Spatiotemporal aspects of pulsed electrical stimuli on the responses of rabbit retinal ganglion cells

Targeted Stimulation of Retinal Ganglion Cells in Epiretinal Prostheses: A Multiscale Computational Study

Retinal prostheses aim at restoring partial sight to patients that are blind due to retinal degenerative diseases by electrically stimulating the surviving healthy retinal neurons. Ideally, the electrical stimulation of the retina is intended to induce localized, focused, percepts only; however, some epiretinal implant subjects have reported seeing elongated phosphenes in a single electrode stimulation due to the axonal activation of retinal ganglion cells (RGCs). This issue can be addressed by properly devising stimulation waveforms so that the possibility of inducing axonal activation of RGCs is minimized. While strategies to devise electrical stimulation waveforms to achieve a focal RGCs response have been reported in literature, the underlying mechanisms are not well understood. This article intends to address this gap; we developed morphologically and biophysically realistic computational models of two classified RGCs: D1-bistratified and A2-monostratified. Computational results suggest that the sodium channel band (SOCB) is less sensitive to modulations in stimulation parameters than the distal axon (DA), and DA stimulus threshold is less sensitive to physiological differences among RGCs. Therefore, over a range of RGCs distal axon diameters, short-pulse symmetric biphasic waveforms can enhance the stimulation threshold difference between the SOCB and the DA. Appropriately designed waveforms can avoid axonal activation of RGCs, implying a consequential reduction of undesired strikes in the visual field.

Stimulation strategies for selective activation of retinal ganglion cell soma and threshold reduction

OBJECTIVE

Retinal prosthetic implants restore partial vision to patients blinded due to outer retinal degeneration, using a camera-guided multielectrode array (MEA) that electrically stimulates surviving retinal neurons. Commercial epi-retinal prostheses use millisecond-scale charge-balanced, symmetric, cathodic-first biphasic pulses to depolarize retinal ganglion cells (RGCs) and bipolar cells (BCs), frequently creating oblong perceptions of light related to axonal activation of RGCs. Stimulation strategies that avoid axonal stimulation and decrease the threshold of targeted neurons may significantly improve prosthetic vision in terms of spatial resolution and power efficiency.

APPROACH

We developed a virus-transduced genetically encoded calcium indicator (GECI) GCaMP6f and microscopy platform for calcium imaging to record the neural activity from RGCs at single-cell resolution in wholemount retinas. Multiple stimulation paradigms were applied through a microelectrode array (MEA) with transparent indium tin oxide electrodes. The evoked neuronal activities were converted to corresponding 2D calcium imaging transient pattern and spatial threshold map to identify the ideal focal response which corresponds to optimal percept in patient.

MAIN RESULTS

The proposed optical system with GCaMP6f is capable of recording from population of mouse RGCs in real time during electrical stimulation with precise location information relative to the stimulation sites. Optimal duration and phase order of pulse were identified to avoid axonal stimulation and selectively activate targeted RGC somas, without requiring a significant increase in stimulation charge. Additionally, we show that reduced stimulus threshold can be achieved with the special design of asymmetric anodic-first pulse.

SIGNIFICANCE

Our findings support the possibility of manipulating the responses of RGCs through varying the stimulation waveform. Focal response can be achieved with relative short duration (⩽120 μs) pulses, and can be improved by reversing the standard phase order. The RGCs threshold can be significantly reduced by 33.3%-50% in terms of charge through applying hyperpolarizing pre-pulses with a 20:1 ratio (pre-pulse:stimulus pulse). The results support the future retinal prosthesis design that potentially forms more ideal shape perception with higher spatial resolution and power efficiency.

Differential electrical responses in retinal ganglion cell subtypes: effects of synaptic blockade and stimulating electrode location

OBJECTIVE

Visual prostheses have shown promising results in restoring visual perception to blind patients. The ability to differentially activate retinal ganglion cell (RGC) subtypes could further improve the efficacy of these medical devices.

APPROACH

Using whole-cell patch clamp, we investigated membrane potential differences between ON and OFF RGCs in the mouse retina when their synaptic inputs were blocked by synaptic blockers, and examined the differences in stimulation thresholds under such conditions. By injecting intracellular current, we further confirmed the relationship between RGC stimulation thresholds and resting membrane potentials (RMPs). In addition, we investigated the effects of stimulating electrode location on the differences in stimulation thresholds between ON and OFF RGCs.

MAIN RESULTS

With synaptic blockade, ON RGCs became significantly more hyperpolarized (from  -61.8  ±  1.4 mV to  -70.8  ±  1.6 mV), while OFF RGCs depolarized slightly (from  -60.5  ±  0.7 mV to  -58.6  ±  0.9 mV). RGC stimulation thresholds were negatively correlated with their RMPs (Pearson r value:  -0.5154; p-value: 0.0042). Thus, depriving ON RGCs of synaptic inputs significantly increased their thresholds (from 14.7  ±  1.3 µA to 22.3  ±  2.1 µA) over those of OFF RGCs (from 13.2  ±  0.7 µA to 13.1  ±  1.1 µA). However, with control solution, ON and OFF RGC stimulation thresholds were not significantly different. Finally, placement of the stimulating electrode away from the axon enhanced differences in stimulation thresholds between ON and OFF RGCs, facilitating preferential activation of OFF RGCs.

SIGNIFICANCE

Since ON and OFF RGCs have antagonistic responses to natural light, achieving differential RGC activation could convey more natural visual information, leading to better visual prosthesis outcomes.

Improving the spatial resolution of epiretinal implants by increasing stimulus pulse duration

Retinal prosthetic implants are the only approved treatment for retinitis pigmentosa, a disease of the eye that causes blindness through gradual degeneration of photoreceptors. An array of microelectrodes triggered by input from a camera stimulates surviving retinal neurons, with each electrode acting as a pixel. Unintended stimulation of retinal ganglion cell axons causes patients to see large oblong shapes of light, rather than focal spots, making it difficult to perceive forms. To address this problem, we performed calcium imaging in isolated retinas and mapped the patterns of cells activated by different electrical stimulation protocols. We found that pulse durations two orders of magnitude longer than those typically used in existing implants stimulated inner retinal neurons while avoiding activation of ganglion cell axons, thus confining retinal responses to the site of the electrode. Multielectrode stimulation with 25-ms pulses can pattern letters on the retina corresponding to a Snellen acuity of 20/312. We validated our findings in a patient with an implanted epiretinal prosthesis by demonstrating that 25-ms pulses evoke focal spots of light.

Temporal properties of network-mediated responses to repetitive stimuli are dependent upon retinal ganglion cell type

OBJECTIVE

To provide artificially-elicited vision that is temporally dynamic, retinal prosthetic devices will need to repeatedly stimulate retinal neurons. However, given the diversity of physiological types of retinal ganglion cells (RGCs) as well as the heterogeneity of their responses to electric stimulation, temporal properties of RGC responses have not been adequately investigated. Here, we explored the cell type dependence of network-mediated RGC responses to repetitive electric stimulation at various stimulation rates.

APPROACH

We examined responses of ON and OFF types of RGCs in the rabbit retinal explant to five consecutive stimuli with varying inter-stimulus intervals (10-1000 ms). Each stimulus was a 4 ms long monophasic sinusoidal cathodal current, which was applied epiretinally via a conical electrode. Spiking activity of targeted RGCs was recorded using a cell-attached patch electrode.

MAIN RESULTS

ON and OFF cells had distinct responses to repetitive stimuli. Consistent with earlier studies, OFF cells always generated reduced responses to subsequent stimuli compared to responses to the first stimulus. In contrast, a new stimulus to ON cells suppressed all pending/ongoing responses from previous stimuli and initiated its own response that was remarkably similar to the response from a single stimulus in isolation. This previously unreported 'reset' behavior was observed exclusively and consistently in ON cells. These contrasts between ON and OFF cells created a range of stimulation rates (4-7 Hz) that maximized the ratio of the responses arising in ON versus OFF cells.

SIGNIFICANCE

Previous clinical testing reported that subjects perceive bright phosphenes (ON responses) and also prefer stimulation rates of 5-7 Hz. Our results suggest that responses of ON cells are weak at high rates of stimulation (> ∼7 Hz) due to the reset while responses of OFF cells are strong at low rates (< ∼4 Hz) due to reduced desensitization, both reducing the ratio of ON to OFF responses. In combination with previous results indicating that responses in ON cells more closely match physiological patterns (Im and Fried 2015 J. Physiol. 593 3577-96), our results offer a potential reason for the user preference of intermediate rates (5-7 Hz).

The Retinal Response to Sinusoidal Electrical Stimulation

Rectangular electrical pulses are the primary stimulus waveform used in retinal prosthetics as well as many other neural stimulation applications. Unfortunately, the utility of pulsatile stimuli is limited by the inability to avoid the activation of passing axons, which can result in the distortion of the spatial patterns of elicited neural activity. Because avoiding axons would likely improve clinical outcomes, the examination of alternate stimulus waveforms is warranted. Here, we studied the response of rabbit retinal ganglion cells (RGCs) to sinusoidal electrical stimulation applied at frequencies of 5, 10, 25, and 100 Hz. Targeted RGCs were restricted to 4 common types: OFF-Brisk Transient, OFF-Sustained, ON-Brisk Transient, and ON-Sustained. Interestingly, response patterns varied between different types; the most notable difference was the relatively weak response of ON-Sustained cells to low frequencies. Calculation of total spike counts per trial revealed that lower frequencies are more charge efficient than high frequencies. Finally, experiments utilizing synaptic blockers revealed that 5 and 10 Hz activate photoreceptors while 25 and 100 Hz activate RGCs. Taken together, our results suggest that while sinusoidal electrical stimulation may provide a useful research tool, its clinical utility may be limited.

A high-density microelectrode-tissue-microelectrode sandwich platform for application of retinal circuit study

Background

Microelectrode array (MEA) devices are frequently used in neural circuit studies, especially in retinal prosthesis. For a high throughput stimulation and recording paradigm, it is desirable to obtain the responses of multiple surface RGCs initiated from the electrical signals delivered to multiple photoreceptor cells. This can be achieved by an high density MEA-tissue-MEA (MTM) sandwich configuration. However, the retina is one of the most metabolically active tissues, consumes oxygen as rapidly as the brain. The major concern of the MTM configuration is the supply of oxygen.

Methods

We aimed to develop a high density MTM sandwich platform which consists of stacks of a stimulation MEA, retinal tissue and a recording MEA. Retina is a metabolically active tissue and the firing rate is very sensitive to oxygen level. We designed, simulated and microfabricated porous high density MEAs and an adjustable perfusion system that electrical signals can be delivered to and recorded from the clipped retinal tissue.

Results

The porous high-density MEAs linked with stimulation or recording devices within a perfusion system were manufactured and the MTM platform was assembled with a retina slice inside. The firing rate remained constant between 25 and 55 min before dramatically declined, indicating that within certain period of time (e.g. 30 min after habituation), the retina condition was kept by sufficient oxygen supply via the perfusion holes in the MEAs provided by the double perfusion system.

Conclusions

MTM sandwich structure is an efficient platform to study the retinal neural circuit. The material and arrangement of high density microelectrodes with porous design make this MEA appropriate for sub-retina prosthesis. Finding ways to prolong the recording time and reduce the signal-to-noise ratio are important to improve our MTM prototype.

Simultaneous recording of mouse retinal ganglion cells during epiretinal or subretinal stimulation

We compared response patterns and electrical receptive fields (ERF) of retinal ganglion cells (RGCs) during epiretinal and subretinal electrical stimulation of isolated mouse retina. Retinas were stimulated with an array of 3200 independently controllable electrodes. Four response patterns were observed: a burst of activity immediately after stimulation (Type I cells, Vision Research (2008), 48, 1562-1568), delayed bursts beginning >25ms after stimulation (Type II), a combination of both (Type III), and inhibition of ongoing spike activity. Type I responses were produced more often by epiretinal than subretinal stimulation whereas delayed and inhibitory responses were evoked more frequently by subretinal stimulation. Response latencies were significantly shorter with epiretinal than subretinal stimulation. These data suggest that subretinal stimulation is more effective at activating intraretinal circuits than epiretinal stimulation. There was no significant difference in charge threshold between subretinal and epiretinal configurations. ERFs were defined by the stimulating array surface area that successfully stimulated spikes in an RGC. ERFs were complex in shape, similar to receptive fields mapped with light. ERF areas were significantly smaller with subretinal than epiretinal stimulation. This may reflect the greater distance between stimulating electrodes and RGCs in the subretinal configuration. ERFs for immediate and delayed responses mapped within the same Type III cells differed in shape and size, consistent with different sites and mechanisms for generating these two response types.

High-fidelity reproduction of spatiotemporal visual signals for retinal prosthesis

Natural vision relies on spatiotemporal patterns of electrical activity in the retina. We investigated the feasibility of veridically reproducing such patterns with epiretinal prostheses. Multielectrode recordings and visual and electrical stimulation were performed on populations of identified ganglion cells in isolated peripheral primate retina. Electrical stimulation patterns were designed to reproduce recorded waves of activity elicited by a moving visual stimulus. Electrical responses in populations of ON parasol cells exhibited high spatial and temporal precision, matching or exceeding the precision of visual responses measured in the same cells. Computational readout of electrical and visual responses produced similar estimates of stimulus speed, confirming the fidelity of electrical stimulation for biologically relevant visual signals. These results suggest the possibility of producing rich spatiotemporal patterns of retinal activity with a prosthesis and that temporal multiplexing may aid in reproducing the neural code of the retina.

Activation of ganglion cells in wild-type and P23H rat retinas with a small subretinal electrode

Electronic retinal prostheses are being developed for people who become blind due to loss of photoreceptors from the disease retinitis pigmentosa. Previously, we reported on the responses of RGCs in the P23H rat (a model of retinitis pigmentosa) and the Sprague-Dawley (SD) rat to stimulation with a 400-μm diameter electrode (Jensen and Rizzo, 2011). With recent clinical trials now utilizing smaller (50-200 μm) electrodes, I sought to investigate the electrically evoked responses of RGCs in P23H and SD rat retinas with a smaller (125-μm diameter) electrode. Here, I report on the electrically evoked spike activity from RGCs that arose from stimulation of the retinal neural network. With biphasic current pulses of 1 ms per phase, the thresholds for activation of SD rat RGCs ranged from 0.52 to 2.8 μA; thresholds of P23H rat RGCs ranged from 1.2 to 7.8 μA. Median thresholds of RGCs were 1.4 μA in SD rats and 2.5 μA in P23H rats. These thresholds measurements were obtained with the recording electrode placed over the stimulating electrode. I also examined how thresholds of RGCs change as a function of distance (100-500 μm) from the center of the stimulating electrode. The median threshold currents of RGCs were much higher in P23H rats for all distances. What was striking was that the thresholds for activation of RGCs in P23H rat retinas rose much more rapidly. When the recording electrode was only 100-200 μm from the center of the stimulating electrode, the median threshold current of P23H rat RGCs rose by 460%. In contrast, the median threshold current of SD rat RGCs increased only 29%. I also investigated the contribution of photoreceptors to the electrically evoked responses of ON-center RGCs in SD rat retinas by examining the change in RGC thresholds when photoreceptor input to ON bipolar cells was blocked with the mGluR6 antagonist CPPG. At 500-600 μm, CPPG suppressed the light-evoked responses of the RGCs and at the same time increased the amount of current needed to generate an electrically evoked response. Similar to what was observed with SD rat RGCs, CPPG suppressed the light-evoked responses of ON-center P23H rat RGCs. However, the stimulation thresholds were not significantly altered. In conclusion, the data show that the threshold currents for indirect stimulation of both SD and P23H rat RGCs with a 125-μm diameter electrode are much lower than what we found previously with a 400-μm diameter electrode. To achieve high resolution vision with a multielectrode array, the spread of activation of RGCs needs to be limited. Our findings indicate that the spread of activation of RGCs is more confined in the degenerate retina. Lastly, my findings indicate that photoreceptors contribute to the lower stimulation thresholds of RGCs in normal, healthy retinas.

Multiple components of ganglion cell desensitization in response to prosthetic stimulation

Retinal prostheses aim to restore functional vision to those blinded by outer retinal diseases using electric stimulation of surviving neurons. Previous work indicates that repetitive stimulation with stimuli that activate the synaptic network reduces the sensitivity of retinal neurons to further stimulation. Such desensitization may contribute to the fading of visual percepts over time reported by human subjects. Here, we show that desensitization may be more complex than previously considered. We recorded spike trains from rabbit retinal ganglion cells and found that desensitization persists in the presence of inhibitory blockers (strychnine and picrotoxin), indicating amacrine cell inhibition is not solely responsible for reducing sensitivity in response to electric stimulation. The threshold for direct activation of the ganglion cell changes little during the simultaneous desensitization of the synaptically mediated response, indicating that desensitization likely occurs upstream of the spike generator. In addition to rapid desensitization acting over hundreds of milliseconds (τ = 176.4 ± 8.8 ms), we report the presence of slow acting desensitization with a time course of seconds (τ = 14.0 ± 1.1 s). The time courses of the two components of desensitization that we found are similar to the two phases of brightness fading seen in human subjects. This suggests that the reduction in ganglion cell firing due to desensitization may be responsible for the fading of visual percepts over time in response to prosthetic stimulation.

Blind subjects implanted with the Argus II retinal prosthesis are able to improve performance in a spatial-motor task

BACKGROUND/AIMS

To determine to what extent subjects implanted with the Argus II retinal prosthesis can improve performance compared with residual native vision in a spatial-motor task.

METHODS

High-contrast square stimuli (5.85 cm sides) were displayed in random locations on a 19″ (48.3 cm) touch screen monitor located 12″ (30.5 cm) in front of the subject. Subjects were instructed to locate and touch the square centre with the system on and then off (40 trials each). The coordinates of the square centre and location touched were recorded.

RESULTS

Ninety-six percent (26/27) of subjects showed a significant improvement in accuracy and 93% (25/27) show a significant improvement in repeatability with the system on compared with off (p<0.05, Student t test). A group of five subjects that had both accuracy and repeatability values <250 pixels (7.4 cm) with the system off (ie, using only their residual vision) was significantly more accurate and repeatable than the remainder of the cohort (p<0.01). Of this group, four subjects showed a significant improvement in both accuracy and repeatability with the system on.

CONCLUSION

In a study on the largest cohort of visual prosthesis recipients to date, we found that artificial vision augments information from existing vision in a spatial-motor task. Clinical trials registry no NCT00407602.