Kwon T , Tang Q , Bargiello TA .

GM46889 NIGMS NIH HHS , R01 GM046889 NIGMS NIH HHS , R01 GM098584 NIGMS NIH HHS

	Figure 1. Side view of Cx26 (A) and Cx32*43E1 N2E ΔCT (B) hemichannels after equilibration by all atom molecular dynamics in a fully hydrated POPC membrane. The side chains of residues 108 (green balls), 109 (blue balls), and 56 (orange balls) are shown. The backbone of the parahelix (residues 42–51) is depicted by the red ribbon. The C terminus of Cx32 was not included in the homology model. The Cx32*43E1 N2E hemichannel was incorporated into a fully hydrated POPC membrane as described previously (Kwon et al., 2011). A tetragonal periodic boundary box, 103 × 103 × 109 à including the protein, lipid membrane, TIP3 waters, and 150 mM KCl, was constructed with a program contained in CHARMM-GUI. All protein atoms in this system used the CHARMM 22 force field with CMAP corrections. The CHARMM 36 force field was used for lipid molecules (Feller and MacKerell, 2000). The system was equilibrated for 1 µs at 310 K using NPnT dynamics in Desmond using an Anton computer (Shaw et al., 2007). NPnT is the ensemble name for constant number of particles (N), pressure in normal direction (Pn), and temperature (T).
	Figure 2. Cadmium stabilizes the loop-gate closed state of Cx32*43E1 N2E E109C hemichannels. (A) A segment of a continuous current trace evoked by a train of alternating voltage polarizations from −10 mV (10 s duration) to −70 mV (10 s duration) with different [Cd2+] indicated by the colored solid bars. (B) Expansion of steady-state current traces shown in A. Current reductions at −70 mV (upward, positive going current relaxations) reflect closure of loop-gates. Increase in currents at −10 mV (downward, negative going current relaxations) reflect opening of loop-gates. (C) Steady-state peak currents measured at −70 mV are plotted against [Cd2+]. Half-maximal current reduction is obtained at [Cd2+] ∼10 µM. (D) Plots of normalized fitted currents of steady-state current relaxations at −70 mV. The time constants of current relaxations are shortened as [Cd2+] is increased. Current relaxations were well fitted by standard exponential function with two terms in Clampfit 9.0. (E) Plots of normalized fitted currents of steady-state current relaxations at −10 mV. Current relaxations in 0 and 10 µM Cd2+ were well fitted by single exponential functions. Current relaxations obtained with 10-s voltage applications became linear in higher [Cd2+] and could not be fitted to an exponential function.
	Figure 3. MTSEA modification of E109C attenuates Cd2+ stabilization of the loop-gate closed state. (A) A segment of a continuous current trace evoked by a train of alternating voltage polarizations between −10 mV (15 s duration) and −70 mV (5 s duration). The colored solid bars indicate the onset and duration of specified treatments. Capacitive transients were reduced but not fully eliminated by 100× data reduction in Clampfit. (B) Expansion of steady-state current traces shown in A as indicated by arrows. Peak currents obtained at steady-state at −70 mV are marked by arrows.
	Figure 4. Endogenous cysteines in the CT are not required for Cd2+ coordination by E109C. (A) A segment of a continuous current trace evoked by a train of alternating voltage polarizations from −10 mV (10 s duration) to −70 mV (10 s duration). 20 µM Cd2+ was applied for the time indicated by the red bar. (B) Plots of normalized fitted currents of steady-state current relaxations at −70 mV corresponding to loop-gate closure. (C) Plots of normalized fitted currents of steady-state current relaxations at −10 mV corresponding to loop-gate opening.
	Figure 5. Cd2+ does not substantially alter Vj-gating of Cx32*43E1 N2E 109C hemichannels. (A) Segments of a continuous current trace evoked by a train of alternating voltage polarizations from 10 mV (15 s duration) to 50 mV (5 s duration). The current relaxations elicited at 50 mV correspond to the closure of V-gates. The time course of currents elicited at 10 mV is nearly linear and suggests that the time course of Vj-gate opening is fast at this potential. The colored solid bars indicate the onset and duration of the specified treatments. (B) Expansion of steady-state current traces shown in A. Peak currents at 50 mV are marked by arrows. (C) Plots of normalized fitted currents of steady-state current relaxations at 50 mV. Current elicited by a 40-mV polarization. The oocyte was perfused with bath solution containing 10 µM CdCl2 at the time indicated by the arrow. (D) A continuous trace of Cx32*43E1 E109C hemichannel evoked by a 40-mV polarization from a holding potential of 10 mV. The observed current relaxation reflects closure of Vj-gates. 10 µM Cd2+ was perfused at the time indicated by the arrow.
	Figure 6. Cadmium stabilizes a closed state but has no effect on the open state of Cx43*43E1 L109C hemichannels. (A) A segment of a continuous current trace evoked by a train of alternating voltage polarizations from 10 mV (10 s duration) to −70 mV (10 s duration) in bath solution containing 1.8 mM MgCl2. The time of perfusion with 60 µM Cd2+ is depicted by the red bar. (B) Same oocyte as in A, but voltage was stepped between −10 and −70 mV. (C) Plot of normalized fitted currents at steady-state of loop-gate closure at −70 mV for the current trace shown in B. Steady currents were fitted to an exponential function with two terms. Current relaxations reach steady-state faster in the presence of Cd2+ (red), indicating a shortening of the time constants of channel closure. (D) Plot of normalized fitted currents at steady-state of loop-gate opening at −10 mV. Steady currents were fitted to an exponential function with one term. Current relaxations reach steady-state slower in the presence of Cd2+, which indicates a lengthening of time constant of channel opening. (E) Time series of current traces evoked by a train of alternating voltage polarizations from 10 mV (10 s duration) to 50 mV (10 s duration) in bath solution containing 1.8 mM MgCl2. The time of perfusion with 60 µM Cd2+ is depicted by the red bar.
	Figure 7. Cadmium stabilizes the loop-gate but not Vj-gate closed state of Cx32*43E1 N2E L108C hemichannels. (A) A segment of a continuous current trace evoked by a train of alternating voltage polarizations from −10 mV (15 s duration) to −70 mV (5 s duration). The time of perfusion with 10 µM Cd2+ is depicted by the red bar. (B) Same as A only with perfusion of 20 µM Cd2+. (C) Expanded segments of the current trace shown in B at steady-state. Initial currents before perfusion of Cd2+ (black trace), steady-state current after perfusion of Cd2+ (red trace), and steady-state current after wash (blue trace). Arrows mark the level of peak current. Capacitive transients were not removed. (D) Plot of normalized fitted currents at steady-state of loop-gate closure at −70 mV. Steady currents were fitted to an exponential function with two terms. Current relaxations reach steady-state faster in the presence of Cd2+, which indicates a shortening of the time constants of channel closure. (E) Plot of normalized fitted currents at steady-state of loop-gate opening at −10 mV. Steady currents were fitted to an exponential function with one term. Current relaxations reach steady-state slower in the presence of Cd2+, which indicates a lengthening of time constant of channel opening. (F) A segment of a continuous current trace elicited by a polarization from 10 to 30 mV. Current relaxation reflects closure of Vj-gates. 20 µM Cd2+ was perfused at the time indicated by the arrow followed by washing with Cd2+-free bath solution. Cd2+ had no effect on the time course to reach steady-state.
	Figure 8. Cadmium does not stabilize closed conformations of Cx32*43E1 Q56C hemichannels. (A) A segment of a continuous current trace evoked by a train of alternating voltage polarizations from −10 mV (15 s duration) to −70 mV (5 s duration) in bath solution containing 1.8 mM MgCl2. The time of perfusion with 20 µM Cd2+ is depicted by the red bar, 60 µM Cd2+ by the green bar. (B) Same as in A only the bath solution contained 1.8 mM CaCl2. (C) Plot of normalized fitted currents at steady-state of loop-gate closure at −70 mV in 1.8 mM MgCl2 bath solutions with no added Cd2+ (black trace) and 60 µM Cd2+ (green trace). Steady currents were fitted to an exponential function with two terms. Current relaxations reach steady-state more slowly in 60 µM Cd2+, which indicates that the reductions in peak current shown in A do not result from the stabilization of closed states by Cd2+ coordination of thiol groups. (D) Plot of normalized fitted currents at steady-state of loop-gate closure at −70 mV in 1.8 mM CaCl2 bath solutions with no added Cd2+ (black trace) and 60 µM Cd2+ (green trace). Steady currents were fitted to an exponential function with one term. The similarity of the time course of current relaxations in the presence and absence of Cd2+ indicate that the reductions in peak current shown in B do not result from the stabilization of closed states by Cd2+ coordination of thiol groups.
	Figure 9. Schematic representation of open and loop-gate closed state of the Cx32*43E1 hemichannel. Residues reported in this study are indicated in red. Residues in black are those reported in Tang et al. (2009). The position of the 45th and 50th residues in the closed state are based on results from Cx50 hemichannels reported in Verselis et al. (2009) and unpublished studies of Cx32*43E1 hemichannels (see also Bargiello et al., 2012).

Abrams, Gap junctions in inherited human disorders of the central nervous system. 2012, Pubmed

Abrams, Gap junctions in inherited human disorders of the central nervous system. 2012, Pubmed
Abrams, Voltage opens unopposed gap junction hemichannels formed by a connexin 32 mutant associated with X-linked Charcot-Marie-Tooth disease. 2002, Pubmed , Xenbase
Abrams, Mutations in connexin 32: the molecular and biophysical bases for the X-linked form of Charcot-Marie-Tooth disease. 2000, Pubmed
Akabas, Acetylcholine receptor channel structure probed in cysteine-substitution mutants. 1992, Pubmed , Xenbase
Allen, Atomic force microscopy of Connexin40 gap junction hemichannels reveals calcium-dependent three-dimensional molecular topography and open-closed conformations of both the extracellular and cytoplasmic faces. 2011, Pubmed
Arslan, Cytosolic Ca2+ homeostasis in Ehrlich and Yoshida carcinomas. A new, membrane-permeant chelator of heavy metals reveals that these ascites tumor cell lines have normal cytosolic free Ca2+. 1985, Pubmed
Bargiello, Voltage-dependent conformational changes in connexin channels. 2012, Pubmed , Xenbase
Barrio, Gap junctions formed by connexins 26 and 32 alone and in combination are differently affected by applied voltage. 1991, Pubmed , Xenbase
Bennett, Gap junctions: new tools, new answers, new questions. 1991, Pubmed
Bicego, Selective defects in channel permeability associated with Cx32 mutations causing X-linked Charcot-Marie-Tooth disease. 2006, Pubmed
Bukauskas, Gap junction channel gating. 2004, Pubmed
Dobrowolski, Connexin-caused genetic diseases and corresponding mouse models. 2009, Pubmed
Ebihara, Effect of external magnesium and calcium on human connexin46 hemichannels. 2003, Pubmed , Xenbase
Eggermont, Calcium-activated chloride channels: (un)known, (un)loved? 2004, Pubmed
Eswar, Comparative protein structure modeling using Modeller. 2006, Pubmed
Gómez-Hernández, Molecular basis of calcium regulation in connexin-32 hemichannels. 2003, Pubmed , Xenbase
Harris, Emerging issues of connexin channels: biophysics fills the gap. 2001, Pubmed
Hartzell, Calcium-activated chloride channels. 2005, Pubmed
Holmgren, The activation gate of a voltage-gated K+ channel can be trapped in the open state by an intersubunit metal bridge. 1998, Pubmed
Karlin, Substituted-cysteine accessibility method. 1998, Pubmed
Khalili-Araghi, Molecular dynamics investigation of the ω-current in the Kv1.2 voltage sensor domains. 2012, Pubmed
Khalili-Araghi, Calculation of the gating charge for the Kv1.2 voltage-activated potassium channel. 2010, Pubmed
Kr zel, Coordination of heavy metals by dithiothreitol, a commonly used thiol group protectant. 2001, Pubmed
Kwon, Molecular dynamics simulations of the Cx26 hemichannel: evaluation of structural models with Brownian dynamics. 2011, Pubmed , Xenbase
Kwon, Molecular dynamics simulations of the Cx26 hemichannel: insights into voltage-dependent loop-gating. 2012, Pubmed
Laird, The gap junction proteome and its relationship to disease. 2010, Pubmed
Lee, Connexin mutations causing skin disease and deafness increase hemichannel activity and cell death when expressed in Xenopus oocytes. 2009, Pubmed , Xenbase
Maeda, Structure of the connexin 26 gap junction channel at 3.5 A resolution. 2009, Pubmed
Moreno, Role of the carboxyl terminal of connexin43 in transjunctional fast voltage gating. 2002, Pubmed , Xenbase
Müller, Conformational changes in surface structures of isolated connexin 26 gap junctions. 2002, Pubmed
Oh, Changes in permeability caused by connexin 32 mutations underlie X-linked Charcot-Marie-Tooth disease. 1997, Pubmed , Xenbase
Oh, Stoichiometry of transjunctional voltage-gating polarity reversal by a negative charge substitution in the amino terminus of a connexin32 chimera. 2000, Pubmed , Xenbase
Oh, Determinants of gating polarity of a connexin 32 hemichannel. 2004, Pubmed , Xenbase
Pfahnl, A chimeric connexin forming gap junction hemichannels. 1997, Pubmed , Xenbase
Revilla, Molecular dissection of transjunctional voltage dependence in the connexin-32 and connexin-43 junctions. 1999, Pubmed , Xenbase
Rubin, Molecular analysis of voltage dependence of heterotypic gap junctions formed by connexins 26 and 32. 1992, Pubmed , Xenbase
Sánchez, Differentially altered Ca2+ regulation and Ca2+ permeability in Cx26 hemichannels formed by the A40V and G45E mutations that cause keratitis ichthyosis deafness syndrome. 2010, Pubmed , Xenbase
Seki, Modifications in the biophysical properties of connexin43 channels by a peptide of the cytoplasmic loop region. 2004, Pubmed
Shibayama, Effect of charge substitutions at residue his-142 on voltage gating of connexin43 channels. 2006, Pubmed , Xenbase
Sosinsky, Structural organization of gap junction channels. 2005, Pubmed
Tang, Conformational changes in a pore-forming region underlie voltage-dependent "loop gating" of an unapposed connexin hemichannel. 2009, Pubmed , Xenbase
Thimm, Calcium-dependent open/closed conformations and interfacial energy maps of reconstituted hemichannels. 2005, Pubmed
Trexler, Voltage gating and permeation in a gap junction hemichannel. 1996, Pubmed , Xenbase
Unwin, Calcium-mediated changes in gap junction structure: evidence from the low angle X-ray pattern. 1983, Pubmed
Unwin, Two configurations of a channel-forming membrane protein. , Pubmed
Vargas, In search of a consensus model of the resting state of a voltage-sensing domain. 2011, Pubmed
Verselis, Loop gating of connexin hemichannels involves movement of pore-lining residues in the first extracellular loop domain. 2009, Pubmed , Xenbase
Verselis, Opposite voltage gating polarities of two closely related connexins. 1994, Pubmed , Xenbase
Verselis, Divalent cations regulate connexin hemichannels by modulating intrinsic voltage-dependent gating. 2008, Pubmed , Xenbase
Yellen, The moving parts of voltage-gated ion channels. 1998, Pubmed
Zonta, Permeation pathway of homomeric connexin 26 and connexin 30 channels investigated by molecular dynamics. 2012, Pubmed