Improvements for recording retinal function with Microelectrode Arrays.

A microelectrode array (MEA) is a configuration of multiple electrodes that enables the concurrent targeting of multiple sites for extracellular recording and stimulation. By utilizing light pulses or electrical stimulations, MEA recordings unveil the complex patterns of electrical activities that arise from the signaling processes within the retinal network. Here, we present a stepwise approach for using microelectrode arrays (MEAs) for recording action potentials from the mouse retina in response to electrical and light stimuli. We provide detailed techniques accompanied by description of a custom optical system, example recordings, troubleshooting guidelines, and data processing methods including spike sorting and code resources for analyzing light and electrical responses. The comprehensive nature of this paper aims to guide researchers in utilizing MEAs effectively for investigating retinal functionality. In particular, it can be easy to have a MEA experiment fail, but hard to identify the source of the failure. This paper is meant to demystify that process. It includes:•A description of MEA setup, recording, and spike data validation.•A troubleshooting guide for common failure modes in MEA recordings from mouse retina.•Spike detection and sorting to precisely extract distinctive action potential.

Classification of pseudocalcium visual responses from mouse retinal ganglion cells.

Recently, a detailed catalog of 32 retinal ganglion cell (RGC) visual response patterns in mouse has emerged. However, the 10,000 samples required for this catalog-based on fluorescent signals from a calcium indicator dye-are much harder to acquire from the extracellular spike train recordings underlying our bionic vision research. Therefore, we sought to convert spike trains into pseudocalcium signals so that our data could be directly matched to the 32 predefined, calcium signal-based groups. A microelectrode array (MEA) was used to record spike trains from mouse RGCs of 29 retinas. Visual stimuli were adapted from the Baden et al. study; including moving bars, full-field contrast and temporal frequency chirps, and black-white and UV-green color flashes. Spike train histograms were converted into pseudocalcium traces with an OGB-1 convolution kernel. Response features were extracted using sparse principal components analysis to match each RGC to one of the 32 RGC groups. These responses mapped onto of the 32 previously described groups; however, some of the groups remained unmatched. Thus, adaptation of the Baden et al. methodology for MEA recordings of spike trains instead of calcium recordings was partially successful. Different classification methods, however, will be needed to define clear RGC groups from MEA data for our bionic vision research. Nevertheless, others may pursue a pseudocalcium approach to reconcile spike trains with calcium signals. This work will help to guide them on the limitations and potential pitfalls of such an approach.

Spike-triggered average electrical stimuli as input filters for bionic vision-a perspective.

Bionic retinal implants are gaining acceptance in the treatment of blindness from degenerative diseases including retinitis pigmentosa and macular degeneration. OBJECTIVE: A current obstacle to the improved performance of such implants is the difficulty of comparing the results of disparate experiments. Another obstacle is the current difficulty in selectively activating the many different retinal ganglion cell types that are used as separate pathways for visual information to the brain. APPROACH: To address these obstacles, we propose a modelling framework based on white noise stimulation and reverse correlation. In this perspective, we first outline early developments in visual retinal physiology leading up to the implementation of white noise stimuli and spike-triggered averaging. We then review recent efforts to adapt the white noise method for electrical stimulation of the retina and some of the nuances of this approach. MAIN RESULTS: Based on such white noise methods, we describe a modelling framework whereby the effect of any arbitrary electrical stimulus on a ganglion cell's neural code can be better understood. SIGNIFICANCE: This framework should additionally disentangle the effects of stimulation on photoreceptor, bipolar cell and retinal ganglion cell-ultimately supporting selective stimulation of specific ganglion cell types for a more nuanced bionic retinal implant. Finally, we point to upcoming considerations in this rapidly developing domain of research.

Correspondence between visual and electrical input filters of ON and OFF mouse retinal ganglion cells.

OBJECTIVE: Over the past two decades retinal prostheses have made major strides in restoring functional vision to patients blinded by diseases such as retinitis pigmentosa. Presently, implants use single pulses to activate the retina. Though this stimulation paradigm has proved beneficial to patients, an unresolved problem is the inability to selectively stimulate the on and off visual pathways. To this end our goal was to test, using white noise, voltage-controlled, cathodic, monophasic pulse stimulation, whether different retinal ganglion cell (RGC) types in the wild type retina have different electrical input filters. This is an important precursor to addressing pathway-selective stimulation. APPROACH: Using full-field visual flash and electrical and visual Gaussian noise stimulation, combined with the technique of spike-triggered averaging (STA), we calculate the electrical and visual input filters for different types of RGCs (classified as on, off or on-off based on their response to the flash stimuli). MAIN RESULTS: Examining the STAs, we found that the spiking activity of on cells during electrical stimulation correlates with a decrease in the voltage magnitude preceding a spike, while the spiking activity of off cells correlates with an increase in the voltage preceding a spike. No electrical preference was found for on-off cells. Comparing STAs of wild type and rd10 mice revealed narrower electrical STA deflections with shorter latencies in rd10. SIGNIFICANCE: This study is the first comparison of visual cell types and their corresponding temporal electrical input filters in the retina. The altered input filters in degenerated rd10 retinas are consistent with photoreceptor stimulation underlying visual type-specific electrical STA shapes in wild type retina. It is therefore conceivable that existing implants could target partially degenerated photoreceptors that have only lost their outer segments, but not somas, to selectively activate the on and off visual pathways.

Tickling the retina: integration of subthreshold electrical pulses can activate retinal neurons.

OBJECTIVE: The field of retinal prosthetics has made major progress over the last decade, restoring visual percepts to people suffering from retinitis pigmentosa. The stimulation pulses used by present implants are suprathreshold, meaning individual pulses are designed to activate the retina. In this paper we explore subthreshold pulse sequences as an alternate stimulation paradigm. Subthreshold pulses have the potential to address important open problems such as fading of visual percepts when patients are stimulated at moderate pulse repetition rates and the difficulty in preferentially stimulating different retinal pathways. APPROACH: As a first step in addressing these issues we used Gaussian white noise electrical stimulation combined with spike-triggered averaging to interrogate whether a subthreshold sequence of pulses can be used to activate the mouse retina. MAIN RESULTS: We demonstrate that the retinal network can integrate multiple subthreshold electrical stimuli under an experimental paradigm immediately relevant to retinal prostheses. Furthermore, these characteristic stimulus sequences varied in their shape and integration window length across the population of retinal ganglion cells. SIGNIFICANCE: Because the subthreshold sequences activate the retina at stimulation rates that would typically induce strong fading (25 Hz), such retinal 'tickling' has the potential to minimize the fading problem. Furthermore, the diversity found across the cell population in characteristic pulse sequences suggests that these sequences could be used to selectively address the different retinal pathways (e.g. ON versus OFF). Both of these outcomes may significantly improve visual perception in retinal implant patients.

Spectral features control temporal plasticity in auditory cortex.

Cortical responses are adjusted and optimized throughout life to meet changing behavioral demands and to compensate for peripheral damage. The cholinergic nucleus basalis (NB) gates cortical plasticity and focuses learning on behaviorally meaningful stimuli. By systematically varying the acoustic parameters of the sound paired with NB activation, we have previously shown that tone frequency and amplitude modulation rate alter the topography and selectivity of frequency tuning in primary auditory cortex. This result suggests that network-level rules operate in the cortex to guide reorganization based on specific features of the sensory input associated with NB activity. This report summarizes recent evidence that temporal response properties of cortical neurons are influenced by the spectral characteristics of sounds associated with cholinergic modulation. For example, repeated pairing of a spectrally complex (ripple) stimulus decreased the minimum response latency for the ripple, but lengthened the minimum latency for tones. Pairing a rapid train of tones with NB activation only increased the maximum following rate of cortical neurons when the carrier frequency of each train was randomly varied. These results suggest that spectral and temporal parameters of acoustic experiences interact to shape spectrotemporal selectivity in the cortex. Additional experiments with more complex stimuli are needed to clarify how the cortex learns natural sounds such as speech.