J Gen Physiol
February 1, 1997;
Role of the S3-S4 linker in Shaker potassium channel activation.
Structural models of voltage-gated channels suggest that flexibility of the S3-S4 linker region may be important in allowing the S4 region to undergo large conformational changes in its putative voltage-sensing function. We report here the initial characterization of 18 mutations in the S3-S4 linker of the Shaker channel, including deletions, insertions, charge change, substitution of prolines, and chimeras replacing the 25-residue Shaker linker with 7- or 9-residue sequences from Shab, Shaw, or Shal. As measured in Xenopus oocytes with a two-microelectrode voltage clamp, each mutant construct yielded robust currents. Changes in the voltage dependence of activation were small, with activation voltage shifts of 13 mV or less. Substitution of linkers from the slowly activating Shab and Shaw channels resulted in a three-to fourfold slowing of activation and deactivation. It is concluded that the S3-S4 linker is unlikely to participate in a large conformational change during channel activation. The linker, which in some channel subfamilies has highly conserved sequences, may however be a determinant of activation kinetics in potassium channels, as previously has been suggested in the case of calcium channels.
J Gen Physiol
[+] show captions
References [+] :
Figure 7. Activation and deactivation in chimeric channels.
(A) Time courses of activation
during a −20-mV depolarization
for ShΔ, ShΔ/Shal, ShΔ/Shaw,
and ShΔ/Shab currents, and for
the cysteine insertion mutant
ShΔ/Cys. Currents in the mutant
constructs were scaled by factors
of 0.9, 5.0, 1.0, and 2.17, respectively, to match the ShΔ currents.
(B) Deactivation time courses at −60 mV after a prepulse to +40 mV. Currents from ShΔ/Shal, ShΔ/Shaw, and ShΔ/Shab were scaled by
factors of 1.5, 0.78, 1.45, and 1.25, respectively. (C) Voltage dependence of the predominant time constant of deactivation, τd1, as a function of test potential for ShΔ and the chimeric channels. Plotted are the geometric means ± SE (n = 6).
Figure 2. Mutations of the Shaker S3-S4 region. The wild-type
Shaker sequence (top) is compared with the sequences of mutations studied here. Mutations included the deletion and charge-reversal of acidic residues at the NH2-terminal end, deletions at the
COOH-terminal end, and substitutions of each of the three proline residues with an alanine. In the insertion mutations (ShΔ/Cys,
ShΔ/FLAG, ShΔ/HA, ShΔ/Xa) the underlined sequences were
inserted following proline 344, as indicated by the upward arrows.
In the final three mutations, the entire S3-S4 linker was replaced
with shorter segments from Shab11, Shaw2, or Shal2, respectively
(Butler et al, 1990). Dashes indicate amino acid identity to the
Shaker sequence and blank spaces indicate gaps.
Figure 3. Whole-oocyte voltage-clamp protocols and currents
from oocytes expressing the “wild-type” ShΔ and the ShΔ/Shaw
chimeric channels. (A) Activation protocol. Depolarizations of
100-ms duration are given from a holding potential of −80 mV.
(B) Deactivation protocol. After a 20-ms prepulse to +40 mV, tail
currents are measured during 100-ms test pulses at potentials between −100 and +40 mV. Potentials are varied in steps of 10 mV,
and pulses are applied at 5-s intervals.
Figure 4. Voltage dependence
of activation and time constants.
Normalized conductance-voltage (G-V) curves, and voltage dependence of activation time constant are shown for ShΔ and various mutants: (A) deletions at the
NH2-terminal end of the linker;
(B) charge reversals at the NH2-terminal end; (C) deletions at
the COOH-terminal end. G-V
curves (top) are shown for individual oocytes, chosen to have Va
and ka values near the medians of
4–10 recordings. The conductance values were computed
from the average current during
the second half of 100-ms depolarizations to the given voltages
and are shown fitted to the
fourth power of a Boltzmann
function. Time constant τa1 of
the predominant component of
the activation time course (bottom) is plotted as the geometric
mean ± SE for n = 5–7 determinations as a function of depolarizing voltage. Lines connect the
values for ShΔ (filled circles).
Figure 5. Comparison of the time course of activation of ShΔ
and the ΔA and ΔAMS mutants. Depolarizations to −20 mV were
applied from a holding potential of −80 mV; the entire time
courses are shown in the inset. The ΔA and ΔAMS current recordings were scaled by factors of 1.57 and 2.54 to match the steady-state current in the ShΔ recording.
Figure 6. Voltage dependence
of activation and time constants
of mutants. (A) Proline substitutions; (B) insertions; (C) linker
sequence substitutions. G-V
curves (top) and time constants
(bottom) are plotted as in Fig. 4.
Error bars represent standard errors with n = 5–7.
Figure 8. Determination of apparent gating valence zapp. (A)
Conductance-voltage curve. Mean current from ShΔ channels during the second half of 400-ms depolarizations from −80 mV were
converted to conductance and plotted as open circles as a function
of voltage; conductances obtained from shorter (100 ms), larger
depolarizations are plotted as filled circles. (B) Values of zapp, computed as the logarithmic derivative of G with respect to voltage according to Eq. 2 are plotted as a function of G as in Zagotta et al.
(1994). Because of growing errors at lower conductances, we took
the estimate of zapp to be the value at 1% of the maximal conductance, as indicated by the arrow.
Contribution of the S4 segment to gating charge in the Shaker K+ channel.