Taking Apart the Gating of Voltage-Gated K + Channels

Abstract

Though the precise mechanism of C-type inactivation is unknown, several observations suggest that the mechanism involves a partial collapse of the selectivity filter in the pore (for review, see Yellen 1998xYellen, G. Q. Rev. Biophys. 1998; 31: 239–295Crossref | PubMed | Scopus (343)See all ReferencesYellen 1998). More recent evidence suggests that C-type inactivation may occur in two steps. In the first step, termed P-type inactivation, the outer mouth of the pore closes. In the second step, the closure of the outer gates is stabilized by an interaction with the extruded S4 segment to reach what generally has been described before as the C-type inactivated state (18xOlcese, R., Latorre, R., Toro, L., Bezanilla, F., and Stefani, E. J. Gen. Physiol. 1997; 110: 579–589Crossref | PubMed | Scopus (133)See all References, 16xLoots, E. and Isacoff, E.Y. J. Gen. Physiol. 1998; 112: 377–389Crossref | PubMed | Scopus (138)See all References).In this issue of Neuron, Gandhi et al. 2000xGandhi, C.S., Loots, E., and Isacoff, E.Y. Neuron. 2000; 27: 585–595Abstract | Full Text | Full Text PDF | PubMedSee all ReferencesGandhi et al. 2000 explored the potential interaction between S4 and the pore domain further by tracking gating motions in Shaker K+ channels under voltage clamp. This was accomplished by systematically measuring the fluorescence changes of covalently linked fluorophores at numerous sites in S4 and the pore domain and looking for emerging patterns. A couple observations reinforce the notion that the S4 segment interacts with the pore during slow inactivation. First, based on the kinetics of ΔFon, positions along S4 can be divided into three zones that run vertically in stripes along S4. This pattern is consistent with an outward movement of S4 during channel activation. Second, at many positions in S4, the ΔFons and the ΔFoffs are asymmetric, a phenomenon they refer to as fluorescence hysteresis. They convincingly demonstrate that this originates from an interaction between S4 and the inactivating pore. Overall, their results are consistent with S4 segment undergoing a “helical screw” motion during channel activation. Following P-type inactivation, they propose that the S4 segment tilts inward and stabilizes the inactivated pore.What type of rearrangements occurs within the pore domain itself? Larsson and Elinder 2000xLarsson, P. and Elinder, F. Neuron. 2000; 27: 573–583Abstract | Full Text | Full Text PDF | PubMedSee all ReferencesLarsson and Elinder 2000, also in this issue of Neuron, describe the use of cysteine mutagenesis to demonstrate the role of a conserved glutamate at the end of S5. First, they demonstrate that this side chain, E418, is likely to hydrogen bond with another residue in the pore domain. Using the KcsA structure to identify candidate interacting partners, they introduced cysteines into paired positions and induced the formation of disulfide bonds. This worked for two sets of mutations: E418C/G452C and E418C/V451C. Both of these interacting partners are located near the end of the P–S6 linker. Surprisingly, a disulfide bridge between E418C and G452C stabilizes the open state, while a disulfide bridge between E418C and V451C stabilizes the inactivated state. The opposing effects of disulfide bridges in two neighboring positions suggest that during slow inactivation the P–S6 loop rotates. This leads to a model of how the pore constricts during P-type inactivation. In KcsA, the selectivity filter is held open by a network of tryptophans that interact between subunits to form an aromatic cuff that encircles the pore (Doyle et al. 1998xDoyle, D.A., Morais Cabral, J., Pfuetzner, R.A., Kuo, A., Gulbis, J.M., Cohen, S.L., Chait, B.T., and MacKinnon, R. Science. 1998; 280: 69–77Crossref | PubMed | Scopus (4615)See all ReferencesDoyle et al. 1998). The side chain of a conserved proline (P450 in Shaker) is perched just above where two tryptophans meet. A rotation of the P–S6 loop would cause this proline to rotate outward and allow the tryptophans to move closer together. This could tighten the aromatic cuff and cause the selectivity filter to narrow. As the authors point out, this model agrees well with experiments showing that a cysteine in P450 in Shaker becomes more exposed to the extracellular solution during slow inactivation (for review, see Yellen 1998xYellen, G. Q. Rev. Biophys. 1998; 31: 239–295Crossref | PubMed | Scopus (343)See all ReferencesYellen 1998).We are becoming increasingly familiar with the parts that comprise a voltage-gated K+ channel. The next major objective will be to understand how these different parts—the selectivity filter, the gates, the voltage sensor, the T1 domain, and others—operate together. Once these long-standing questions are “settled,” we will have a completed picture of how a voltage-gated ion channel works.*To whom correspondence should be addressed (e-mail: gkw@itsa.ucsf.edu).