Hydrophobic plug functions as a gate in voltage-gated proton channels

Adam Chamberlin; Feng Qiu; Santiago Rebolledo; Yibo Wang; Sergei Y. Noskov; H. Peter Larsson

doi:10.1073/pnas.1318018111

New Research In

Edited by Ramon Latorre, Centro Interdisciplinario de Neurociencias, Universidad de Valparaíso, Valparaíso, Chile, and approved December 4, 2013 (received for review October 1, 2013)

Significance

Voltage-gated proton (Hv1) channels play important roles in various physiological processes, such as the innate immune response. However, the mechanism by which this channel closes and opens its proton permeation pathways is unknown, due to the lack of structural information about the closed and open states of the channel. This study uses both simulation and experimental approaches to develop models of the closed and open states of the Hv1 channel. These models suggest a mechanism for how the channel closes and opens. The models also suggest a mechanism explaining why a blocker only binds to the open state of the channel. These structural models will be essential for future investigations of this channel and the development of new pharmacological blockers.

Abstract

Voltage-gated proton (Hv1) channels play important roles in the respiratory burst, in pH regulation, in spermatozoa, in apoptosis, and in cancer metastasis. Unlike other voltage-gated cation channels, the Hv1 channel lacks a centrally located pore formed by the assembly of subunits. Instead, the proton permeation pathway in the Hv1 channel is within the voltage-sensing domain of each subunit. The gating mechanism of this pathway is still unclear. Mutagenic and fluorescence studies suggest that the fourth transmembrane (TM) segment (S4) functions as a voltage sensor and that there is an outward movement of S4 during channel activation. Using thermodynamic mutant cycle analysis, we find that the conserved positively charged residues in S4 are stabilized by countercharges in the other TM segments both in the closed and open states. We constructed models of both the closed and open states of Hv1 channels that are consistent with the mutant cycle analysis. These structural models suggest that electrostatic interactions between TM segments in the closed state pull hydrophobic residues together to form a hydrophobic plug in the center of the voltage-sensing domain. Outward S4 movement during channel activation induces conformational changes that remove this hydrophobic plug and instead insert protonatable residues in the center of the channel that, together with water molecules, can form a hydrogen bond chain across the channel for proton permeation. This suggests that salt bridge networks and the hydrophobic plug function as the gate in Hv1 channels and that outward movement of S4 leads to the opening of this gate.

The best-studied function of voltage-gated proton (Hv1) channels is in the immune system, where the activity of Hv1 channels has been shown to play a key role in charge compensation for the electron extrusion by NADPH oxidase during the respiratory burst in phagocytes (1, 2). In addition, this channel is also found in many other cell types including neurons, sperm, and lung airway epithelia cells, where it has been implicated in acid extrusion, male fertility, and the pathology of asthma, respectively (3, 4). During stroke, the activity of Hv1 channels exacerbates neuronal death (5). In 2006, two independent groups identified the genes coding for the Hv1 channel in humans (hHv1) (6) and mice and Ciona intestinalis (voltage-sensing only protein; we call it “Ci-Hv1” in this paper) (7). The sequences of Hv1/(VSOP) are homologous to the sequences of the voltage-sensing domain (VSD) of other voltage-gated ion channels, such as voltage-gated sodium, potassium, and calcium channels (Nav, Kv, and Cav channels), and the voltage sensor-containing phosphatase VSP (6, 7). Hydropathy analysis of the Hv1 sequence suggests the existence of four transmembrane (TM) segments (Fig. 1A), designated S1 to S4, similar to the four TM segments of the VSD of Kv channels (Fig. S1). However, Hv1 channels lack the last two TM segments of Kv channels that make up the pore-forming domain in Kv channels. In Kv channels, this pore-forming domain contains the gate that controls the flow of ions through the pore. Therefore, it is unclear how Hv1 channels gate and control the flow of protons. Here we performed both experimental and computational studies to better understand the gating process in Hv1 channels.

Fig. 1.

Hv1 channel topology and principle of thermodynamic mutant cycle analysis. (A) Schematic of the Hv1 channel. (B) The relative positions of the charged residues in the four TM segments of one subunit. (C and D) Principle of thermodynamic mutant cycle analysis: (C) if mutant 1 (M₁) and mutant 2 (M₂) are not interacting with each other, the free-energy change caused by double-mutant M_1,2 would be equal to the sum of the free-energy changes by mutant 1 and mutant 2 alone; (D) if they are interacting with each other, the free-energy change due to the double mutation of 1 and 2 (M_1,2) would be different from the sum of the free-energy changes by mutant 1 and mutant 2 alone.

Hv1 channels form dimers mediated by a coiled-coil formation of the two C termini (Fig. 1A) (8, 9). The removal of the C terminus generates monomeric Hv1 channels that still function as voltage-gated proton channels, indicating that the proton permeation pathway is intrinsic to a single VSD of the Hv1 channel. Ramsey et al. (10) and Wood et al. (11) constructed a homology model of the Hv1 channel in the open state. Molecular dynamics (MD) simulations of this model suggested a Grotthuss mechanism of the proton hopping along a water wire for proton conduction in Hv1 (10, 11). Subsequently, a conserved and unique aspartate (Asp112 in hHv1) was identified as the selectivity filter for the permeation of protons in Hv1 channels (12). Berger and Isacoff (13) proposed an interaction between Asp112 and the third charged residue in S4 in the open conformation of the channel and that this interaction was important for proton selectivity. Recently, 2-guanidobenzimidazole (2-GBI), a blocker of Hv1 channels, was identified that blocks the Hv1 channel when applied on the intracellular side of the channel (14), and 2-GBI only accesses its binding site in open channels, suggesting that the gate in Hv1 channels is located on the cytosolic side of the channel and that the closed gate thereby prevents intracellular access to the binding site in the closed state (14). Despite these advances, the structural information of how Hv1 channels open and close their permeation pathways remains to be elucidated.

Three positively charged residues in a conserved motif in S4 of Hv1 channels have been shown to be involved in voltage sensing (15, 16). Accessibility experiments of cysteines introduced into S4 of Hv1 channels suggest that S4 functions as the voltage sensor and that there is an outward movement of S4 upon membrane depolarization (15, 16). The three positively charged residues contribute most, if not all, of the gating charges in Hv1 channels (15, 16). Mutations of each of these three residues caused different effects on channel function, indicating that the individual basic S4 residues play different roles, in addition to being gating charges, during channel activation (6, 7, 10). This is most likely due to the different local environments experienced by each residue or different structural interactions with other residues in the closed versus the open state (Fig. 1B). For example, it has been shown that S4 residues in Kv and Nav channels are stabilized by forming ion pairs with acidic residues in the other TM domains of the channels and that these interactions change between closed and open states (17, 18).

Here, we made single and double mutants of pairs of Hv1 residues with opposite charges and tested the effects of these mutations to identify possible electrostatic interactions between these residues. We used the position of the conductance–voltage [G(V)] curve, which reflects the relative stability of the closed and open conformations of the channel protein, to evaluate the effects of the mutations. We used thermodynamic mutant cycle analysis to measure the interactions between residues in the Hv1 channels to obtain structural information about the closed and open states of Hv1 channels. If a single-residue mutation in the channel changes the stability of the closed state relative to the open state, there will be a change in the voltage dependence of activation, G(V), that can be used to estimate the change in the free-energy difference between the closed and open states caused by the mutation. If the effects of two mutations on channel gating are independent, the free-energy changes caused by the two single mutations will be additive; the sum of the free-energy changes caused by the single mutations is equal to the free-energy change of the double mutation (Fig. 1C). In contrast, if the two residues are interacting with each other, there will be a difference between the sum of the free-energy changes of the two single mutations and the free-energy change of the double mutation (Fig. 1D). This difference is defined as the coupling energy, indicating the strength of interactions between these two residues. Usually, a coupling energy of greater than 1.5 kT (∼0.89 kcal/mol) represents a significant interaction between two residues (19). We modeled both the open and closed structures of a single subunit of Ci-Hv1 based on the crystal structure (for open state, see ref. 20) and model (for closed state, see ref. 21) of the homologous Kv1.2–2.1 chimeric channel. Our closed- and open-state models are consistent with our thermodynamic mutant cycle analysis and clarify several aspects of the molecular mechanism underlying voltage-dependent gating of Hv1 channels.

Results

Several groups have suggested different possible alignments of the VSDs of Hv1 and Kv channels (10, 11, 22). However, unambiguous alignment of the VSD of Kv channels and Hv1 channels remains challenging, especially for the S4 helix (23). We adopted an alignment proposed by Musset et al. (12) with the modification that the first arginine in S4, Arg255, aligns with Arg312 rather than with Arg315 in the Kv1.2–2.1 channel (Fig. S1B). This is analogous to the sequence alignment reported by Gonzalez et al., based on our previous accessibility data on cysteines introduced in S4 (15).

Generation of Open- and Closed-State Models.

The open state is modeled using the Kv1.2–2.1 chimeric channel chain B [Protein Data Bank (PDB) ID code 2R9R] (20) as the template. The open Hv1 monomer model exhibits considerable conformational flexibility in the course of 100-ns MD simulations. The average root mean squared fluctuation values of heavy atoms peak at 3.3 Å and have a minimum of 1.5 Å (Fig. S2A). We find several significant and long-lived contacts between residues, particularly involving the arginines in S4 and a cluster of negative residues in the external half of S1 and S3. Accounting for the considerable flexibility of the side chains involved in intrasubunit salt bridging, we opt to assess average interaction energies and occupancy times. We here describe the main interactions, from the external to the internal part of the protein. Glu167 in the external half of S1 interacts with the two most external S4 arginines, Arg255 and Arg258, (but not with Arg261), while there is an absence of interaction between Asp171 at the external end of S1 and any of the S4 arginines (Fig. 2A). There is a significant stabilization by the S3–S4 interactions of Asp233 in the external half of S3 with Arg255, Arg258, and Arg261 (Fig. 2A). Asp160 located in the middle of S1 interacts with Arg258 and Arg261 (but not with Arg255) (Fig. 2A). In the open-state model, not only are S4 interactions significant, but there are also significant S2–S3 interactions between the positively charged Lys205 in the internal half of S2 and the negatively charged Glu219 and Asp222 in S3 near the intracellular S2–S3 loop (Fig. S2B).

Fig. 2.

Models of the closed and open states of the Hv1 channel. The (A) open and (B) closed state of the monomer with the helices labeled (left) and the residues involved in the internal salt bridges labeled (right).

The closed state is modeled on the closed-state model for the Kv1.2 channel of Pathak et al. (21). In our closed-state model, S4 clearly moves into the intracellular solution, as supported by Gonzalez et al. (15). In the closed state, the TM segments of the channel appear to align along the normal axis of the membrane, whereas in the open state S1 and S2 form a stable antiparallel helix pair and S3 and S4 form another pair at an angle with the normal axis of the membrane (Fig. 2 A and B). The strength and number of contacts changes significantly upon closing. Most notably, the strong salt bridges present in the open state between the three S4 arginines and the negative residues in the external half of S1 (Glu167 and Asp160) and S3 (Asp233) are broken and instead the two most external S4 arginines (Arg255 and Arg258) form new interactions with a cluster of negative residues in the internal half of S2 (Glu201) and S3 (Glu219 and Asp222). The third arginine, Arg261, is pulled far enough into the cytosolic milieu that it interacts solely with the surrounding waters, consistent with our accessibility data (15) (Fig. 2B). The S1–S4 interactions in the open state break during the transition to the closed state due to the large inward movement of S4 toward the intracellular phase, a smaller displacement of S1 outwards toward the extracellular phase, and rotation of the S1–S2 TM segments to align with the normal axis of the membrane. In the closed state, Asp160 in the middle of S1 swings out toward the aqueous phase to bind with Lys173 in the S1–S2 loop (Fig. S2C). The Asp160–Lys173 interaction stabilizes the S1–S2 loop and prevents Asp160’s putative involvement in the proton conduction pathway in the closed state (Discussion, Implications for the Proton Conduction Pathway and ref. 12). In the closed state, Glu201, which in the open state is engaged in weak binding with His145 modulated by an electrostatic repulsion due to nearby carboxylates of Glu219 (Fig. S2B), binds strongly to Arg255 and Arg258 in S4 (Fig. 2B). This change in interactions occurs while maintaining a strong network of bonds between Lys205 in S2 and Glu219 and Asp222 in S3, which has moved very little relative to S2 between the open- and closed-state models. In the closed state, Asp222 interacts with both Arg255 and Arg258, whereas Glu219 interacts solely with Arg258.

Shifts in Voltage Dependence of Activation by Charge Neutralizations.

Next, we experimentally assessed the interactions between pairs of residues by measuring the voltage dependence of activation of the wild type (WT), single-mutant, and double-mutant Hv1 channels from C. intestinalis. The mutant and WT channels are expressed in Xenopus oocytes and the conductance–voltage [G(V)] relationship is deduced from the tail current–voltage relationship. Neutralization of either one of two basic residues in S4, Arg255Cys and Arg258Cys, shifted the voltage dependence of gating in the hyperpolarized direction (WT, V_1/2 = +53.1 ± 2.1 mV; Arg255Cys, V_1/2 = +16.3 ± 1.4 mV; Arg258Cys, V_1/2 = +43.1 ± 1.4 mV) (Fig. 3 and Table 1). Similarly, the neutralization of two acidic residues Glu201Cys (in S2) and Asp222Ala (in S3) shifted the G(V) in the hyperpolarized direction (Glu201Cys, V_1/2 = −47.8 ± 2.2 mV; Asp222Ala, V_1/2 = −58 ± 2.4 mV) (Fig. 3 A and B). One possible interpretation is that the acidic residues Glu201 and Asp222 stabilize S4 in the closed state by electrostatic interactions with the two basic residues Arg255 and Arg258 in S4. Therefore, we measured the activation of the double mutants Arg255Cys:Glu201Cys and Arg258Cys:Asp222Ala. In contrast to the leftward shifts in the G(V) relationships of the single-mutant channels, the double mutation Arg258Cys:Asp222Ala shifts the V_1/2 by ∼45 mV in the depolarized direction (right) relative to WT (Arg258Cys:Asp222Ala, V_1/2 = +100.8 ± 0.34 mV) (Fig. 3B) and the G(V) of the double mutant Arg255Cys:Glu201Cys is close to the G(V) of the single mutant Arg255Cys (Arg255Cys:Glu201Cys, V_1/2 = +10.1 ± 0.8 mV) (Fig. 3A). Many other mutations of these charged residues, e.g., Glu201Gln and Glu201Ala, did not generate functional channels. To rule out the possibility that the double cysteine mutation Arg255Cys:Glu201Cys causes any disulfide bond formation, we tested the effect of dithiothreitol (DTT) and tris(2-carboxyethyl) phosphine (TCEP) on the Arg255Cys:Glu201Cys mutant. DTT and TCEP did not induce any detectable changes in the G(V) relationship, suggesting that Arg255Cys and Glu201Cys do not form disulfide bonds (Fig. S3 A and B).

Fig. 3.

Electrostatic interactions are important for stabilizing the open and closed states of Hv1. (A–D) G(V) curves of WT (red), mutant 1 (black), mutant 2 (dark cyan), and the double mutant (blue). Mutant 1 and mutant 2: (A) Glu201Cys and Arg255Cys, (B) Asp222Ala and Arg258Cys, (C) Glu167Ala and Arg255Cys, and (D) Asp233Ala and Arg258Cys. (E) ΔG_coupling (kilocalories per mole) for 10 double-mutant residue pairs. Experimental values are shown as black bars and the values calculated from MD as gray bars. The blue bar is an experimental data estimated from ref. 13. Error bars, SEM.

View this table:

Table 1.

Thermodynamic mutant cycle analysis of electrostatic interactions

Thermodynamic Mutant Cycle Analysis Suggests Charge–Charge Interactions.

Our data indicates that mutations of Arg258 and Asp222 are clearly not independent. Both single mutations shift the G(V) in the negative direction, whereas the double mutation shifts the G(V) in the positive direction compared with the WT channel (Fig. 3B). To estimate the free-energy coupling between Arg255 or Arg258 and Glu201 or Asp222 quantitatively, we apply thermodynamic mutant cycle analysis (18, 19). Recent theoretical analysis shows that the free-energy differences between the open and closed states is more accurately extracted from the gating charge versus voltage curves than from G(V) curves (24). However, no gating currents have been measured from Hv1 channels, most likely due to the slow gating kinetics of Hv1 channels (15, 16). Therefore, we use G(V) curves to estimate the free-energy coupling between residues. Although the above approach assumes that the structure of the channel is not significantly changed by mutations and it is known that some mutations do interfere with the structure, several papers on double-mutant analysis involving charge neutralization in channels have proven very successful (18, 25, 26). The coupling energy between Arg258 and Glu201 is about 3.7 kcal/mol (Materials and Methods), whereas the interaction between Arg255 and Glu201 yields the slightly smaller value of 2.4 kcal/mol (Fig. 3E). These results suggest that Glu201 interacts with both Arg255 and Arg258 in the closed state, but more strongly with Arg258. The thermodynamic mutant cycle analysis on Asp222 with Arg258 yields similar results showing that Asp222 interacts with Arg258 (Fig. 3E). The signs of these interactions are such that these interactions tend to keep S4 from moving outward and thereby stabilize the channel in the closed state. Our previous cysteine accessibility studies showed that the three S4 arginines are in, or close to, the cytosolic solution in the closed state (15, 16). Because Glu201 and Asp222 are in the cytosolic half of S2 and S3 (Fig. 1B), we propose that Arg255 and Arg258 interact with Glu201 and Asp222 in the closed state. Neutralization of the third basic residue in S4, Arg261, shifts the G(V) in the depolarized direction compared with WT (Fig. S3E). The double mutant Arg261Cys:Glu201Cys similarly shifts the G(V) in the depolarized direction compared with Glu201Cys (Fig. S3E), as if there were no interaction between Glu201 and Arg261.

We also investigated whether there are any countercharge(s) that interact with Arg255 and Arg258 when S4 has moved outwards and the channel is open. The mutant cycle analysis of Arg258 showed weak coupling energies of 1.4 kcal/mol and 1 kcal/mol with the more extracellularly located Asp233 and Glu167 (Fig. 3D and Fig. S3F). The coupling energy between Glu167 and Arg255 is around 1.5 kcal/mol (Fig. 3 C and E). The interactions of Arg255 and Arg258 with Asp233 and Glu167 are such that they tend to stabilize the channel in the open state. The electrostatic interactions of Arg255 and Arg258 with negatively charged residues in the open state of the Hv1 channel are relatively weaker than those experienced in the closed state (Fig. 3E). Berger and Isacoff showed that an interaction between Asp112 and R3 in hHv1 (corresponding to Asp160 and Arg261 in Ci-Hv1) stabilizes the open state by around 3.5 kcal/mol (13).

The interactions determined by the thermodynamic mutant cycle analyses are qualitatively consistent with our molecular model of the closed and open states of Hv1 channels. To quantitatively compare our model with experimental observations, we computed coupling free energies of residues in the closed and open states of our Hv1 model, using a scheme developed earlier by Eriksson and Roux (27). We find that Glu201 and Asp222 interact with Arg255 and Arg258 in the closed state (Fig. 3E). However, the interaction between Glu201 and Arg258 is weaker than the others. The double-mutant cycle calculations on Glu167 and Asp233 with Arg255 and Arg258 show positive coupling energies (Fig. 3E), which suggests that those coupling interactions stabilize the open state. Meanwhile, there are no significant coupling interactions for Glu201 with Arg261 and Asp171 with Arg255. The calculated results from our model are similar to the experimental analysis of coupling interactions (Fig. 3E), supporting our model of the closed and open states of the Hv1 channel.

The Model Explains a State-Dependent Block by 2-GBI.

To further validate our model, we tested whether our open- and closed-state models could explain the state dependence of the binding of the guanidine analog 2-GBI from the cytosolic side of the channel (14). Previously, it has been shown that 2-GBI had no effect on closed hHv1 channels, suggesting that the binding site for 2-GBI is either at the gate or located extracellularly to the gate (14). Mutations of Phe198 in S2 were shown to affect the binding of 2-GBI (14). We docked 2-GBI to our open-state model of Hv1. The most populated cluster of docked 2-GBI is centered around Phe198, consistent with the experimental data (14). The most favorable orientations of 2-GBI allow for interactions between the positively charged guanidine moiety of 2-GBI and the negatively charged residues Glu201, Asp222, and Glu219, whereas the aromatic ring of 2-GBI is stabilized by residues in the proximity of Phe198 (Fig. 4 and Fig. S4). It is important to note that the state dependence of 2-GBI binding to Hv1 channels is easily explained in our model. In the closed-state model, Glu201, Asp222, and Glu219, are occupied by the guanidine moieties of Arg255 and Arg258 from S4 (Fig. 2B), thereby preventing 2-GBI access to its binding site. However, when the channel is in the open state, all of the S4 arginines move away from Glu201, Asp222, or Glu219, allowing the guanidine moiety of 2-GBI to bind to these acidic residues (Fig. 4). Previously, the state dependence of the binding of 2-GBI to Hv1 channels was interpreted as due to an intracellular gate in Hv1 channels that prevents access of 2-GBI to its binding site in the closed state (14). Our model suggests a different mechanism for the state dependence of 2-GBI binding: in the closed state, the binding site is occupied by the S4 charges and the hydrophobic gate is instead located extracellularly of the 2-GBI binding site. We conducted further experiments to test our proposed 2-GBI binding site. However, 2-GBI binding is coordinated by more than one amino acid and changing a single amino acid did not affect the binding affinity significantly. This is reasonable in light of the observation that in the simulations three residues appear to be competing for two sites on the guanidinuim moiety. Thus, removal of a single residue could easily be masked by improved interactions of 2-GBI with the remaining two residues. Unfortunately, changing two or more amino acids did not give functional channels when expressed exogenously, which is unsurprising because these amino acids are also important for channel gating and structure.

Fig. 4.

Simulation of 2-GBI binding in the open state. (A) The Hv1 channel in the open state with 2-GBI docked in a binding pocket near F198. (B and C) Magnified view of the binding pocket for 2-GBI; 2-GBI in Hv1 is localized at the intracellular part of the protein when the channel is open. The positively charged guanidine head group is in contact with the negatively charged residues in S2 and S3. The detailed contact information is shown in the Fig. S4.

A Tentative Gating Path from Targeted MD Simulations.

To identify a tentative mechanism of gating path between the closed and open states, we performed nine independent 100-ns long simulations using targeted MD (TMD) simulations (28). A similar strategy was used recently with considerable success to study conformational transitions in a number of secondary amino acid transporters and the Shaker K⁺ channel (29, 30). An obvious advantage of using TMD simulations with a rmsd-based reaction coordinate is the relative simplicity and bias-free selection of a collective variable that propagates the system along the path, e.g., no need to chose a specific distance or specific angle. Although it is very difficult to show rigorously that the rms constraint will accurately reproduce transition pathways for large conformational changes, within the context of VSD simulations it has been shown to reproduce results from long and unbiased simulations. Most notably Schow et al. (31) reported TMD application for studies of the gating of the bacterial voltage-gated potassium channel KvAP. TMD simulations propagating the VSD gating provided an impressive qualitative agreement with Jensen et al. (32) who ran unconstrained simulations on the channel under hyperpolarized conditions and observed a similar gating transition. This brute-force application of atomistic MD simulation involved upwards of 250-μs-long simulations. Furthermore, TMD was used by Mashl and Jakobsson (33) to confirm and explain the empirical observation that the highly conserved glycine of the S6 hinge region in a potassium channel is needed for the gating transition, which was later confirmed by Denning and Woolf (34) using an alternative transition-searching technique known as “dynamic importance” sampling. A similar conclusion was reached by Zhao and Noskov (35) who used a combination of transition path sampling (swarm-of-trajectories/string method) to investigate the gating dynamics of secondary transporters.

The gating transition from the closed to the open state in our Hv1 model occurs in two phases (Fig. 5 A–G and Movie S1). In the first phase, occurring at roughly 30 ns, the C-terminal end of S4 moves perpendicular to the plane of the membrane, largely independent of motions by the other TM regions. During this phase, S4 moves into place adjacent to S3 and forms new bonds. S4 also becomes somewhat tilted relative to the axis of the membrane. The S1–S2 pair conversely move inwards slightly as they begin to form new salt bridges with the S4 domain. In the second phase, occurring at rough 60 ns, S1–S2 and S3–S4 pairs both tilt slightly relative to the axis of the membrane. Key binding interactions between TM domains during these phases are shown in Fig. 5 A–F.

Fig. 5.

The interaction energy of selected salt bridges between the S1–S4 segments during gating. (A–F) Salt bridge binding of key residues during the TMD computation of the closed-to-open transition. The TMD was performed over a 100-ns run, which is not intended to mimic the actual time-scale of the transition, but rather intended to be slow enough to estimate the free energies associated with the transition pathway. (G) The position of the residue side chains of Asp160, Arg255, Arg258, and Arg261 along the z-axis during a 100-ns TMD computation of the closed-to-open transition. The center of the membrane was set to be 0 for the z-axis. (H) Cumulative gating charge (black) and individual gating charge (blue) transported across the membrane during the closed-to-open transition.

The salt bridging between S1 and S2 does not change throughout the closed-to-open transition (Fig. 5A), indicating that S1 and S2 move together throughout the process. S1–S3 interactions show a clear change at about 60 ns (Fig. 5B), corresponding to the coupled motion of the S1–S2 and S3–S4 pairs. The interactions between S1 and S4 are slightly more complex. During the first phase, the interaction between Asp160 and Arg255 becomes significantly stronger as S4 moves across the membrane and Arg255 moves toward and past Asp160 (Fig. 5C). Eventually, the interaction of Asp160 with Arg255 disappears. Instead, Asp160 interacts first with Arg258 and subsequently with Arg261 during the second phase. Around 60 ns, we see further formation of salt bridges between S1 and S4, as Glu167 binds to Arg255 and Arg258. Fig. 5G shows the position of Asp160 in S1 and Arg255, Arg258, and Arg261 in S4 relative to the center of the membrane during the transition from the closed state to the open state. During both the first and second phases, there are sudden changes in the positions of the S4 arginines. In contrast, Asp160 moves mainly during the first phase. The interactions between S2, S3, and S4, although complex, also reflect this two-step transition. Interactions between Arg211 and Glu219 in or near the loop between S2 and S3 break around 60 ns as the S3–S4 pair move (Fig. 5D). Instead, Glu219 in S3 forms a salt bridge with Lys205 near the center of S2 (Fig. 5D), stabilizing the angled conformation of the S1–S2 and S3–S4 pairs in the open state. In the closed state, Glu201 in S2 and Asp222 in S3 stabilize Arg255 and Arg258 in S4. However, these bridges are broken in the first phase and replaced with salt bridges between Glu201 and Asp222 with Arg271 in the C-terminal end of S4 that remains during the second phase (Fig. 5 E and F). After the salt bridges between Arg255 and Arg258 with Glu201 and Asp222 break in the first phase, Arg255 and Arg258 instead form interactions with Asp233 in the outer end of S3 (Fig. 5F).

The Computed Gating Charge Is Consistent with Experimental Data.

The gating charge is calculated from an ensemble of structures (5,000 for each state) sampled from evenly distributed points over the 100-ns runs for both the open and closed states. The gating charges are computed using the Poisson–Boltzmann voltage equation as previously described (36, 37). The transition of the closed-state monomer to the open-state monomer transports ∼1.9 e^– across the membrane per subunit. This result is in agreement with the estimates of 2- to 3-e^– effective gating charges per Hv1 subunit (15, 38). This gating charge is mainly mediated by the motion of the charged arginines in S4 across the membrane. Fig. 5H shows the contribution of Arg255, Arg258, and Arg261 to the cumulative gating charge. There are some minor contributions from other groups, such as Asp160, Glu167, and Lys173 (Fig. 5H).

Hydrophobic Clusters in the Closed State.

To understand how the channel gating results in a permeation pathway shutdown in the closed state, we first identify the changes in water occupancy in the channel between the open and closed states. The hydration profiles of the Hv1 channel indicate a region of dry present within the pore region both in the open and closed states (Fig. 6 A and B). The width of the dry area in the closed state is larger than that in the open state (Fig. 6B and Fig. S5A).

Fig. 6.

The hydrophobic plug and salt bridge network function as the gate(s) for the Hv1 channel. (A) Average number of waters in the open and closed states along the z-axis. (B) Possibility of being dry in the open and closed states along the z-axis. The graphs are based on a histogram analysis along the z-axis composed of bins of 3-Å width with z = 0 is set at residue F198. The averaging was done from the last 60 ns of a 100-ns simulation. The probability of being dry is the fraction of frames out of the total analyzed in which no waters were present in a given bin. (C) Side (Left) and top (Right) views of the two barriers (circles) in the closed states. (D) Residues that form the two barriers in the closed state are moved out of the center of the subunit in the open state. S4 is labeled magenta for reference.

In the closed state, there are two plugs in the center of the subunit that exclude water molecules (Fig. 6C and Fig. S5B). The first dry region is formed by a cluster of hydrophobic residues right below the narrowest part of the channel (Fig. 6C and Fig. S5B). Upon closing, hydrophobic groups of S1 (e.g., Val157) and S4 (e.g., Val252) move into the center of the subunit to form a hydrophobic plug that prevents water from penetrating between the TM segments (upper circle in Fig. 6C). Furthermore, upon closing, the aromatic side chain of Phe198 appears to rotate with the phenyl group lying more perpendicular to the primary axis of the channel (Fig. S6), thereby contributing to the hydrophobic plug. Interestingly, a second dry area is formed by a ring of charged residues located at the mouth of the pore on the cytosolic side (bottom circle in Fig. 6C). These residues are the same ones that form the strong salt bridges that stabilize the closed state (Fig. 2B). Upon closing, Glu201 forms interactions with Arg255 and Arg258 in S4 (Fig. 2B) that appears to pull S2 and S4 toward one another, causing the S1–S4 and S2–S3 pairs to move closer to each other. These bonds also pinch off a small cavity of water located between these bonds and the hydrophobic plug (Fig. S5B). These two dry areas are two barriers for proton permeation, which might function as the gate(s) of Hv1 channels in the closed state. These two barriers are broken apart once S4 moves out (Fig. 6D). In the open state, none of the three S4 charges is located close to Glu201 and most of the hydrophobic residues have moved out of the center of the subunit.

Water Wire Dynamics.

There has been a debate regarding the mechanism of proton transport through the central pore (3). One of the most contentious issues centers on whether there is a continuous water wire present in the open state, allowing water transport via a Grotthuss-type mechanism. A connectivity analysis of the hydration profiles does not seem to fully agree with the idea of a continuous water wire present for a long time across the open Hv1 channel. The longest-living continuous water wire existed for 200–300 ps at most in the course of our 100-ns simulation. To fully assure that the channel was fully hydrated, we considered only the latter 60 ns (the first continuous water wire forms around 40 ns) of the 100-ns run when analyzing hydration properties. The average lifetime of a continuous water wire in the open state was ∼6 ps, which was so short that it was difficult to count as the sole mechanism for proton permeation in Hv1 channels. Whereas our modeling does not rule out water wires completely, it suggests a possible mixture of mechanisms responsible for effective proton conductance.

In mixed mechanism proton transport, the protons are passed through a membrane protein via a combination of hydrogen-bonded chains formed by suitable positioning of hydrogen-bonding side groups of the protein residues (39) and water molecules in the channel. The water wire in this case does not have to be a continuous connection between both sides of the membrane. In our open-state model, there are protonatable residues in the dry area that can contribute to form a proton-hopping wire without a continuous water wire (Fig. 7B). Therefore, a continuous water wire does not seem indispensable for proton permeation. The average lifetimes of hydrogen bonds between individual water molecules in the closed and open channels are ∼2.8 and ∼2.5 ps, respectively, whereas several of the residues in the open state exhibit hydrogen bonds with water with 50-ps lifetimes (Fig. S7). The residues that putatively form a hydrogen bond chain in the open state are scattered about the intra- and extracellular part of the channel in the closed state (Fig. 7A), preventing these residues from forming a hydrogen-bonded chain in the closed state. It is important to note that in the closed state, the ionizable residues are located such that there are no alternative routes by which the channel could transport protons in the closed state. One should note that, in this paper, we have not directly studied the proton permeation mechanisms (which would require quantum mechanic simulations), but instead we focused on the modeling of gating transitions. The dynamics, formation, and even stability or water wires would be greatly perturbed by the presence of excess protons in the wire, as when proton permeation occurs.

Fig. 7.

Protonatable residues in the permeation pathway. Protonatable and deprotonatable residues in the closed (A) and open (B) states.

State-Dependent Protonation of Asp160 in the Selectivity Filter.

Asp160 was identified by Musset et al. as part of the selectivity filter in Hv1 channels (12). We evaluated the protonation of Asp160 in the closed and open states of our model, using a series of free-energy simulations. In the open state, Asp160 is likely to be ionized (ΔΔG⁰ = +4.4 ± 2.3 kcal/mol). In contrast, in the closed state the Asp160 residue is very likely to be protonated (ΔΔG⁰ = −10.6 ± 1.4 kcal/mol), which would render the side chain virtually unavailable for proton transport. This result is due in part to the change in salt-bridging interactions and local hydration. In the open state, Asp160 interacts with S4 arginines (Fig. 2A). In the closed state, Asp160 forms a weaker salt bridge with the Lys173 and becomes surrounded by five to seven water molecules, as it is exposed to the external aqueous crevice (Fig. S2C).

Discussion

In this work, we construct and test structural models of both the open and closed states of the Ci-Hv1 channel. Our results from thermodynamic mutant cycle analysis indicate that electrostatic interactions between charged residues stabilize the Hv1 channel in both the open and closed states. Using free-energy differences estimated from G(V) curves for thermodynamic mutagenesis cycle analysis has many caveats (24, 40, 41). However, the interactions we identified are consistent with the crystal structures and models of the VSD of homologous Kv-channels (20, 21) and are consistent with cysteine accessibility data of the Ci-Hv1 channel in the closed and open states (15, 16). In addition, our structural model and the thermodynamic mutagenesis cycle analysis are internally consistent (Fig. 3E).

Comparisons with Previous Studies.

The sequence similarity between Hv1 and the VSD of Kv1.2 is very limited, which makes it difficult to generate a homology model based only on sequence alignment, especially for the positioning of the three arginines in the S4 segment of Hv1 (10, 11, 22). Three other groups have proposed structural models of the open state of Hv1 channel. In 2010, Ramsey et al. (10) showed an aqueous H⁺ permeation pathway in an Hv1 model structure based on the KvAP and Kv1.2–2.1 chimera crystal structures (Fig. S8 A and B). However, cysteine accessibility data indicates that the S4 segment in the Hv1 channel does not move as far out into to extracellular space during channel opening as the S4 segment in the KvAP channel structure (21) (Fig. S8A). Therefore, the continuous aqueous pathway in this model might be caused by the farther outward positioning of the S4 segment. Wood et al. (11) suggested two different open-state models (R1-Hv1 and R2-Hv1) with different S4 alignments (Fig. S8 C and D). Our open-state model is closer to the R2-Hv1 model. However, in the R2-Hv1 model, the third arginine (Arg261 in Ci-Hv1) makes salt bridges with the intracellular acidic cluster in S2 and S3, whereas in our model Arg261 makes a salt bridge with Asp160 in the center of the channel (Fig. S8C). Our mutagenesis cycle data are not consistent with a salt bridge between Arg261 and the intracellular acidic cluster. In addition, Arg261 would, in this position, interfere with the binding of 2-GBI to the acidic cluster in the open state. The R1-Hv1 model by Wood et al. (11) is even more inconsistent with our data (Fig. S8D). Kulleperuma et al. (22) suggested two different open-state models (R2D and R3D), in which the second or third S4 arginine (Arg258 and Arg261 in Ci-Hv1) interact with Asp160 in the selectivity filter (Fig. S8 E and F). The R2D model is inconsistent with our data for the same reasons as the R2-Hv1 model from Wood et al. (11) (Fig. S8E). The R3D model is the most similar to our open-state model of all of the different Hv1 models. However, in the R3D model, Arg261 is above Asp160 and Arg258 does not interact with Asp160, in contrast to our model where both Arg258 and Arg261 interact with Asp160 (Fig. S8F). So our model is somewhat in between the R2D and R3D models of Kulleperuma et al. (22). It should be stressed that in addition to small differences in alignments chosen in these different studies, there are also many methodological differences, such as the modeling of side-chain packing and the combination of de novo ROSETTA-membrane and homology modeling in an iterative routine. Our approach (42, 43) was tested rigorously on models of voltage-gated potassium hERG channels (42) and secondary transporters (44) and was found to be consistent with independent experimental data.

It has been shown that all three charges in S4 contribute to the total gating charge in Hv1 channels (15, 16). By calculating the total gating charge movement from the closed to the open state considering all charged residues, we find that the three charges in S4 contribute the majority of the gating charge in Hv1 channels. This is in agreement with the observations of Khalili-Aragi et al. (37) who observed that the equivalent residues (R2, R3, and R4 in their notation) carried the majority of the gating charge in the VSD of the Shaker channel. The reason we observe additional small contributions from the charged residues in S1 is likely due to the additional charge of Asp160 and larger movements of S1 in Hv1 channels compared with Kv1.2 channels, which may be a result of either the absence of the pore domain in Hv1 channels or due to differences between the VSDs of Hv1 and Kv1.2 channels. We did not observe a rigid-body movement of an entire helical bundle, but rather a sequential rearrangement: the gating transition from closed to open is achieved by a translocation of S4 by ∼10–12 Å, followed by a coupled rotation of S3–S4 relative to S1–S2. During the closed-to-open transition, the three arginines of S4 crossed the hydrophobic plug and moved from the intracellular to the external aqueous environment. The gating charges are not exposed to the lipid phase during the gating transition, nor do they interact with the head groups of the lipids. We also see no evidence of a 3₁₀ helix in the S4 TM segment (Fig. S9). The gating mechanism partially resembles a combination of the revised paddle model (45), as the S3–S4 pair rotates together to form an angle with the S1–S2 pair, and the helical screw/sliding helix model (46), as Asp160 engages in sequential interactions with the S4 arginines. The proposed gating transition agrees with the conclusions of Pathak et al. (21) regarding the gating of the VSDs of Kv1.2.

Implications for the Proton Conduction Pathway.

The precise nature and location of the proton conduction pathway in Hv1 channels is still under debate. It was suggested in recent studies that this pathway is constructed by two water vestibules at the intra- and extracellular portion of the channel connected by a narrow bottleneck where the proton selectivity occurs (10, 12, 13). Musset et al. showed that the negatively charged residue Asp160 in S1 (Asp112 in hHv1) acts as a proton-selectivity filter (12). Berger and Isacoff suggested that the bottleneck is composed of the interaction of a negatively charged residue in the S1 (D112 in hHv1 and Asp160 in Ci-Hv1) and the third arginine in the S4 (Arg211 in hHv1 and Arg261 in Ci-Hv1) (13). Our MD simulations showed that both the closed- and open-state models contain a dry region, although a continuous water wire forms sporadically in the open state. In our model, the lifetime of the continuous water wire, even in the open state, is very short: a mean lifetime of 6 ps, with a maximum lifetime of 320 ps. In addition, a continuous water wire was only present ∼15% of time in the open-state model.

In the open state, water wires can form, but they are fairly rare and short lived. Instead, we propose that proton permeation occurs by a mixed mechanism, where both protonatable residues and water molecules form the permeation pathway in open Hv1 channels. Asp160 has been suggested to form one of these protonatable residues and to form part of the selectivity filter in Hv1 channels. Our simulations suggest that Asp160 is strongly protonated and not available for hydrogen bonding in the closed state, but that Asp160 is deprotonated and readily forms hydrogen bonds in the open state. At the narrowest part of the channel, the presence of a water molecule is so rare that this region is mainly in a dehydrated state. The absence of a continuous water wire suggests that nearby titratable residues contribute to the proton permeation through the narrowest part of the channel. Indeed, in our open-state model, many titratable residues are located in this region, thereby providing a plausible hydrogen bond chain for proton permeation through Hv1 channels in the open state. Although the formation of a water wire is rare and short lived, we cannot rule out a contribution from water wire conduction to proton permeation in Hv1 channels. Further quantum mechanical simulations are needed to determine the contributions of water wires and protonatable residues to the proton permeation mechanism in Hv1 channels.

Proposed Role of the Hydrophobic Plug as a Gate.

A cluster of hydrophobic residues in the VSD of voltage-gated potassium and sodium channels was earlier shown to be important for preventing ion leakage in the VSD of these channels (20, 25, 47 ⇓–49). In our closed-state model of Hv1 channels, there are two putative barriers to proton permeation in the intracellular part of proton channel: one is the hydrophobic plug formed by hydrophobic residues in S1 and S2, such as Phe198, and the other is the network of strong salt bridges formed by interactions of basic residues in S4 and acidic residues in S2 and S3 at the cytosolic end of the pore. We propose that the strong interactions in the salt bridge network contribute to keeping the TM segments close together such that the cluster of hydrophobic residues capped by Phe198 form a tight plug that prevents the flow of protons both by blocking the access of water to the central pore and by not providing any ionizable side chains in the central cavity. During membrane depolarization, the outward movement of S4 and the reorientation of the TM segments lead to the breakup of the hydrophobic plug, which is replaced by protonatable residues that can contribute to a hydrogen bond chain through narrowest part of the channel. The salt bridge network of S4 arginines interacting with the acidic residues in S2 and S3 in the closed state is replaced in the open state by weak contacts of Asn264 with these acidic residues. This may explain the small conductance of the Asn264Arg mutant (8, 50): if Asn264 is mutated to an arginine, the positive charge of Asn264Arg could interact with the negative charges in S2 and S3 to reform a strong salt bridge network at the cytosolic end of the channel that functions as a barrier in the open state of Asn264Arg Hv1 channels. In most other voltage-gated cation channels, there are more positive charges in the S4 (51), so that there is always an arginine interacting with the acidic residues in lower S2 and S3. These additional S4 charges may maintain this salt bridge network as a barrier and prevent proton leakage through the VSD in all states of S4 in other voltage-gated cation channels.

Conclusions

A combination of theoretical and experimental studies was used to develop and validate models of open and closed states of the Hv1, as well as to propose a gating-transition mechanism from over 1 μs of equilibrium and nonequilibrium MD simulations. We report a cross-validated structure of the closed state of the Hv1 channel. Our mutagenesis data and molecular model indicate that electrostatic interactions are important in stabilizing both the open and closed states of Hv1 channels. We find that the developed models offer excellent agreement with the thermodynamic mutant cycle analysis and gating charge measurements, as well as offer a structural mechanism for the state-dependent binding of 2-GBI. We propose that the intracellular charge network and the hydrophobic cluster form two barriers that function as the gate of the Hv1 channel. The free-energy simulations of ionization dynamics at the proposed selectivity filter (Asp160) showed that Asp160 is likely to be protonated in the closed state and that the propensity of Asp160 to retain a proton is state dependent.

Materials and Methods

Molecular Biology.

Mutations in Ci-Hv1 were introduced using QuikChange site-directed mutagenesis kit (Qiagen) and were fully sequenced. In vitro transcription of cRNA was performed using the mMessageMachineSP6 RNA Transcription Kit (Ambion). Fifty nanoliters of cRNA were injected into the Xenopus oocytes 2 d before measuring.

Electrophysiology.

We performed two-electrode voltage clamp (TEVC) recordings as described earlier (15). Solutions for TEVC contained in mM: 88 NaCl, 1 KCl, 1 MgCl₂, 1 CaCl₂, and 100 Hepes (pH 7.4). We injected oocytes with 50 nL of 1 M Hepes (pH 7.0) to minimize pH changes due to the proton currents. This results in ∼100 mM Hepes in the cytosol. The voltage dependence of channel activation G(V) was obtained by using fits of A2 + (A1 − A2)/(1 + exp((V – V_1/2)/k)) to G(V), where V_1/2 is the voltage at which there is half-maximum activation, and k is a slope factor equal to RT/zF; R is the gas constant, T is absolute temperature, z is the apparent gating charge, and F is Faraday’s number. Data were normalized between the A1 and A2 values of the fit.

Thermodynamic Mutant Cycle Analysis.

The amount of free energy required to shift the channel from the open to closed state was calculated as . The perturbation in free energy by a single residue mutation relative to the WT was calculated as . For each interaction pair, the coupling energy between the two residues was calculated as .

Molecular Modeling.

We created models of the closed and open state of Hv1 channels, using the structures (20) and models (21, 25, 52, 53) of the VSD of the Kv1.2 channel as templates and our experimental data as constraints for these homology models. The open state was modeled using Kv1.2–2.1 chimeric channel chain B (PDB ID code 2R9R) (20) as the template, and the closed state was modeled with several reported closed-state models (21). The modeling protocol was similar to that used in state-dependent modeling of hERG channel by Durdagi et al. (42).

MD Simulations.

The refined structures from iterative ROSETTA-membrane/homology modeling routines were equilibrated in a DMPC (1,2-dimyristoyl-sn-glycero-3-phosphocholine) lipid bilayer solvated by 150 mM of KCl. The TIP3P water model and Chemistry at Harvard Molecular Mechanics (CHARMM)–27 ion parameters were used for all MD simulations. The solvent–protein–membrane systems were built using the graphical user interface for Chemistry at Harvard Molecular Mechanics, CHARMM-GUI and coworkers (54). Most notably, the monomer was positioned such that the majority of the helices were centered in the membrane with the long axis of the monomer perpendicular to the membrane surface. Consequently, the C-terminal end of the S4 helix extends further into the cytosolic aqueous region than any other TM segments, resulting in a box with a wider water region than on the other. Each system was equilibrated at 303.15 K with NPaT ensemble for 10 ns using periodic boundary conditions in a tetragonal box 72 × 72 × 100 Å. This was followed with simulation runs of 100 ns. All MD simulations were performed with a program suite Not Just Another Molecular Dynamics program (NAMD) Version 2.9 (55). Subsequent analysis of the system was performed using CHARMM program suite (35b1r1) (56). Further details for analysis are discussed in SI Materials and Methods.

Acknowledgments

We thank Dr. Laura Perisinotti and Dr. Valentina Corradi for her help with docking. S.Y.N. was supported by operating funds from the Canadian Institutes of Health Research (MOP-186232) and Heart and Stroke Foundation of Alberta and NWT Grant-In-Aid funding. H.P.L. is funded by a grant from National Heart, Lung, and Blood Institute (R01-HL095920). S.Y.N. is an Alberta Heritage Foundation for Medical Research Scholar.

Footnotes

↵¹A.C. and F.Q. contributed equally to this work.
↵²To whom correspondence may be addressed. E-mail: PLarsson{at}med.miami.edu or snoskov{at}ucalgary.ca.

Author contributions: A.C., F.Q., S.Y.N., and H.P.L. designed research; A.C., F.Q., S.R., and Y.W. performed research; A.C., F.Q., S.R., Y.W., and S.Y.N. analyzed data; and A.C., F.Q., S.Y.N., and H.P.L. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1318018111/-/DCSupplemental.

View Abstract

References

↵
1. Henderson LM,
2. Chappell JB,
3. Jones OT
(1987) The superoxide-generating NADPH oxidase of human neutrophils is electrogenic and associated with an H+ channel. Biochem J 246(2):325–329.
OpenUrl PubMed
↵
1. DeCoursey TE,
2. Morgan D,
3. Cherny VV
(2003) The voltage dependence of NADPH oxidase reveals why phagocytes need proton channels. Nature 422(6931):531–534.
OpenUrl CrossRef PubMed
↵
1. Decoursey TE
(2003) Voltage-gated proton channels and other proton transfer pathways. Physiol Rev 83(2):475–579.
OpenUrl Abstract/FREE Full Text
↵
1. Lishko PV,
2. Botchkina IL,
3. Fedorenko A,
4. Kirichok Y
(2010) Acid extrusion from human spermatozoa is mediated by flagellar voltage-gated proton channel. Cell 140(3):327–337.
OpenUrl CrossRef PubMed
↵
1. Wu LJ,
2. et al.
(2012) The voltage-gated proton channel Hv1 enhances brain damage from ischemic stroke. Nat Neurosci 15(4):565–573.
OpenUrl CrossRef PubMed
↵
1. Ramsey IS,
2. Moran MM,
3. Chong JA,
4. Clapham DE
(2006) A voltage-gated proton-selective channel lacking the pore domain. Nature 440(7088):1213–1216.
OpenUrl CrossRef PubMed
↵
1. Sasaki M,
2. Takagi M,
3. Okamura Y
(2006) A voltage sensor-domain protein is a voltage-gated proton channel. Science 312(5773):589–592.
OpenUrl Abstract/FREE Full Text
↵
1. Tombola F,
2. Ulbrich MH,
3. Isacoff EY
(2008) The voltage-gated proton channel Hv1 has two pores, each controlled by one voltage sensor. Neuron 58(4):546–556.
OpenUrl CrossRef PubMed
↵
1. Koch HP,
2. et al.
(2008) Multimeric nature of voltage-gated proton channels. Proc Natl Acad Sci USA 105(26):9111–9116.
OpenUrl Abstract/FREE Full Text
↵
1. Ramsey IS,
2. et al.
(2010) An aqueous H+ permeation pathway in the voltage-gated proton channel Hv1. Nat Struct Mol Biol 17(7):869–875.
OpenUrl CrossRef PubMed
↵
1. Wood ML,
2. et al.
(2012) Water wires in atomistic models of the Hv1 proton channel. Biochim Biophys Acta 1818(2):286–293.
OpenUrl CrossRef PubMed
↵
1. Musset B,
2. et al.
(2011) Aspartate 112 is the selectivity filter of the human voltage-gated proton channel. Nature 480(7376):273–277.
OpenUrl CrossRef PubMed
↵
1. Berger TK,
2. Isacoff EY
(2011) The pore of the voltage-gated proton channel. Neuron 72(6):991–1000.
OpenUrl CrossRef PubMed
↵
1. Hong L,
2. Pathak MM,
3. Kim IH,
4. Ta D,
5. Tombola F
(2013) Voltage-sensing domain of voltage-gated proton channel Hv1 shares mechanism of block with pore domains. Neuron 77(2):274–287.
OpenUrl CrossRef PubMed
↵
1. Gonzalez C,
2. Koch HP,
3. Drum BM,
4. Larsson HP
(2010) Strong cooperativity between subunits in voltage-gated proton channels. Nat Struct Mol Biol 17(1):51–56.
OpenUrl CrossRef PubMed
↵
1. Gonzalez C,
2. Rebolledo S,
3. Perez ME,
4. Larsson HP
(2013) Molecular mechanism of voltage sensing in voltage-gated proton channels. J Gen Physiol 141(3):275–285.
OpenUrl Abstract/FREE Full Text
↵
1. Papazian DM,
2. et al.
(1995) Electrostatic interactions of S4 voltage sensor in Shaker K+ channel. Neuron 14(6):1293–1301.
OpenUrl CrossRef PubMed
↵
1. DeCaen PG,
2. Yarov-Yarovoy V,
3. Zhao Y,
4. Scheuer T,
5. Catterall WA
(2008) Disulfide locking a sodium channel voltage sensor reveals ion pair formation during activation. Proc Natl Acad Sci USA 105(39):15142–15147.
OpenUrl Abstract/FREE Full Text
↵
1. Ranganathan R,
2. Lewis JH,
3. MacKinnon R
(1996) Spatial localization of the K+ channel selectivity filter by mutant cycle-based structure analysis. Neuron 16(1):131–139.
OpenUrl CrossRef PubMed
↵
1. Long SB,
2. Tao X,
3. Campbell EB,
4. MacKinnon R
(2007) Atomic structure of a voltage-dependent K+ channel in a lipid membrane-like environment. Nature 450(7168):376–382.
OpenUrl CrossRef PubMed
↵
1. Pathak MM,
2. et al.
(2007) Closing in on the resting state of the Shaker K(+) channel. Neuron 56(1):124–140.
OpenUrl CrossRef PubMed
↵
1. Kulleperuma K,
2. et al.
(2013) Construction and validation of a homology model of the human voltage-gated proton channel hHV1. J Gen Physiol 141(4):445–465.
OpenUrl Abstract/FREE Full Text
↵
1. Smith SME,
2. et al.
(2011) Voltage-gated proton channel in a dinoflagellate. Proc Natl Acad Sci USA 108(44):18162–18167.
OpenUrl Abstract/FREE Full Text
↵
1. Chowdhury S,
2. Chanda B
(2013) Free-energy relationships in ion channels activated by voltage and ligand. J Gen Physiol 141(1):11–28.
OpenUrl CrossRef PubMed
↵
1. Yarov-Yarovoy V,
2. et al.
(2012) Structural basis for gating charge movement in the voltage sensor of a sodium channel. Proc Natl Acad Sci USA 109(2):E93–E102.
OpenUrl Abstract/FREE Full Text
↵
1. Paldi T,
2. Gurevitz M
(2010) Coupling between residues on S4 and S1 defines the voltage-sensor resting conformation in NaChBac. Biophys J 99(2):456–463.
OpenUrl CrossRef PubMed
↵
1. Eriksson MAL,
2. Roux B
(2002) Modeling the structure of agitoxin in complex with the Shaker K+ channel: A computational approach based on experimental distance restraints extracted from thermodynamic mutant cycles. Biophys J 83(5):2595–2609.
OpenUrl CrossRef PubMed
↵
1. Schlitter J,
2. Engels M,
3. Kruger P,
4. Jacoby E,
5. Wollmer A
(1993) Targeted molecular-dynamics simulation of conformational change - Application to the T[–]R transition in insulin. Mol Simul 10(2-6):291–308.
OpenUrl CrossRef
↵
1. Vargas E,
2. Bezanilla F,
3. Roux B
(2011) In search of a consensus model of the resting state of a voltage-sensing domain. Neuron 72(5):713–720.
OpenUrl CrossRef PubMed
↵
1. Zhao CF,
2. et al.
(2012) Ion-controlled conformational dynamics in the outward-open transition from an occluded state of LeuT. Biophys J 103(5):878–888.
OpenUrl CrossRef PubMed
↵
1. Schow EV,
2. Freites JA,
3. Nizkorodov A,
4. White SH,
5. Tobias DJ
(2012) Coupling between the voltage-sensing and pore domains in a voltage-gated potassium channel. Biochim Biophys Acta 1818(7):1726–1736.
OpenUrl
↵
1. Jensen MO,
2. et al.
(2012) Mechanism of voltage gating in potassium channels. Science 336(6078):229–233.
OpenUrl Abstract/FREE Full Text
↵
1. Mashl RJ,
2. Jakobsson E
(2008) End-point targeted molecular dynamics: Large-scale conformational changes in potassium channels. Biophys J 94(11):4307–4319.
OpenUrl CrossRef PubMed
↵
1. Denning EJ,
2. Woolf TB
(2010) Cooperative nature of gating transitions in K(+) channels as seen from dynamic importance sampling calculations. Proteins 78(5):1105–1119.
OpenUrl CrossRef PubMed
↵
1. Zhao C,
2. Noskov SY
(2013) The molecular mechanism of ion-dependent gating in secondary transporters. PLOS Comput Biol 9(10):e1003296.
OpenUrl CrossRef PubMed
↵
1. Jogini V,
2. Roux B
(2007) Dynamics of the Kv1.2 voltage-gated K+ channel in a membrane environment. Biophys J 93(9):3070–3082.
OpenUrl CrossRef PubMed
↵
1. Khalili-Araghi F,
2. et al.
(2010) Calculation of the gating charge for the Kv1.2 voltage-activated potassium channel. Biophys J 98(10):2189–2198.
OpenUrl CrossRef PubMed
↵
1. Fujiwara Y,
2. et al.
(2012) The cytoplasmic coiled-coil mediates cooperative gating temperature sensitivity in the voltage-gated H(+) channel Hv1. Nat Commun 3:816.
OpenUrl CrossRef PubMed
↵
1. Nagle JF,
2. Morowitz HJ
(1978) Molecular mechanisms for proton transport in membranes. Proc Natl Acad Sci USA 75(1):298–302.
OpenUrl Abstract/FREE Full Text
↵
1. Yifrach O,
2. MacKinnon R
(2002) Energetics of pore opening in a voltage-gated K(+) channel. Cell 111(2):231–239.
OpenUrl CrossRef PubMed
↵
1. Shem-Ad T,
2. Yifrach O
(2013) Using hierarchical thermodynamic linkage analysis to study ion channel gating. J Gen Physiol 141(4):507–510.
OpenUrl FREE Full Text
↵
1. Durdagi S,
2. Deshpande S,
3. Duff HJ,
4. Noskov SY
(2012) Modeling of open, closed, and open-inactivated states of the hERG1 channel: Structural mechanisms of the state-dependent drug binding. J Chem Inf Model 52(10):2760–2774.
OpenUrl CrossRef PubMed
↵
1. Subbotina J,
2. et al.
(2010) Structural refinement of the hERG1 pore and voltage-sensing domains with ROSETTA-membrane and molecular dynamics simulations. Proteins 78(14):2922–2934.
OpenUrl CrossRef PubMed
↵
1. Henry LK,
2. et al.
(2011) A conserved asparagine residue in transmembrane segment 1 (TM1) of serotonin transporter dictates chloride-coupled neurotransmitter transport. J Biol Chem 286(35):30823–30836.
OpenUrl Abstract/FREE Full Text
↵
1. Ruta V,
2. Chen JY,
3. MacKinnon R
(2005) Calibrated measurement of gating-charge arginine displacement in the KvAP voltage-dependent K+ channel. Cell 123(3):463–475.
OpenUrl CrossRef PubMed
↵
1. Vargas E,
2. et al.
(2012) An emerging consensus on voltage-dependent gating from computational modeling and molecular dynamics simulations. J Gen Physiol 140(6):587–594.
OpenUrl Abstract/FREE Full Text
↵
1. Gosselin-Badaroudine P,
2. Delemotte L,
3. Moreau A,
4. Klein ML,
5. Chahine M
(2012) Gating pore currents and the resting state of Na(v)1.4 voltage sensor domains. P Natl Acad Sci 109(47):19250–19255.
OpenUrl Abstract/FREE Full Text
↵
1. Tombola F,
2. Pathak MM,
3. Gorostiza P,
4. Isacoff EY
(2007) The twisted ion-permeation pathway of a resting voltage-sensing domain. Nature 445(7127):546–549.
OpenUrl CrossRef PubMed
↵
1. Campos FV,
2. Chanda B,
3. Roux B,
4. Bezanilla F
(2007) Two atomic constraints unambiguously position the S4 segment relative to S1 and S2 segments in the closed state of Shaker K channel. Proc Natl Acad Sci USA 104(19):7904–7909.
OpenUrl Abstract/FREE Full Text
↵
1. Sakata S,
2. et al.
(2010) Functionality of the voltage-gated proton channel truncated in S4. Proc Natl Acad Sci USA 107(5):2313–2318.
OpenUrl Abstract/FREE Full Text
↵
1. Bezanilla F
(2000) The voltage sensor in voltage-dependent ion channels. Physiol Rev 80(2):555–592.
OpenUrl Abstract/FREE Full Text
↵
1. Khalili-Araghi F,
2. Tajkhorshid E,
3. Roux B,
4. Schulten K
(2012) Molecular dynamics investigation of the ω-current in the Kv1.2 voltage sensor domains. Biophys J 102(2):258–267.
OpenUrl CrossRef PubMed
↵
1. Yarov-Yarovoy V,
2. Baker D,
3. Catterall WA
(2006) Voltage sensor conformations in the open and closed states in ROSETTA structural models of K(+) channels. Proc Natl Acad Sci USA 103(19):7292–7297.
OpenUrl Abstract/FREE Full Text
↵
1. Jo S,
2. Lim JB,
3. Klauda JB,
4. Im W
(2009) CHARMM-GUI Membrane Builder for mixed bilayers and its application to yeast membranes. Biophys J 97(1):50–58.
OpenUrl CrossRef PubMed
↵
1. Phillips JC,
2. et al.
(2005) Scalable molecular dynamics with NAMD. J Comput Chem 26(16):1781–1802.
OpenUrl CrossRef PubMed
↵
1. Brooks BR,
2. et al.
(2009) CHARMM: The biomolecular simulation program. J Comput Chem 30(10):1545–1614.
OpenUrl CrossRef PubMed

Article Alerts

Email Article

Citation Tools

Request Permissions

Proceedings of the National Academy of Sciences: 111 (2)

Table of Contents

Submit

Article Classifications

Biological Sciences
Neuroscience

[1] ↵
Henderson LM,
Chappell JB,
Jones OT
(1987) The superoxide-generating NADPH oxidase of human neutrophils is electrogenic and associated with an H+ channel. Biochem J 246(2):325–329.
OpenUrl PubMed

[2] Henderson LM,

[3] Chappell JB,

[4] Jones OT

[5] ↵
DeCoursey TE,
Morgan D,
Cherny VV
(2003) The voltage dependence of NADPH oxidase reveals why phagocytes need proton channels. Nature 422(6931):531–534.
OpenUrl CrossRef PubMed

[6] DeCoursey TE,

[7] Morgan D,

[8] Cherny VV

[9] ↵
Decoursey TE
(2003) Voltage-gated proton channels and other proton transfer pathways. Physiol Rev 83(2):475–579.
OpenUrl Abstract/FREE Full Text

[10] Decoursey TE

[11] ↵
Lishko PV,
Botchkina IL,
Fedorenko A,
Kirichok Y
(2010) Acid extrusion from human spermatozoa is mediated by flagellar voltage-gated proton channel. Cell 140(3):327–337.
OpenUrl CrossRef PubMed

[12] Lishko PV,

[13] Botchkina IL,

[14] Fedorenko A,

[15] Kirichok Y

[16] ↵
Wu LJ,
et al.
(2012) The voltage-gated proton channel Hv1 enhances brain damage from ischemic stroke. Nat Neurosci 15(4):565–573.
OpenUrl CrossRef PubMed

[17] Wu LJ,

[18] et al.

[19] ↵
Ramsey IS,
Moran MM,
Chong JA,
Clapham DE
(2006) A voltage-gated proton-selective channel lacking the pore domain. Nature 440(7088):1213–1216.
OpenUrl CrossRef PubMed

[20] Ramsey IS,

[21] Moran MM,

[22] Chong JA,

[23] Clapham DE

[24] ↵
Sasaki M,
Takagi M,
Okamura Y
(2006) A voltage sensor-domain protein is a voltage-gated proton channel. Science 312(5773):589–592.
OpenUrl Abstract/FREE Full Text

[25] Sasaki M,

[26] Takagi M,

[27] Okamura Y

[28] ↵
Tombola F,
Ulbrich MH,
Isacoff EY
(2008) The voltage-gated proton channel Hv1 has two pores, each controlled by one voltage sensor. Neuron 58(4):546–556.
OpenUrl CrossRef PubMed

[29] Tombola F,

[30] Ulbrich MH,

[31] Isacoff EY

[32] ↵
Koch HP,
et al.
(2008) Multimeric nature of voltage-gated proton channels. Proc Natl Acad Sci USA 105(26):9111–9116.
OpenUrl Abstract/FREE Full Text

[33] Koch HP,

[34] et al.

[35] ↵
Ramsey IS,
et al.
(2010) An aqueous H+ permeation pathway in the voltage-gated proton channel Hv1. Nat Struct Mol Biol 17(7):869–875.
OpenUrl CrossRef PubMed

[36] Ramsey IS,

[37] et al.

[38] ↵
Wood ML,
et al.
(2012) Water wires in atomistic models of the Hv1 proton channel. Biochim Biophys Acta 1818(2):286–293.
OpenUrl CrossRef PubMed

[39] Wood ML,

[40] et al.

[41] ↵
Musset B,
et al.
(2011) Aspartate 112 is the selectivity filter of the human voltage-gated proton channel. Nature 480(7376):273–277.
OpenUrl CrossRef PubMed

[42] Musset B,

[43] et al.

[44] ↵
Berger TK,
Isacoff EY
(2011) The pore of the voltage-gated proton channel. Neuron 72(6):991–1000.
OpenUrl CrossRef PubMed

[45] Berger TK,

[46] Isacoff EY

[47] ↵
Hong L,
Pathak MM,
Kim IH,
Ta D,
Tombola F
(2013) Voltage-sensing domain of voltage-gated proton channel Hv1 shares mechanism of block with pore domains. Neuron 77(2):274–287.
OpenUrl CrossRef PubMed

[48] Hong L,

[49] Pathak MM,

[50] Kim IH,

[51] Ta D,

[52] Tombola F

[53] ↵
Gonzalez C,
Koch HP,
Drum BM,
Larsson HP
(2010) Strong cooperativity between subunits in voltage-gated proton channels. Nat Struct Mol Biol 17(1):51–56.
OpenUrl CrossRef PubMed

[54] Gonzalez C,

[55] Koch HP,

[56] Drum BM,

[57] Larsson HP

[58] ↵
Gonzalez C,
Rebolledo S,
Perez ME,
Larsson HP
(2013) Molecular mechanism of voltage sensing in voltage-gated proton channels. J Gen Physiol 141(3):275–285.
OpenUrl Abstract/FREE Full Text

[59] Gonzalez C,

[60] Rebolledo S,

[61] Perez ME,

[62] Larsson HP

[63] ↵
Papazian DM,
et al.
(1995) Electrostatic interactions of S4 voltage sensor in Shaker K+ channel. Neuron 14(6):1293–1301.
OpenUrl CrossRef PubMed

[64] Papazian DM,

[65] et al.

[66] ↵
DeCaen PG,
Yarov-Yarovoy V,
Zhao Y,
Scheuer T,
Catterall WA
(2008) Disulfide locking a sodium channel voltage sensor reveals ion pair formation during activation. Proc Natl Acad Sci USA 105(39):15142–15147.
OpenUrl Abstract/FREE Full Text

[67] DeCaen PG,

[68] Yarov-Yarovoy V,

[69] Zhao Y,

[70] Scheuer T,

[71] Catterall WA

[72] ↵
Ranganathan R,
Lewis JH,
MacKinnon R
(1996) Spatial localization of the K+ channel selectivity filter by mutant cycle-based structure analysis. Neuron 16(1):131–139.
OpenUrl CrossRef PubMed

[73] Ranganathan R,

[74] Lewis JH,

[75] MacKinnon R

[76] ↵
Long SB,
Tao X,
Campbell EB,
MacKinnon R
(2007) Atomic structure of a voltage-dependent K+ channel in a lipid membrane-like environment. Nature 450(7168):376–382.
OpenUrl CrossRef PubMed

[77] Long SB,

[78] Tao X,

[79] Campbell EB,

[80] MacKinnon R

[81] ↵
Pathak MM,
et al.
(2007) Closing in on the resting state of the Shaker K(+) channel. Neuron 56(1):124–140.
OpenUrl CrossRef PubMed

[82] Pathak MM,

[83] et al.

[84] ↵
Kulleperuma K,
et al.
(2013) Construction and validation of a homology model of the human voltage-gated proton channel hHV1. J Gen Physiol 141(4):445–465.
OpenUrl Abstract/FREE Full Text

[85] Kulleperuma K,

[86] et al.

[87] ↵
Smith SME,
et al.
(2011) Voltage-gated proton channel in a dinoflagellate. Proc Natl Acad Sci USA 108(44):18162–18167.
OpenUrl Abstract/FREE Full Text

[88] Smith SME,

[89] et al.

[90] ↵
Chowdhury S,
Chanda B
(2013) Free-energy relationships in ion channels activated by voltage and ligand. J Gen Physiol 141(1):11–28.
OpenUrl CrossRef PubMed

[91] Chowdhury S,

[92] Chanda B

[93] ↵
Yarov-Yarovoy V,
et al.
(2012) Structural basis for gating charge movement in the voltage sensor of a sodium channel. Proc Natl Acad Sci USA 109(2):E93–E102.
OpenUrl Abstract/FREE Full Text

[94] Yarov-Yarovoy V,

[95] et al.

[96] ↵
Paldi T,
Gurevitz M
(2010) Coupling between residues on S4 and S1 defines the voltage-sensor resting conformation in NaChBac. Biophys J 99(2):456–463.
OpenUrl CrossRef PubMed

[97] Paldi T,

[98] Gurevitz M

[99] ↵
Eriksson MAL,
Roux B
(2002) Modeling the structure of agitoxin in complex with the Shaker K+ channel: A computational approach based on experimental distance restraints extracted from thermodynamic mutant cycles. Biophys J 83(5):2595–2609.
OpenUrl CrossRef PubMed

[100] Eriksson MAL,

[101] Roux B

[102] ↵
Schlitter J,
Engels M,
Kruger P,
Jacoby E,
Wollmer A
(1993) Targeted molecular-dynamics simulation of conformational change - Application to the T[–]R transition in insulin. Mol Simul 10(2-6):291–308.
OpenUrl CrossRef

[103] Schlitter J,

[104] Engels M,

[105] Kruger P,

[106] Jacoby E,

[107] Wollmer A

[108] ↵
Vargas E,
Bezanilla F,
Roux B
(2011) In search of a consensus model of the resting state of a voltage-sensing domain. Neuron 72(5):713–720.
OpenUrl CrossRef PubMed

[109] Vargas E,

[110] Bezanilla F,

[111] Roux B

[112] ↵
Zhao CF,
et al.
(2012) Ion-controlled conformational dynamics in the outward-open transition from an occluded state of LeuT. Biophys J 103(5):878–888.
OpenUrl CrossRef PubMed

[113] Zhao CF,

[114] et al.

[115] ↵
Schow EV,
Freites JA,
Nizkorodov A,
White SH,
Tobias DJ
(2012) Coupling between the voltage-sensing and pore domains in a voltage-gated potassium channel. Biochim Biophys Acta 1818(7):1726–1736.
OpenUrl

[116] Schow EV,

[117] Freites JA,

[118] Nizkorodov A,

[119] White SH,

[120] Tobias DJ

[121] ↵
Jensen MO,
et al.
(2012) Mechanism of voltage gating in potassium channels. Science 336(6078):229–233.
OpenUrl Abstract/FREE Full Text

[122] Jensen MO,

[123] et al.

[124] ↵
Mashl RJ,
Jakobsson E
(2008) End-point targeted molecular dynamics: Large-scale conformational changes in potassium channels. Biophys J 94(11):4307–4319.
OpenUrl CrossRef PubMed

[125] Mashl RJ,

[126] Jakobsson E

[127] ↵
Denning EJ,
Woolf TB
(2010) Cooperative nature of gating transitions in K(+) channels as seen from dynamic importance sampling calculations. Proteins 78(5):1105–1119.
OpenUrl CrossRef PubMed

[128] Denning EJ,

[129] Woolf TB

[130] ↵
Zhao C,
Noskov SY
(2013) The molecular mechanism of ion-dependent gating in secondary transporters. PLOS Comput Biol 9(10):e1003296.
OpenUrl CrossRef PubMed

[131] Zhao C,

[132] Noskov SY

[133] ↵
Jogini V,
Roux B
(2007) Dynamics of the Kv1.2 voltage-gated K+ channel in a membrane environment. Biophys J 93(9):3070–3082.
OpenUrl CrossRef PubMed

[134] Jogini V,

[135] Roux B

[136] ↵
Khalili-Araghi F,
et al.
(2010) Calculation of the gating charge for the Kv1.2 voltage-activated potassium channel. Biophys J 98(10):2189–2198.
OpenUrl CrossRef PubMed

[137] Khalili-Araghi F,

[138] et al.

[139] ↵
Fujiwara Y,
et al.
(2012) The cytoplasmic coiled-coil mediates cooperative gating temperature sensitivity in the voltage-gated H(+) channel Hv1. Nat Commun 3:816.
OpenUrl CrossRef PubMed

[140] Fujiwara Y,

[141] et al.

[142] ↵
Nagle JF,
Morowitz HJ
(1978) Molecular mechanisms for proton transport in membranes. Proc Natl Acad Sci USA 75(1):298–302.
OpenUrl Abstract/FREE Full Text

[143] Nagle JF,

[144] Morowitz HJ

[145] ↵
Yifrach O,
MacKinnon R
(2002) Energetics of pore opening in a voltage-gated K(+) channel. Cell 111(2):231–239.
OpenUrl CrossRef PubMed

[146] Yifrach O,

[147] MacKinnon R

[148] ↵
Shem-Ad T,
Yifrach O
(2013) Using hierarchical thermodynamic linkage analysis to study ion channel gating. J Gen Physiol 141(4):507–510.
OpenUrl FREE Full Text

[149] Shem-Ad T,

[150] Yifrach O

[151] ↵
Durdagi S,
Deshpande S,
Duff HJ,
Noskov SY
(2012) Modeling of open, closed, and open-inactivated states of the hERG1 channel: Structural mechanisms of the state-dependent drug binding. J Chem Inf Model 52(10):2760–2774.
OpenUrl CrossRef PubMed

[152] Durdagi S,

[153] Deshpande S,

[154] Duff HJ,

[155] Noskov SY

[156] ↵
Subbotina J,
et al.
(2010) Structural refinement of the hERG1 pore and voltage-sensing domains with ROSETTA-membrane and molecular dynamics simulations. Proteins 78(14):2922–2934.
OpenUrl CrossRef PubMed

[157] Subbotina J,

[158] et al.

[159] ↵
Henry LK,
et al.
(2011) A conserved asparagine residue in transmembrane segment 1 (TM1) of serotonin transporter dictates chloride-coupled neurotransmitter transport. J Biol Chem 286(35):30823–30836.
OpenUrl Abstract/FREE Full Text

[160] Henry LK,

[161] et al.

[162] ↵
Ruta V,
Chen JY,
MacKinnon R
(2005) Calibrated measurement of gating-charge arginine displacement in the KvAP voltage-dependent K+ channel. Cell 123(3):463–475.
OpenUrl CrossRef PubMed

[163] Ruta V,

[164] Chen JY,

[165] MacKinnon R

[166] ↵
Vargas E,
et al.
(2012) An emerging consensus on voltage-dependent gating from computational modeling and molecular dynamics simulations. J Gen Physiol 140(6):587–594.
OpenUrl Abstract/FREE Full Text

[167] Vargas E,

[168] et al.

[169] ↵
Gosselin-Badaroudine P,
Delemotte L,
Moreau A,
Klein ML,
Chahine M
(2012) Gating pore currents and the resting state of Na(v)1.4 voltage sensor domains. P Natl Acad Sci 109(47):19250–19255.
OpenUrl Abstract/FREE Full Text

[170] Gosselin-Badaroudine P,

[171] Delemotte L,

[172] Moreau A,

[173] Klein ML,

[174] Chahine M

[175] ↵
Tombola F,
Pathak MM,
Gorostiza P,
Isacoff EY
(2007) The twisted ion-permeation pathway of a resting voltage-sensing domain. Nature 445(7127):546–549.
OpenUrl CrossRef PubMed

[176] Tombola F,

[177] Pathak MM,

[178] Gorostiza P,

[179] Isacoff EY

[180] ↵
Campos FV,
Chanda B,
Roux B,
Bezanilla F
(2007) Two atomic constraints unambiguously position the S4 segment relative to S1 and S2 segments in the closed state of Shaker K channel. Proc Natl Acad Sci USA 104(19):7904–7909.
OpenUrl Abstract/FREE Full Text

[181] Campos FV,

[182] Chanda B,

[183] Roux B,

[184] Bezanilla F

[185] ↵
Sakata S,
et al.
(2010) Functionality of the voltage-gated proton channel truncated in S4. Proc Natl Acad Sci USA 107(5):2313–2318.
OpenUrl Abstract/FREE Full Text

[186] Sakata S,

[187] et al.

[188] ↵
Bezanilla F
(2000) The voltage sensor in voltage-dependent ion channels. Physiol Rev 80(2):555–592.
OpenUrl Abstract/FREE Full Text

[189] Bezanilla F

[190] ↵
Khalili-Araghi F,
Tajkhorshid E,
Roux B,
Schulten K
(2012) Molecular dynamics investigation of the ω-current in the Kv1.2 voltage sensor domains. Biophys J 102(2):258–267.
OpenUrl CrossRef PubMed

[191] Khalili-Araghi F,

[192] Tajkhorshid E,

[193] Roux B,

[194] Schulten K

[195] ↵
Yarov-Yarovoy V,
Baker D,
Catterall WA
(2006) Voltage sensor conformations in the open and closed states in ROSETTA structural models of K(+) channels. Proc Natl Acad Sci USA 103(19):7292–7297.
OpenUrl Abstract/FREE Full Text

[196] Yarov-Yarovoy V,

[197] Baker D,

[198] Catterall WA

[199] ↵
Jo S,
Lim JB,
Klauda JB,
Im W
(2009) CHARMM-GUI Membrane Builder for mixed bilayers and its application to yeast membranes. Biophys J 97(1):50–58.
OpenUrl CrossRef PubMed

[200] Jo S,

[201] Lim JB,

[202] Klauda JB,

[203] Im W

[204] ↵
Phillips JC,
et al.
(2005) Scalable molecular dynamics with NAMD. J Comput Chem 26(16):1781–1802.
OpenUrl CrossRef PubMed

[205] Phillips JC,

[206] et al.

[207] ↵
Brooks BR,
et al.
(2009) CHARMM: The biomolecular simulation program. J Comput Chem 30(10):1545–1614.
OpenUrl CrossRef PubMed

[208] Brooks BR,

[209] et al.

Main menu

User menu

Search

New Research In

Hydrophobic plug functions as a gate in voltage-gated proton channels

Significance

Abstract

Results

Generation of Open- and Closed-State Models.

Shifts in Voltage Dependence of Activation by Charge Neutralizations.

Thermodynamic Mutant Cycle Analysis Suggests Charge–Charge Interactions.

The Model Explains a State-Dependent Block by 2-GBI.

A Tentative Gating Path from Targeted MD Simulations.

The Computed Gating Charge Is Consistent with Experimental Data.

Hydrophobic Clusters in the Closed State.

Water Wire Dynamics.

State-Dependent Protonation of Asp160 in the Selectivity Filter.

Discussion

Comparisons with Previous Studies.

Implications for the Proton Conduction Pathway.

Proposed Role of the Hydrophobic Plug as a Gate.

Conclusions

Materials and Methods

Molecular Biology.

Electrophysiology.

Thermodynamic Mutant Cycle Analysis.

Molecular Modeling.

MD Simulations.

Acknowledgments

Footnotes

References

Citation Manager Formats

Article Classifications

You May Also be Interested in

Similar Articles

Articles

PNAS Portals

Information

Main menu

User menu

Search

New Research In

Physical Sciences

Featured Portals

Articles by Topic

Social Sciences

Featured Portals

Articles by Topic

Biological Sciences

Featured Portals

Articles by Topic

Hydrophobic plug functions as a gate in voltage-gated proton channels

Significance

Abstract

Results

Generation of Open- and Closed-State Models.

Shifts in Voltage Dependence of Activation by Charge Neutralizations.

Thermodynamic Mutant Cycle Analysis Suggests Charge–Charge Interactions.

The Model Explains a State-Dependent Block by 2-GBI.

A Tentative Gating Path from Targeted MD Simulations.

The Computed Gating Charge Is Consistent with Experimental Data.

Hydrophobic Clusters in the Closed State.

Water Wire Dynamics.

State-Dependent Protonation of Asp160 in the Selectivity Filter.

Discussion

Comparisons with Previous Studies.

Implications for the Proton Conduction Pathway.

Proposed Role of the Hydrophobic Plug as a Gate.

Conclusions

Materials and Methods

Molecular Biology.

Electrophysiology.

Thermodynamic Mutant Cycle Analysis.

Molecular Modeling.

MD Simulations.

Acknowledgments

Footnotes

References

Citation Manager Formats

Sign up for Article Alerts

Article Classifications

Jump to section

You May Also be Interested in

Similar Articles

Articles

PNAS Portals

Information