Abstract

Sometimes the best way to solve a problem is to look at it from a different perspective. That is certainly the case in ion channel research. Hodgkin and Huxley must have realized that 50 years ago when they looked at membrane currents with a voltage clamp, and Sakmann and Neher realized it 20 years ago when they measured single-channel currents with the patch-clamp technique. The cloning and mutagenesis of ion channel cDNAs over the past decade has also provided a new perspective into the molecular mechanisms of ion channel function. Most recently, however, it has been structural determination of ion channels that has been providing the new perspectives. In this issue of Neuron, MacKinnon and his colleagues provide another look at the structure of a K+ channel, this time an intracellular domain of the E. coli K+ channel, which is also present in large conductance Ca2+-activated K+ channels (BK channels) (Jiang et al., 2001xJiang, Y., Pico, A., Cadene, M., Chait, B.T., and MacKinnon, R. Neuron. 2001; 29: 593–601Abstract | Full Text | Full Text PDF | PubMed | Scopus (210)See all References)(Jiang et al., 2001). This domain, following the inner helix of the pore, is in a region of the channel thought to couple intracellular ligand binding to the conformational change that opens the channel pore. This structure promises to yield new insights into the molecular mechanisms of gating.Channels that are gated by intracellular ligands are beginning to take their rightful place beside the voltage-gated and neurotransmitter-gated channels. These channels include the large and small conductance Ca2+-activated K+ channels, the cyclic nucleotide-gated (CNG) channels, and the K(ATP) channels. They are all members of the P region–containing family of channels where the pore is formed by the last two transmembrane segments and intervening P region of each of the four subunits. The structure of this region is typified by the structure of another bacterial K+ channel, KcsA from S. lividans (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 (4570)See all References)(Doyle et al., 1998). In KcsA, the inner helices of each of the four subunits are arranged like an inverted teepee with a “smokehole” at the intracellular end where the ions enter and exit, and a selectivity filter at the extracellular end formed by the P region. This structure has already yielded unprecedented insights into the mechanism for ion selective conduction in K+ channels. Furthermore, it has been hypothesized that movement of these inner helices, and perhaps widening of the “smokehole,” underlies gating (Liu et al., 1997xLiu, Y., Holmgren, M., Jurman, M.E., and Yellen, G. Neuron. 1997; 19: 175–184Abstract | Full Text | Full Text PDF | PubMed | Scopus (379)See all References)(Liu et al., 1997). In fact, movement of the inner helices in KcsA has been directly observed with EPR spectroscopy (Perozo et al., 1999xPerozo, E., Cortes, D.M., and Cuello, L.G. Science. 1999; 285: 73–78CrossRef | PubMed | Scopus (456)See all References)(Perozo et al., 1999). Consistent with this idea, virtually all channels gated by intracellular ligands contain domains following the inner helix of the pore that are involved in ligand binding or the coupling of ligand binding to channel opening. This suggests a simple model for channel activation where conformational changes in a carboxy-terminal ligand binding domain are directly coupled to movement of the inner helices and opening of the pore.The BK channel is particularly interesting because it is gated by both voltage and intracellular Ca2+. The voltage gating is presumed to result from movement of the S4 segments, while the Ca2+ appears to bind to an intracellular domain in the carboxy-terminal region referred to as the Ca2+ bowl (Schreiber and Salkoff, 1997xSchreiber, M. and Salkoff, L. Biophys. J. 1997; 73: 1355–1363Abstract | Full Text PDF | PubMedSee all References)(Schreiber and Salkoff, 1997). Both the movement of the S4 segments and the binding of Ca2+ to each subunit are thought to modulate the same opening conformational change in the pore (Horrigan et al. 1999xHorrigan, F.T., Cui, J., and Aldrich, R.W. J. Gen. Physiol. 1999; 114: 277–304CrossRef | PubMed | Scopus (157)See all References, Rothberg and Magleby 1999xRothberg, B.S. and Magleby, K.L. J. Gen. Physiol. 1999; 114: 93–124CrossRef | PubMed | Scopus (93)See all References). This highlights the conservation in the mechanisms of voltage-dependent gating and ligand-dependent gating. One domain, between the inner helix and the Ca2+ bowl, is particularly intriguing given its location and strong conservation. This domain, now referred to as the RCK (for regulating conductance of potassium) domain, is found not only in BK channels, but also in a majority of prokaryotic K+ channels and in TrkA, an intracellular component of the Trk system of prokaryotic K+ transporters. In some of these proteins, RCK domains may bind NAD; however, the NAD binding motif is not conserved in most of the K+ channels, indicating that the RCK domain may serve a function other than binding a nucleotide cofactor.In this issue of Neuron, Jiang et al. present the X-ray crystal structure of the RCK domain of the E. coli K+ channel. The structure contains some reoccurring themes and some novel features. The core of the RCK domain forms a Rossmann fold, an α/β structure with a six-stranded parallel β sheet sandwiched between two sets of α helices. Rossmann-fold domains have been found in a variety of proteins where they exhibit diverse functions. In many of these proteins, though, the Rossmann fold serves as a binding site for a substrate or cofactor, suggesting the possibility that the RCK domain of the E. coli K+ channel may serve a ligand binding function.One particularly intriguing finding is the presence of RCK dimers in the crystals. One dimer involves a helix-strand-helix structure extending from one edge of the Rossmann fold, and the other involves conserved hydrophobic resides on the external face of helix D. These dimers exhibit a 2-fold rotational symmetry, in striking contrast to the 4-fold rotational symmetry predicted for the pore-lining segments. This suggests the possibility that, while the pore may form as a 4-fold symmetric tetramer as in KcsA, the RCK domains may form as a dimer of dimers in the intact channel (see Figure 1Figure 1) . A similar symmetry has been seen for the glutamate binding domains of the GluR2 channel (Armstrong et al., 1998xArmstrong, N., Sun, Y., Chen, G.Q., and Gouaux, E. Nature. 1998; 395: 913–917CrossRef | PubMed | Scopus (531)See all References)(Armstrong et al., 1998) and has been proposed for the cyclic nucleotide binding domain of CNG channels (Liu et al., 1996xLiu, D.T., Tibbs, G.R., and Siegelbaum, S.A. Neuron. 1996; 16: 983–990Abstract | Full Text | Full Text PDF | PubMed | Scopus (144)See all References)(Liu et al., 1996), suggesting that dimer symmetry may be a reoccurring theme in this family of ligand-gated channels.Figure 1Possible Symmetries of Tetrameric ChannelView Large Image | View Hi-Res Image | Download PowerPoint SlideOne of the novel features of the E. coli RCK domain structure is a salt bridge between a lysine and aspartic acid residue located 27 amino acids apart in the primary sequence. These residues are conserved in all RCK domains, but are not found in other Rossmann-fold structures. Using a double mutant cycle analysis, Jiang et al. show that this salt bridge is also present in the RCK domain of BK channels. These results attest to the applicability of this RCK domain structure to the RCK domain of BK channels, for which much more functional information is known. Furthermore, the large effects of the charge-reversal mutations on the gating of BK channels points to the involvement of the RCK domain in the coupling of ligand binding to channel opening. These mutations represent only the tip of the iceberg to investigations of the molecular movements associated with gating of BK channels, and this structure opens the door to many more such studies.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call