Gourgy-Hacohen O et al. (2014), Capturing distinct KCNQ2 channel resting states...

XB-ART-50950

J Gen Physiol 2014 Dec 01;1446:513-27. doi: 10.1085/jgp.201411221.

Show Gene links Show Anatomy links

Capturing distinct KCNQ2 channel resting states by metal ion bridges in the voltage-sensor domain.

Gourgy-Hacohen O , Kornilov P , Pittel I , Peretz A , Attali B , Paas Y .

???displayArticle.abstract???
Although crystal structures of various voltage-gated K(+) (Kv) and Na(+) channels have provided substantial information on the activated conformation of the voltage-sensing domain (VSD), the topology of the VSD in its resting conformation remains highly debated. Numerous studies have investigated the VSD resting state in the Kv Shaker channel; however, few studies have explored this issue in other Kv channels. Here, we investigated the VSD resting state of KCNQ2, a K(+) channel subunit belonging to the KCNQ (Kv7) subfamily of Kv channels. KCNQ2 can coassemble with the KCNQ3 subunit to mediate the IM current that regulates neuronal excitability. In humans, mutations in KCNQ2 are associated with benign neonatal forms of epilepsy or with severe epileptic encephalopathy. We introduced cysteine mutations into the S4 transmembrane segment of the KCNQ2 VSD and determined that external application of Cd(2+) profoundly reduced the current amplitude of S4 cysteine mutants S195C, R198C, and R201C. Based on reactivity with the externally accessible endogenous cysteine C106 in S1, we infer that each of the above S4 cysteine mutants forms Cd(2+) bridges to stabilize a channel closed state. Disulfide bonds and metal bridges constrain the S4 residues S195, R198, and R201 near C106 in S1 in the resting state, and experiments using concatenated tetrameric constructs indicate that this occurs within the same VSD. KCNQ2 structural models suggest that three distinct resting channel states have been captured by the formation of different S4-S1 Cd(2+) bridges. Collectively, this work reveals that residue C106 in S1 can be very close to several N-terminal S4 residues for stabilizing different KCNQ2 resting conformations.

???displayArticle.pubmedLink??? 25385787
???displayArticle.pmcLink??? PMC4242811
???displayArticle.link??? J Gen Physiol

Species referenced: Xenopus laevis
Genes referenced: dio3 kcnq2 kcnq3

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

	Figure 1. Effects of DTT on the KCNQ2 S4 mutants S195C, R198C, and R201C expressed in CHO cells. (A) Representative current traces (out of seven) of mutant S195C in the absence or presence of 100 ÂµM DTT. Currents were evoked from a holding potential of â90 mV by depolarizing steps from â110 to 30 mV in 10-mV increments and repolarized to â60 mV. (B) Representative current traces (out of eight) of mutant R198C in the absence or presence of 100 ÂµM DTT. Currents were evoked as in A. (C) Representative current traces (out of seven) of mutant R201C in the absence or presence of 100 ÂµM DTT. Currents were evoked as in A.
	Figure 2. Effects of Cd2+ ions on the KCNQ2 S4 mutants S195C, R198C, and R201C expressed in CHO cells. (A) Representative current traces of mutant S195C in the absence (top) and presence of 100 ÂµM Cd2+ (middle), and after washout (bottom). Currents were evoked as in Fig. 1 A. (B) Normalized currentâvoltage relations of mutant S195C in the absence and presence of 100 ÂµM Cd2+ (n = 7). (C) Representative current traces of mutant R198C in the absence (top) and presence of 100 ÂµM Cd2+ (middle), and after washout (bottom). (D) Normalized currentâvoltage relations of mutant R198C in the absence and presence of 100 ÂµM Cd2+ (n = 7). (E) Representative current traces of mutant R201C in the absence (top) and presence of 100 ÂµM Cd2+ (middle), and after washout (bottom). (F) Normalized currentâvoltage relations of mutant R201C in the absence and presence of 100 ÂµM Cd2+ (n = 6).
	Figure 3. Effects of Cd2+ ions on the KCNQ2 S4âS5 and S4âS1 double mutants expressed in CHO cells. (A) From top to bottom rows, representative current traces of the indicated KCNQ2 S4âS5 and S4âS1 double mutants in the absence (left) or presence of 100 ÂµM Cd2+ (right), and (B) their corresponding normalized currentâvoltage relations; n = 5â7. Currents were evoked as in Fig. 1 A. Error bars show mean Â± SEM.
	Figure 4. State dependence of disulfide bond formation in mutant S195C expressed in Xenopus oocytes. (A) Representative current traces of WT KCNQ2 in the absence or presence of 100 ÂµM Cu-Phen; n = 10. At a holding potential of â80 mV, Xenopus oocytes were depolarized from â90 to 45 mV in 15-mV increments and repolarized at â60 mV. (B) Representative current traces of mutant S195C before or after Cu-Phen application in the channel closed state and after washout with 1 mM DTT. Oocytes were incubated for 2 min at â80 mV in ND96 solution containing 100 ÂµM Cu-Phen, washed with ND96 for another 5 min at â80 mV, and then depolarized as in A; subsequently, the oocytes were washed out in the presence of 1 mM DTT and subjected to the same depolarizing protocol as in A; n = 8. (C) Representative current traces of mutant S195C before or after Cu-Phen application in the channel open state. From a holding potential of â80 mV, oocytes were subjected to a depolarizing step to 30 mV. 2 s after the beginning of the test pulse, fast application of 100 ÂµM Cu-Phen was applied for up to 2 min; n = 8.
	Figure 5. State dependence of disulfide bond formation in mutants R198C and R201C expressed in Xenopus oocytes. (A) Representative current traces of mutant R198C before or after Cu-Phen application in the channel closed state and after washout with 1 mM DTT. Oocytes were incubated for 2 min at â80 mV in ND96 solution containing 100 ÂµM Cu-Phen, washed with ND96 for another 5 min at â80 mV, and then depolarized as in Fig. 4 A; subsequently, the oocytes were washed out in the presence of 1 mM DTT and subjected to the same depolarizing protocol; n = 7. (B) Representative current traces of mutant R198C before or after Cu-Phen application in the channel open state. From a holding potential of â80 mV, oocytes were subjected to a depolarizing step to 0 mV. 2 s after the beginning of the test pulse, fast application of 100 ÂµM Cu-Phen was applied for up to 2 min; n = 7. (C) Representative current traces of mutant R201C before or after Cu-Phen application at â80 mV and after washout with 1 mM DTT. Oocytes were treated and currents were evoked as in Fig. 4 A; n = 7. (D) Representative current traces of mutant R201C before or after Cu-Phen application in the channel open state. Oocytes were treated and currents were evoked as in B; n = 10.
	Figure 6. Sensitivity of S4 and S1 mutants to Cd2+ ions in KCNQ2/KCNQ3 heteromers expressed in CHO cells. (A) From top to bottom rows, representative current traces of the indicated KCNQ channel assemblies, in the absence (left) or presence of 100 ÂµM Cd2+ (right) and (B) their corresponding normalized currentâvoltage relations; n = 6â23. Currents were evoked as in Fig. 1 A. Error bars show mean Â± SEM.
	Figure 7. Effects of Cd2+ ions on concatenated tetrameric KCNQ2 mutants expressed in CHO cells. (A; right) Scheme of the KCNQ2 concatenated tetrameric channel construct, where subunits D1, D2, D3, and D4 are connected by flexible linkers. Each linker (eight glycines) harbors a unique restriction site. (Left) Western blot showing lysates of HEK 293 cells transfected with empty vector (Mock) and KCNQ2 monomeric and concatenated tetrameric constructs. (B) Conductanceâvoltage relations of WT monomeric and WT concatenated tetrameric KCNQ2 constructs. Data were fitted to a single Boltzmann function. (C and D) Representative current traces of KCNQ2 concatemer D1 R198C mutant in the absence or presence of 100 ÂµM Cd2+ and corresponding current densityâvoltage relations; n = 7. (E and F) Representative current traces of KCNQ2 concatemer D1 R198C, D2 C106A in the absence or presence of 100 ÂµM Cd2+ and corresponding current densityâvoltage relations; n = 7. (G and H) Representative current traces of KCNQ2 concatemer D1 R198C, C106A in the absence or presence of 100 ÂµM Cd2+ and corresponding current densityâvoltage relations; n = 7. Error bars show mean Â± SEM.
	Figure 8. VSDs of KCNQ2 open- and closed-state models. Ribbon diagrams of individual VSDs and their stabilizing interactions, respectively, in: (A and B) the deep resting state C3, (C and D) the intermediate resting state C2, (E and F) the superficial resting state C1, and (G and H) the open state. The S4 Î± helix is colored in magenta, and the 310-helix portion is colored in dark magenta. The bottom panels show the cysteine pairs that are sufficiently close to each other to form Cd2+ bridges (dashed black lines) and the potential interactions between the S4 and S2 segments. The distances provided inside the panels are in angstrom.
	Figure 9. Homology models of KCNQ2 channel open and closed states. (Left) Side views of ribbon diagrams of two facing subunits showing the pore-forming region and the VSD for: (A) the deep resting state C3, (B) the intermediate resting state C2, (C) the superficial resting state C1, and (D) the open state. The rare and frontal subunits were removed for clarity. (Right) Stick models of the channel pore region (black) and the corresponding Voronoi diagrams (bluish-green), as generated for the pore molecular surface from the predicted place of the uppermost K+ ion down to the bottom of the pore (prepared by MOLE 2.0). The diameter of the pore at the activation gate (level of A317, as indicated in A by a gray arrow) was calculated by MOLE 2.0 as (A) 3.0 Ã, (B) 3.6 Ã, and (C) 4.5 Ã. For the open state (D), the diameter was calculated as 11.4 Ã at the level of G310.

References [+] :

Armstrong, Charge movement associated with the opening and closing of the activation gates of the Na channels. 1974, Pubmed