br Results br Discussion Using primarily an electrophysiolog
Discussion Using primarily an electrophysiological analysis of dysbindin-deficient hippocampal neurons in cultures and slices, we have uncovered a previously unreported role for dysbindin in enhancing CA3–CA1 AMPAR-mediated transmission. An earlier study, however, has reported a reduction in CA3–CA1 neurotransmission in the sandy mouse mutant attributed to a reduction in the probability of glutamate release (Chen et al., 2008). Furthermore, consistent with another study (T.T. Tang et al., 2009), we found no changes in PPF, or in mEPSC frequency, suggesting no changes in release probability. The disparity in results may be due to differences in the hippocampal sub-region studied. While their study investigated the ventral hippocampus, ours examined slices from dorsal and mid-septotemporal hippocampus. An interesting hypothesis is that loss of dysbindin expression could modulate synaptic transmission at either the pre- and post-synaptic locus depending on the anatomical location within the hippocampus. Fractionation of purified hippocampal synapses has indeed revealed that dysbindin is present in both pre- and post-synaptic compartments (Talbot et al., 2004, Talbot et al., 2011). Similar to our findings, another group has also reported an enhancement of both CA3–CA1 glutamatergic transmission and LTP in the sandy mutant. However, in scd1 with our study, these effects were attributed to enhanced surface expression of NR2A-containing NMDA receptors (T.T. Tang et al., 2009). Thus, it remains unclear why dysbindin affected NMDAR function in their study and AMPAR function in ours, but could be due to the different genetic background strains used, C57Bl/6J and DBA/2J, respectively. It has indeed been found that strain background strongly affects the phenotype in the sandy mouse which can lead to behavioral defects in the opposite direction (Cox et al., 2009, Hattori et al., 2008, Ji et al., 2009, Takao et al., 2008). We report compelling evidence that dysbindin modulates AMPAR-mediated transmission. The precise mechanism of such modulation remains undefined. Dysbindin has been found in the nucleus and appears to affect gene transcription (Fei et al., 2010). However, we found no changes in the hippocampal expression levels of GluA1–4. Studies have shown that loss or reduced dysbindin expression lead to an increase in the surface expression of the NR2A-containing NMDA receptor and the D2R due to protein mis-trafficking (Iizuka et al., 2007, Ji et al., 2009, Marley and von Zastrow, 2010, Tang et al., 2009b). Moreover, a mutation in the DTNBP1 gene leads to Hermansky–Pudlak syndrome in humans, characterized by defects in blood clotting and pigmentation (Li et al., 2003). This is due to defects in a specialized class of lysosome-related organelles, important for the storage of molecules to be secreted (Di Pietro and Dell'Angelica, 2005, Di Pietro et al., 2006). Therefore, one mechanism for the enhanced AMPAR response seen here may involve the mis-trafficking of AMPARs from intracellular endosomal stores (Lee et al., 2001) to the membrane surface of synapses (Park et al., 2004). Several studies have found that dysbindin interacts with components of the exocyst complex (Camargo et al., 2007, Gokhale et al., 2012, Mead et al., 2010) which has been shown to be involved in the trafficking of the AMPAR to the membrane surface (Gerges et al., 2006, Mao et al., 2010). An alternative functional partner may involve another susceptibility gene for schizophrenia, disrupted in schizophrenia (DISC) 1 (St Clair et al., 1990). It was reported that a reduction of DISC1 led to an increase in expression of GluR1 (Wang et al., 2011). An interaction between dysbindin and DISC1 has been reported (Ottis et al., 2011) suggesting that these two could modulate AMPAR trafficking through a common mechanism. Because of dysbindin's link to schizophrenia, our data points to a potential role for AMPAR-mediated synaptic transmission in the disorder. Post-mortem studies of schizophrenia patients have generally found a reduction of AMPARs in the hippocampus (Kerwin et al., 1990). However, analysis of schizophrenia patient brain tissue revealed an increase of GluA1 in early endosomes, and an increase in the expression of GRIP1 and SAP97, two proteins involved in AMPAR trafficking, as well as changes in the expression of regulatory proteins of AMPAR function (Drummond et al., 2013, Hammond et al., 2010). Based on these data, the authors concluded that an abnormality in forward trafficking of AMPARs to the synapse was associated with the disease pathophysiology. Further evidence for a role of AMPAR in schizophrenia can be seen in a mouse model which does not express GluA1 subunit and displays behavioral phenotypes related to schizophrenia (Wiedholz et al., 2008). Based on these studies it appears that proper modulation of AMPAR-mediated synaptic transmission by dysbindin is an integral component in the prevention of schizophrenia.