Mannuzzu LM and Isacoff EY (2000), Independence and cooperativity in rearrangement...

XB-ART-11429

J Gen Physiol 2000 Mar 01;1153:257-68.

Show Gene links Show Anatomy links

Independence and cooperativity in rearrangements of a potassium channel voltage sensor revealed by single subunit fluorescence.

Mannuzzu LM, Isacoff EY.

???displayArticle.abstract???
Voltage-gated potassium channels are composed of four subunits. Voltage-dependent activation of these channels consists of a depolarization-triggered series of charge-carrying steps that occur in each subunit. These major charge-carrying steps are followed by cooperative step(s) that lead to channel opening. Unlike the late cooperative steps, the major charge-carrying steps have been proposed to occur independently in each of the channel subunits. In this paper, we examine this further. We showed earlier that the two major charge-carrying steps are associated with two sequential outward transmembrane movements of the charged S4 segment. We now use voltage clamp fluorometry to monitor these S4 movements in individual subunits of heterotetrameric channels. In this way, we estimate the influence of one subunit's S4 movement on another's when the energetics of their transmembrane movements differ. Our results show that the first S4 movement occurs independently in each subunit, while the second occurs cooperatively. At least part of the cooperativity appears to be intrinsic to the second S4 charge-carrying rearrangement. Such cooperativity in gating of voltage-dependent channels has great physiological relevance since it can affect both action potential threshold and rate of propagation.

???displayArticle.pubmedLink??? 10694254
???displayArticle.pmcLink??? PMC2217208

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

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

	Figure 1. Subunit stoichiometry in heterotetrameric channels. Cartoon shows expected subunit arrangement of labeled (shaded) and unlabeled (unshaded) subunits that have either wild-type (blank) or voltage-shifted (V and S) S4 gating. The identity of each subunit shown in A and B, and the corresponding abbreviations used in this paper, are shown in C. (A) Homo- and heterotetrameric channels (middle) expressed in the experiments involving the L382V mutant (V). Heterotetrameric channels with one labeled subunit were favored in coinjection of pairs of cRNAs encoding labelable and unlabelable subunits (e.g., C* + V) (top), or defined by injection of a single cRNA encoding four linked subunits (bottom). (B) Subunit stoichiometry in the experiments involving expression of the R365S mutant (S). For expression of heterotetrameric channels, subunits were covalently linked in all experiments to constrain their stoichiometry.
	Figure 2. Dependence of fluorescence on the number of labeled subunits. Possible interactions between fluorophore bound to neighboring subunits does not distort the fluorescence of 352C-TMRM (C), since the f-v of 352C channels labeled to saturation is very little affected by bleaching to a degree that decreases Δf by 80%. Solid lines represent fit of the data points to a Boltzmann function: y = B/{1 + exp[−z (x − v)/25]}. C not bleached: V = −34.7 ± 0.9, z = 2.17 ± 0.09. C* bleached: v = −39.6 ± 1.5, z = 1.95 ± 0.05. n = 10 for both.
	Figure 3. Voltage-dependent gating charge movement (Q) and fluorescence change (F) of TMRM-labeled S352C (C). (A) Fluorescence recording (top) during voltage steps (bottom) from a holding potential of −80 to +20 mV in 10-mV increments. Each fluorescence trace is the average of five sweeps. (B) Normalized q-v (▴) and f-v (▵) relations show a tight correlation, indicating that the f-v is a good indicator of the activation state of a C subunit's S4. Solid lines represent the fit of the data points to a Boltzmann function (q:v = −34.6 ± 1.8 mV, z = 2.26 ± 0.05; f:v = −31.3 ± 2.26 mV, z = 2.21 ± 0.12, n = 12). Note that only fluorescence increase is seen across the measured voltage range.
	Figure 5. Off-gating currents recorded from oocytes expressing either wild type (W, top), labeled double mutant 352C/382V (C/V, bottom), or linked heterotetramer C/V-W-W-W (middle) channels. The gating currents were evoked by steps from −80 mV to voltages ranging from −100 to 0 mV in 10- mV increments, followed by repolarization to −60 mV. Vertical dashed lines correspond to the time constant of the slower component from a two-exponential fit of the off-gating current following the step to −20 mV (left dashed line = C/V's time constant = 1.7 ms; middle dashed line = C/V-W-W-W time constant = 6.0 ms; right dashed line = W time constant = 16.0 ms).
	Figure 4. Cooperativity in wild-type S4 movement. (A) Normalized f-v from channels expressed in oocytes injected with cRNA encoding S352C (C, ▵), or coinjected with a 1.5:8.5 ratio of S352C:L382V cRNA (C+V, ○), and the normalized q-v of L382V channels (V, ♦). The presence of a majority of V subunits produces a rightward shift in the C* subunit f-v toward the even further rightward-shifted depolarized component of the V subunit q-v. Note the absence of a leftward shift toward the V subunit q-v at the foot of the C* f-v. Solid lines represent the fit of the data points to a single Boltzmann (C+V: V = −18.8 ± 0.9 mV, z = 1.54 ± 0.05, n = 6) or to the sum of two Boltzmann functions (V: v1 = −42.6 ± 1.9, z1 = 1.84 ± 0.04, v2 = +33.0 ± 2.0, z2 = 1.38 ± 0.23, n = 5). (B) Normalized f-v of S352C (C, ▵) and of the linked tetramer S352C-L382V-L382V-L382V (C-V-V-V, ○: V = −20.6 ± 2.4 mV, z = 1.74 ± 0.11, n = 11). The linked heterotetramer with only one labelable subunit closely resembles the coinjection (A) that favors channels with zero or one labelable subunit, of which only the latter contribute to the fluorescence signal. (C) S4 gating of the labeled double-mutant S352C-TMRM/L382V (C/V, ▵) is not changed by coassembly with three wild-type subunits (W, ♦), as shown by the comparison of the normalized f-v of the homotetramer C/V to that of the heterotetramer with a 1:3 ratio of C/V to W subunits (obtained by coinjection of C/V and W cRNA in a 1:8.5 ratio (C/V + W in the figure, ○) or by injection of the C/V-W-W-W–linked cRNA, □). (C/V: v = −18.8 ± 0.8, z = 1.70 ± 0.07; C/V + W: v = −16.7 ± 4.2, z = 1.62 ± 0.07; C/V-W-W-W: v = −22.9 ± 1.0, z = 1.68 ± 0.16, n = 5). The normalized q-v of the unlabeled wild type is also shown (W; see Table for the results of the Boltzmann relation fit).
	Figure 7. Subunit interaction in the C/S-W-W-W heterotetramer. (A) Both the q-v (▪) and f-v (□) of S352C/R365S (C/S) have two components. The estimates from the fit of the q-v to the sum of two Boltzmann functions (v1 = −94.0 ± 7.3 mV, z1 = 0.98 ± 0.05, v2 = −7.4 ± 2.5 mV, z2 = 2.46 ± 0.15, mean ± SEM, n = 6) well fits the f-v (solid line) when v1/2 and slope values are fixed, but the amplitudes and polarity of the two components are allowed to vary. This indicates that C* reports on both of S4's major outward activating movements. (B) Assembly of one C/S subunit with three W subunits alters C/S S4 movement. Graph includes f-v's of the linked tetramers C/S-W-W-W (with C/S = S352C-TMRM/R365S, and W = wild type; ⋄) and C/S-S-S-S (with S = R365S; ▵), and of C/S homotetramers (□), as well as the q-v of W homotetramers (•). See Table for the results of the Boltzmann relation fits.
	Figure 6. Cooperativity in late activation transitions. Diagram illustrates the main results of the L382V experiments. (Top) S4 gating of the labeled reporter subunit (C) in heterotetramers containing three L382V subunits (V), obtained either by coinjection (middle) or by expression of the linked tetramer (right), is different from that of the same subunit in the homotetramer C (left). (Bottom) The labeled L382V subunit (C/V) behaves independently of the subunits with which it is assembled, since gating of C/V coassembled with three W subunits, both in monomer coinjection (C/V+W) and in linked tetramer (C/V-W-W-W) experiments, is not different from gating of the C*/V homotetramer. The apparent uncoupling of a subunit from the cooperative influence of its neighbors by the L382V mutation suggests that a wild-type–like labeled subunit coassembled with three L382V subunits is voltage shifted because of the loss of wild-type cooperative interactions.
	Figure 8. Fluorescence changes of the linked tetramers C/S-S-S-S (left) and C/S-W-W-W (right). (A) Fluorescence changes in a double-pulse experiment (five episodes, each composed of five epochs, sequentially in mV: −80/−140/−80/−30/−80, with 20-mV increments of the second and fourth epochs). Note that, for both channels, only a single component is present from very negative voltages up until −30 mV. With further depolarization, only C/S-S-S-S (left) shows two kinetic components both in the “on” and in the off fluorescence: an upward transient, which rapidly decays to a steady state level. These components are the kinetic expressions of S4's three gating states (resting, intermediate, and activated), which are well separated in their voltage dependence by the mutation R365S. As a result of the cooperative interaction in q2, coassembly of one C/S subunit with three W subunits restores the direction of the f2 fluorescence change to that of C* (compare right with Fig. 3 A). (B) Fluorescence changes in response to voltage steps from −80 to −30 and +50 mV. Arrows points to the on and off transients shown by C/S-S-S-S, but absent in C/S-W-W-W, as they are in wild type (Fig. 3 A).
	Figure 9. The single C/S subunit S4 moves independently in q1, but is strongly influenced by the three W subunits in q2. (A) Free fit of the C/S-W-W-W's f-v to the sum of two Boltzmann functions (solid line). See Table for estimates of the voltage parameters. (B) The f-v of C/S-W-W-W is closely approached by a constrained fit using the f1 voltage parameters of C/S and the q2 voltage parameters of W (solid line), but not by constrained fits with the q1 of W and f2 of C/S (dotted line), or the f1 and f2 of C/S-S-S-S, with f2 polarity inverted (dashed line). Note that the foot of the heterotetramer's f-v does not show the shift toward more depolarized potentials expected for subunit interactions in q1 (dotted line).

References [+] :

Aggarwal, Contribution of the S4 segment to gating charge in the Shaker K+ channel. 1996, Pubmed, Xenbase

	Figure 1. Subunit stoichiometry in heterotetrameric channels. Cartoon shows expected subunit arrangement of labeled (shaded) and unlabeled (unshaded) subunits that have either wild-type (blank) or voltage-shifted (V and S) S4 gating. The identity of each subunit shown in A and B, and the corresponding abbreviations used in this paper, are shown in C. (A) Homo- and heterotetrameric channels (middle) expressed in the experiments involving the L382V mutant (V). Heterotetrameric channels with one labeled subunit were favored in coinjection of pairs of cRNAs encoding labelable and unlabelable subunits (e.g., C* + V) (top), or defined by injection of a single cRNA encoding four linked subunits (bottom). (B) Subunit stoichiometry in the experiments involving expression of the R365S mutant (S). For expression of heterotetrameric channels, subunits were covalently linked in all experiments to constrain their stoichiometry.
	Figure 2. Dependence of fluorescence on the number of labeled subunits. Possible interactions between fluorophore bound to neighboring subunits does not distort the fluorescence of 352C-TMRM (C), since the f-v of 352C channels labeled to saturation is very little affected by bleaching to a degree that decreases Δf by 80%. Solid lines represent fit of the data points to a Boltzmann function: y = B/{1 + exp[−z (x − v)/25]}. C not bleached: V = −34.7 ± 0.9, z = 2.17 ± 0.09. C* bleached: v = −39.6 ± 1.5, z = 1.95 ± 0.05. n = 10 for both.
	Figure 3. Voltage-dependent gating charge movement (Q) and fluorescence change (F) of TMRM-labeled S352C (C). (A) Fluorescence recording (top) during voltage steps (bottom) from a holding potential of −80 to +20 mV in 10-mV increments. Each fluorescence trace is the average of five sweeps. (B) Normalized q-v (▴) and f-v (▵) relations show a tight correlation, indicating that the f-v is a good indicator of the activation state of a C subunit's S4. Solid lines represent the fit of the data points to a Boltzmann function (q:v = −34.6 ± 1.8 mV, z = 2.26 ± 0.05; f:v = −31.3 ± 2.26 mV, z = 2.21 ± 0.12, n = 12). Note that only fluorescence increase is seen across the measured voltage range.
	Figure 5. Off-gating currents recorded from oocytes expressing either wild type (W, top), labeled double mutant 352C/382V (C/V, bottom), or linked heterotetramer C/V-W-W-W (middle) channels. The gating currents were evoked by steps from −80 mV to voltages ranging from −100 to 0 mV in 10- mV increments, followed by repolarization to −60 mV. Vertical dashed lines correspond to the time constant of the slower component from a two-exponential fit of the off-gating current following the step to −20 mV (left dashed line = C/V's time constant = 1.7 ms; middle dashed line = C/V-W-W-W time constant = 6.0 ms; right dashed line = W time constant = 16.0 ms).
	Figure 4. Cooperativity in wild-type S4 movement. (A) Normalized f-v from channels expressed in oocytes injected with cRNA encoding S352C (C, ▵), or coinjected with a 1.5:8.5 ratio of S352C:L382V cRNA (C+V, ○), and the normalized q-v of L382V channels (V, ♦). The presence of a majority of V subunits produces a rightward shift in the C* subunit f-v toward the even further rightward-shifted depolarized component of the V subunit q-v. Note the absence of a leftward shift toward the V subunit q-v at the foot of the C* f-v. Solid lines represent the fit of the data points to a single Boltzmann (C+V: V = −18.8 ± 0.9 mV, z = 1.54 ± 0.05, n = 6) or to the sum of two Boltzmann functions (V: v1 = −42.6 ± 1.9, z1 = 1.84 ± 0.04, v2 = +33.0 ± 2.0, z2 = 1.38 ± 0.23, n = 5). (B) Normalized f-v of S352C (C, ▵) and of the linked tetramer S352C-L382V-L382V-L382V (C-V-V-V, ○: V = −20.6 ± 2.4 mV, z = 1.74 ± 0.11, n = 11). The linked heterotetramer with only one labelable subunit closely resembles the coinjection (A) that favors channels with zero or one labelable subunit, of which only the latter contribute to the fluorescence signal. (C) S4 gating of the labeled double-mutant S352C-TMRM/L382V (C/V, ▵) is not changed by coassembly with three wild-type subunits (W, ♦), as shown by the comparison of the normalized f-v of the homotetramer C/V to that of the heterotetramer with a 1:3 ratio of C/V to W subunits (obtained by coinjection of C/V and W cRNA in a 1:8.5 ratio (C/V + W in the figure, ○) or by injection of the C/V-W-W-W–linked cRNA, □). (C/V: v = −18.8 ± 0.8, z = 1.70 ± 0.07; C/V + W: v = −16.7 ± 4.2, z = 1.62 ± 0.07; C/V-W-W-W: v = −22.9 ± 1.0, z = 1.68 ± 0.16, n = 5). The normalized q-v of the unlabeled wild type is also shown (W; see Table for the results of the Boltzmann relation fit).
	Figure 7. Subunit interaction in the C/S-W-W-W heterotetramer. (A) Both the q-v (▪) and f-v (□) of S352C/R365S (C/S) have two components. The estimates from the fit of the q-v to the sum of two Boltzmann functions (v1 = −94.0 ± 7.3 mV, z1 = 0.98 ± 0.05, v2 = −7.4 ± 2.5 mV, z2 = 2.46 ± 0.15, mean ± SEM, n = 6) well fits the f-v (solid line) when v1/2 and slope values are fixed, but the amplitudes and polarity of the two components are allowed to vary. This indicates that C* reports on both of S4's major outward activating movements. (B) Assembly of one C/S subunit with three W subunits alters C/S S4 movement. Graph includes f-v's of the linked tetramers C/S-W-W-W (with C/S = S352C-TMRM/R365S, and W = wild type; ⋄) and C/S-S-S-S (with S = R365S; ▵), and of C/S homotetramers (□), as well as the q-v of W homotetramers (•). See Table for the results of the Boltzmann relation fits.
	Figure 6. Cooperativity in late activation transitions. Diagram illustrates the main results of the L382V experiments. (Top) S4 gating of the labeled reporter subunit (C) in heterotetramers containing three L382V subunits (V), obtained either by coinjection (middle) or by expression of the linked tetramer (right), is different from that of the same subunit in the homotetramer C (left). (Bottom) The labeled L382V subunit (C/V) behaves independently of the subunits with which it is assembled, since gating of C/V coassembled with three W subunits, both in monomer coinjection (C/V+W) and in linked tetramer (C/V-W-W-W) experiments, is not different from gating of the C*/V homotetramer. The apparent uncoupling of a subunit from the cooperative influence of its neighbors by the L382V mutation suggests that a wild-type–like labeled subunit coassembled with three L382V subunits is voltage shifted because of the loss of wild-type cooperative interactions.
	Figure 8. Fluorescence changes of the linked tetramers C/S-S-S-S (left) and C/S-W-W-W (right). (A) Fluorescence changes in a double-pulse experiment (five episodes, each composed of five epochs, sequentially in mV: −80/−140/−80/−30/−80, with 20-mV increments of the second and fourth epochs). Note that, for both channels, only a single component is present from very negative voltages up until −30 mV. With further depolarization, only C/S-S-S-S (left) shows two kinetic components both in the “on” and in the off fluorescence: an upward transient, which rapidly decays to a steady state level. These components are the kinetic expressions of S4's three gating states (resting, intermediate, and activated), which are well separated in their voltage dependence by the mutation R365S. As a result of the cooperative interaction in q2, coassembly of one C/S subunit with three W subunits restores the direction of the f2 fluorescence change to that of C* (compare right with Fig. 3 A). (B) Fluorescence changes in response to voltage steps from −80 to −30 and +50 mV. Arrows points to the on and off transients shown by C/S-S-S-S, but absent in C/S-W-W-W, as they are in wild type (Fig. 3 A).
	Figure 9. The single C/S subunit S4 moves independently in q1, but is strongly influenced by the three W subunits in q2. (A) Free fit of the C/S-W-W-W's f-v to the sum of two Boltzmann functions (solid line). See Table for estimates of the voltage parameters. (B) The f-v of C/S-W-W-W is closely approached by a constrained fit using the f1 voltage parameters of C/S and the q2 voltage parameters of W (solid line), but not by constrained fits with the q1 of W and f2 of C/S (dotted line), or the f1 and f2 of C/S-S-S-S, with f2 polarity inverted (dashed line). Note that the foot of the heterotetramer's f-v does not show the shift toward more depolarized potentials expected for subunit interactions in q1 (dotted line).