Desensitization and binding properties determine distinct a1b2c2 and a3b2c2 GABA A receptor-channel kinetic behavior Andrea Barberis, 1 Jerzy W. Mozrzymas, 1,3 Pavel I. Ortinski 1,2 and Stefano Vicini 1,2 1 Department of Physiology and Biophysics and 2 Interdisciplinary Program in Neuroscience, Georgetown University School of Medicine, 225 Basic Research Building, 3900 Reservoir Road, NW Washington, DC 20007, USA 3 Laboratory of Neuroscience, Department of Biophysics, Wroclaw Medical University, Wroclaw, Poland Keywords: GABA, kinetics, patch clamp, rat, subunit Abstract GABA A receptor subtypes comprising the a1 and a3 subunits change with development and have a specific anatomical localization in the adult brain. These receptor subtypes have been previously demonstrated to greatly differ in deactivation kinetics but the underlying gating mechanisms have not been fully elucidated. Therefore, we expressed rat a1b2c2 and a3b2c2 receptors in human embryonic kidney 293 cells and recorded current responses to ultrafast GABA applications at macroscopic and single-channel levels. We found that the slow deactivation of a3b2c2-mediated currents is associated with a relatively small rate and extent of apparent desensitization. In contrast, responses mediated by a1b2c2 receptors had faster deactivation and stronger desensitization. a3b2c2 receptors had faster recovery in the paired-pulse agonist applications than a1b2c2 channels. The onset of currents mediated by a3b2c2 receptors was slower than that of a1b2c2 for a wide range of GABA concentrations. Single-channel analysis did not reveal differences in the opening ⁄ closing kinetics of a1b2c2 and a3b2c2 channels but burst durations were longer in a3b2c2 receptors. Simulation with a previously reported kinetic model was used to explore the differences in respective rate constants. Reproduction of major kinetic differences required a smaller desensitization rate as well as smaller binding and unbinding rates in a3b2c2 compared with a1b2c2 receptors. Our work describes the mechanisms underlying the kinetic differences between two major GABA A receptor subtypes and provides a framework to interpret data from native GABA receptors. Introduction GABA A receptors (GABA A Rs) are pentameric structures and as many as 20 subunits (a1–6, b1–4, c1–3, d, q1–3, e, p and h) have been cloned so far (Fritschy & Brunig, 2003). The kinetics and pharmacology of GABA A Rs strictly depend on their subunit composition (Whiting, 2003). a subunits are critical determinants of ligand binding and of activation, deactivation and desensitization kinetics of GABA-induced responses of native and recombinant GABA A Rs (Verdoorn, 1994; Gingrich et al., 1995; Lavoie et al., 1997; McClellan & Twyman, 1999; Bohme et al., 2004). Such a subunit dependence of GABA A R function appears to be associated with differential localization and the specific functions of these receptors in the central nervous system (Pirker et al., 2000; for review see Rudolph & Mohler, 2004). Receptors containing the a3 subunit are characterized by slow deactivation, slow desensitization onset and low affinity (Verdoorn, 1994; Gingrich et al., 1995). Such a peculiar kinetic pattern has been shown to play an important role in the regulation of the network temporal resolution in the early development stage (Ortinski et al., 2004). a3-containing receptors have been shown to be abundant in serotoninergic neurons in Raphe nuclei (Gao et al., 1993) and Browne et al. (2001) attributed the slow GABAergic current kinetics observed in thalamic neurons to the presence of the a3 subunit. Although the physiological relevance of the spatial and temporal differential expression of a3-containing receptors in the brain is still unclear, it is likely that the a3 subunit could efficiently serve as a prolonged and sustained synaptic GABAergic control. Thus, it is convenient to investigate the kinetic features of a3b2c2 receptors in comparison to the a1b2c2 receptor subtype that has been extensively studied and is believed to be most abundantly expressed among all GABA A Rs in the central nervous system (Whiting, 2003). Verdoorn (1994) and Gingrich et al. (1995) have analysed the currents elicited by exogenous GABA applications and found profound differences in activation, deactivation and desensitization kinetics between currents mediated by a1 and a3 subunit-containing receptors. Although Verdoorn (1994) did not address a detailed determination of receptor gating, Gingrich et al. (1995) adapted the model of Twyman et al. (1990) and concluded that the differences between binding and unbinding rates were nearly sufficient to account for the functional differences between a1b2c2 and a3b2c2 receptors observed in their experiments. However, a limited temporal resolution of these record- ings (application time < 30 ms, Gingrich et al., 1995) might have precluded detection of the fastest current components. For instance, in some neuronal and recombinant GABA A Rs, the onset of rapid desensitization occurs within a millisecond time scale (Jones & Correspondence: Dr Stefano Vicini, 1 Department of Physiology and Biophysics, as above. E-mail: svicin01@georgetown.edu Received 1 September 2006, revised 1 March 2007, accepted 9 March 2007 European Journal of Neuroscience, Vol. 25, pp. 2726–2740, 2007 doi:10.1111/j.1460-9568.2007.05530.x ª The Authors (2007). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd