Ionotropic glutamate receptors GluRs are ligand gated
Ionotropic glutamate receptors (GluRs) are ligand-gated ion channels that mediate information processing at the majority of excitatory synapses in the brain and participate in such physiological processes as learning and memory, development and maintenance of cellular connections, and pain perception (Dingledine et al., 1999). Dysfunctional GluRs have also been implicated in numerous neurodegenerative and psychiatric disorders (Doble, 1999). The GluR channel shares a common design with a K+ channel, although it is inverted in the membrane (for recent reviews see Kuner et al., 2003, Wollmuth and Sobolevsky, 2004). The major pore-lining domains in GluR channels, the M2 loop and the M3 segment (Fig. 1), are structurally similar to the P loop and the inner helix (TM2 in KcsA or MthK) in K+ channels (Kuner et al., 1996, Kuner et al., 2001, Panchenko et al., 2001, Sobolevsky et al., 2003). Additionally, the GluR M3 segment, like the homologous TM2 domain in K+ channels, is extensively involved in channel gating (Kohda et al., 2000, Jones et al., 2002, Sobolevsky et al., 2002, Sobolevsky et al., 2003).
To study the electrostatic potential in GluRs, we took advantage of substituted cysteines located at different levels in the α-amino-3-hydroxy-5-methylisoxazole-4-proprionic Closantel Sodium receptor (AMPAR) channel pore and measured the voltage dependence of the rate of their modification by externally applied methanethiosulfonate (MTS) reagents. The voltage dependence was distinctly state dependent. In the presence of glutamate, the voltage dependence became gradually stronger for positions located deeper in the pore suggesting that the electrostatic potential drops fairly uniformly across the pore in the open state. Surprisingly, we find that a much greater portion of the transmembrane electric field drops across the narrow region of the pore (intracellular vestibule) in the closed than in the open state. We suggest that this state-dependent change in the electrostatic potential arises from a differential distribution of charges within the pore during gating. Structurally, this state-dependent charge distribution may be due to a movement of the M2 α-helix dipoles during gating with the negative (C-terminal) poles of these dipoles pointed toward the center of the pore in the open state and away from Genotype in the closed state.
Results To characterize the electrostatic potential across the pore of the AMPAR channel, we measured the voltage dependence of the rate of reactivity of MTS reagents with substituted cysteines. We focused on cysteines introduced at five positions in the pore-forming M2 and M3 domains that are accessible to extracellularly applied MTS reagents both in the presence (closed and open states) and absence (closed state) of glutamate (Kuner et al., 2001, Sobolevsky et al., 2003). Four of these positions are located in the M3 segment (L−5, T+2, L+5, and F+8) and one (D586) is in the M2 loop (Fig. 1). Labels and a description of mutant GluR-A subunits containing cysteine substitutions at these five positions are listed in Table 1. Initially, we measured modification rates of cysteine-substituted AMPAR channels by MTS reagents in the presence or absence of glutamate.
Discussion We used the voltage dependence of the rate of modification of substituted cysteines as a tool to probe the electrostatic potential across the pore of the GluR channel. The interpretation of our results is limited by the assumptions of the substituted cysteine accessibility method (Karlin and Akabas, 1998). For example, we assume that the cysteine (as well as other) substitutions do not alter greatly the structure of the protein. Additionally, we assume that the MTS reagents themselves do not change significantly the distribution of the electrostatic potential inside the pore. Although this assumption may not be completely correct for the absolute values of zδ, it seems likely that the corresponding errors would be minimized or canceled out when comparing zδ values for different positions in the pore as well as considering the state-dependent differences (Δzδ).