Cholinergic Amacrine Cell Varicosities are Sites of Synaptic Release

Abstract

Purpose: : Amacrine cells of many different subtypes make conventional synapses with ganglion cells and bipolar cells. These synaptic sites are more ambiguous than those of photoreceptor and bipolar cell terminals which contain distinctive electron dense ribbons, and the analysis of amacrine cell contacts is more difficult. Furthermore, amacrine cell dendrites may cofasciculate, sharing the same physical space. Previous work suggests that the varicosities of S1/S2 amacrine cells make synaptic contacts (Zhang et al., 2002). Here, we used the starburst amacrine cell (SAC) as a model system to test the hypothesis that amacrine cell varicosities are sites of synaptic release.

Methods: : ON SACs were dye injected with Neurobiotin in wholemount rabbit retina. After fixation, the retina was labeled with antibodies against the vesicular acetylcholine transporter (VAChT) and the vesicular GABA transporter (VGAT) as pre-synaptic markers and an antibody against the GABA_A receptor as a post-synaptic marker. This material was analyzed by triple-label confocal microscopy.

Results: : Injected SACs displayed the typical symmetric dendritic morphology. There were distinct annular regions containing fine proximal dendrites arising from the soma and more distal dendrites studded with varicosities. The vesicular markers VAChT and VGAT were coexpressed within starburst varicosities. Quantitative image analysis confirmed these results, whereas randomizing the VAChT label by a 90° rotation completely eliminated the colocalization pattern. This indicates that these labeling patterns do not occur by chance. Colabeling for GABA_A receptors as a postsynaptic marker demonstrated peripheral labeling around the vesicular transporter markers and varicosities, a requirement expected for postsynaptic receptor labeling.

Conclusions: : These data provide clear evidence that the varicosities located on SACs coexpress presynaptic markers and are apposed by postsynaptic receptors. This suggests that starburst varicosities are the sites of synaptic release. These results confirm previous work which showed that S1/S2 varicosities are synaptic sites. The identification of pre- and post-synaptic specializations will permit a more reliable analysis of synaptic contacts between amacrine cells and their post-synaptic targets.

Keywords: retina: proximal (bipolar, amacrine, and ganglion cells) • retinal connections, networks, circuitry • immunohistochemistry