The B2P6 chimera was completely insensitive to ion exchanges (peak currents, relative to NaCl: 94% ± 1% for CsCl; 95% ± 5%, NaNO3, n = 3, not shown). Glutamate (10 mM) activated a large steady state current at the B2P6 chimera
(21% ± 1% of peak, n = 15 patches), reminiscent of the Willardiine series of partial agonists (Jin et al., 2003). To check that glutamate remains a full agonist at the B2P6 chimera, we estimated open probability using noise analysis (Figures S2A–S2D). Wild-type GluA2 receptors have a high peak open probability (77% ± 7%, n = 5 patches), and the peak open probability of the B2P6 chimera was not significantly different (65 ± 5%, n = 5 patches; p > 0.05, randomization test). Weighted single-channel conductance was also similar (WT A2: 18 ± 3 pS; B2P6: Talazoparib in vivo 16 ± 1 pS). Additionally, we checked if quisqualate, which activates a larger current than glutamate in GluA2 mutants where domain closure is hindered (Robert et al., 2005), could activate c-Met inhibitor bigger responses than glutamate at the B2P6 construct (Figures S2E and S2F). Currents activated by quisqualate (2 mM) and glutamate (10 mM) were similar in amplitude (Quis. peak current: 92% ± 8% of that evoked by 10 mM glutamate, n = 5 patches), suggesting that domain closure in
the B2P6 channel is normal. These results exclude spurious partial agonism as an explanation for fast recovery, and suggest that the large steady state current in the B2P6 chimera is due to recovery that is even faster than wild-type GluA2. Our recordings of the B2P6 and B6P2 chimeras displayed striking features that we reasoned could constrain parameters in simulations of receptor kinetics and thereby provide insight to the molecular mechanisms determining recovery from desensitization. Our aim was to identify if individual
kinetic transitions could explain the observed behavior and, by comparing with existing biophysical studies (Robert et al., 2005 and Zhang et al., 2008), pinpoint the region of the LBD most likely to control recovery. Using a simplified (-)-p-Bromotetramisole Oxalate model of GluR activation (see Figure 2A and Supplemental Experimental Procedures), we tried three scenarios to account for changes in recovery rate. First, we varied the lifetime of a deep desensitized state (AD2), from which agonist dissociation was very slow. Second, alterations in the bound lifetime of glutamate might change recovery, and we simulated this on two backgrounds, with initially slow and fast recovery, respectively. Finally, we tested the hypothesis that the equilibrium between resting (AR) and desensitized states (AD) differs between AMPA and kainate receptors. The rate of recovery from desensitization was sensitive to rate changes in each case.