Human Biology and Disease
Human Biology and Disease BIOL 1210
Popular in Course
Popular in Biology
This 7 page Class Notes was uploaded by Nedra Gulgowski on Monday September 21, 2015. The Class Notes belongs to BIOL 1210 at University of Virginia taught by Staff in Fall. Since its upload, it has received 18 views. For similar materials see /class/209632/biol-1210-university-of-virginia in Biology at University of Virginia.
Reviews for Human Biology and Disease
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 09/21/15
review article The voltagegated potassium channels and their relatives Gary Yellen Department ofNeurobiology Harvard Medical School 220 Longwood Avenue Boston Massachusetts 02115 USA The voltagegated potassium channels are the prototypical members of a family of membrane signalling proteins These protein based machines have p ores that pass millions of ions per second 0 t e membra gates snap open and shut in milliseconds as they sense changes in voltage or ligand concentration The architectural module acr ss h ne with astonishing selectivity and their s an d functional components of these sophisticated signalling molecules are becoming clear but some important links remain to be elucidated lectricity plays an unavoidable role in biology Whenever solutes like phosphate compounds amino acids or inorganic ions are transported across membranes the movement of their charge constitutes an electrical current that produces a voltage difference across the membrane Beyond managing to keep this accumulation of charge from getting too far out of balance all living cells have developed the ability to exploit a transmembrane electrical potential as an intermediate in the storage of energy and the synthesis of ATP However a specialized class of animal cells neurons muscle cells and endo crine cellsknown collectively as the excitable cells have made the management and production of electrical signals into a high art Fast electrical signalling is made possible by the slow and steady homeostatic mechanisms that establish the standard environment and content of animal cells high Na concentration Na in the blood and extracellular uid and high K but low Na and Ca2 in the cytoplasm The plasma membrane separates these l TL 1 t L L 4 and maintained by active transporters and pumps which prepares the way for rapid changes in membrane voltage to be produced by passive transport through ion channels These pore forming proteins allow ions to ow only downhill as dictated by their electrochemical gradients but they do it rapidly and selectively Opening a Na selective channel permits Na to ow down its gradient into a cell makin the intracellular voltage more positive Opening a K selective channel permits K to ow out of a cell and restores the voltage to a negative value The dimensions of a typical cell and its membrane allow the voltage to be changed rapidly back and forth many times with relatively small changes in concentration This is essentially how all cellular electrical signalling is produced The ultimate specialization of electrical signalling is the action potential a stereotyped millisecond long electrical signal that is capable of propagating at rates of metres per second along a nerve bre As Hodgkin and Huxley showed this signal is made possible by a rapid feedback process involving a direct action of voltage on ion channels Na channels activated opened by positive voltage con stituting the positive feedback process followed by K channels activated more slowly by positive voltage the negative feedback Basic channel architecture The voltage gated K channels are the prototypical voltage gated channels At their simplest they are homotetrameric channels with each subunit containing a voltage sensor and contributing to the central pore The standard K4r channel subunit contains six trans membrane regions with both amino and carboxy termini on the intracellular side of the membrane a tetrameric 6TM architecture see Box 1 The pore forming subunits of voltage gated Na and Ca2 channels contain four non identical repeats of this motif strung together on a single polypeptide There is enormous variety within each of these channel families voltage gated K channels alone are made by at least 22 different genes in mammals with additional variety produced by alternative splicing and heteromultimerization Some close relatives ofthe voltage gated K4r channels share the tetrameric 6TM architecture but differ in key functional features These include the sensory channels of the photoreceptors and olfactory neurons which are non selective Na and K channels that are cyclic nucleotide gated the CNG channels and insensitive to voltage and the pacemaker channels found in heart muscle and neurons the HCN channels that are controlled by both cyclic nucleotides and voltage Each of these signalling proteins performs three main functions Ion permeation is the central function and the movement of ions through the pore must be fast and ion selective This permeation is then regulated by opening and closing of the pore a set of conformational changes called gating Finally the gating is coupled to a sensing mechanism which detects transmembrane voltage in the core members ofthe family but can also be geared to sense Ca2 cyclic nucleotides and perhaps other cellular signals From approxi mately 40 years of electrophysiology and biophysics a little over a decade of cloning followed by mutagenesis and physiology on cloned channels and from a few important crystal structures of some bacterial relatives of these channels we now have a relatively clear picture of how these functions are accomplished Exmim te selectivity with high throughput K channels are extremely selective in which ions they allow to pass yet they allow transport rates close to the aqueous diffusion limit Selectivity and speed are both crucial for the biological function of the channels As discussed above in neurons and other cells the ion speci city of a channel for example Na or K determines the direction of current ow Na and Ca ions ow inward and thus carry positive charge into the cell because most voltage activated channels are activated by positive voltage this further activates channels and produces regenerative excitation K ions owin outward or Cl ions owing inward reduce the positive charge inside the cell and terminate or prevent excitation High ow rates are essential for producing these voltage changes rapidly A typical action potential in a mammalian neuron requires millions of ions to ow in a millisecond To accomplish this without using millions of proteins requires that each one have a very high throughput while maintaining high selectivity What are the energetic and structural requirements for rapid permeation The reason that ions require assistance to cr review article Box 1 The tetrame 39 he K channel lamin The TyprcaT voTTage gaTed Tlt channeT Ts an assembTyoHouT TdemTcaT or subunTT has sTxTransmemorane crossTngs ST 561 erh ooTh N and c Termrnron Tcrar re r The narrowesT parT oTThe pore r e seTecTTvTTy oeTween 5 and so The voTTage sensor TncTudes The SA regTon erh TTs mLTHTpTe posTTTve charges The oacTerTaT KcsA Tlt channeT Top rTghT paneT Tn cross secTTon Ts The proToType Tor The pore Th h T H h T pTaTTorrh Tor aTTachrhenT oT The opTTonan supunTTs ancT Tor oTher proTeTn channeT acTTvrTy and may aTso prodee an avenue TorThe channeTs To rgnar uTT Tr T TTTTTu war The exTrerheN oaTT39 rcr see TexT The OTeranT green have no oovrous domaTn sTrucTure Tn ThTs biamTTy aTThough There Ts commonTy a PDZ andTng moTTTTT ThaT deTeranes The physTcaT TocaTTzaTTon oTThe channeT and TTs assocTaTTon The Tnner heTTxTo so The TnT n r erh secondary sTchuTe shown as rrooons Tor Three oT The Tour subunusr and The waTer exposed surTace oTThe proTeTn Ts grey The n rr WWW pr rcr mon andTng STTeS These are TypTcaTTy occupTed by aTTernaTTng KT Tons and T Tam Owe rrr rnr The absence oT aTT domaTn and The presence oT a sensor domaTn Tn The c TeTanus Tower rTghT paneT These TncTude The voTTage gaTed KONG and eagerg channeTs as weTT 032 gaTeg channeTs and channeTs gaTeg echusTveTy by TnTTaceHuTaT TTgands E E g m o g m E E er m s e at The bottom The bund e c TeTTTTTTTaT en or Tossmgquot uT To whereasm oTher bundTe crossrng Ts The waTer TTTTed cavuy casesThey mayTuncTTon as a dTmeT onTmers 2 ATThe NTeTanus some T T e cnanner crganr Tng Ten The sTrucTuraT domaTns oTThe channeTs ATT have The cenTraT cTM addTTTonaT sensor domaTn The auxTTTaTy suounTTs known To be Tm TTTV P MTRD comprTsTng ST SA The core TltvT x Ten par Terr has aT TTs N TeTanus a TeTTameTTZaTTon domaTn Tm pTTrpTeu N ThaT deTeTanes The specTTTcTTy oTsuounTT assembTy and aTso serves as a KvaAx appear ar re depTace The voTTage sensor domaTn or TnTeracTerh TT dTTecTTy membranes is that they have an extremely favourable interaction with water In the vicinity of a positively charged ion like K the the ion In lipid or even in protein there is no analogous electro static accommodation to the presence of the ion these substances a e said to have a relatively low didectric constant compared with ater One option for making the Kquot ion stable inside a protein pore ii i n i i i i review article water molecules Water is extremely exible and accommodating and easily adapts the size and con 39 mmodate ions of different sizes such as Na and K The selectivity lter of the potassium chann 39 water structure around a Kquot i the presumably more compact structu originally proposed by Armstrong Each K 39 lter is surrounded by two groups of four oxygen atoms just as in i i ii i i A i i i i A M exit the pore as required for a high throughput This is not the mechanism used by K channels Instead the Kt channel proteins have four specific architectural features that keep the ion almost exactly as stable as it is in water pgti i ii tip h than maintaining a narrow pore ofatomic dimension through the broad and contains a large amount of water see Fig 1 This explanation was proposm many years ago particularly as an urpla nation of the very large conductance of certain Ca activated Kquot channels and is borne 39 L rriimir of fact the backbone carbonyl oxygens of the selectivity filter loops from the four subunits Finally a long known feature ofpotassium channels is that K ions pass through in single file with simultaneous occupancy by multiple ions The mutual electrostatic repulsion b een adiacent K ions spaced about 7A apart destabilizes ions in the pore permitting the other favourable interactions to produce ion electivity without producing overly tight binding that would imp air rapid permeation The structural and architectural features of potassium channels are thus perfectly adapted to fit their function They solve the much more stable than they are in water by using plenty of water out in L the pores oftwo different bacterial K channels KcsA and MthK3 5 Secondthe K quot 39 39 Every oi helix has a dipole moment owing to the alignment ofthe dipoles ofits hydrogen bonds and the intracellular vestibule ofK its heart3 It has been proposed that these helix dipoles produce a preferential stabilization of cations near the entrance to the narrow selectivity lter This concept is mirrored in the quite different i i i i i i i i i i chloride ions these have the positive ends of multiple helices pointed towards the central ion si ex A third feature ofthe K channels approach to permeation and ion selectivity is creation of a series of customized polar oxygen cages As a K ion diffuses through water itself it is more or less constantly surrounded by a cage of polar oxygen atoms from the m Cm aa ind nanp u Mum Ion munm Figure 1 Architectural features of K cnannels important for ion permeation Approxlmam cross sectlon oran open Kcnannel based on tliecrystal structure ortne MtiilltciiannelS Gail Tne Wide p nignlignted Tne diagram on tne rignt is a VlBW derived from tne nign resolution structure or tne KcsA cnannel snoWing a narrowerand probably closed access to a water tilled Cavity in tne middle ortne membrane protein A trapped W ion purple resolved tne backbone oxygens ortiie seiectivty riiter red spneres provide a good lnn ent within the membrane Furthermore they solve the problem of stabilizing potassium in preference to sodium by precisely matching the con guration of oxygen atoms around a solvated potassium ion Emmy in pan with cnnfnnnarional Inutlnns To use these transmembrane pores for electrical signalling or even or regulation of ion transport it is necessa to re ulate their permeability On the slowest timescales it is possible to control the manufacture of channel proteins by a tra mechanism on a more rapid timescale it is possible to insert multiple channels into the membrane or withdraw them by use of a vesicle fusion mechanism But the very rapid signalling in nscriptional regulatory There arethree established mechanisms by which the voltage gated channels can close two of them involve a conformational constriction of the permeation pathway and one involves con ditional plugging of the pore by an auto inhibitory part of the channel protein lasting 31 no so ninlle nrmmu F39 the close by pinching shut at the intracellular entrance This intracellular or So gate obstructs entrance from the cyt s ic o ewater lled cavity in opla m surface t th the centre of the annel p i he So tran membrane region corresponds to th inner hel of the bacterial KcsA a nel etop extracell ar ch n ds ofthe four so helices form a whereas the bottom ends converge to a right handed bundle crossing below the cavity This bundle crossing corresponds to the functionally determined f at least some of the voltage gated Kquot channels The So transmembrane region and 39ts intracellular extension show very igh sequence conservation wi in the principal families of voltage gated K channels and at the bundle crossing there is a i r u in the bacterial K channels Studies ofthis region indicate that it is likely to bend the so in both the closed and open channels 16 The probable nature of the so opening motion has now been 4 L h d 39 39 ofthe structure ofa second bacterial K channel MthK a tetrameric 2TM that can be opened 2 n i i W ions nas an important role in selectiiityandotner properties ottne channel5 0 2 r 7 y H m n i iiaiiiii A L L i i to 72 selectiveforNa orCa ions L i ii 1 1 J L review article selectivity lter Comparing this with the presumably closed structure of KcsA gives a plausible picture for the gating m tion Fig 2 top and centre panels the 5 helices swing open from an pparent hinge located at a highly conserved glycine residue The H quot Llargi 39 39 7 It is still unkn wn whether the voltage gated channels may use PXP sequence as a second or alternative hinge or whether the PXP sequence provides a xed bend that allows the 55 gate of these quot rail i than to the intracellular calcium sensor used by MthK Figure 2 Tne Conformatlollal clianges tnat gate tne K cliannel poreTlie tnree pest understood c ntormational clianges torclosing potassium cliannels are sliown nere llie centre diagram s owst e core open K on tne pasisottlie MtllK structures Tne selecthity tilter green s sliown Witli a cutaway to reveal tne narrow An important consequence of the 85 gating mechanism is its tt itii 11 ii 1 local anaesthetic like molecules act by binding to sites within the cavity Because of the 55 gate these blockers an enter the channel only after the channel has bem opened by voltage or another stimuli 4 U i be trapped in the cavitywhen the 5 gate closesW The early discovery of these phenomena by Armstrong inspired the initial hypothesis of an intracdlular gate i Blocker gating and tra 39 39 39 anism for state gt8 v n v a n a n r gt8 v a n v a n 8 S o B a a L E E n 7 g r a at is often crucial to the therapeutic value ofchannel drugs for instance anticonvulsants that must inhibit excessive electrical activity without affecting normal lower levels of activity Ballandchain gating i ii it i protein Fig 2 lower right panel Similar to the small organic blockers studied by Armstrongs the N terminus of shaker K Li L impriili to the cavity Because this interaction occurs only after the 85 gate has opened some voltage gated K r channels conduct only transi quotH i i i i i i N terminal peptide blocks the channels electrophysiologists call this N type inactivation N type inactivation also called the ball and chain mechanism 39 gated K 39 l either the N terminus ofthe principal at subunit ofthe channel or the N terminus of an associated 8 subunit It can be disrupted by enzymatic or genetic removal of the N terminal sequence and intracellular bathing solution There is little sequence conserva of inactivation although both positive charge and hydrophobic character are important for the interaction with the open chan nel The peptide appears to inhibit the channel by a simple ii i i ii iiL rr nel blockers is expelled by potassium ow from the opposite side of the membrane and is speci cally aitected by mutation of residues within the cavity The known form ofball and chain inactivation requires an intact 85 gating mechanism but it is easy to imagine some interesting alt rnari Fnrin tan L U l might not itself restrict ion movements but might nevertheless regulate binding ofthe inhibitory peptide This would produce a channel that would close whenever the 55 gate is in the open position Alternatively instead of using the 85 gate to regulate 39 f 39 channel protein might regulate the availability of the N terminus or limit its access to the inner mouth of the channel For instance in one K r channel the inactivation process can be eliminated by disulphide formation between the N terminal ball and a unidenti ed cysteine effectively tying the ball out of the way Channel sequences that are antagonistic to ball and chain inacti NW vation may unction as decoy binding sites luring the peptide quotW quotW quotWW 9 Wquot W away from its real site of action 8 subunits might regulate the tnirtiire Tnemndianra L snows a closed w on tne pasis ot tne KCSA structure ine tourse transmemprane oproducease lose ca lie diagrams illustrate tne open and close states respec c osed at tne SBlBCtlllty tilter lower tan auto innipitory peptide ower lert diagram wnen Witn Weptap viewer nttpwwwaccelryscom Acan nttpwnwac3dorg and POV Ray htlpMwwpovrayorg as as yet unknown NAD dependent enzymatic redox function Finally both at and a subunit N termini as well as permeant ions must approach the pore through windows betwe domain and the transmembrane pore domain 32 see Box 1 Access to and through these windows might be restricted by conformational changes or by the binding ofprotein partners min at m dedlvlv mm A third mechanism for closing the pore is to pinch shut at the e inactivation a alternati ctivation for ty l 7 top right panel This mechanismwas rst recognized in the form of C typ 39 n 39ve to N type ina producing transient K conductance by closin chann in spite of a maintained stimulus C type inactivation persists in Shaker channels when the inactivation ball is deleted and it is quite sensitive to extracellular K1 and to mutations at the r a ntryway to the pore Closure o this gate can be utuu quot quot 39 inn to the pore once closed this inactivation gate can be latched shut by a metal ion that bridges cysteines introduced at the outer mouth in ach of the four subunits35 3 39Ihis inactivation process normally slower than N type inactivation probably as a function in regulating repetitive electrical activity its sensitivity to W i u i i i i Figure 3 Two sensor uornains tnat govern gating in tne w cnannei family lne top panel snows tne proposed sensor gating mechanismquot tortne Calcium gaeu MthK t i tne cnannel supunit one producedasasoluble proteinlneopen structure on tne left oornain anu tne sensor domain is not resolvedThe Closed structure on tne ngnt is proposed on tne basis of tne Closed cnannel strutture of nest1 and the related RCK 1 V Mldllllt RCK uirners union would be produced in MM by Ca dissociationcouldclose tne cnannel by rotating tne uirners anucontratting tne gating ring39lie pott pan l alc u Kcnannel tlne two Tne two supunitson w pinuing of Ca2 ns re to 0am plue produces uirnenzation of tne post so 0 tenninal tails snown alternately in greyanu yellow g E g o m i review article of extracellular K refs 3738 The structural details of c type inactivation probably resemble L 39 L 39 39 39 39 h u t uu channels are crystallized with very low K the selectivity lter contains fewer K ions and instead of pointing towards the central axis the carbonyl oxygens of the lter project obliquely with a partial collapse ofthe lter In some voltage gated K channels the inactivated selectivity lter can continue to conductNa ions when all K is removed suggesting a slightly different collapsed form where the carbonyl oxygens can now comfortably accommodate a 39 Na ion Single channel measurements in voltage gated K channels have suggested that even during the normal activation process which at least in Shaker involves the 5 gate there may be i i i h u i i y have an even more prominent role in other relatives of the voltage gated K r channels In CN channels but not HCN channels the selectivity lter apparently acts as the principal L t uiiuaul quot solution to the 85 region even in the closed state of the channel though there is some change in size restriction with gatingquot and there are clear conformational changes in 86 ref Overall it seems probable that the basic functioning of the 85 gate and the sdectivity lter gate are conserved among different chan 5 gate moves in response to a transmembrane voltage ensor domain and the selectivity lter may react to the change in 55 by opening or closing The conserved glycine that acts as a gating hinge in the MthK channel probably helps to decouple these two motions but not completely Depend i in 1L i ri iwm the Shaker and HCN channels or may remain aiar even in the closed state as inthe CNG andsmall conductance Ca activated K 5K channels Also depending on the details the sdectivity opening or closing or may respond directly to voltage sensor ions Gating sansnrs fur mlmgn and liganns Control of these gating motions by voltage or in the case of other family members by ligand binding is essential for the cha nels f nction Before considering control by voltage I will two sensor mechanisms that are better understood at the el Both of these involve intracellular sensor domains 39 alend ofthe 55 region a otif common to ma ily Box 1 The MthK channel a bacterial ZTM channel alrmdy mentioned i i i u s2 3 v has recently given us a new insight into one possible sensor mech 39 sed by the tetrameric 5TM channels Similar t other channels in the ligand gated branch of the family Box 1 lower right pan MthK channel has a c domain Th regulator of K cond ance do i e here is found also in other K channels such a an Escherichia 011 6TM channel an in a mammalian high con uctance Ca2 activated K BK channel The original nding that RCK domains form homodimers t with the general impression that the four 3 terminal sensor domains ofa tetrameric channel might 39 439 a dimer ofdimers pattern But the MthK channel was a surprise a functional channel contains eight RCK domains one on each channel subunit and one made in a soluble A tight and invariant dimer interface connects each channel domain with a sou e domain and these four dimers each displayed as a single block in Fig 3 top panel then form a ring with four fold symmetry A second interface between the nei bouring dimers has a different appearance in the two known crystal forms of the RCK domains suggesting a remarkable and 39u r 39 review article elegant transduction mechanism By changing from one form of the secondary interface to another the gating ring could contract and expand prying open the S6 gate of the attached channel domain Fig 3 top panel Other ligand gated channels in the family are likely to use multimeric sensor domains The CNG channels use a C terminal nucleotide binding domain that is closely related to the dimeric catabolite activator protein of bacteria The distantly related glutamate receptors upside down K channels with their trans membrane domains ipped and the ligand sensor on the extra cellular side seem to have an important dimer relationship between their sensor domains5 Structural evidence for a possible dimer of dimers sensor mech anism comes from another t e of Ca2 activated channel that appears to signal by Ca induced association of sensors connected to the S6 regions These SK channels have a calmodulin binding domain CaMBD located immediately adjacent to each S6 calmo ulin CaM is stably bound here even in the absence of Ca2 Studies of the detached CaM CaMBD complex show that it behaves lik nomer in the absence of Ca2 but dimerizes in the presence of Ca2 The structure of this Ca induced dimer shows an intimate reciprocal relationship between two CaM CaMBD structures with each CaM embracing both CaMBDs Fig 3 bottom As in any dimer of dimers arrangement formation of such dimers between neighbouring subunits would break the four fold symmetry of the transmembrane channel somehow this dimerization would produce a conformational change in the channel domain that opens the pore Alternatively with a modest change in the structure and using the same protein interfaces seen in the crystal dimer structure the four CaM CaMBD complexes might form a four fold symmetric tetramer upon Ca2 binding The elusive voltage sensor e voltage sensor is obviously central to the function of the voltage gated K channels but its three dimensional structure is still unknown There is however a great deal of functional infor mation about voltage sensing The transmembrane voltage must be sensed by a structure that moves charge at least partially across the membrane The energetics of voltage gating together with direct electrical measurements of the gating charge movement show that as many as four charges per channel subunit are effectively translocated across the membrane as the channel switches between closed and open states53 The expectation that transmembrane charges would be respon sible for voltage sensing made it relatively easy to recognize the probable voltage sensor in the rst channels cloned and sequenced The only sequence motif conserved across all vol tage gated ion channels Na Ca2 K is in the fourth trans membrane region S4 which at every third position has a positively charged arginine or lysine residue Mutations of these charges have complicated effects on voltage gating but the ultimate interpret ation of these experiments is that the S4 charges are indeed the critical gating chargesss These charges in the transmembrane region are energetically disfavoured as discussed above for ion permeation two possible mechanisms to compensate for this are negative countercharges in the S2 transmembrane region 57 and water lled canals 58 at each end of the S4 transmembrane region The canals also focus the voltage drop across a thin septum allowing a small physical movement to have large electrostatic conse quences Translocation ofthese S4 charges from one side ofthe membrane to the other can be con rmed biochemically by replacing a charged residue t i d U i i i i one side or another59 5 A position accessible from the intracellular side at negative voltage can become accessible from the outside at positive voltage Remarkably little of the S4 is actually buried in t e membrane when one charge position is accessible from the 40 2002 Nature Publishing Group outside a site only six residues away in the sequence is accessible from the inside This too is consistent with the idea of water lled access pathways at each end Altogether the four segments S1 S4 are presumed to form the voltage sensor domain On the basis of the pattern of sequence conservation cysteine accessibility and tolerance for substitution all are probably mainly 4x helical6 54 The pattern of the charged residues at every third position led to the early proposal of a helical screw motions the S4 would advance screw like to move each charge to the position of the next charge or the next and so on thereby maintaining the charge distribution in the core while producing an overall translocation of charge across the membrane any physical measurements have been made on uorescence probes attached to cysteines placed in or near the S4 in an attem t to infer the motion of the voltage sensor The pattern of uorescence changes and the pattern of distances estimated from resonance energy transfer have roughly agreed with the idea of a rotation Yet all of the measured distance changes particularly from the measurements with the highest precision are very small about 1 A and seem inconsistent with the popular proposal of a substantial rotation There are many other ways in which the effective translo cation of charge could be accomplished such as the relative lateral movement of crossed helices or a rocking motion at the interface between two domains Solving this particular puzzle would be greatly helped by some knowledge of the three dimensional structure The coupling problem Unlike the ligand binding domains attached directly to S6 the transmembrane voltage sensor does not have an obvious speci c mechanism for coupling to the channel gates Two possibilities have come to light recently One is a direct coupling between the voltage sensor and the selectivity lter gate in Shaker K4r channels activation gating produces a close approach between the extracellu lar part of the S4 and the region anking the selectivity lters This then produces an allosteric effect on slow inactivation involving the selectivity lter gate At the intracellular side of the channel a different site has been proposed7D for coupling between S4 move ment and the S6 activation gate some sort of tight association between the S4 S5 linker and the intracellular projection of S6 On the basis ofthe two best known sensor mechanisms described in the previous section it seems probable that a consistent theme of coupling will be that changes in quaternary structure of a sensor such as ring expansioncontraction or domain multimerization are transduced to the gating motions of the pore It may therefore be tricky to identify a speci c functional link between sensor and pore a e ma comprise a small number of tie rods with identi able importance or it may consist of an entire domain inter ace Dissecting these coupling mechanisms will of course be easier w en we w t e whole structure of a voltage gated channel however particularly in this case structure w e enough and careful mutagenesis and functional measurements will probably be required too The fascination of the coupling problem is enhanced when one remembers that certain channels for example the HCN channels L n i i it etween move ment and S6 gating it appears that the same S6 gate used by normal voltage gated K channels closes at positive voltages rather than negative voltages43 although S4 has the normal positive charges and presumably the normal direction of motion Outlook A 1 of functional probing of channel proteins and recent structural advances have given clear pictures for many of the essential functions of the voltage gated K channels and their relatives The remarkable optimizations of these channels for permitting rapid and selective ion ow across the hostile barrier NATURE lVDL 419 l 5 SEPTEMBER 2002 lwwmaturecomnature ing details of the transmembrane voltage sensor itself Many challenges remain in learning how the intricate machinery is integrated into a functional whole and into the larger function of cellular signalling D dol 01038nature00978 l H r l V F eonduetron and exertatron in nerve I Phynal Land 117 500544 1952 1naet1vat1on meehanrsmsNeurt7r1 15 9sle95 9 w 1 V1quot 1 w L Pardo L A etal Chapman M L an ngen H review article V l M F I gating Neurtm 16 859467 1995 attraeellularw speer eallymodulates a rat brain K ehannel Prat NatlAtatl Set ehannel durrng l g 0 i 0 wrtz T amp Yellen G Modulation of Kt eurrent by frequeney and external Kt a tale of two 0 1995 ehannels Braphys I 76 253e253 1999 v Do M ampVanDongen A M Activationedepmdent subeonduetanee levels in the dxkl llt ehannel suggest a subunrtbasrs for ion permeation and gating Emphyx I 72 705419 1997 h l n I Gert Phyxml 110101e117 1997 111 p tr W10 1 L quot1quotquotin e ated h 1 N 0 9e59s2001 12 rhl n h w Phal P ampYellen G J 2 1 t 12 M11 4 1 Mrrrlrr 111 l 1 11730 i r 1 J Gm WWI 1198amp91lt2002 1983 44 Johnson 1 P Ir amp Zagotta w N Rotational movement during eyelre nudeotldeegated ehannel 3 J a N 41 91 79212001 and selectivity Serertte230 59e77 199s 451mg 1 1e 12 4 zhouYMorars ahral 1 14 w t 12 4 011 K1 1 r i i p 1 N 9 9 gm L 1 L 39 39 r olutror wtr 41443432001 2001 s Morlse 11 h M V P 1 nT n1 12 5151 p 11 Kt seleetrvrty lter Nature 414 3742 2001 i 1 i w r 1 5 liang Y etal The open pore conformation ofpotassmm ehannels Nature 417 523425 2002 23544811998 7 12 u M 1e 12 r 39 1lt 39 Almgyhml t 1 1 1 1M 11 stabilization of monovalent eatrons Sneute 205100102 1999 515522 zggzy Dutzler R Campbell E 13 Cadene M chart B TampM 4 WeberIT r a lquot 4 1 MP ae1ltrnnon R Xeray strueture of a ClC M 41 72942002 9 Armstrong c M lome pores gates and gating eurrents Q Rev Emphys 7179e210 1975 10 HodglonA L arKeynes R D 1 1 er l Land 12115148 1955 I quotII R h 409442 1978 12 Spassova M 1 ehannel Gert Phynal 112211e221 1998 1 14 l M 1 M F 11 4 4 ehannelNeurar1 19 175484 1997 d 1 n Yellen G is 1 ay re ned at 25A resolution I Mal B101 198 311e325 1957 A9 l1 3 M M V F Xra X M e 1 J N 402 8174521 Sun Y etal Meehanrsm of glutamate reeeptor desensitization Nature 417 24se253 2002 r 1 r 1 u ehannels Nature 395 s03es07 199s sehumaeher M A rvard A Baehrnger H P tar Adelman 1 P strueture of the gating domain of a Ca aetrvated K ehannel eomplexed with CaHealmodulrn Nature 410 11201124 2001 srgworth r 1 Voltage gating of ion ehannels Q Rev Braphys 27140 1994 4 Noda M etal 1 r i i sequenee Nature 312121e12719s4 14 r i llt ehannel Newml 32 54e5s5 2001 15 del Camrno 0 Holmgren M Lin Y tar Yellen G t 39 4 llt ehannel and its struetural implieations Nature 403 321e325 2000 15141 M b r eller n 39 eleanbe trapped in the open state by an intersubunrt metal bridge Neurtm 21 517e521 1998 P F r r n M 11 1 gating Serertte2s5 7345 1999 r 1 grantaxon Gen Phynal 58413e437 1971 I9 H l M 41 DY ll learle h n P r n M n 11 ofdle Shaker K hannelNewm1 16 115971167 1996 l w M w 12 L ehannel Newml 16 1159e1177 1995 FF 1lt 1 M b 11 1 rnteraetrons rnthe Shaker K ehannel I Gen Phyxml 115 123e13s2000 llaln ronh11 pt tnaut Pltyxlol 117 59es92001 oodr IGeu J 4 59 Yang N Geor e A L Ir amp Horn R Moleeular basis of eharge movement in voltageegated sodium Kt ehannels evidenee for a trap doormeehanrsm of aetrvatron gating I Gen Phynal 109 527es3s ehatmels Ne an 16 113e122 1995 1997 01 quotPub 0 m n 1 cm 20 HoshrTZagotta N 1d h 12 b 1 4 N 153s7e3971995 potassium ehannel rnaetrvatronSerertte250s33es3s1990 1 n 14 H 12 1 J I Rem I r l J L and other Kt ehannels in open elosed and rnaetrvated conformations I Strutt Bral 121 253214 Nature 369 259e294 1994 1998 22 Zagotta NHoshlTampAldnd1R 5 Monk N 1 1 M11 1 1 L ehannels by apeptrde derived from ShBSaer1e2250558e571 1990 segment in the Shaker Kt ehannelI Gen Phystal 113 415423 1999 23 Murrellel d 12 n M 39 h 12 391 53 LleSmenn Y H m V 1 L with aporebloelongsrte studredwrth synthetre peptides Gen Phynal 102 94w97s 1993 sensing domains ofa K rhannel I Gen Phynal 115 33es0 2000 24Murrellel d 12m M hr 414 KM h 1 H1 1758 peptides Gen Phynal 102 977e1003 1993 2000 25h 11 hw ell 553L141 1P 1 1 L 3913 Ntl meehanrsms lnvoltageeactlvated Kt ehannelsPrat NatlAtatl Set USA 335092es09s 1991 Atatl Set 1154 3350se512 1955 5 n 11 n h 55 CatterallwA 39 blocker Neurtm 7 743e7s3 1991 Trends Neuraser 9 H0 1955 h M M 39eCameLH M M r R it t the 57 laur r K Marmu 111 M dh rnaetrvatron gate and quaternary amine inhibitors Nature 411 557e551 2001 sensor movement in the Shaker potassium ehannel Nature 402 513417 1999 ernrhrlPtl 39 68l1 vrl F 1 P12 12 111 eysterne olcldatlon Nature 352 711e714 1991 region in a potassium ehannel measured via spectrosmpy Nature 402 509413 1999 l N 59 F r F M 9 Ram r r 391 390393 1995 30 11 1 M M M Cell 97 943e952 1999 31 Kobertzw 11 1 M11 gated llt hannelBmchem15ry 3910347e10352 2000 32 Sokolova 0 Kolmakovaepanensky L amp 311 one N Threeeddmensmnal strueture of avoltagee at 25nm resolution Snmmve Cami 9 215e220 2001 wot 1I D V H aetrvatron in a K ehannelI Gen Physml 1181e102001 70 T r J M Chen I 8r Sangurnettr M C lnteraetrons between 5455 linker and S 5 transmembrane domain modulate gating of HERG K ehannels I Bml Chem 277 1899449000 2002 g tham L Abramson T tar MadGlmon R A functional eonneetron between the pores of distantly related ron ehannels as revealed by mutant Kt ehannels Sereute 250 11524155 1 DH 44h 1 T Zagotta w N ampAldrreh R w T ypes oflnactlvatlon in shaker llt ehannels effects of 72 94 e237s 2001 alterations in th carboxyete39rmmal IglonNemm1 7 s47ess5 1991 b w m 1 r n b 1 34 pezeBameo 1 Hoshr T Hernemann s H ampAldnehRW lute haul ehannel Nnr 9 94 e94s1995 P F 7 74 ulhr 1 M b M M M 1e 12 L 1 51e71 1993 assembly ofvoltageede pendent Kt ehannels Sneute 239123e127 2000 35 I r l h l1 Te l M F 75 Shel1 NV Wamn V Pl 4 a 1lt ehannel allows rnaetrvatron to be modulated by metal binding B11pr I 66 10554075 1994 416 N 14 525e533 199s NATURE VDL 419 l S EPTEMBER