The functional role of intracellular calcium signaling in retinal amacrine cells, particularly with respect to computation of visual stimuli
Final Report Abstract
Retinal starburst amacrine cells (SACs) perform a specific dendritic computation: they detect the direction of visual motion and relay this information to direction selective (DS) ganglion cells. Dendritic Ca2+ activity in SAGs reflects this compulation. The goal of this project was to characterize the subcellular distribution of Ca2+ signals in SAC dendrites and to find out how they are generated. In particular, we wanted to know, if intracellular signaling mechanisms, such as Ca2+-induced Ca2+ release (CIGR), are involved. Our main hypothesis was that CICR acts as an "amplifier" that translates the result of the dendritic computation (=larger depolarization for stimuli moving centrifugally along the dendrite) into a robust synaptic output signal (=[Ca2+]-triggered transmitter release). This project provided three lines of evidence to support this hypothesis: 1) We found light-evoked Ca2+ signals to be localized to "hotspots", which were heterogeneously distributed along SAC dendrites. Moreover, that neighboring "hotspots", located a few tens of microns apart, could display different directional tuning, suggests that processing occurs at a local scale. This is consistent with a ClGC-based mechanism. 2) During whole cell recordings, when intracellular Ca2+ signaling in SACs appears to be constricted, voltage stimulation evoked a much more widespread Ca2+ activity than with light stimulation under conditions, when intracellular Ca2+ signaling was intact. This suggest that with light stimulation only a fraction of the available voltage-gated Ca2+ channels (VGCCs) were recruited. However, the Ca2+ signal amplitude in "hotspots" is similar for both stimulation types, therefore it is likely that light-evoked Ca2+ signals are complemented with Ca2+ from other sources, such as internal stores. 3) Light-evoked Ca2+ signals were blocked or strongly reduced by inhibitors of IP3 receptor-mediated CICR, consistent with an involvement of Ca2+ stores in SACs. Nonetheless, with these drugs we also observed a reduction in SAC input, therefore it is possible that the reduction in Ca2+ signal by the CICR inhibitors was partially due to an effect on the presynaptic network. In conclusion, an involvement of CICR in the generation of SAC Ca2+ signals seems very likely, consistent with findings in other amacrine cells. We are convinced that our ongoing experiments (i.e. effects of CICR inhibitors under conditions when presynaptic effects can be neglected) will help to validate our hypothesis. Such an ClCR-based amplification mechanism would explain how SACs are able to use their Ga2+ channels both to perform a spatio-temporal dendritic computation and to initiate transmitter release: VGCCss just need to provide the trigger, while the Ca2+ levels required for efficient transmitter release are provided from internal stores. Moreover, such a mechanism allows to define more complex relationships between electrical events (e.g. resulting from synaptic input) and local synaptic output. This is particularly useful in interneurons, such as many types of amacrine cells, that use their dendrites both as input and output structures.
