Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 54 (2), 205-18

In Vivo Light-Induced Activation of Neural Circuitry in Transgenic Mice Expressing channelrhodopsin-2


In Vivo Light-Induced Activation of Neural Circuitry in Transgenic Mice Expressing channelrhodopsin-2

Benjamin R Arenkiel et al. Neuron.


Channelrhodopsin-2 (ChR2) is a light-gated, cation-selective ion channel isolated from the green algae Chlamydomonas reinhardtii. Here, we report the generation of transgenic mice that express a ChR2-YFP fusion protein in the CNS for in vivo activation and mapping of neural circuits. Using focal illumination of the cerebral cortex and olfactory bulb, we demonstrate a highly reproducible, light-dependent activation of neurons and precise control of firing frequency in vivo. To test the feasibility of mapping neural circuits, we exploited the circuitry formed between the olfactory bulb and the piriform cortex in anesthetized mice. In the olfactory bulb, individual mitral cells fired action potentials in response to light, and their firing rate was not influenced by costimulated glomeruli. However, in piriform cortex, the activity of target neurons increased as larger areas of the bulb were illuminated to recruit additional glomeruli. These results support a model of olfactory processing that is dependent upon mitral cell convergence and integration onto cortical cells. More broadly, these findings demonstrate a system for precise manipulation of neural activity in the intact mammalian brain with light and illustrate the use of ChR2 mice in exploring functional connectivity of complex neural circuits in vivo.


Figure 1
Figure 1. The Thy1 Promoter Drives Transgenic Expression of ChR2-YFP in Subsets of Neurons in the Central Nervous System
(A) Diagram of the Thy1-ChR2-YFP transgene construct. (B) Whole-brain image of ChR2-YFP expression in Thy1 transgenic mouse brain, illustrating the coronal planes of section shown in (C)–(E). (C) Coronal section through the caudal end of the Thy1-ChR2-YFP transgenic mouse brain. CB, cerebellum; BS, brainstem. (D) Midcoronal section through the Thy1-ChR2-YFP transgenic mouse brain. CTX, cortex; HP, hippocampus. (E) Immunoenhanced section using an anti-GFP antibody to amplify ChR2-YFP detection in the mitral cells of the olfactory bulb. The asterisk illustrates apical dendrites extending from mitral cell bodies, forming tufts in the glomerular layer. (F) DAPI stain of the section shown in (E), highlighting the cell layers of the olfactory bulb. GL, glomerular layer; EPL, external plexiform layer; ML, mitral cell layer. (G) Serial z sections taken at 2 μm intervals through the olfactory bulb of Thy1-ChR2-YFP mice showing the localization of ChR2-YFP at the plasma membrane of mitral cells. Top panels show ChR2-YFP expression in serial z planes of section. Bottom panels show corresponding planes of section with colabeling of mitral cell bodies and lateral dendrites with an antibody directed against the mitral cell-expressed 5-HT2A receptor (purple) and ChR2-YFP (green). The arrowhead points to ChR2-YFP expression on the membrane of a lateral dendrite. Arrows point to ChR2-YFP expression on the membrane of mitral cell bodies. EPL, External plexiform layer; ML, mitral cell layer; GC, granule cell layer. (H–J) Within the olfactory bulb, ChR2-YFP is selectively expressed in mitral cells. (H) High-magnification view of ChR2-YFP expression in the mitral cell layer. (I) Colocalization of membrane-bound ChR2-YFP (green) and 5-HT2A receptor (blue), a marker of mitral cells. (J) ChR2-YFP is not expressed in GAD67-positive granule cells (red). Note the presence of membrane-localized ChR2-YFP on the mitral cell (long arrows) with no ChR2-YFP present on the smaller GAD67-positive granule cell (arrowheads). MC, mitral cell; GC, granule cell. Scale bars, 2.5 mm (B); 1 mm (C and D); 50 μm (E and F); 10 μm (G); 10 μm (H–J).
Figure 2
Figure 2. Transgenically Expressed ChR2 Produces Rapid, Light-Evoked, Cell Type-Specific Responses in Acute Brain Slices
(A) Illumination with 488 nm light generates rapid photocurrents with both transient and sustained components in a voltage-clamped mitral cell from a main olfactory bulb slice. Currents are shown in response to graded illumination intensities of 2 (trace 1), 4 (trace 2), and 20 (trace 3) mW/mm2. (B) Brief 5 ms illumination produces reproducible EPSP-like events in a mitral cell recorded in current clamp, and these subthreshold responses were graded with light intensity (black and dark gray traces). Higher light intensity produces short latency action potential firing (light gray traces, truncated). The prominent afterhyperpolarization (AHP) observed following the spike is typical of mitral cells. Black traces, low intensity light (2 mW/mm2); dark gray, medium intensity (4 mW/mm2); light gray, high intensity (20 mW/mm2). Three repeated responses to illumination at each intensity are overlaid. (C) Rapid photocurrent kinetics (black, bottom traces) permit precise and reproducible control of action potential firing (gray, top traces). Left, overlay of six single spikes evoked by 5 ms light pulses. Right, overlay of three spike trains evoked by a train of pulses (5 ms, 40 Hz). Thick gray bars indicate illumination periods. (D) (a) Mitral/tufted (M/T) cells in the main olfactory bulb consistently show ChR2-mediated photocurrents (black, middle panel) that drive robust neuronal firing (gray, top panel). (b and c) Light-evoked activity is absent from MOB periglomerular (PG) and granule cell (GC) interneurons (top, middle panels), although spiking is still readily driven by current injection (bottom panel). (E) Peak light-evoked current amplitudes for different neuronal types in the MOB. ChR2 photocurrents are observed only in mitral/tufted (M/T) cells in the MOB. PG, periglomerular cells; GC, granule cells. Bars show mean ± SEM.
Figure 3
Figure 3. In Vivo Photoactivation of Layer 5 Cortical Neurons Expressing ChR2
(A) Illustration of the in vivo light delivery and electrophysiological recording system. (B–D) Example traces recording spike activity of a single unit in response to different light stimulus patterns. (B) Spike responses to three 500 ms exposures spaced 500 ms apart. (C) Spike responses to a single 4 s light exposure. (D) Spike responses to three 20 Hz stimulations. Note the precise temporal correspondence between delivered illumination and recorded spikes.
Figure 4
Figure 4. ChR2 Generates Temporally Precise Light-Evoked Neuronal Activity In Vivo
(A) Brief light pulses generate rapid and consistent firing of cortical pyramidal neurons. Top, extracellular spikes from five stimulus presentations. Middle, raster plot of the same neuron’s response to 100 repeated presentations. Bottom, histogram of spike times aligned to light onset shows reliability of timing across trials. Gray bars indicate time of illumination. (B) Histogram of light-induced response latencies from 11 neurons. The mean time to first spike is 9.9 ± 2.2 ms (±SD). (C) Individual neurons reliably follow stimulus trains across a wide range of frequencies. Left, raster plots of responses to trains of stimuli repeated at frequencies ranging from 10 to 50 Hz (10 trials each). Right, spike generation is very reliable for stimulation frequencies from 10 to 40 Hz (n = 8; n = 5 at 50 Hz). (D and E) Continuous illumination generates sustained high-frequency firing over extended time periods. (D) A 500 ms light presentation evoked a mean firing rate of 64 ± 17 Hz. n = 4 neurons. (E) Top, spike train from a cell illuminated for 1 s. Bottom, firing rates plotted for different periods of illumination. Lines in the graph represent different durations of illumination. Solid line, 100 ms; dotted line, 200 ms; dashed line, 500 ms; wide-dashed line, 1 s. Bars in (C) and (D) show ± SEM.
Figure 5
Figure 5. Illuminating the Dorsal Surface of the Olfactory Bulb Drives Mitral Cell Spike Activity in ChR2-YFP Mice
(A) Exposed dorsal olfactory bulb imaged for ChR2-YFP emission under broad illumination. (B) Example electrophysiological trace from a mitral cell activated by light, highlighting the respiration-coupled spike activity that is significantly increased during illumination. (C) Image of exposed dorsal olfactory bulb showing the area of light stimulation achieved using a 100 μm diameter fiber optic. (D) Spike response of a single mitral unit following four 250 ms light pulses. (E) Image of exposed dorsal olfactory bulb showing the area of light stimulation achieved using a 300 μm diameter fiber optic bundle. Scale bars, 1 mm (A, C, and E). (F) Spike responses from two distinct units recorded from the same site as in (D). The recruitment of another cell’s activity (green spike trace) does not influence the original cell’s firing properties (black spike trace, seen also in [D]). (G and H) Traces represent spike shape profiles used for sorting and unit identification. The black waveform represents the first unit initially isolated under narrow light, whereas the overlaid green waveform corresponds to the second unit recruited upon broad light stimulation.
Figure 6
Figure 6. Individual Mitral Cell Responses are Unaltered by Increasing the Area of Olfactory Bulb Photostimulation
(A and B) Peristimulus time histograms (PSTHs) assembled from multitrial light stimulations of the same cell under narrow 100 μm diameter (A) or broad 300 μm diameter (B) illumination of the dorsal olfactory bulb. Raster sweeps used to generate the PSTHs are shown at the top of each panel. (C) Distribution plot of cell responses under either narrow or broad light stimulation. Shown are 20 independent trials for both narrow and broad stimulations overlaid for each independent cell analyzed. The two red asterisks denote cells that were found to be significantly different in narrow versus broad illumination responses using the Kolmogorov-Smirnov nonparametric two-sample test constrained to a p value of less than 0.05.
Figure 7
Figure 7. Broad Area Light Stimulation of the Olfactory Bulb Is Required to Drive Activity in the Piriform Cortex
(A–C) Images of exposed dorsal olfactory bulb showing the relative areas of light stimulation using (A) 600 μm, (B) 300 μm, or (C) 100 μm spot sizes for illumination. Scale bars, 1 mm (A, B, and C). (D) Data represent mean ± SEM of the fold change in piriform cortical cell firing rate above baseline induced by stimulating different areas of the bulb (600 μm diameter spot, 7.01-fold ± 0.78-fold; 300 μm, 2.40-fold ± 0.24-fold; 100 μm 1.26 ± 0.14; n = 20 for each). *p < 0.001, one-tailed t test. (E) Individual cortical cell responses using 600 μm, 300 μm, or 100 μm diameter fiber optics. Baseline firing rates are indicated. (F–H) Representative PSTHs (lower panels) and action potential raster plots (top panels) showing responses of piriform cortical neurons to different areas of bulb illumination at time zero using (F) 600 μm, (G) 300 μm, or (H) 100 μm diameter fiber optics.

Similar articles

See all similar articles

Cited by 310 PubMed Central articles

See all "Cited by" articles

Publication types

MeSH terms

LinkOut - more resources