Espinosa F et al. (2001), Dynamic interaction of S5 and S6 during voltage...

XB-ART-8637

J Gen Physiol 2001 Aug 01;1182:157-70.

Show Gene links Show Anatomy links

Dynamic interaction of S5 and S6 during voltage-controlled gating in a potassium channel.

Espinosa F , Fleischhauer R , McMahon A , Joho RH .

???displayArticle.abstract???
A gain-of-function mutation in the Caenorhabditis elegans exp-2 K(+)-channel gene is caused by a cysteine-to-tyrosine change (C480Y) in the sixth transmembrane segment of the channel (Davis, M.W., R. Fleischhauer, J.A. Dent, R.H. Joho, and L. Avery. 1999. Science. 286:2501-2504). In contrast to wild-type EXP-2 channels, homotetrameric C480Y mutant channels are open even at -160 mV, explaining the lethality of the homozygous mutant. We modeled the structure of EXP-2 on the 3-D scaffold of the K(+) channel KcsA. In the C480Y mutant, tyrosine 480 protrudes from S6 to near S5, suggesting that the bulky side chain may provide steric hindrance to the rotation of S6 that has been proposed to accompany the open-closed state transitions (Perozo, E., D.M. Cortes, and L.G. Cuello. 1999. Science. 285:73-78). We tested the hypothesis that only small side chains at position 480 allow the channel to close, but that bulky side chains trap the channel in the open state. Mutants with small side chain substitutions (Gly and Ser) behave like wild type; in contrast, bulky side chain substitutions (Trp, Phe, Leu, Ile, Val, and His) generate channels that conduct K(+) ions at potentials as negative as -120 mV. The side chain at position 480 in S6 in the pore model is close to and may interact with a conserved glycine (G421) in S5. Replacement of G421 with bulky side chains also leads to channels that are trapped in an active state, suggesting that S5 and S6 interact with each other during voltage-dependent open-closed state transitions, and that bulky side chains prevent the dynamic changes necessary for permanent channel closing. Single-channel recordings show that mutant channels open frequently at negative membrane potentials indicating that they fail to reach long-lasting, i.e., stable, closed states. Our data support a "two-gate model" with a pore gate responsible for the brief, voltage-independent openings and a separately located, voltage-activated gate (Liu, Y., and R.H. Joho. 1998. Pflügers Arch. 435:654-661).

???displayArticle.pubmedLink??? 11479343
???displayArticle.pmcLink??? PMC2233824

???displayArticle.grants??? [+]

Species referenced: Xenopus
Genes referenced: kcna3 kcnb1 shb tbx2

???attribute.lit??? ???displayArticles.show???

	Figure 1. S5-Pore-S6 alignment and pore model of EXP-2. (A) Sequence alignment for the S5-pore-S6 region of several K+ channels from different Kv subfamilies. The side chains highlighted by asterisks above the EXP-2 sequence are drawn as sticks in B. (â-â indicates identity to the sequence of Kv2.1; â.â indicates a gap required to maintain alignment. Conserved amino acids are shown by letters on the top; residues that are absolutely conserved in Kv channels are in bold.) (B) Top and side views of the S5-pore-S6 region of EXP-2. The four conserved side chains in S5 that are marked by asterisks (M414, L418, G421, and F425) are on the same side of the S5 helix and project toward S6. The side chain at position 480 in S6 points toward the conserved G421 that is flanked by L418 and F425 in S5. The model suggests possible interaction between G421 in S5 and the side chain at position 480 in S6. In case of an aromatic side chain at position 480, there may be aromaticâaromatic interaction with the invariant F425 in S5.
	Figure 2. Bulky side chains at position 480 in S6 trap EXP-2 channels in an open state. Bulky side chains prevent EXP-2 channels from fully closing, indicated by large steady-state inward currents at â120 mV. Oocytes (in 100 mM KCl containing Ringer) were held at â120 mV and subjected to 1-s depolarizing pulses in 20-mV increments from â80 to 40 mV (only the beginning part of the 1-s pulse is shown), followed by 600-ms pulses to â120 mV. Neither wild-type nor mutant channels show measurable outward currents, indicating fast inactivation (Ïinact < 1 ms at 20 mV for wild-type EXP-2). When small side chains (Gly, Ser, and Cys) occupy position 480, channels show tail currents that approach zero after a few hundred milliseconds. For bulky side chain substitutions (Phe, Tyr, and Trp), channels incompletely deactivate at â120 mV, which is indicated by substantial tail currents that do not reach zero.
	Figure 3. Bulky, aromatic side chains prevent channel closing. The ratio of steady-state current at â120 mV (ISS) to maximal tail current (Ipeak) increases with side chain volume and aromaticity at position 480 in S6. The side chain volumes are from ZÃ¼hlke et al. 1994.
	Figure 4. Voltage dependence of S6 mutants. (A) G-V relationships are shown for those S6 mutations (Gly, Ser, Thr, Val, and Leu) that do not lead to large steady-state currents (ISS) at â120 mV. (B) G-V relationships are shown for S6 mutations (His, Ile, Trp, Tyr, and Phe) that lead to increasingly larger ISS/Ipeak ratios at â120 mV. The G-V curve for wild-type C480 is shown as a dotted line in A and B. Oocytes were held at â120 mV, and 1-s test pulses to potentials from â80 to 40 mV were applied in 10-mV increments, followed by 200-ms pulses to â120 mV. The peaks of the resulting tail currents were normalized to the maximal current. The G-V relationships were fit to a Boltzmann equation, and the midpoints of activation and the slope factors k were calculated (listed in Table ). The inset in B shows the G-V curves when ISS at â120 mV is subtracted. The symbols show means Â± SEM (three to seven oocytes).
	Figure 5. Kinetics of activation and recovery from inactivation and deactivation. (A) Oocytes were held at â120 mV, and test pulses to 20 mV were applied lasting from 1 to 5,000 ms. The time dependence of activation was determined from the peak tail currents during the subsequent pulses to â120 mV. (B) EXP-2 channels were activated/inactivated by 1-s prepulses to 20 mV, and then 600-ms test pulses to â120 mV were applied to initiate recovery from inactivation followed by deactivation. Tail currents were normalized (0 corresponds to ISS at â120 mV; â1 corresponds to the peak of the tail current). The inset shows small differences in the kinetics of recovery from inactivation.
	Figure 6. Mutations in S5 trap the channel in an active state. (A) Oocytes (in 100 mM KCl containing Ringer) were held at â120 mV and subjected to 1-s depolarizing pulses in 20-mV increments from â80 mV to 40 mV, followed by 600-ms pulses to â120 mV (as described Fig. 2 legend). Replacement of the conserved G421 in S5 with larger side chains prevents EXP-2 channels from closing similar to the mutations at position 480 in S6. (B) The ISS/Ipeak ratio at â120 mV increases with the side chain volume at position 421 in S5.
	Figure 7. Single-channel activity of wild-type C480 and mutants C480I and C480F. Channels were activated-inactivated from a holding potential of â80 mV by 1-s pulses to 20 mV; the voltage was then returned to â80 mV to allow recovery from inactivation followed by deactivation. The 10 consecutive current traces on the left represent the first 200 ms after repolarization to â80 mV. The expanded traces (10-ms duration) show similar kinetics and single-channel currents for the wild-type and the mutant channels. On the right, 10 consecutive 200-ms current traces were recorded 3 s after the beginning of repolarization when the tail currents had reached near steady-state levels (<1% for wild-type EXP-2, â¼5% for C480I and â¼75% of the peak tail current for C480F; Fig. 3 A). As expected from macroscopic current measurements, wild-type EXP-2 shows mainly null traces. In contrast, the C480F mutant shows single-channel activity similar to the activity seen at the peak of the tail current. The C480I mutant shows behavior between wild-type and C480F. Currents were low-pass filtered at 2 and 10 kHz, and digitized at 25 and 50 kHz for the recordings shown in the 200-ms and 10-ms traces, respectively.
	Figure 8. Open and closed times are not significantly affected by mutations in S6. Histograms of open and closed times show no differences between C480 wild-type and C480I and C480F mutant channels. Open-time distributions were fit with single exponentials, closed-time distributions with two exponentials. The mean open and closed times in milliseconds (mean Â± SEM) and the number of experiments are shown for each channel.
	Figure 9. S6 Mutations in close proximity have opposite effects on channel gating. (A) The sequence alignment shows the similarity between Kv2.1, Shaker B (ShB) and EXP-2. The cysteine at position 393 in Kv2.1 (in bold) influences open-state stability and K+/Rb+ permeation (Liu and Joho 1998); the nearest residue in Shaker (Hoshi et al. 1991) and in Kv1.3 (Panyi et al. 1995) determines the kinetics of C-type inactivation (in bold in the ShB sequence). The position of the EXP-2 mutations that prevent the channel from reaching long-lived closed states is three residues COOH-terminal from the Kv2.1 position that influences open-state stability (in bold in EXP-2 sequence). The approximate position of the voltage-activated gate in ShB (Liu et al. 1997) is shown by two downward arrowheads. (B) Top and side views of the C480Y EXP-2 mutant (modeled after KcsA) suggest how S6 residues that are only three positions apart may independently influence the voltage-activated gate and the voltage-independent pore gate. The bulky side chain Y480 prevents S6 rotation that is required for the channel to reach a long-lived closed state. In the open state, A477 (C393 in Kv2.1) is close to the conserved T454 (T370 in Kv2.1), which is located in the narrow ion conduction pathway (part of the K+ channel signature sequence; Heginbotham et al. 1994). In Kv2.1, small hydrophilic side chains at position 393 in S6 (A477 in EXP-2) stabilize the open state and affect ion selectivity (Liu and Joho 1998) presumably by interacting with T370 (T454 in EXP-2). In this model, substitutions for C393 in Kv2.1 (A477 in EXP-2) would only affect the stability of the open state as long as S6 is in the appropriate orientation for the side chain at position 393 to interact with T370. In the wild-type Kv2.1 channel, the interaction between T370 (T454 in EXP-2) and C393 (A477 in EXP-2) is broken when S6 rotates counterclockwise at negative membrane potentials. This rotation cannot occur in EXP-2 mutants with aromatic side chains at position 480 âtrappingâ the mutant channels in an open state without affecting open state stability (mean open time) or ion selectivity.

References [+] :

Burley, Aromatic-aromatic interaction: a mechanism of protein structure stabilization. 1985, Pubmed