Abstract
Purpose :
That mechanical stimulation of the retina induces visual percepts has been known for millennia but how mechanical deformation produces these responses remains unclear. We were interested in determining whether this might offer an alternative approach to electrical, chemical or optical stimulation for a visual prosthesis. The goal was to determine whether retinal ganglion cells (RGCs) could be stimulated mechanically to evoke responses similar to those generated through visual stimulation and to investigate the underlying mechanism.
Methods :
Dark-adapted retinas from 3- to 5-week wild-type Hooded Long-Evans rats (N = 12) were placed in a multielectrode array from which RGC spike trains were recorded in oxygenated Ames medium at room temperature. The retinas were subjected to visual (20 repetitions of a full-field flash – 2 s of light on followed by 2 s light off – at a mean illuminance of 5 lm/m2) and mechanical stimulation (boluses of Ames medium injected through a micropipette directed at specific retinal locations where robust RGC responses to visual stimulation had been achieved). Our standard stimulation protocol involved 30 injections at an interpulse duration of 3 s and injection time of 100 ms. In some experiments, ruthenium red, a nonselective blocker of the transient receptor potential vanilloid (TRPV) channel family, was used to investigate its role in generating mechanically induced responses.
Results :
Injections of Ames medium elicited spatially localized (200 μm median spread) responses from RGCs. The mechanically induced responses were significantly (p < 0.001) larger in spike rate amplitude and had significantly (p < 0.001) faster response latencies than light-evoked responses. Pharmacological blockade of TRPV channels with ruthenium red significantly (p << 0.001) reduced the amplitude of mechanically induced RGC responses but did not eliminate them entirely. Two-sided Wilcoxon rank-sum tests (α = 0.05) were used to examine the differences between mechanical and visual stimulation and a two-sided Wilcoxon signed-rank test (α = 0.05) was used to compare the spike rate amplitudes of mechanical responses before and after incubation with ruthenium red.
Conclusions :
RGCs can be driven mechanically suggesting that this could be an alternative approach for their stimulation in a visual prosthesis. Currents through TRPV channels are responsible for much of the effect.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.