Paul Werginz1, Vineeth Raghuram, Shelley I Fried. 1. Institute for Analysis and Scientific Computing, Vienna University of Technology, Vienna, Austria. Department of Neurosurgery, Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States of America. Author to whom any correspondence should be addressed.
Abstract
OBJECTIVE: Retinal prostheses strive to restore vison to patients that are blind from retinal degeneration by electrically stimulating surviving retinal ganglion cells (RGCs). The quality of elicited percepts remains limited however and it is desirable to develop improved stimulation strategies. Here, we examine how the anatomical and biophysical properties of RGCs influence activation thresholds, including the effects of variations found naturally. APPROACH: Detailed reconstructions were made of a large number of mouse α RGCs and were used to create an array of model cells; the models were used to study the effects of individual anatomical features on activation threshold to electric stimulation. Stimulation was delivered epiretinally from a point-source or disk electrode and consisted of monophasic or biphasic rectangular pulses. MAIN RESULTS: Modeling results show that the region of minimum threshold always is within the axon initial segment (AIS). The properties of this region as well as the absolute value of the minimum threshold are dependent on the length of the AIS as well as on the relative composition of sodium channels within the AIS. Other morphological features, including cell size, dendritic field size and the distance between the AIS and the soma had only a minimal influence on thresholds. Introducing even a small number of low-threshold Nav1.6 channels into the AIS was sufficient to lower minimum thresholds substantially although further increases in Nav1.6 had diminishing effects. The distance between the AIS and the electrode affects threshold levels while alignment of the electrode with the axon or dendritic parts of the RGC can result in lower thresholds, even if the distance to the cell remains the same. SIGNIFICANCE: Intrinsic morphological features can influence activation thresholds with the AIS having the strongest influence. However, the combined influence remains limited and may not be large enough to allow for selective activation between different RGC types.
OBJECTIVE: Retinal prostheses strive to restore vison to patients that are blind from retinal degeneration by electrically stimulating surviving retinal ganglion cells (RGCs). The quality of elicited percepts remains limited however and it is desirable to develop improved stimulation strategies. Here, we examine how the anatomical and biophysical properties of RGCs influence activation thresholds, including the effects of variations found naturally. APPROACH: Detailed reconstructions were made of a large number of mouse α RGCs and were used to create an array of model cells; the models were used to study the effects of individual anatomical features on activation threshold to electric stimulation. Stimulation was delivered epiretinally from a point-source or disk electrode and consisted of monophasic or biphasic rectangular pulses. MAIN RESULTS: Modeling results show that the region of minimum threshold always is within the axon initial segment (AIS). The properties of this region as well as the absolute value of the minimum threshold are dependent on the length of the AIS as well as on the relative composition of sodium channels within the AIS. Other morphological features, including cell size, dendritic field size and the distance between the AIS and the soma had only a minimal influence on thresholds. Introducing even a small number of low-threshold Nav1.6 channels into the AIS was sufficient to lower minimum thresholds substantially although further increases in Nav1.6 had diminishing effects. The distance between the AIS and the electrode affects threshold levels while alignment of the electrode with the axon or dendritic parts of the RGC can result in lower thresholds, even if the distance to the cell remains the same. SIGNIFICANCE: Intrinsic morphological features can influence activation thresholds with the AIS having the strongest influence. However, the combined influence remains limited and may not be large enough to allow for selective activation between different RGC types.
Authors: Maarten H P Kole; Susanne U Ilschner; Björn M Kampa; Stephen R Williams; Peter C Ruben; Greg J Stuart Journal: Nat Neurosci Date: 2008-01-20 Impact factor: 24.884
Authors: Lauren H Jepson; Pawel Hottowy; Keith Mathieson; Deborah E Gunning; Wladyslaw Dabrowski; Alan M Litke; E J Chichilnisky Journal: J Neurosci Date: 2013-04-24 Impact factor: 6.167
Authors: Lauren N Ayton; Peter J Blamey; Robyn H Guymer; Chi D Luu; David A X Nayagam; Nicholas C Sinclair; Mohit N Shivdasani; Jonathan Yeoh; Mark F McCombe; Robert J Briggs; Nicholas L Opie; Joel Villalobos; Peter N Dimitrov; Mary Varsamidis; Matthew A Petoe; Chris D McCarthy; Janine G Walker; Nick Barnes; Anthony N Burkitt; Chris E Williams; Robert K Shepherd; Penelope J Allen Journal: PLoS One Date: 2014-12-18 Impact factor: 3.240