Gating currents

• Dr. Clay M. Armstrong, Department of Physiology, University of Pennsylvania, Philadelphia, PA

Figure 1: Gating and sodium current during a voltage-clamp step (after subtraction of capacitive and leakage current). Outward current is up. I_Na has been reduced 20X by lowering Na⁺ concentration.

Gating currents are generated by the movement of specific charges in the transmembrane helices of V(oltage)-gated ion channel proteins. When these charge-containing helices move in response to a change of membrane voltage, they force conformational changes in the protein, opening or closing a gate that controls ion flow through a channel (or pore) in the protein.

V-gated ion channels are found in many cells, notably in nerve and muscle cells which must conduct action potentials over long distances. Message transmission through a Trans-Atlantic cable serves as a good analogy to their function. A message, coded in the form of voltage pulses and put into the cable in London, weakens as it travels under the Atlantic due to losses through the cable’s insulation. To reach New York intact it must be strengthened at regularly spaced booster stations, which detect the weakened incoming pulses and boost them back to full strength. Similarly, an action potential in an axon or muscle fiber weakens with distance unless boosted by regularly spaced, V-gated, Na⁺ selective (Na_V) channels. In an axon the insulator is the surface membrane (aided by myelin in some fibers), through which ions can flow only via ion channels. Na_V channels combine the functions of detecting an incoming signal and amplifying it. Although an action potential is only about 0.12 volt, the field within the membrane can be 200,000 V/cm or more. It is easy for the charges built into the Na_V channel to detect the large changes in the field caused by an action potential upstream. Upon detection, the incoming action potential is amplified when the gate of each channel opens, allowing Na⁺ ions to flow through the channel (and the membrane) into the axon. This drives the internal voltage (V_m) positive, strengthening the signal, which spreads on downstream. To terminate the action potential promptly in preparation for another action potential, the Na_V channels “inactivate”, and K_V channels (selective for K⁺ ions) open, allowing K⁺ ions to pass out of the axon, driving the voltage back to rest (roughly -70 mV).

Detection and amplification thus require the movement of charged voltage sensor/effectors built into the channel proteins. As these structures move, they generate a small, transient gating current (I_g) that flows outward as the channel gate opens, inward as it closes. Hodgkin and Huxley (1952) predicted gating current in the days when ion channels (or carriers) were concepts rather than known proteins, and they made it clear that any channel with a voltage-operated gate must have such current. Actual measurement of these small currents occurred only 20 years later (Armstrong and Bezanilla, 1973).

Fig.1 is an early example of I_g. Also shown is I_Na, the current generated by the flow Na⁺ ions through the channels opened by the gating current. Three traces are shown: I_g + I_Na ; I_g isolated by adding tetrodotoxin, a very selective toxin that blocks Na⁺ flow through the channels without disturbing the gate; and a trace showing I_Na alone, obtained by subtraction. I_g begins before I_Na, as expected because the gates must open before Na⁺ can flow through the channels. I_g continues until all the V-gates are open. In the figure I_Na has already decreased substantially in amplitude before I_g falls to zero because a second gate built into the protein, the I-(inactivation) gate, has closed some of the channels. The I-gate, as explained below, is not directly voltage dependent, and generates no gating current. Functionally, occluding the channel by closure of the I-gate makes it easy to restore membrane voltage (V_m) to the resting level.

K⁺ channels are similar in both structure and function to the Na⁺ channel, but simpler, in part because many of them do not have I-gates to complicate the kinetics. Further, the channel protein is much smaller (four identical subunits rather than a single large peptide), and much better known structurally than Na⁺ channels, for which no x-ray structure is yet available. Consequently most effort has been devoted to studying gating properties of K⁺ rather than Na⁺ channels. For this reason we turn first to the K⁺ channel.

K⁺ channel structure

Figure 2: Partial Structure of a V-Gated K+ Channel.

Fig. 2 (derived from Long et al., 2005) shows some of the membrane-spanning segments of a K⁺ channel, with large parts of the channel omitted for clarity. The channel is formed of four identical subunits, each with transmembrane segments S1-S6. Viewed from the outside (Fig. 2A) the pore through which ions flow is in the center, lined with carbonyl oxygens (red). Surrounding the pore is the central core region, formed by two intertwined transmembrane helices (S5 and S6), and, in the segment linking S5 to S6, the “selectivity filter” (better seen in Fig. 2C). Peripheral to the core are the S4 segments, each of which contains seven evenly spaced positive amino acids (arginines and lysines). Fig. 2B shows these elements in side view, with the approximate outer and inner surface of the membrane marked. A section through the central pore (Fig. 2C) shows the selectivity filter, the “vestibule”, and the “gating region” with the gate in open configuration. Two dehydrated K⁺ ions (yellow) are in the selectivity filter, and a hydrated K⁺ ion in the vestibule, surrounded by water molecules (blue).

A side view in strand version (Fig. 2D) explains how the gate may operate. The positively charged amino-acids spiral around the S4 segment and are shown in blue. They are labeled with their conventional numbers, R1 (arginine 1) through K7 (lysine 7). At the bottom of S4 is the S4-5 linker, which is in close proximity to the gate region of S6. No closed-gate x-ray structure is yet available, but the arrows indicate possible inward motion of the S4 (at negative voltage) and movement of the S4-5 linker, which may lock together the lower part of the S6 helices, keeping the gate closed. A much simpler K⁺ channel, Kcsa, which has only the central core, no S1- S4 segments, and no voltage-gating, provides the only present model of the closed state (Doyle et al. 1998). In Kcsa, the equivalent of the S6 segments are pinched together in the gating region, occluding the pore (see arrows in Fig. 2C and D).

Gating current is generated by the movement of the charged S4 segment through the membrane. Outward S4 motion, it is thought, moves the S4-5 linker out of the way so the gate can open, and inward motion puts it in locking position when the gate is closed.

Coupling of S4 movement to the channel gate

The inner vestibule of a K⁺ channel is the first stop for a K⁺ ion moving outward through the channel (Fig 2C). This site can also be occupied by tetraethylammonium ion (TEA⁺), which is about the same size as a K⁺ ion with a single hydration shell. When the V-gate of a K⁺ channel is open TEA⁺ can enter the vestibule. In certain K_V channels the V-gate cannot close when TEA⁺ is in the vestibule, a foot in the door effect. In these channels TEA⁺ occupation of the vestibule is known to block the inward S4 movement that normally closes the channel: the gate can close, and the S4's can move inward, only after TEA⁺ has left the vestibule (Armstrong, 1971) . Some K_V channels are inactivated after being open for a few milliseconds. In this case a ‘ball’ attached to the N terminus of the K⁺ channel peptide diffuses into the vestibule, producing what is called N-type inactivation (Hoshi et al., 1990). Like TEA⁺, the ball holds the V-gate open, and paralyzes inward S4 movement (Demo and Yellen, 1991).

The action of 4-aminopyridine (4-AP) is of further help in understanding the coupling of S4 to gating. Like TEA⁺, 4-AP enters the vestibule when the V-gate is open (Armstrong and Loboda, 2001). Unlike TEA⁺, it pulls the gate closed behind it, rather than holding it open, and glues the gate shut with roughly the energy of a hydrogen bond. Though the gate remains glued shut, gating current on depolarization is only slightly affected: total charge movement is reduced by about 5%, and the Q-V (charge-voltage) curve is displaced slightly to the right on the voltage axis (Loboda and Armstrong, 2001). Overall, 95% of the charge moves almost as freely as normally. The small effect of 4-AP on the Q-V curve shows that the locked gate does not appreciably restrain the S4. Reciprocally, the S4 must not pull forcefully on the gate. Thus the main action of the S4 is probably to lock the gate shut at rest, rather than to pull it open during activity (Armstrong, 2003). This seems in good agreement with the K_V1.2 structure (Long et al., 2005), where the S4-5 linker is in potential blocking position relative to the gating region of the S6 helix: in the closed state S4-5 may keep the gate pinched shut. Consistent with this, there is (so far) no structural evidence to suggest a direct pulling action of the S4 on the gate.

Research questions

It is clear that the S4 segment is the main controller of the V-gate of ion channels, but there are still many questions. Gating current is one type of evidence helpful in detailing the mechanics of V-gating, but it must be reconciled with other evidence. Some outstanding questions about V-gating mechanisms are the following.

(i) How can so many charges be in the low dielectric medium of the membrane? Each arginine inserted in a low dielectric region must have a counter ion with which to form a salt bridge (see Radzicka and Wolfenden, 1988, regarding the very high energy cost of inserting arginine into a low dielectric solvent). A recent crystallographic model (Long et al., 2007) of an open K channel, schematized in Fig. 3

Figure 3: Charge pairing within the membrane.

shows the outer two arginines (R1, R2) in the external solution, while the third through the sixth charges (R3, R4, K5, R6) are within the membrane, and form salt bridges (two-dash lines) with negative glutamates in transmembrane segments S1- S3. From the energetic point of view this seems to be a satisfactory arrangement, as all of the charges within the membrane are energetically stabilized by salt bridges.

(ii) How far does each S4 charge move during gating? Measurements of total charge transfer give estimates of 12 to 14 charges per K channel transferred all the way across the membrane field (Aggarwal and MacKinnon, 1996; Islas and Sigworth, 1999). Piecing together the evidence, in the open state R1 and R2 are in the external solution (Long et al., 2007); while in the closed state labeling experiments (Larsson et al., 1996) suggest that R3 through K7 are accessible from inside (the red trace in Fig. 3 labeled closed?). The membrane field may be concentrated over a distance of about 10 A in the middle of the membrane (Long et al., 2007), as shown by the dashed line labeled V_m. To account for total charge movement, R1 through R4 would necessarily move almost all the way across the field in the transition from open to closed, transferring a maximum of 16 electronic charges for the whole channel, close to the estimated 12 to 14. A clear problem with the speculative red trace in Fig. 3is that E183 and E226 are not charge paired. Perhaps they bind ions from the external solution. R1, R2, and R3 in the closed state may be charge paired to E236, D239, and E154.

(iii) The scheme in the preceding paragraph (ii) implies that each S4 moves four steps in going from closed to fully activated. This would have consequences on both the Q-V curve and on I_g kinetics. If the four charges moved as a unit in a single jump, the Q-V curve could be described as a single Boltzmann curve with charge of four. The experimental curve is much less steep (2.2 e from averaging values in the literature), which suggests that S4 movement is not a single step, but several. Kinetically, several steps of S4 movement are consistent with the observed kinetics of I_g.

(iv) Do the charges in an S4 segment move as a rigidly linked unit? Normally most gating charge moves between -70 mV and 0 mV. Mutation of certain amino acids in or just below the S4 helix (Smith-Maxwell et al., 1998; Schoppa and Sigworth, 1998) seems to split charge movement into two parts. With one of these mutations (Smith-Maxwell et al., 1998), about 85% of the charge seems to be unleashed from the remainder and to move in the abnormally negative range -100 to -70 mV. This presumably reflects movement of the top part of the S4, moving independently of the lower part. The remainder of the charge, presumably in the lower part of S4, moves in the range +50 to +150 mV, the approximate voltage range of the actual gate opening in this mutant. These observations raise the possibility that even the normal S4 charge moves in pieces rather than as a coherent unit, with complete movement of all pieces required for gate opening. It is interesting that Long et al. (2007) identify a portion of the S4 that has an extended (3₁₀ rather than α-helical) configuration: does this part of S4 stretch from α-helical to a 3₁₀ helix during activation?

(v) Why are the positive charges spirally arranged in S4? A possible answer is to help stabilize the S4 in the membrane. Were the charges in a stripe down one side of the S4, the helix would tend to lie at the membrane surface, with hydrophobic stripe in the membrane and charges in the solution.

Gating current and the sodium channel

Much remains unknown about the Na⁺ channel. The following section synthesizes some of the available evidence, but is necessarily speculative in its conclusions.

The Na channel was probably evolved by stitching together the four subunits of a K⁺ channel to form a single very large protein with four domains, D1-D4. Each domain has six transmembrane segments, like a subunit of a K⁺ channel, with multiple positive charges in the S4 segment of each domain (Noda et al., 1984). Gradually, one supposes, the domains differentiated, to form the present day Na⁺ channel. The general conformation is almost certainly like that of a K⁺ channel, with a central core containing a selectivity filter that is rather weakly tuned to prefer Na⁺ over K⁺; a vestibule that is wider than the filter, and wide enough to admit, e.g., a local anesthetic molecule; and, protecting the inner end of the vestibule, a V-gate, formed by the convergence of the S6 segments. A significant difference from K⁺ channels is that the number of positive charges contained in the S4 segments differs in the four domains, having, in the skeletal muscle channel, 4, 5, 6 and 8 positive charges. The significance of this is not clear, but the model below speculates that S4 of D4 (S4/D4) is longer than the other S4 segments, helping to explain its special relationship to inactivation (Horn et al., 2000). Similar to the inactivation ‘ball’ found on some K⁺ channels, Na⁺ channels contain an intrinsic ‘inactivation particle’ (Armstrong et al., 1973) which is located cytoplasmically, on the segment of protein that links domains 3 and 4 (Kellenberger et al., 1997). This inactivation particle or gate (I-gate) occludes the channel by blocking its inner mouth after it has conducted for about a millisecond. Inactivation of the Na⁺ channel makes it relatively easy for the more slowly opening K⁺ channels to drive V_m negative.

A model for four of the most important gating states of a sodium channel is shown in Fig. 4 (c.f. Armstrong, 2006).

Figure 4: Na+ channel activation and inactivation, a model.

At Rest, the V-gate is locked closed by the S4 segments, which are pulled firmly inward by the internally negative V_m. The inactivation particle at Rest is prevented from reaching its blocking site by the inner end of the S4/D4, or the attached S4-5 linker. During an action potential, the S4s of all domains move outward stochastically in one or more quick steps, unlocking the V-gate (all of the S4s must move to unlock it). The gate then opens (Open1), allowing Na⁺ ions to flow through the channel. In state Open1, S4/D4 has moved outward enough to unlock the V-gate, but still prevents the inactivation particle from occluding the channel. Propelled by positive V_m, S4/D4 moves relatively slowly a further step outward, driving the channel into state Open2. This movement generates a slow component of I_g. In state Open2 the channel continues to conduct Na⁺ ions until the inactivation particle, no longer hindered by S4/D4, diffuses into blocking position (arrow Fig. 4upper right). The Inactivated state is relatively stable, and persists until V_m returns to a negative level.

As the drawing implies, when the inactivation particle is in blocking position the inward motion of S4/D4 (and also of S4/D3) is prevented, even when V_m becomes negative after the action potential. This immobilizes about two-thirds of the gating charge for as long as the inactivation particle remains stuck to its binding site. S4/ D1 and S4/D2, however, are free to move, and their inward motion in the steps labeled ‘recovery 1’ closes the V-gate ( to Closed&Inactivated), and generates the component of I_g that is not immobilized. Functionally, closing the V-gate is useful, because it prevents Na⁺ flux through the channel during recovery from inactivation. Failure of the recovery 1 step to occur properly has serious consequences, and may be the basis for some paralyses of muscle fibers.

Recovery from inactivation continues in step ‘recovery 2’. Inward pressure on S4/D4 by the now negative V_m helps dislodge the inactivation particle from its binding site, and the channel returns to Rest. This step in theory should be reversible, and experimentally it is found that a fraction of the channels inactivate without having opened their V-gates. This ‘closed state inactivation’ happens mainly during small, slow depolarizations in experimental conditions.

References

Aggarwal, S.K. and R. MacKinnon. 1996. Contribution of the S4 segment to gating charge in the Shaker K+ channel. Neuron 16:1169-77

Armstrong, C.M. 1971. Interaction of tetraethylammonium ion derivatives with the potassium channels of giant axons. J. Gen. Physiol. 59:413-437

Armstrong, C.M. 2003. Voltage-gated K channels. Science STKE. 2003(188):RE10.

Armstrong, C.M. 2006. Na channel inactivation from open and closed states. PNAS 103:17991-6

Armstrong, C.M. and A. Loboda. 2001. A model for 4-aminopyridine action on K channels. Similarities to TEA+ action. Biophys. J. 81:895-904.

Armstrong, C.M. and F. Bezanilla. 1973. Currents related to movement of the gating particles of the sodium channels. Nature 242:459-461

Armstrong, C.M., F.M Bezanilla and E. Rojas. 1973. Destruction of sodium conductance inactivation in squid axons perfused with pronase. J. Gen. Physiol. 62:375-391

Demo, S.D. and G. Yellen. 1991. The inactivation gate of the Shaker K+ channel behaves like an open-channel blocker. Neuron 7:743-753.

Doyle, D.A., J.M. Cabral, R.A. Pfuetzner, A. Kuo, J.M. Gullbis, S.L. Cohen, B. T. Chait, and R. M. MacKinnon. 1998. The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280:69-77.

Hodgkin, A.L. and A.F Huxley. 1952. A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol. (London) 117:500-544

Horn, R., S. Ding, and H.J. Gruber H.J. 2000. Immobilizing the moving parts of voltage-gated ion channels. J. Gen. Physiol.116:461-76

Hoshi, T., W. N. Zagotta, and R.W. Aldrich. 1990. Biophysical and molecular mechanisms of Shaker potassium channel inactivation. Science 250:533-538.

Islas, L., Sigworth, F. 1999. Voltage sensitivity and gating charge in Shaker and Shab family potassium channels. J. Gen. Physiol. 114:723-741

Kellenberger, S., J.W. West, T. Scheurer, and W.A. Catterall. 1997. Molecular analysis of the putative inactivation particle in the inactivation gate of brain type IIA Na+ channels. J Gen Physiol. 109:589-605

Larsson, H.P, O.S. Baker, D.S. Dhillon, and E.Y. Isacoff. 1996. Transmembrane movement of the Shaker K+ channel S4. Neuron 16:387-397

Loboda, A. and C.M. Armstrong. 2001. Resolving the gating charge movement associated with late transitions in K channel activation. Biophys J. 81:905-16

Long, B., E.B. Campbell, and R. MacKinnon. 2005. Crystal Structure of a Mammalian Voltage-Dependent Shaker Family K+ Channel. Science 309:897-903

Long, S.B., X. Tao, E.B. Campbell, and R. MacKinnon. 2007. Atomic structure of a voltage-dependent K+ channel in a lipid membrane-like environment. Nature 450:376-82.

Noda, M., Shimuzu, S., Tanabe, T. Takai, T. Kayano,T., Ikeda, T., Takahashi, H., Nakayama, H., Kanaoka, Y., Minamino, N., Numa, S. 1984. Primary structure of Electrophorus electricus sodium channel deduced from cDNA sequence. Nature312:121-7

Radzicka, A., and R. Wolfenden. 1988. Comparing the polarities of the amino-acids: side-chain distribution coefficients between the vapor phase, cyclohexanone, 1-octanol, and neutral aqueous solution. Biochemistry 27:1670-1677

Schoppa, N.E. and F.J. Sigworth. 1998. Activation of Shaker potassium channels. II. Kinetics of the V2 mutant channel. J Gen Physiol. 111:295-311

Smith-Maxwell, C.J., J.L. Ledwell, and R.W. Aldrich. 1998. Uncharged S4 residues and cooperativity in voltage-dependent potassium channel activation. J Gen Physiol. 111:421-39

Internal references

• Jan A. Sanders (2006) Averaging. Scholarpedia, 1(11):1760.

• Bertil Hille (2008) Ion channels. Scholarpedia, 3(10):6051.

• Rodolfo Llinas (2008) Neuron. Scholarpedia, 3(8):1490.

• Philip Holmes and Eric T. Shea-Brown (2006) Stability. Scholarpedia, 1(10):1838.

See also

Electrophysiology, Hodgkin-Huxley model, Neurons

Sponsored by: Eugene M. Izhikevich, Editor-in-Chief of Scholarpedia, the peer-reviewed open-access encyclopedia

Reviewed by: Anonymous

Reviewed by: Dr. Rick Aldrich, University of Texas, Austin, TX

Accepted on: 2008-10-13 21:46:33 GMT