Department of Ophthalmology, Henry Ford Health System, 1 Ford Place, Detroit, MI 48202, United States of America. Department of Anatomy and Cell Biology, Wayne State University School of Medicine, 540 East Canfield Street, Detroit, MI 48201, United States of America. Department of Electrical and Computer Engineering, Wayne State University College of Engineering, 5050 Anthony Wayne Drive, Detroit, MI 48202, United States of America. Department of Neurosurgery, Massachusetts General Hospital, Harvard Medical School, 50 Blossom Street, Boston, MA 02114, United States of America.
J Neural Eng. 2018 Jun;15(3):036010. doi: 10.1088/1741-2552/aaadc1. Epub 2018 Feb 8.
To improve the quality of artificial vision that arises from retinal prostheses, it is important to bring electrically-elicited neural activity more in line with the physiological signaling patterns that arise normally in the healthy retina. Our previous study reported that indirect activation produces a closer match to physiological responses in ON retinal ganglion cells (RGCs) than in OFF cells (Im and Fried 2015 J. Physiol. 593 3677-96). This suggests that a preferential activation of ON RGCs would shape the overall retinal response closer to natural signaling. Recently, we found that changes to the rate at which stimulation was delivered could bias responses towards a stronger ON component (Im and Fried 2016a J. Neural Eng. 13 025002), raising the possibility that changes to other stimulus parameters can similarly bias towards stronger ON responses. Here, we explore the effects of changing stimulus duration on the responses in ON and OFF types of brisk transient (BT) and brisk sustained (BS) RGCs.
We used cell-attached patch clamp to record RGC spiking in the isolated rabbit retina. Targeted RGCs were first classified as ON or OFF type by their light responses, and further sub-classified as BT or BS types by their responses to both light and electric stimuli. Spiking in targeted RGCs was recorded in response to electric pulses with durations varying from 5 to100 ms. Stimulus amplitude was adjusted at each duration to hold total charge constant for all experiments.
We found that varying stimulus durations modulated responses differentially for ON versus OFF cells: in ON cells, spike counts decreased significantly with increasing stimulus duration while in OFF cells the changes were more modest. The maximum ratio of ON versus OFF responses occurred at a duration of ~10 ms. The difference in response strength for BT versus BS cells was much larger in ON cells than in OFF cells.
The stimulation rates preferred by subjects during clinical trials are similar to the rates that maximize the ON/OFF response ratio in in vitro testing (Im and Fried 2016a J. Neural Eng. 13 025002). Here, we determine the stimulus duration that produces the strongest bias towards ON responses and speculate that it will further enhance clinical effectiveness.
为了提高视网膜假体产生的人工视觉质量，重要的是要使电诱发的神经活动更符合健康视网膜中正常产生的生理信号模式。我们之前的研究报告称，间接激活产生的与 ON 视网膜神经节细胞 (RGC) 的生理反应更接近，而与 OFF 细胞的反应则不接近（Im 和 Fried，2015 年，《生理学杂志》，第 593 卷，第 3677-96 页）。这表明，选择性地激活 ON RGC 可以使整个视网膜反应更接近自然信号。最近，我们发现，改变刺激的传递速率可以使反应偏向更强的 ON 成分（Im 和 Fried，2016a 年，《神经工程杂志》，第 13 卷，第 025002 页），这增加了改变其他刺激参数同样可以偏向更强的 ON 反应的可能性。在这里，我们探讨了改变刺激持续时间对 ON 和 OFF 类型的快速瞬态 (BT) 和快速持续 (BS) RGC 反应的影响。
我们使用细胞贴附式膜片钳技术在离体兔视网膜上记录 RGC 放电。首先通过光反应将目标 RGC 分类为 ON 或 OFF 型，然后通过它们对光和电刺激的反应进一步分类为 BT 或 BS 型。在电脉冲刺激下，记录目标 RGC 的放电情况，电脉冲的持续时间从 5 到 100ms 不等。在所有实验中，通过调整刺激幅度来保持总电荷量不变。
我们发现，改变刺激持续时间对 ON 与 OFF 细胞的反应有不同的调制作用：在 ON 细胞中，随着刺激持续时间的增加，放电计数显著减少，而在 OFF 细胞中，变化则较为温和。在 ~10ms 的持续时间下，ON 与 OFF 反应的最大比值出现。ON 细胞中 BT 与 BS 细胞之间的反应强度差异明显大于 OFF 细胞。
临床试验中受试者偏爱的刺激率与体外测试中最大程度地提高 ON/OFF 反应比值的刺激率相似（Im 和 Fried，2016a 年，《神经工程杂志》，第 13 卷，第 025002 页）。在这里，我们确定了产生最强 ON 反应偏向的刺激持续时间，并推测它将进一步提高临床效果。
