We previously described structural changes in photoreceptor ribbons of the hibernating retina. We investigated the functional consequences of such structural alteration and identified a significant role of the ribbon in facilitating replenishment of the readily releasable pool of synaptic vesicles, thereby ensuring high frequency synaptic transmission. This finding at the synaptic level corresponds with those at the systemic level, considering that electroretinogram (ERG) recordings have revealed a significant decrease in their ability to detect high frequency flicker. We also explored the potential molecular mechanism of this ribbon plasticity and discovered that the ratio of NADH+/NAD in the photoreceptor, which reflects its cellular metabolic state, regulates the interaction of ribeye, the main ribbon protein, with bassoon, another cytoplasmic active zone (CAZ) protein. Reduced interaction between ribeye and bassoon likely contributes to the destabilization of the ribbon structure. In the previous report, I proposed to further investigate functional roles played by the ribbon in 1) producing spontaneous vesicle release, and 2) producing multivesicular release. In the last year, we have refined and expanded the work summarized above and advanced significantly on proposed projects. We set out to examine the impact of ribbon structural change on the amplitude and/or frequency of spontaneous excitatory postsynaptic currents (sEPSCs). We monitored sEPSCs (mediated by AMPA receptors) in postsynaptic b2 Off cone bipolar cells in slices taken from hibernating and awake ground squirrel retinas as a readout of vesicle release from presynaptic cones2, 3. Recordings from retinas of hibernating animals exhibited a pronounced decrease in sEPSC frequency and amplitude. To rule out some other factors, we examined both pre- and post-synaptic elements. We found no change in resting membrane potential of cones, nor an obvious change in the distribution of postsynaptic glutamate receptor localization. The presynaptic cone calcium currents were found to be slightly smaller in the hibernating tissues, and there was no change in current-voltage relationship. Nonetheless, even a small difference in calcium current can have large effects on neurotransmitter release rates. To test whether calcium current underlies the effects of hibernation on sEPSC frequency, we used 100 M Co2+ to modestly reduce calcium channel currents in awake animals. Reduction of the calcium current by approximately 20% resulted in a reduction of sEPSC frequency by approximately 25% and a slight shift in the amplitude distribution of sEPSC, as expected for the decrease in frequency. In contrast, hibernating animals exhibited sEPSC frequencies that were 90% smaller than in awake animals. Evidently, the reduction in calcium current cannot account for the reduction in sEPSC frequency. Moreover, we were not able to recover the frequency of sEPSCs with presynaptic cone depolarization, higher extracellular Ca2+ concentration, or treatment with Bay K, indicating that the large portion of the ribbon structure removed from the membrane during hibernation is critical for maintaining sEPSC frequency. Another important feature of vesicle release from the ribbon synapse is multi-vesicular release (MVR) bulk/coordinated release of transmitter contents from multiple vesicles at a single synapse. We observed that, on average, the amplitude of the sEPSCs were 60% smaller in hibernating squirrel retinas, with no change in the kinetics of the individual events. An amplitude histogram revealed that the decrease in amplitude is the result of the specific loss of large amplitude events. To test whether the effect on sEPSC size was secondary to the change in frequency and whether the remaining ribbon is capable of generating MVR, we next depolarized cones in paired cone bipolar recordings to raise the frequency of sEPSCs in bipolar cells in hibernating squirrel retinas. Indeed, we observed an emergence of larger events in the amplitude histogram. When compared with awake tissues that were treated with a low concentration of Co2+ in order to match the frequencies of depolarized hibernating cones, the amplitude distributions were essentially the same. Evidently, the remaining small ribbons, which are limited to the very bottom compartment, are sufficient to support multivesicular release, but are unable to support normal frequencies of sEPSCs. These data, taken together, lead us to conclude that 1) the upper portion of the ribbon (which is eradicated during hibernation) plays an important role in spontaneous vesicle release, and 2) the bottom portion of the ribbon is sufficient to generate MVR.
