In the retina, gap junctions are established by many cell types and may participate in retinal circuitry in a variety of ways. Gap junctions are composed of connexins, encoded by a >20-member family of highly related genes. Only one of these, connexin36 (Cx36), has been conclusively demonstrated in retinal neurons. Using a Cx36 knockout (KO) mouse incorporating a histochemical reporter, we found Cx36 to be present in most or all All amacrine cells, many photoreceptors, subsets of ON and OFF cone bipolar cells and a low number of ganglion cells. However, horizontal cells and many ganglion cells establish gap junctions yet do not express Cx36. Using RT-PCR, we have found Cx45 and Cx57 to be prominently expressed in mouse retina. We propose to characterize the cellular distribution of these connexins and determine if other gap junction channel-forming proteins may be present in retinal neurons. Many lines of evidence indicate that rod photoreceptor signals utilize multiple pathways to reach ganglion cells. In our Cx36 KO, ON ganglion cell responses to scotopic stimuli are completely eliminated, indicating that Cx36 is required in all pathways contributing to rod ON signaling. We showed that gap junctional coupling between AII and cone ON bipolar cells, the presumed 'primary' pathway for rod photoreceptor signaling, was abolished in the KO. Since a proposed 'alternative' pathway involved gap junction between rods and cones, and since we found Cx36 expression in photoreceptors, our findings were consistent with the proposed 'alternative' pathway. However, recent studies of retinas from mice genetically altered to lack cones suggest that rod-cone coupling does not contribute to rod photoreceptor signaling. In addition, it has not been technically feasible to measure junctional coupling between photoreceptors in mouse retinas. Therefore, to resolve the discrepancy between current models of the 'alternative' pathway, we will 1) develop a method to directly assess rod-cone coupling in WT and Cx36 KO mice and 2) produce conditional knockouts that eliminate Cx36 specifically from either cones or from AII amacrine cells. Then, ganglion cell responses to light can be assessed in retinas where either primary or alternative pathways are ablated.
Showing the most recent 10 out of 21 publications
