrenberg, A.B., Baier, H., and Isacoff, E.Y. (2012). Emergence of
patterned activity in the developing zebrafish spinal cord. Curr. Biol. 22,
93–102.
Yamada, Y., Michikawa, T., Hashimoto, M., Horikawa, K., Nagai, T., Miyawaki,
A., Ha¨ usser, M., and Mikoshiba, K. (2011). Quantitative comparison of genetically
encoded Ca indicators in cortical pyramidal cells and cerebellar Purkinje
cells. Front Cell Neurosci. 5, 18.
Yang, G., Pan, F., Parkhurst, C.N., Grutzendler, J., and Gan, W.B. (2010).
Thinned-skull cranial window technique for long-term imaging of the cortex
in live mice. Nat. Protoc. 5, 201–208.
Yasuda, R., Nimchinsky, E.A., Scheuss, V., Pologruto, T.A., Oertner, T.G.,
Sabatini, B.L., and Svoboda, K. (2004). Imaging calcium concentration
dynamics in small neuronal compartments. Sci. STKE 2004, pl5.
Young, P., Qiu, L., Wang, D., Zhao, S., Gross, J., and Feng, G. (2008).
Single-neuron labeling with inducible Cre-mediated knockout in transgenic
mice. Nat. Neurosci. 11, 721–728.
Zariwala, H.A., Borghuis, B.G., Hoogland, T.M., Madisen, L., Tian, L.,
De Zeeuw, C.I., Zeng, H., Looger, L.L., Svoboda, K., and Chen, T.W. (2012).
A Cre-dependent GCaMP3 reporter mouse for neuronal imaging in vivo.
J. Neurosci. 32, 3131–3141.
Zhao, S., Ting, J.T., Atallah, H.E., Qiu, L., Tan, J., Gloss, B., Augustine, G.J.,
Deisseroth, K., Luo, M., Graybiel, A.M., and Feng, G. (2011a). Cell type–
specific channelrhodopsin-2 transgenic mice for optogenetic dissection of
neural circuitry function. Nat. Methods 8, 745–752.
Zhao, Y., Araki, S., Wu, J., Teramoto, T., Chang, Y.F., Nakano, M., Abdelfattah,
A.S., Fujiwara, M., Ishihara, T., Nagai, T., and Campbell, R.E. (2011b). An
expanded palette of genetically encoded Ca2+
indicators. Science 333,
1888–1891.
Neuron
Thy1-GCaMP Transgenic Mice
308 Neuron 76, 297–308, October 18, 2012 ª2012 Elsevier Inc.