Multiple-color optical activation, silencing, and desynchronization of neural activity, with single-spike temporal resolution

Abstract

The quest to determine how precise neural activity patterns mediate computation, behavior, and pathology would be greatly aided by a set of tools for reliably activating and inactivating genetically targeted neurons, in a temporally precise and rapidly reversible fashion. Having earlier adapted a light-activated cation channel, channelrhodopsin-2 (ChR2), for allowing neurons to be stimulated by blue light, we searched for a complementary tool that would enable optical neuronal inhibition, driven by light of a second color. Here we report that targeting the codon-optimized form of the light-driven chloride pump halorhodopsin from the archaebacterium Natronomas pharaonis (hereafter abbreviated Halo) to genetically-specified neurons enables them to be silenced reliably, and reversibly, by millisecond-timescale pulses of yellow light. We show that trains of yellow and blue light pulses can drive high-fidelity sequences of hyperpolarizations and depolarizations in neurons simultaneously expressing yellow light-driven Halo and blue light-driven ChR2, allowing for the first time manipulations of neural synchrony without perturbation of other parameters such as spiking rates. The Halo/ChR2 system thus constitutes a powerful toolbox for multichannel photoinhibition and photostimulation of virally or transgenically targeted neural circuits without need for exogenous chemicals, enabling systematic analysis and engineering of the brain, and quantitative bioengineering of excitable cells.

Figure 1. Millisecond-timescale Halo-mediated neuronal hyperpolarization, elicited by pulses of yellow light.

(A) A representative cultured hippocampal neuron expressing mammalian codon-optimized N. pharaonis halorhodopsin (abbreviated Halo) fused to GFP, under the CaMKII promoter. Scale bar, 20 µm. (B) Neuronal currents elicited by optical activation of Halo. Left, representative outward currents elicited by two 1-second pulses of yellow (560±27.5 nm) light (∼10 mW/mm²) in a voltage-clamped neuron held at −70mV. Right, population data for n = 22 neurons. In this and subsequent figures, gray bars represent mean ± standard deviation unless otherwise indicated. Yellow bars in this and subsequent figures represent the period of yellow light exposure. (C) Kinetic properties of yellow light-elicited, Halo-mediated currents from voltage-clamped neurons. (i), 15–85% current onset time; (ii), 85–15% offset time. For each measurement, data is presented from neurons held at −70 mV, −30 mV and+10 mV (left to right). In this panel, gray bars represent mean ± standard error of the mean (S.E.M.). (D) Neuronal hyperpolarizations elicited by optical activation of Halo. Left, representative membrane voltage hyperpolarizations elicited by two 1-second pulses of yellow light, in a current-clamped neuron held at resting membrane potential. Right, population data for n = 19 neurons. (E) Kinetic properties of yellow light-elicited, Halo-mediated hyperpolarizations from current-clamped neurons, including both 15–85% voltage change onset time and 85–15% offset time.

Figure 2. Safety of Halo in cultured hippocampal neurons.

(A) Membrane resting potential of Halo-expressing vs. wild-type neurons (n = 19 each). (B) Membrane resistance of Halo-expressing vs. wild-type neurons. (C) Fraction of Halo-expressing (n = 22) vs. wild-type (n = 308) cultured neurons whose nuclei stained positive for the membrane-integrity assessing DNA stain ethidium homodimer-1.

Figure 3. Halo-mediated naturalistic trains of inhibitory events.

(A) Three voltage traces of a representative current-clamped hippocampal neuron, exposed to a Poisson train of yellow light pulses. Each light pulse lasts 10 ms, and the Poisson train has a mean inter-pulse interval of λ = 100 ms. (B) Voltage traces of three different representative current-clamped neurons exposed to the same Poisson train of light pulses (λ = 100 ms). (C) Properties of hyperpolarization events elicited by Poisson trains with inter-pulse interval λ = 100 ms (i, ii) and λ = 200 ms (iii, iv), plotted versus onset time of each light pulse. Plots (i) and (iii) show the peak amplitude of each hyperpolarization event (black symbols), as well as the across-trials standard deviation of these amplitude values across ten trials (gray symbols). Plots (ii) and (iv) show the latency between the onset time of the light pulse and the time of the hyperpolarization peak (black symbols), as well as the across-trials standard deviation of these timing values across ten trials (gray symbols). In this panel, plotted points are across-neuron mean±S.E.M. (n = 5 neurons). (D) Comparison of the peak hyperpolarization (i) and the time-to-peak (ii) data between the beginning (first5) and end (last5) of each Poisson train, for the n = 5 neurons described in Fig. 3C. In (i): for each neuron, the average of the first 5 or last 5 hyperpolarization peaks (black) or the across-trials standard deviation of these amplitude values (gray) was first computed, then the across-neuron mean±S.E.M. was plotted. In (ii): for each neuron, the average of the first 5 or last 5 times-to-peak (black) or the across-trials standard deviation of these times-to-peak (gray) were first computed, then the across-neuron mean±S.E.M. was plotted. For (ii), the gray bars were stacked on top of the black ones for ease of visualization.

Figure 4. Halo-mediated silencing of neuronal spiking.

(A) Light-driven spike blockade, demonstrated for a representative hippocampal neuron. Top (“I-injection”), neuronal firing of 20 spikes at 5 Hz, induced by pulsed somatic current injection (∼300 pA, 4 ms). Middle (“Light”), membrane hyperpolarization induced by two periods of yellow light, timed so as to be capable of blocking spikes 7–11 and spike 17, out of the train of 20 spikes. Bottom (“I-injection±Light”), yellow light drives Halo to block neuron spiking (note absence of spikes 7–11 and of spike 17), while leaving spikes elicited during periods of darkness largely intact. (B) Population data (n = 6 neurons) for light-driven, Halo-mediated spike blockade, showing high spike probability during periods of darkness (spikes 1–6, 12–16, and 18–20), and low spike probability during periods of yellow light illumination (spikes 7–11 and spike 17). Error bars (S.E.M.) are smaller than the points plotted.

Figure 5. Blue light facilitates optimal Halo function.

(A) (i) Timecourse of Halo-mediated hyperpolarizations in a representative current-clamped hippocampal neuron during 15 seconds of continuous yellow light, followed by four 1-second test pulses of yellow light (one every 30 seconds, starting 10 seconds after the end of the first 15-second period of yellow light). (ii) Timecourse of Halo-mediated hyperpolarization for the same cell exhibited in (i), but when Halo function is facilitated by a 400-ms pulse of blue light in between the 15-second period of yellow light and the first 1-second test pulse. (B) Population data for blue-light facilitation of Halo recovery (n = 8 neurons). Plotted are the hyperpolarizations elicited by the four 1-second test pulses of yellow light, normalized to the peak hyperpolarization induced by the original 15-second yellow light pulse. Dots represent mean±S.E.M. Black dots represent experiments when no blue light pulse was delivered (as in Fig. 5Ai.). Open blue dots represent experiments when 400 ms of blue light was delivered to facilitate recovery (as in Fig. 5Aii.).

Figure 6. Bi-directional optical control of voltage with blue and yellow light pulses.

(A) Action spectrum for ChR2 (blue, adapted from [71]), overlaid with absorption spectrum for N. pharaonis halorhodopsin (orange, adapted from [32]). Each spectrum is normalized to its own peak, for ease of comparison. Gray bars denote bandpass windows of commercially-available Chroma HQ450/50× and HQ590/55× filters, which may be useful for high-fidelity fast-wavelength switching between ChR2 photostimulation and Halo photoinhibition. (B) Co-expression of ChR2-mCherry (left) and Halo-GFP (middle) in a single representative cell expressing both (right, overlay). Scale bar, 20 µm. (C) Hyperpolarization and depolarization events elicited in a single representative neuron, by two interleaved 2.5 Hz trains of yellow and blue light pulses (50 ms duration), denoted by bars of respective coloration. (D) Hyperpolarization and depolarization events induced in a representative neuron by a Poisson train (mean inter-pulse interval λ = 100 ms) of alternating pulses of yellow and blue light (10 ms duration), denoted by appropriate colors.

Figure 7. Multichannel optical disruption of precise spike timing, without alteration of spike rate.

(A) Optical disruption of spike timing, without alteration of spike rate, for a representative neuron expressing both ChR2 and Halo. (i), stimulus traces showing subsegments of the somatically injected filtered Gaussian white noise current used in all these experiments (top), as well as of the Poisson train (mean inter-pulse interval λ = 100 ms) of alternating yellow and blue light pulses (bottom). (ii), twenty-trace overlays of voltage responses to the somatically injected white noise current, either with no light (top, black traces) or with delivery of a Poisson train of yellow and blue light pulses (bottom, green traces). (iii), spike raster plots for the traces shown in Fig. 7Aii. (iv), spike-timing histograms (bin size: 500 µs) for the rasters shown in Fig. 7Aiii. (B) Spike rates of neurons (n = 7) injected with filtered Gaussian white noise current, either with no light (left) or with concurrent delivery of a Poisson train of yellow and blue light pulses (right). Plotted is mean±S.E.M. (C) Cross-correlation between spike trains elicited from the same filtered Gaussian white noise current injection, played twice, when either both current injections were performed in the dark (black curve), or when one of the current injections was performed with concurrent delivery of a Poisson train of yellow and blue light pulses (green trace). Data is plotted as mean±S.E.M (averaged across n = 7 neurons). See Methods for details.