br Method br Results br Discussion There were two major
Discussion There were two major sources of fluorescent signals of NADH in the mouse retina. One was a product derived from glutamate and the other was the elevation of intrinsic NADH. In the present study, we observed significant contamination from intrinsic NADH in “the outer retina”, while there was no significant contamination from intrinsic NADH in “the inner retina”. Therefore, the elevation of the fluorescent intensity in “the inner retina” primarily reflects the actual glutamate release. On the other hand, we have to be careful when monitoring glutamate release in “the outer retina”, since contamination from fluorescent signals derived from intrinsic NADH might lead to overestimation of glutamate release in “the outer retina”. In rat hippocampal-entorhinal Deacetylase Inhibitor Cocktail slices, depolarizing stimuli such as glutamate or electrical stimulation increase NADH autofluorescence in mitochondria (Schuchmann et al., 2001). Thus, one explanation for the elevation of intrinsic NADH is that high K stimulation-induced depolarization triggered the elevation of intramitochondrial NADH in the inner and outer segments of photoreceptors where a lot of intrinsic NADH is used for photo transduction. Therefore, it is possible to speculate that the difference in the magnitude of the ΔF/F in the OS might reflect the different activities among samples. However, we have to consider the possibility that spillover glutamate from the OPL is somehow affecting the elevation of the fluorescent intensity in “the outer retina”, since high K stimulation (without TBOA with NAD) induced elevation of the fluorescent signals in “the outer retina”. In the retina, glutamate release occurs in the synaptic layers (the OPL and IPL). The elevation of the fluorescent intensity in the IPL reflects the release of glutamate from the axon terminals of bipolar cells, while the elevation of the fluorescent intensity in the OPL reflects the release of glutamate from the axon terminals of photoreceptors. To modulate the glutamate release from these glutamatergic neurons, we applied Cd2+ or a cocktail of picrotoxin and strychnine. The action of Cd2+ or a cocktail of picrotoxin and strychnine on glutamate release was detected in the IPL, but not to the same extent in the OPL. There are several possible explanations for the different responses between the OPL and IPL. One possibility is that the different synaptic structures between the OPL and IPL affected the diffusion of glutamate. In general, synapses in the OPL are invaginated synapses, while synapses in the IPL are not invaginated synapses (Dowling, 2012). Therefore, invaginated synapses in the OPL might hamper the spillage of glutamate from the synaptic cleft. However, this is not likely the case since a modelling study showed that invaginated synapses cannot become barriers for the spillover of glutamate (Rao-Mirotznik et al., 1998). A second possibility is that the activity of inhibitory circuits in the slice preparations might be altered. In general, GABAergic feedback is provided by neurons with long dendrites (horizontal cells in the OPL and wide-field amacrine cells in the IPL), while glycinergic feedback is supported by neurons with relatively compact dendrites (narrow-field amacrine cells in the IPL) (Dowling, 2012). Therefore, possible damage to the dendrites of inhibitory neurons, especially GABAergic neurons, during preparation of the retinal slices might strongly affect the activity of inhibitory circuits in the OPL. Alternatively, GABAergic feedback to cones might not act as a major component of feedback to cones (Tatsukawa et al., 2005). In addition, we need to consider the possibility that contamination from intrinsic NADH-derived signals in the OPL masked the actual glutamate-mediated elevation of the fluorescent signals. To assess the inhibitory effects of Cd2+ on high K stimulation, we normalized the ΔF/Fs in the OPL and IPL against the ΔF/F in the OS. Following this normalization, we detected the inhibitory actions of Cd2+ in the IPL, but not in the OPL. In this normalization, we assumed the following conditions in the OS. First, there will be no glutamate-induced elevation of the ΔF/F since there is no synaptic contact in the OS. Second, the magnitude of the ΔF/F in the individual sample might reflect the activity of samples (the magnitude of the ΔF/F in the OS). However, it is possible that the variation of the ΔF/F in the OS among samples might be determined by different factors.