GABA and GABA(A) receptor antagonists alter developing cone photoreceptor development in neonatal rabbit retina

Gamma aminobutyric acid (GABA) has been established as an important developmental signal in a number of regions of the central nervous system (CNS), including retina. Our previous studies have shown that GABAergic horizontal cells act as the initial synaptic target for developing cone photoreceptors in neonatal rabbit retina. Since intraocular injections of the GABA(A) receptor antagonists, picrotoxin or bicuculline, disrupt cone synaptogenesis in vivo, GABA released from horizontal cells may provide a necessary signal for cone axon growth and/or synapse formation. In the current report, we have used cultured retinal explants to examine the effects of GABA(A) receptor antagonists on other aspects of developing cones. These include the distribution pattern of cone cell bodies across the outer surface of the retina and the expression of GABA(A) receptors within both cone cell bodies and axonal processes. Peanut agglutinin (PNA), a plant lectin that specifically labels cone plasma membrane and extracellular matrix, was used to monitor cone development, and a GABA(A) receptor antibody against the beta2/3 subunits of the protein was used to label GABA(A) receptors. Results showed that cones maintained in the explant culture express GABA(A) receptors in a temporal and spatial pattern similar to that observed in vivo, namely a low expression of receptors on cone cell bodies at postnatal day 1 (P1), peaking around P3 and diminishing by P7. Neonatal retinal explants exposed to the GABA(A) receptor antagonists, bicuculline (10 microM) or SR95531 (5 microM), for 24 h in culture showed disruption of the normal distribution of cone cell bodies. When GABA (100 microM) was added along with either antagonist, cone cell bodies appeared normal. Neither bicuculline nor SR95531 alone had any effect on the general morphology of other retinal layers, suggesting that these GABA(A) receptor antagonists at the concentrations used were not acting as nonspecific disruption agents. The effects of GABA antagonists were confined to the first week after birth with no disruption seen in P9 or adult explant cultures. These data provide a direct demonstration of the necessity for GABAergic input to cones during active synaptogenesis. As we have previously shown, GABA(A) receptor activation causes a substantial increase in intracellular calcium concentrations in cones and thereby could provide a mechanism by which GABA regulates cone maturation.