Rebuttal from David M. MacLean

Abstract

The main effects of the transmembrane AMPA receptor (AMPAR) auxiliary proteins (TARPs) on AMPAR gating are slowing of deactivation and desensitization, increasing peak Popen and speeding of recovery from desensitization (Priel et al. 2005; Tomita et al. 2005). All of these observations can be reproduced by increasing the rate of recovery from desensitization, γ, and by one of two additional means: either increasing the open rate, β (as proposed by Dr Howe, 2013), or reducing the cleft opening rate, CO, and desensitization rate, δ (as suggested by myself, MacLean, 2013). Moreover, increasing the channel opening rate or decreasing the cleft open rate both result in the reported substantial left-shift in steady-state glutamate EC50 values produced by TARPs (Priel et al. 2005; Tomita et al. 2005; Kott et al. 2007) as well as the smaller left-shift in peak EC50 values (Morimoto-Tomita et al. 2009). Therefore, such concentration–response data are not helpful in distinguishing an effect on gating transitions from an effect on pre-gating transitions such as stabilizing closed-cleft states. However, peak inhibition curves using NBQX and CNQX may provide some support for the hypothesis that TARPs affect pre-gating transitions by promoting states with closed ligand-binding domains (LBDs). The LBDs of AMPARs exist in multiple closed-cleft conformations, both when the agonist is bound and in the apo state (Landes et al. 2011). If the presence of TARPs causes AMPAR LBDs to adopt more closed conformations, then even in the apo state the binding cleft would be, on average, slightly more shut. Consequently, larger ligands such as NBQX or CNQX would have slower binding rates since there would be fewer favourable collisions between these compounds and the more closed LBDs. Such slowed binding rates of larger antagonists would result in right-shifts of peak inhibition curves and this is precisely what is observed experimentally. The peak response IC50 values obtained using rapid perfusion on outside-out patches for both CNQX and NBQX are right-shifted approximately 2-fold by stargazin (MacLean & Bowie, 2011). While this observation is consistent with TARPs stabilizing more closed LBD conformations, more compelling evidence is needed. Ultimately, to discriminate between these mechanisms, direct measurements of both the degree of LBD closure and coupling between LBD closure and channel opening will be required.