Graduation Date

Spring 5-7-2016

Document Type


Degree Name

Doctor of Philosophy (PhD)


Pharmacology and Experimental Neuroscience

First Advisor

Wallace B. Thoreson, Ph.D.


Polarization of the horizontal cell (HC) membrane potential causes changes in the synaptic cleft pH that result in inhibitory feedback from HCs to cone photoreceptors (PRs). HCs average signals from many PRs and so negative feedback onto PR terminals from HCs subtracts the average luminance of the visual scene from the light responses of an individual cone. This feedback operates by changing the voltage-dependence and amplitude of the L-type Ca2+ current (ICa) that regulates synaptic release. Feedback regulation of PR Ca2+ channels involves protons but the mechanism by which this pH change occurs is unclear. We investigated three possible sources for protons in the cone synaptic cleft: 1) extracellular carbonic anhydrase (CA), 2) protons released into the cleft upon exocytosis of synaptic vesicles, and 3) sodium-hydrogen exchangers (NHEs). Using electrophysiological measurements of HC to cone feedback, we found that CA and vesicular protons are not major sources of protons for feedback. Feedback was eliminated by removal of extracellular Na+ and significantly inhibited by an NHE antagonist, cariporide, implicating NHEs as a significant source of protons. While NHEs are a major proton source, they are not known to be voltage-sensitive and thus unlikely to be responsible for changes in extracellular proton levels caused by changes in HC membrane potential. Instead we found that removal of bicarbonate and inhibition of bicarbonate transporters with 500 μM DIDS both eliminated feedback, suggesting that HC polarization changes extracellular pH by altering bicarbonate transport.

To test whether an ephaptic mechanism is involved in mediating feedback, we used paired whole cell recordings to hyperpolarize the HC while cone ICa was active and then measured the kinetics of feedback-induced changes in the cone membrane current. The time constants of the resulting feedback current were slower than the measurement time resolution and not instantaneous as predicted by an ephaptic mechanism.