3. Optrode-Microelectrode Array Device for Optical Stimulation and Recording of Neural Microcircuits in Freely Moving Animals

Studying brain function and its local circuit dynamics requires neural interfaces that can record and stimulate the brain with high spatiotemporal resolution. Optogenetics, a technique that genetically targets specific neurons to express light sensitive channel proteins, provides the capability to control central nervous system (CNS) neuronal activity in mammals with millisecond time precision. This technique enables precise optical stimulation of neurons and simultaneous monitoring of neural response by electrophysiological means, both in the vicinity of and distant to the stimulation site.

We previously demonstrated, in-vitro, the dual capability (optical delivery and electrical recording) while testing a novel hybrid device (optrode-MEA), which incorporates a tapered coaxial optical electrode (‘optrode’) and a 100 element microelectrode array (MEA). Here we report a fully chronic implant of a new version of this device in ChR2-expressing rats, and demonstrate its use in freely moving animals over periods up to eight months. In its present configuration, we show the device delivering optical excitation to a single cortical site while mapping neural response from surrounding 30 channels of the 6×6 element MEA, thereby enabling recording of optically modulated single-unit and local field potential (LFP) activity across several millimeters of the neocortical landscape. 
This work has been published in Journal of Neural Engineering, 'Integrated Device for Combined Optical Neuromodulation and Electrical Recording for Chronic In Vivo Applications', Jing Wang, Fabien Wagner, David A. Borton, Jiayi Zhang, Ilker Ozden, Rebecca D. Burwell, Arto V. Nurmikko, Rick van Wagenen, Ilka Diester, and Karl Deisseroth.


Figure below: Overview of the optrode-MEA. (a) Image of the 6×6 multi-electrode array device with one element being replaced by an optrode (arrow). The spacing between neighbor electrodes is 400 μm and electrode shank length is 1 mm. (b) A close-up view of the optrode shows the laser light emitted from the tip of optrode. (c) Schematic of the optrode-MEA implant, which shows the cannulated tube used to guide the optrode as well as the injection needle.­­­ (d) One of the subjects with the optrode-MEA implanted. The fiber optics and headstage connector are protected by a cone-shaped plastic cylinder. The optical fiber can be coiled and secured inside the cone after each recording session.

Figure below: The ChR2 expressing volume and the optical excitation volume. (a) shows a fluoresence image of a coronal slice that resolves individual EYFP-opsin expressing neurons. The cell nuclei were indicated by DAPI staining. (b) The Monte Carlo simulation of photon counts (proportional to light intensity) distribution in brain tissue. The contours of 10% and 1% of the iso-intensity are indicated by yellow dots. The light intensity reaching the neighbor electrodes is 5 mW/mm2 estimated from the simulation. The output light from optrode is approximated as a point source and has divergent angle of 30 degrees. (c) The enlarged images to display cell bodies (left) and fibers (right) that project to subcortical structures.

Figure below: Representative examples of light activation of single units and LFPs. (a) Raster plots and PSTHs from a light-responsive cell after 5 months of implantation. Blue ticks indicate the pulse train stimulation (473 nm, 8 Hz pulse frequency and 20 ms pulse duration). A sample of overlaid spike waveforms is shown on the right.  Lower panel shows that trial averaged LFP has negative deflections with a positive rebound in response to each pulse stimulation. (b) Power spectrogram and power density plot of optically modulated LFP. Note that the power is significantly enhanced at the light stimulation frequency and its harmonics.  (c) LFP power around the pulse stimulation frequency as a function of estimated light intensity at the recording electrode site. The arrow indicates the light intensity used for neural stimulation in the rest of the article. (d) Optically induced LFP response and comparison with control recordings. “De-sensitization” is shown as the amplitude of negative peak decreases in response to repeatedly optical stimulation.


Figure below: Examples of single unit recordings under light modulation. (a) Raster plots and PSTHs from a neuron activated by the pulse train (blue ticks). The spike count histogram (right panel) shows the time-locked spikes evoked by the stimulation. Samples of selected spike waveforms are plotted in the right. (b) A unit showing inhibitory response to the same stimulation. (c) Examples of activation patterns of single units at selected sites across the microelectrode array (insets indicate locations), in response to 500 ms continuous light stimulation. The last plot is an example of LFP response.

Figure below: Spatially and temporally resolved neuronal activities from a large cortical area.  (a) The mapping of averaged (N=100) spike waveforms on each input channel obtained from a sample recording session. (b) The fractional change of firing rate (see details in the text) of single units as a function of the distance between the recording electrode and stimulating optrode. The firing rate data were collected from same set of isolated units as in (a). (c) Pulse-triggered LFPs at various locations show both proximal and distal field potential in response to the 1ms pulse stimulation. Vertical line indicates the onset of light pulse and red dots indicate the negative peaks. Output power of optrode is estimated to be 1.2mW. The distances between recording electrode and the optrode are labeled on the individual traces. (d) Mapping the full-band power of pulse-triggered LFP ­­over 2.4×2.4 mm2 cortical area. Black colors indicate the sites without recordings. All the PSTH and LFP signals in figure 4 are averaged over 30 trials.