The importance of TARPs for AMPAR expression and function
The importance of TARPs for AMPAR expression and function was revealed initially in the cerebellum, where the lack of γ-2 in the mutant mice waggler and stargazer (stg/stg) was associated with a selective loss of AMPAR-mediated synaptic currents in cerebellar granule Budesonide receptor (Chen et al., 2000; Hashimoto et al., 1999; Tomita et al., 2003). Whereas granule cells contain only CI-AMPARs, a variety of other cerebellar neurons and glia express both CP- and CI-AMPARs (see Fig. 1). Recent studies conducted on the cerebellum of stg/stg mice indicate that the extent of the disruption to AMPAR-mediated currents caused by the absence of γ-2 varies from one cell type to another, and depends both on the other TARP isoforms normally expressed, as well as the subtypes of AMPARs present (Bats et al., 2012; Jackson and Nicoll, 2011; Menuz et al., 2008; Yamazaki et al., 2010). There is now growing evidence for a differential regulation of CI- and CP-AMPARs by TARPs (Bats et al., 2012; Soto et al., 2007, 2009; Yamazaki et al., 2010; Zonouzi et al., 2011). Below we present recent findings and discuss the specific roles of γ-2 and other TARPs in the regulation of CP-AMPAR expression and plasticity.
CP-AMPARs in the cerebellum
Mechanisms underlying CP-AMPAR plasticity in the cerebellum Since the initial identification in cerebellar stellate cells of synaptic plasticity that involves a switch in the expression of AMPAR subtypes, from calcium-permeable to -impermeable (Liu and Cull-Candy, 2000), it has emerged that such dynamic changes in AMPAR GluA2-content occur widely throughout the CNS. Thus, the insertion and activation of CP-AMPARs, at normally ‘CI-AMPAR only’ synapses, seems to play a role in early stages of hippocampal LTP (Asrar et al., 2009; Lu et al., 2007; Plant et al., 2006; Yang et al., 2010). Other studies, using CP-AMPAR blockers, suggest that presence of CP-AMPARs is not required for the induction or maintenance of LTP at Schaffer collateral-to-CA1 pyramidal cell synapses (Adesnik and Nicoll, 2007; Gray et al., 2007). Therefore, this issue remains unresolved. Importantly, changes in AMPAR subtype prevalence can occur during both physiological and pathological events and have been described in the lateral amygdala following fear conditioning (Clem and Huganir, 2010), in cortical neurons in response to sensory stimulation (Clem and Barth, 2006), in the ventral tegmental area of cocaine-treated animals (Bellone and Luscher, 2006), and in post-ischemic forebrain (Liu et al., 2004) and hippocampus (Noh et al., 2005). Our experiments, focussing mainly on cerebellar stellate- and NG2+ cells, have identified some of the mechanisms underlying CP-/CI-AMPAR plasticity and the auxiliary subunits involved in this process.
Role of TARPs in AMPAR trafficking, synaptic transmission, and plasticity in cerebellar cells
Conclusion The dynamic regulation of CP-AMPARs plays a key role in the physiology of various neurons and glial cells in the central nervous system. Based on recent findings in the cerebellum, TARPs are involved not only in canonical forms of plasticity such as hippocampal LTP and LTD (Tomita et al., 2005b), and LTD in Purkinje cells (Nomura et al., 2012), but are also implicated in CP-AMPAR plasticity. While γ-5 appears to play a specific role in CP-AMPAR delivery (Soto et al., 2009; but see Kato et al., 2008), it will be interesting to determine whether the other type II TARP γ-7 shows similar selectivity. However, other important questions also remain to be addressed, including: (1) how do TARPs contribute to the differential expression of CP- and CI-AMPARs beyond the cerebellum, (2) what are the molecular mechanisms accounting for the apparent opposite action of γ-2 in the regulation of CP-AMPAR expression in neurons and glia, (3) how do type I and type II TARPs cooperate to regulate the relative expression of CP- and CI-AMPARs, and (4) are other auxiliary subunits also involved in this process? Indeed, while finalizing this review, a study appeared showing that CNIH-2/-3 (Cornichon-2/-3, another group of AMPAR auxiliary subunits) selectively promotes the trafficking of GluA1-containing receptors to the plasma membrane of hippocampal neurons (Herring et al., 2013). Interestingly, this subunit-specific action appears to rely on the prevention of the interaction of CNIH with non-GluA1 subunits by TARP γ-8 (which is enriched in hippocampus). This shows that a CNIH/TARP interplay is involved in the subunit-specific trafficking of AMPARs, and by extension suggests that interaction between different auxiliary subunits could also be involved the control of the expression of specific AMPAR subtypes, such as CP- and CI-AMPARs.