Red-shifted optogenetic excitation: a tool for fast neural control derived from Volvox carteri

Abstract

The introduction of two microbial opsin-based tools, channelrhodopsin-2 (ChR2) and halorhodopsin (NpHR), to neuroscience has generated interest in fast, multimodal, cell type-specific neural circuit control. Here we describe a cation-conducting channelrhodopsin (VChR1) from Volvox carteri that can drive spiking at 589 nm, with excitation maximum red-shifted approximately 70 nm compared with ChR2. These results demonstrate fast photostimulation with yellow light, thereby defining a functionally distinct third category of microbial rhodopsin proteins.

Figure 1

Identification and characterization of VChR1. (a) Volvox carteri colony showing several thousand small somatic cells and 16 gonidia (scale bar, 125 μm). (b) Photocurrents in VChR1-expressing HEK293 cells under varying light intensity (traces represent averages of six recordings). (c) I-V relationship in HEK293 cells under varying extracellular pH. External and internal [Na⁺] were held constant at 140 mM. (d) RSB structure shows the predicted changes in environmental polarity as partial charges (δ–; top). Sequence comparison of ChR2, ChR1 and VChR1 is shown with residues highlighted that are predicted to define the RSB Schiff base counterion (blue), modulate electrical potential of the RSB binding pocket (red) or modulate RSB potential via long-range coupling (green). (e) Ion flux characteristics of VChR1 in HEK293 cells (n = 4; see Supplementary Methods). (f) Action spectra for VChR1 and ChR2 in oocytes; wavelengths used in later figures are shaded. Values are mean ± s.e.m.

Figure 2

VChR1 permits red-shifted neural photostimulation. (a) Cultured hippocampal neuron expressing VChR1-EYFP. A lentiviral vector carrying the αCaMKII promoter was used to drive neuron-specific expression (scale bar, 75 μm). (b) Inward photocurrent in a voltage-clamped VChR1-EYFP neuron evoked by 1 s of 531- or 589-nm light (colored bars). (c) Summary data of mean current evoked by 531- or 589-nm light over increasing light intensities (n = 10). (d) Voltage traces showing spikes in a current-clamped hippocampal neuron evoked by 5-, 10-, 20- or 30-Hz trains of 5-ms light pulses (orange dashes). (e) Population data summarizing light-spike fidelity in current-clamped neurons illuminated with 531- or 589-nm light pulses (n = 12). (f) Barrage summation of subthreshold synaptic-like events evoked with increasing light-pulse frequency. (g,h) Membrane resistances and resting membrane potentials of neurons expressing VChR1-EYFP (n = 10), not expressing VChR1-EYFP (wild type, n = 10), or expressing VChR1-EYFP and measured 24 h after exposure to activating light (n = 10). Values are mean ± s.e.m.

Figure 3

Separable channels of optogenetic excitation. (a) Voltage traces showing that ChR2 and VChR1 can be selectively activated by 406- (1.2 mW mm⁻²) and 589-nm (2.3 mW mm⁻²) light, respectively. Trains of 5-ms light flashes were delivered at 5 Hz. (b) Population data comparing activation of ChR2 by 406- and 589-nm light at three different light intensities (n = 7). Trains of 5-ms light pulses were delivered at 5 Hz. (c) Population data comparing activation of VChR1 by 406- and 589-nm light at three different light intensities (n = 10). Values are mean ± s.e.m.