| First Author | Tsetsenis T | Year | 2011 |
| Journal | Proc Natl Acad Sci U S A | Volume | 108 |
| Issue | 34 | Pages | 14300-5 |
| PubMed ID | 21844341 | Mgi Jnum | J:175234 |
| Mgi Id | MGI:5285014 | Doi | 10.1073/pnas.1112237108 |
| Citation | Tsetsenis T, et al. (2011) Rab3B protein is required for long-term depression of hippocampal inhibitory synapses and for normal reversal learning. Proc Natl Acad Sci U S A 108(34):14300-5 |
| abstractText | Rab3B, similar to other Rab3 isoforms, is a synaptic vesicle protein that interacts with the Rab3-interacting molecule (RIM) isoforms RIM1alpha and RIM2alpha as effector proteins in a GTP-dependent manner. Previous studies showed that at excitatory synapses, Rab3A and RIM1alpha are essential for presynaptically expressed long-term potentiation (LTP), whereas at inhibitory synapses RIM1alpha is required for endocannabinoid-dependent long-term depression (referred to as 'i-LTD'). However, it remained unknown whether i-LTD also involves a Rab3 isoform and whether i-LTD, similar to other forms of long-term plasticity, is important for learning and memory. Here we show that Rab3B is highly enriched in inhibitory synapses in the CA1 region of the hippocampus. Using electrophysiological recordings in acute slices, we demonstrate that knockout (KO) of Rab3B does not alter the strength or short-term plasticity of excitatory or inhibitory synapses but does impair i-LTD significantly without changing classical NMDA receptor-dependent LTP. Behaviorally, we found that Rab3B KO mice exhibit no detectable changes in all basic parameters tested, including the initial phase of learning and memory. However, Rab3B KO mice did display a selective enhancement in reversal learning, as measured using Morris water-maze and fear-conditioning assays. Our data support the notion that presynaptic forms of long-term plasticity at excitatory and inhibitory synapses generally are mediated by a common Rab3/RIM-dependent pathway, with various types of synapses using distinct Rab3 isoforms. Moreover, our results suggest that i-LTD contributes to learning and memory, presumably by stabilizing circuits established in previous learning processes. |
