Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2017 May 23;114(23):5924–5929. doi: 10.1073/pnas.1703654114

Understanding the essential proton-pumping kinetic gates and decoupling mutations in cytochrome c oxidase

Ruibin Liang a, Jessica M J Swanson a,1, Mårten Wikström b, Gregory A Voth a,1
PMCID: PMC5468613  PMID: 28536198

Significance

Understanding the molecular mechanism that enables the high proton-pumping efficiency of cytochrome c oxidase (CcO) is not only fundamentally important but also potentially useful in providing design principles for bio-inspired materials. To this end, much effort has been devoted to the D-channel mutants of CcO that impair proton pumping. In this study, we have used multiscale reactive molecular dynamics simulations to explicitly characterize the free-energy profiles of proton transport (PT) through the D-channel for several important mutants. Our results show that both the fast forward rate and the slow backward rate of PT in the D-channel are important kinetic gating elements to achieve high proton-pumping efficiency, the latter of which has been largely overlooked to date.

Keywords: proton pump, decoupling mutants, cytochrome c oxidase, proton transport, multiscale

Abstract

Cytochrome c oxidase (CcO) catalyzes the reduction of oxygen to water and uses the released free energy to pump protons against the transmembrane proton gradient. To better understand the proton-pumping mechanism of the wild-type (WT) CcO, much attention has been given to the mutation of amino acid residues along the proton translocating D-channel that impair, and sometimes decouple, proton pumping from the chemical catalysis. Although their influence has been clearly demonstrated experimentally, the underlying molecular mechanisms of these mutants remain unknown. In this work, we report multiscale reactive molecular dynamics simulations that characterize the free-energy profiles of explicit proton transport through several important D-channel mutants. Our results elucidate the mechanisms by which proton pumping is impaired, thus revealing key kinetic gating features in CcO. In the N139T and N139C mutants, proton back leakage through the D-channel is kinetically favored over proton pumping due to the loss of a kinetic gate in the N139 region. In the N139L mutant, the bulky L139 side chain inhibits timely reprotonation of E286 through the D-channel, which impairs both proton pumping and the chemical reaction. In the S200V/S201V double mutant, the proton affinity of E286 is increased, which slows down both proton pumping and the chemical catalysis. This work thus not only provides insight into the decoupling mechanisms of CcO mutants, but also explains how kinetic gating in the D-channel is imperative to achieving high proton-pumping efficiency in the WT CcO.

Cytochrome c oxidase (CcO) is the terminal enzyme of the respiratory electron transfer chain in the inner membrane of mitochondria and the plasma membrane of bacteria. It catalyzes the oxidation of cytochrome c molecules and reduction of O2 to H2O. For the aa3-type CcO, found in mitochondria and many bacteria, the free energy available from the chemical reaction is used to pump one proton across the membrane per electron transferred to O2, generating the transmembrane electrochemical proton gradient necessary for ATP synthesis (1–3). Thus, four “substrate” protons are taken up from the mitochondrial matrix or bacterial cytoplasmic side (negatively charged; N-side) of the membrane and consumed in the reduction of O2, whereas another four “pumped” protons are transported to the intermembrane (or extracellular in bacteria) space (positively charged; P-side) of the membrane (Fig. 1). All four pumped protons and at least two of the substrate protons are taken up from solution on the N-side of the membrane by the D-channel, which begins with amino acid residue D132 and ends at residue E286 two-thirds of the way into the protein (4, 5). The remaining substrate protons are taken up by the K-channel (not shown in Fig. 1, but extensively discussed in ref. 6). Next, the protons are either transferred to the binuclear center (BNC) to react with O2 or to a pump loading site (PLS) to be pumped to the solution on the P-side of the membrane. Both of these proton transport (PT) pathways are influenced by electron transfer (ET) from heme a to the BNC (7, 8).

Fig. 1.

Fig. 1.

Open in a new tab

(A) Overview of the PT pathways (blue arrows) and ET (red arrow) in CcO, highlighting the heme groups and important residues in licorice color; D132, E286, PRDa3, and the PLS in yellow; and the BNC CuB in orange. (B) Reaction scheme during the PR → F transition with labeled redox groups and protonatable groups. Arrows show the PT processes involved in the chemical reaction (blue), pumping (black), and the E286 reprotonation (blue). The solid arrows indicate PT events that actually happen. The dashed gray arrows indicate potential proton back-leakage pathways that would decouple pumping from the chemical reaction.

To understand the molecular mechanism that enables the high proton-pumping efficiency of CcO, much effort has been devoted to the studies of the so-called “decoupling mutants” of CcO. The decoupling mutants maintain the O2 reduction reaction [often at the same rate as the wild-type (WT) CcO], but fail to couple the chemical reaction to proton pumping (9–20). Mutagenesis studies have further revealed several residues in the D-channel that are essential for efficient proton pumping. One of them is the amino acid residue N139, which forms double hydrogen bonds with the residue N121 and interrupts the network of hydrogen-bonded water molecules in the D-channel (Fig. 1 and Fig. S1A). It is the mutation of N139 (into various amino acid alternatives) that decouples proton pumping from chemical reaction most often (9–11, 14–18). However, how the mutation of N139 influences the key pumping and chemical reaction PT steps, which occur above E286 more than 25 Å away (Fig. 1), is unknown. It is speculated that replacement with a charged residue, such as N139D, induces decoupling by shifting the pKa of the E286 via long-range electrostatic interactions (e.g., with the negatively charged D139 residue) (20, 21). However, proton pumping can also be decoupled by the replacement of N139 with charge-neutral residues (e.g., N139T, N139C, and N139L), suggesting a more nuanced decoupling mechanism for these mutants.

Fig. S1.

Fig. S1.

Open in a new tab

(A–D) The structure near residues X139 (where X is the amino acid mutation), D132, and N121 in the (A) WT, (B) N139T, (C) N139C, and (D) N139L mutant of CcO. D132 is shown in green and the excess proton in purple. The N139, T139, C139, L139, and N121 residues and water molecules are shown as sticks. The mutant structures are taken from umbrella-sampling windows that have protonated D132 and are on the minimum free-energy pathway of the 2D-PMFs in Fig. 2. The WT structure is similarly taken from the previous work (27).

Of course, understanding the impaired proton-pumping mechanism of D-channel mutants at a molecular level will add to our understanding of the proton-pumping mechanism in the WT CcO. To this end, experimental results can be complemented with molecular-level insight from computer simulations. However, it is challenging to simulate PT in large and complicated biomolecular systems, because it requires an explicit treatment of the charge delocalization and Grotthuss proton shuttling (22) between water molecules and sometimes amino acids associated with the hydrated excess proton. To overcome this challenge, a multiscale reactive molecular dynamics (MS-RMD) method has been extensively developed and applied in our group to study the explicit PT process in aqueous and biological contexts, including CcO (e.g., refs. 23–31).

In this work, we have carried out extensive MS-RMD free-energy simulations to investigate the decoupling mechanism in the charge-neutral mutants N139T, N139C, N139L and S200V/S201V. Our results indicate that the N139T and N139C mutants reduce the proton back-leakage barrier in the D-channel and facilitate the proton back leakage to the N-side bulk, which decouples the proton pumping from the chemical reaction. This is particularly interesting because the importance of the D-channel back-leakage barrier has been largely overlooked in discussions of WT CcO’s proton-pumping mechanism. In contrast, the N139L mutation inhibits the forward PT through the D-channel, which prevents timely reprotonation of E286 and thus eliminates both the proton pumping and the chemical reaction. In the S200V/S201V mutant, the proton affinity of E286 is increased, slowing down both proton pumping and the chemical reaction. Our present work thus provides insights into the impaired proton-pumping mechanism of several important D-channel mutants of CcO, reveals key features of the kinetic gating, and complements our recent findings on the overall kinetic steps in WT CcO (27).

Results and Discussion

Before discussing the results for the mutants, we first briefly summarize our recent conclusions (27) regarding the sequence of the main PT and ET events during the A → PR → F transitions for the WT enzyme. First, ET from heme a to the BNC coupled with PT from the E286 to the PLS leads to the A → PR transition. Second, E286 is rapidly reprotonated through the D-channel in the PR state. Third, PT from E286 to the BNC leads to the PR → F transition. Fourth, E286 is reprotonated for a second time through the D-channel, accompanied by proton release from the PLS to the P-side of the membrane and partial ET from CuA to heme a. Transfer of both the pumped and chemical protons from E286 was found to be thermodynamically driven by ET from heme a to the BNC. However, transfer of the pumped proton (E286 to PLS) is kinetically favored whereas transfer of the chemical proton (E286 to the BNC) is rate limiting for the PR → F transition. This constitutes one kinetic gate that is essential for pumping. A second kinetic gate is the rapid reprotonation of E286 that precedes proton backflow from the PLS to E286 in the PR state. The third kinetic gate, which is extensively discussed below, is the large backflow barrier from E286 to D132 through the D-channel.

N139T and N139C Mutants Increase Proton Back Leakage Through the D-channel.

The 2D free-energy profiles (2D-PMFs) for explicit PT in the D-channel were calculated for the N139T and N139C mutants (Fig. 2 B and C). These 2D-PMFs track both the position of the net positive charge defect associated with the excess proton [the “center of excess charge” (CEC)] (Materials and Methods) and the degree of hydration around residue 139. Due to the dynamic interplay between the charge translocation via excess proton shuttling and the intervening water molecules and amino acids, such 2D-PMFs have recently been shown to be essential for understanding the PT in CcO and in other proteins (27, 30).

Fig. 2.

Fig. 2.

Open in a new tab

(A–E) The 2D-PMFs (free-energy surfaces) for PT from D132 to E286 in the D-channel of (A) WT, (B) N139T, (C) N139C, (D) N139L, and (E) S200V/S201V. The PMFs are calculated as a function of the excess proton charge defect CEC and the water density along the PT pathway. The 2D-PMF for WT was calculated for the F state and was published in ref. 27. The 2D-PMFs for the N139T and N139C mutants were calculated for the F state, whereas those for the N139L and S200V/S201V mutants were calculated for the PR state (as explained in text). The 1D minimum free-energy pathways are depicted as black curves.

The one-dimensional free-energy curves along the minimum free-energy pathways traced out from the 2D-PMFs in Fig. 2 B and C are plotted in Fig. 3. The barriers for proton back leakage from E286 to D132 in the N139T and N139C mutants (11.5 ± 0.5 kcal/mol and 12.7 ± 0.3 kcal/mol, respectively) are smaller than that for WT (16.6 ± 0.6 kcal/mol). The calculated rates (shown as time constants in Table 1) indicate that the proton back leakage through the D-channel is thus faster in the mutants. The proton-pumping barrier from the PLS to the P-side bulk after water formation at the BNC (F state) was estimated to be ∼14 kcal/mol with a time constant of 2.6 ms, in the O → E transition (32, 33). In the PR → F transition, the proton release to P-side bulk in D2O is slowed down to ∼1 ms with a kinetic isotope effect of ∼7 (34). Based on this we estimate that the time constant for proton release from the PLS to the P-side is ∼140 μs during PR → F transition. Therefore, in WT CcO proton pumping is faster than proton back leakage to the N-side bulk (Table 1). For the N139T and N139C mutants, however, proton pumping is now slower than proton back leakage to the N-side bulk (Table 1). This is based on the assumption that the forward proton-pumping barrier (PLS to P-side bulk) is not changed by the mutants, which is a reasonable assumption given the long distance between the mutation site and the proton release pathway. These results suggest that it is the loss of a kinetic gate at N139 that eliminates the proton pumping in these mutants. In the presence of a full transmembrane electrochemical gradient of 0.2 eV [assuming the PLS, E286, and N139 are roughly one-sixth, one-half, and five-sixths of the full membrane width from the P-side bulk, respectively, and following the analysis of Siegbahn and Blomberg (35)] the back-leakage barrier in the D-channel is decreased by ∼1.5 kcal/mol and the pumping barrier is increased by ∼0.8 kcal/mol. Thus, the modified free-energy barriers in the presence of a full electrochemical gradient will further facilitate proton back leakage in the mutants, but in the WT the back-leakage barrier is still slightly higher than the pumping barrier; thus, WT prevents back leakage from happening even in the presence of a transmembrane electrochemical gradient.

Fig. 3.

Fig. 3.

Open in a new tab

The 1D free-energy traces for PT through the D-channel of WT (solid and dashed black curves), N139T (blue curve), N139C (purple curve), N139L (red curve), and S200V/S201V (yellow curve), along the minimum free-energy pathways of the 2D-PMFs in Fig. 2. The free-energy traces for the N139T and N139C mutants are calculated for the F state, whereas those for the N139L and S200V/S201V mutants are calculated in the PR state. For WT, the free-energy trace of both the PR and the F states was calculated and published in ref. 27. Note that the N139T trace ends at ∼31 Å simply because the PT pathway in the D-channel is shorter in this mutant. This is a result of the less-curved pathway through the T139 residue.

Table 1.

Calculated time constants (inverse of rate constants) for PT in the D-channel from D132 to E286 and in the opposite direction in WT, N139T, N139C, and N139L mutants, as well as pKa values

Mutant State E286 → D132, μs D132 → E286, μs Experimental E286 pKa Calculated E286 pKa
WT F (2.8 ± 0.5) × 104 (2 ± 1) × 10−1* 9.4 10.0
N139T F 7 ± 4 (2 ± 1) × 10−4 7.6 9.4
N139C F (5 ± 2) × 101 (1.9 ± 0.3) × 10−4 >11 9.3
N139L PR (5 ± 1) × 107 (1.0 ± 0.7) × 104 NA 8.6
S200V/S201V PR (1.5 ± 0.8) × 1010 5 ± 3 >12 14.4
Open in a new tab

The time constants are compared with experimental time constants for proton pumping from PLS to the P-side bulk (32, 33) and for the PR → F transition (7, 8). WT exp. (PLS → P-side) = 1.40 × 102 (PR → F) and 2.6 × 103 (O → E); WT exp. (PR → F) = 200. Exp., experimental; NA, not analyzed.

*

Ref. 27.

In the N139T mutant, the reduced proton back-leakage barrier in the D-channel can be attributed to the hydroxyl group on the T139 side chain acting as a hydrogen bond acceptor (Fig. S1B), which stabilizes the hydrated excess proton as it shuttles through the T139–N121 gate region. In N139C, the reduced back-leakage barrier can be attributed to the increased hydration level around C139 compared with N139 in WT, even when the excess proton is away from this region. This is evident from the minimum free-energy pathway traced along the 2D-PMFs (black lines in Figs. 2 A and C). Closer inspection of the structure around C139 and N121 (Fig. S1C) also supports this explanation. The C139 side chain rotates away from the N121 side chain, creating more space for water molecules to form a continuous water wire before the deprotonation of D132. In contrast, the water wire is broken by the hydrogen bonds formed between the N139 and the N121 in the WT (Fig. S1A). Thus, the water wire is connected only when the excess proton moves close to N139, which is coupled with increased hydration in this region (27).

It is interesting to note that the N139T and N139C mutants also decrease the forward PT barrier from D132 to E286, increasing the rate for PT from the D132 to the E286 (Table 1). This is consistent with previous experimental work showing that the double-mutant N139T/D132N facilitates proton uptake compared with the single-mutant D132N (36). Moreover, it is also consistent with the experimental observations that the N139T and N139C mutants have similar F → O transition rates compared with WT (15, 18). More specifically, the F → O transition involves ET and PT to the BNC, both of which depend on E286 reprotonation through the D-channel (37). Because reprotonation of E286 is faster in the N139T and N139C mutants, the rate-limiting step for the F → O transition remains the same as that in WT. Therefore, the F → O transition rate in the mutants is similar compared with WT (see below for discussion on the N139L mutant).

Because the pKa of the proton acceptor at the BNC is higher than the pH of neutral N-side bulk solution, the thermodynamic driving force for PT from N-side bulk to the BNC is large. Therefore, the chemical reaction that reduces oxygen to water is not impaired by the N139T and N139C mutations. This leads to the decoupling of transmembrane proton pumping and chemical reaction in these mutants. This explanation for the decoupling mechanism of N139T and N139C mutants is likely generalizable for other N139 decoupling mutants with neutral side chains, such as N139A and N139S (17). Their small size, and the polar hydroxyl group on N139S, likely allows for easy proton back leakage through the D-channel, eliminating a kinetic gating and decoupling proton pumping from the chemical reaction. It is noted here that Siegbahn and Blomberg (38) also hypothesized that a reduced back-leakage barrier could be a possible decoupling mechanism.

N139L Mutant Inhibits Reprotonation of E286 Through the D-channel.

The 2D-PMF for PT through the D-channel of the N139L mutant was also calculated (Fig. 2D). The 1D free-energy curve along the minimum free-energy pathway traced out from the 2D-PMF is plotted in Fig. 3. The bulky L139 side chain causes a large forward barrier (∼17 kcal/mol) compared with that in the WT (∼9 kcal/mol). This slows down the rate of PT from the D132 to the E286 compared with that in the WT (Table 1). This rate of PT from D132 to E286 in N139L is also slower than proton back flow from the PLS to deprotonated E286 in the PR state of the WT system, which has a time constant of 1.2 × 103 μs (27). In our previous work (27), we suggested that without fast reprotonation of E286 in the PR state, the proton loaded at the PLS would flow back to E286 and then to the BNC, forming the F state and leaving behind an empty PLS and deprotonated E286. This would eliminate proton pumping. In the N139L mutant, the proton pumping is essentially eliminated (16), which supports our prediction. Moreover, the slowdown of E286 reprotonation will leave the E286 in its deprotonated form for a longer time in the F state (16). As a result, the subsequent PT from E286 to the BNC that converts the F state to the O state will also be slowed down. This is consistent with the experimental data that the F → O transition is significantly slowed down in this mutant (16). A closer inspection of the 2D-PMF (Fig. 2D) reveals that the hydration of the L139 region constitutes a large part of the overall barrier for the PT through the L139 region. This is because the bulky hydrophobic L139 side chain dehydrates this region to a greater extent than N139 does in the WT system Fig. S1A with Fig. S1D and Fig. 2A with Fig. 2D. Therefore, it is energetically more unfavorable for the excess proton to shuttle past L139 in the mutant than past N139 in WT CcO. (Fig. 2D).

S200V/S201V Mutant Increases the pKa of E286.

For the S200V/S201V mutant, it is found that the free-energy minimum around the E286 is deeper than that in WT (Figs. 2E and 3). Because the S200V and S201V double mutation is far from residue D132, it is likely that the D132 pKa value is not changed significantly. Then based on the PMF we can estimate the pKa of the E286 in this mutant to be increased by ∼4.6 units compared with that of WT (9.4), giving a pKa of ∼14. This is in reasonable agreement with the experimental measurements by Lee et al. (39), which showed that the pKa of E286 is increased by at least 3 units in this mutant. The increased pKa of E286 indicates an increased proton affinity of E286, and as a result both the pumping and the chemical reaction will be slowed down, which agrees well with the experimental findings by Lee et al. (39). A closer examination of the water structure of the V200/V201 region provides insight into why the double mutation leads to deeper free-energy minima around E286 relative to WT. Below E286, the water molecules in the second and third solvation shells form hydrogen bonds with Ser-201 in WT, and there are continuous water wires connecting E286 all of the way down to N139. In contrast, in S200V/S201V, the second and third solvation shells of E286 have an interface with the bulky hydrophobic side chains of V200 and V201. This leads to a narrower region and frequently broken water wires, which makes E286 harder to deprotonate in the backward direction, leading to a deeper free-energy minimum around E286. This observation is also consistent with our previous simulation results using a reduced-system model (39).

Estimated E286 pKa Values in CcO Mutants.

If the pKa of residue D132 is known, then the pKa of E286 can be calculated by the ratio between forward and backward PT rate constants in the D-channel. To estimate the E286 pKa shifts induced by each mutation, here we assume that the mutations do not significantly change the D132 pKa from its WT value, which is assumed to be 4.9 [raised by 1 pH unit (40) relative to that in bulk (41)]. The estimated pKa values are included in Table 1. The E286 pKa for the S200V/S201V mutant is in qualitative agreement with experimental measurements (39), as discussed above. For the N139T and N139C mutants, the calculated pKa values are similar to WT, which differs somewhat from experimental measurements (15, 18). This can perhaps be attributed to the classical force field that is used for C139 and T139, which does not account for charge transfer from the hydrated excess proton to the polar groups on C139 and T139. However, MS-RMD models of C139 and T139 that could account for charge transfer would only further stabilize the excess proton, reducing the back-leakage barrier in the D-channel even more. In addition, the D132 pKa could be shifted in the C139 and T139 mutants. Therefore, although we do not predict the E286 pKa values exactly for all mutants, we do not expect this to change our conclusions on the decoupling mechanisms.

Comparison with the E286 Valve Mechanism.

Previously, the “E286 valve” mechanism (42) was also proposed to explain the high proton-pumping efficiency in WT CcO. In this mechanism, the preferred “down” orientation of deprotonated E286 prevents proton back leakage from the PLS to E286 after water is formed in the BNC. However, other studies have suggested that the orientational preference of deprotonated E286 cannot effectively prevent proton back leakage by itself (43). Regardless of this controversy, our current findings suggest that in the case that the excess proton leaks from the PLS back to E286, the large D-channel back-leakage barrier in the WT CcO prevents further back leakage to the N-side bulk. It is possible that both the barriers in the HC (including the E286 valve and stabilization of the protonated PLS) and those in the D-channel contribute to preventing proton back leakage and achieving high proton-pumping efficiency in the WT system. However, the loss of pumping efficiency in the mutants is explained by the alteration of the D-channel barriers.

Conclusions

Despite receiving significant attention, it has proved challenging in the past to explain the decoupling mechanisms of CcO mutants, including N139 mutants with neutral side chains (such as N139T, N139C, and N139L). Our multiscale simulation results presented here show that the N139T and N139C mutants facilitate solvation and PT through the X139–N121 gate region (where X is the amino acid mutation), thereby enabling proton back leakage through the D-channel to the N-side bulk and eliminating proton pumping. In contrast, the N139L mutant slows down PT from D132 to E286 in the D-channel, which eliminates proton pumping and also impairs the chemical reaction. This work highlights how N139 and N121 in the WT CcO hydrogen bond with each other to function as a kinetic gate in the D-channel. We note that previous studies of the N139 residue focused more on its role of gating the forward PT from D132 to E286 through the D-channel (44, 45). However, our work reveals that the more important role of the N139 gate is to create a large kinetic barrier for proton back leakage to the N-side bulk. Meanwhile, the N139 gate also fine tunes the forward PT barrier through the D-channel such that it does not inhibit timely reprotonation of E286 (27). Other decoupling mutations of N139 will likely also either facilitate proton back leakage to N-side bulk (like N139T and N139C) or slow down E286 reprotonation (like N139L). This explains why the N139 and N121 residues are so important for the high proton-pumping efficiency in the WT enzyme.

These findings build significantly upon our recently published work on the WT CcO proton-pumping mechanism (27). In that study, we discovered how kinetic gating avoids two potential pathways that would decouple proton pumping from the chemical reaction. The first pathway is premature proton uptake to the BNC before PT to the PLS, which would leave behind an empty PLS and bypass proton pumping altogether. In WT CcO this pathway is avoided by PT from the E286 to the PLS being faster than PT from the E286 to the BNC (27). The second pathway is proton back flow from the PLS to the deprotonated E286 before reprotonation through the D-channel and subsequent PT to the BNC. In the WT CcO this pathway is avoided by reprotonation of E286 through the D-channel being faster than proton back flow from the PLS to the E286 in the PR state due to the stabilization of the excess proton in the PLS (27). However, in the N139L mutant this pathway dominates due to the slow D-channel PT.

The present work reveals another potential pathway that could decouple proton pumping from the chemical reaction. The delivery of the chemical proton to the BNC to form the F state makes the proton at the PLS less stable. This pumped proton can either be pumped to the P-side bulk or flow back to E286 and then farther back to the N-side bulk through the D-channel, thus eliminating the proton pumping. Therefore, a third kinetic gate is needed to block back flow to the N-side bulk. Combining the mutant results in the present work and the WT results from our previous work (27), we conclude that this kinetic gate is the large proton back-flow barrier in the D-channel created by the N139–N121 gate. This result adds a unique facet to the delicate interplay of different kinetic gating effects that ensure the high proton-pumping efficiency in CcO.

Materials and Methods

System Setup and Structural Equilibration.

The equilibrated structures for WT CcO in ref. 27 were used as the initial structure for the classical MD equilibration of the N139T, N139C, N139L, and S200V/S201V mutants. For the N139T and N139C mutants, to simulate the back leakage of the pumped proton to the N-side through the D-channel after water formation at the BNC, the WT CcO equilibrated structure in the F state was used, with E286 protonated, the PLS deprotonated, D132 deprotonated, and water bound to the BNC. The C139 residue in the N139C mutant is in the protonated state (15). For the N139L and S200V/S201V mutants, to simulate the E286 reprotonation through the D-channel in the PR state after PT from E286 to the PLS, the WT CcO equilibrated structure in the PR state was used, with the PLS protonated and hydroxide bound to the BNC CuB.

The mutant CcO proteins were embedded in a dimyristoylphosphatidylcholine (DMPC) lipid bilayer and solvated by TIP3P (transferable intermolecular potential 3P) water molecules on each side of the membrane (27, 46). Periodic boundary conditions were applied in all three dimensions. The CHARMM22 (47) and CHARMM36 (48) force fields were used for the protein and lipids, respectively. The cutoffs for Lennard–Jones (LJ) and real-space electrostatic interactions were 12 Å, using a switching function starting at 10 Å for the LJ interactions. Long-range electrostatics were treated by the particle–particle particle–mesh (PPPM) method (49) with an accuracy threshold of 10−4. The integration time step was 1 fs. The temperature was controlled at 308 K by a Nosé–Hoover thermostat with a relaxation time constant of 0.1 ps. The system was equilibrated for 20–30 ns. Harmonic restraints with a force constant of 100 kcal⋅mol−1⋅Å−2 were imposed on the heavy atoms of heme a, heme a3, CuA, and CuB and their ligands throughout the simulations. The structural equilibration was carried out with the LAMMPS MD package (lammps.sandia.gov) (50).

Finding PT Pathways.

MS-RMD simulations using metadynamics (51) were performed to identify the PT pathways in the D-channel of the CcO mutants. More detailed information on performing the MS-RMD metadynamics PT simulations can be found in ref. 27.

MS-RMD Free-Energy Profiles in the HC and the D-channel.

The MS-RMD umbrella sampling simulation in the D-channel of the CcO mutants was carried out, restraining both the excess proton CEC position along the PT pathway and the water density in a predefined box that encompasses the 139 and 121 residues. The MS-RMD models for the E286 and D132 residues in the mutants are the same as those in the WT CcO (27), which is justified by the small perturbation of the electrostatic environment upon mutation of N139 to neutral residues. More detailed information on performing the MS-RMD umbrella sampling and the definition of the excess proton CEC can be found in ref. 27.

Acknowledgments

Research reported in this paper was supported by the National Institute of General Medical Sciences of the National Institutes of Health under Award R01GM053148 (to G.A.V., J.M.J.S., and R.L.) and the Sigrid Jusélius Foundation (M.W.). The researchers used computing facilities provided by the Extreme Science and Engineering Discovery Environment, which is supported by National Science Foundation Grant OCI-1053575, as well as from The University of Chicago Research Computing Center, the Texas Advanced Computing Center at The University of Texas at Austin, and the US Department of Defense High Performance Computing Modernization Program.

Footnotes

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.1703654114/-/DCSupplemental.

References

  • 1.Wikstrom MKF. Proton pump coupled to cytochrome c oxidase in mitochondria. Nature. 1977;266:271–273. doi: 10.1038/266271a0. [DOI] [PubMed] [Google Scholar]
  • 2.Ferguson-Miller S, Babcock GT. Heme/copper terminal oxidases. Chem Rev. 1996;96:2889–2908. doi: 10.1021/cr950051s. [DOI] [PubMed] [Google Scholar]
  • 3.Kaila VR, Verkhovsky MI, Wikström M. Proton-coupled electron transfer in cytochrome oxidase. Chem Rev. 2010;110:7062–7081. doi: 10.1021/cr1002003. [DOI] [PubMed] [Google Scholar]
  • 4.Konstantinov AA, Siletsky S, Mitchell D, Kaulen A, Gennis RB. The roles of the two proton input channels in cytochrome c oxidase from Rhodobacter sphaeroides probed by the effects of site-directed mutations on time-resolved electrogenic intraprotein proton transfer. Proc Natl Acad Sci USA. 1997;94:9085–9090. doi: 10.1073/pnas.94.17.9085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Wikström M, Jasaitis A, Backgren C, Puustinen A, Verkhovsky MI. The role of the D- and K-pathways of proton transfer in the function of the haem-copper oxidases. Biochim Biophys Acta. 2000;1459:514–520. doi: 10.1016/s0005-2728(00)00191-2. [DOI] [PubMed] [Google Scholar]
  • 6.Sharma V, Wikström M. The role of the K-channel and the active-site tyrosine in the catalytic mechanism of cytochrome c oxidase. Biochim Biophys Acta. 2016;1857:1111–1115. doi: 10.1016/j.bbabio.2016.02.008. [DOI] [PubMed] [Google Scholar]
  • 7.Faxén K, Gilderson G, Adelroth P, Brzezinski P. A mechanistic principle for proton pumping by cytochrome c oxidase. Nature. 2005;437:286–289. doi: 10.1038/nature03921. [DOI] [PubMed] [Google Scholar]
  • 8.Belevich I, Verkhovsky MI, Wikström M. Proton-coupled electron transfer drives the proton pump of cytochrome c oxidase. Nature. 2006;440:829–832. doi: 10.1038/nature04619. [DOI] [PubMed] [Google Scholar]
  • 9.Pfitzner U, et al. Tracing the D-pathway in reconstituted site-directed mutants of cytochrome c oxidase from Paracoccus denitrificans. Biochemistry. 2000;39:6756–6762. doi: 10.1021/bi992235x. [DOI] [PubMed] [Google Scholar]
  • 10.Pawate AS, et al. A mutation in subunit I of cytochrome oxidase from Rhodobacter sphaeroides results in an increase in steady-state activity but completely eliminates proton pumping. Biochemistry. 2002;41:13417–13423. doi: 10.1021/bi026582+. [DOI] [PubMed] [Google Scholar]
  • 11.Namslauer A, Pawate AS, Gennis RB, Brzezinski P. Redox-coupled proton translocation in biological systems: Proton shuttling in cytochrome c oxidase. Proc Natl Acad Sci USA. 2003;100:15543–15547. doi: 10.1073/pnas.2432106100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Vakkasoglu AS, Morgan JE, Han D, Pawate AS, Gennis RB. Mutations which decouple the proton pump of the cytochrome c oxidase from Rhodobacter sphaeroides perturb the environment of glutamate 286. FEBS Lett. 2006;580:4613–4617. doi: 10.1016/j.febslet.2006.07.036. [DOI] [PubMed] [Google Scholar]
  • 13.Dürr KL, et al. A D-pathway mutation decouples the Paracoccus denitrificans cytochrome c oxidase by altering the side-chain orientation of a distant conserved glutamate. J Mol Biol. 2008;384:865–877. doi: 10.1016/j.jmb.2008.09.074. [DOI] [PubMed] [Google Scholar]
  • 14.Varanasi L, Hosler J. Alternative initial proton acceptors for the D pathway of Rhodobacter sphaeroides cytochrome c oxidase. Biochemistry. 2011;50:2820–2828. doi: 10.1021/bi102002v. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Johansson AL, et al. Proton uptake and pKa changes in the uncoupled Asn139Cys variant of cytochrome c oxidase. Biochemistry. 2013;52:827–836. doi: 10.1021/bi301597a. [DOI] [PubMed] [Google Scholar]
  • 16.Siletsky SA, Zhu J, Gennis RB, Konstantinov AA. Partial steps of charge translocation in the nonpumping N139L mutant of Rhodobacter sphaeroides cytochrome c oxidase with a blocked D-channel. Biochemistry. 2010;49:3060–3073. doi: 10.1021/bi901719e. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Zhu J, Han H, Pawate A, Gennis RB. Decoupling mutations in the D-channel of the aa(3)-type cytochrome c oxidase from Rhodobacter sphaeroides suggest that a continuous hydrogen-bonded chain of waters is essential for proton pumping. Biochemistry. 2010;49:4476–4482. doi: 10.1021/bi100344x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Lepp H, Salomonsson L, Zhu JP, Gennis RB, Brzezinski P. Impaired proton pumping in cytochrome c oxidase upon structural alteration of the D pathway. Biochim Biophys Acta. 2008;1777:897–903. doi: 10.1016/j.bbabio.2008.04.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Han D, et al. Replacing Asn207 by aspartate at the neck of the D channel in the aa3-type cytochrome c oxidase from Rhodobacter sphaeroides results in decoupling the proton pump. Biochemistry. 2006;45:14064–14074. doi: 10.1021/bi061465q. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Brzezinski P, Johansson AL. Variable proton-pumping stoichiometry in structural variants of cytochrome c oxidase. Biochim Biophys Acta. 2010;1797:710–723. doi: 10.1016/j.bbabio.2010.02.020. [DOI] [PubMed] [Google Scholar]
  • 21.Chakrabarty S, Namslauer I, Brzezinski P, Warshel A. Exploration of the cytochrome c oxidase pathway puzzle and examination of the origin of elusive mutational effects. Biochim Biophys Acta. 2011;1807:413–426. doi: 10.1016/j.bbabio.2011.01.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Knight C, Voth GA. The curious case of the hydrated proton. Acc Chem Res. 2012;45:101–109. doi: 10.1021/ar200140h. [DOI] [PubMed] [Google Scholar]
  • 23.Swanson JMJ, et al. Proton solvation and transport in aqueous and biomolecular systems: Insights from computer simulations. J Phys Chem B. 2007;111:4300–4314. doi: 10.1021/jp070104x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Knight C, Lindberg GE, Voth GA. Multiscale reactive molecular dynamics. J Chem Phys. 2012;137:A525. doi: 10.1063/1.4743958. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Yamashita T, Peng Y, Knight C, Voth GA. Computationally efficient multiconfigurational reactive molecular dynamics. J Chem Theory Comput. 2012;8:4863–4875. doi: 10.1021/ct3006437. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Liang R, Li H, Swanson JMJ, Voth GA. Multiscale simulation reveals a multifaceted mechanism of proton permeation through the influenza A M2 proton channel. Proc Natl Acad Sci USA. 2014;111:9396–9401. doi: 10.1073/pnas.1401997111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Liang R, Swanson JMJ, Peng Y, Wikström M, Voth GA. Multiscale simulations reveal key features of the proton-pumping mechanism in cytochrome c oxidase. Proc Natl Acad Sci USA. 2016;113:7420–7425. doi: 10.1073/pnas.1601982113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Liang R, et al. Acid activation mechanism of the influenza A M2 proton channel. Proc Natl Acad Sci USA. 2016;113:E6955–E6964. doi: 10.1073/pnas.1615471113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Lee S, Swanson JM, Voth GA. Multiscale simulations reveal key aspects of the proton transport mechanism in the ClC-ec1 antiporter. Biophys J. 2016;110:1334–1345. doi: 10.1016/j.bpj.2016.02.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Lee S, Mayes HB, Swanson JMJ, Voth GA. The origin of coupled chloride and proton transport in a Cl(-)/H(+) antiporter. J Am Chem Soc. 2016;138:14923–14930. doi: 10.1021/jacs.6b06683. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Lee S, Liang R, Voth GA, Swanson JMJ. Computationally efficient multiscale reactive molecular dynamics to describe amino acid deprotonation in proteins. J Chem Theory Comput. 2016;12:879–891. doi: 10.1021/acs.jctc.5b01109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Wikström M, Verkhovsky MI. Mechanism and energetics of proton translocation by the respiratory heme-copper oxidases. Biochim Biophys Acta. 2007;1767:1200–1214. doi: 10.1016/j.bbabio.2007.06.008. [DOI] [PubMed] [Google Scholar]
  • 33.Belevich I, Bloch DA, Belevich N, Wikström M, Verkhovsky MI. Exploring the proton pump mechanism of cytochrome c oxidase in real time. Proc Natl Acad Sci USA. 2007;104:2685–2690. doi: 10.1073/pnas.0608794104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Salomonsson L, Faxén K, Adelroth P, Brzezinski P. The timing of proton migration in membrane-reconstituted cytochrome c oxidase. Proc Natl Acad Sci USA. 2005;102:17624–17629. doi: 10.1073/pnas.0505431102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Siegbahn PEM, Blomberg MRA. Energy diagrams and mechanism for proton pumping in cytochrome c oxidase. Biochim Biophys Acta. 2007;1767:1143–1156. doi: 10.1016/j.bbabio.2007.06.009. [DOI] [PubMed] [Google Scholar]
  • 36.Johansson AL, Högbom M, Carlsson J, Gennis RB, Brzezinski P. Role of aspartate 132 at the orifice of a proton pathway in cytochrome c oxidase. Proc Natl Acad Sci USA. 2013;110:8912–8917. doi: 10.1073/pnas.1303954110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Karpefors M, Adelroth P, Zhen Y, Ferguson-Miller S, Brzezinski P. Proton uptake controls electron transfer in cytochrome c oxidase. Proc Natl Acad Sci USA. 1998;95:13606–13611. doi: 10.1073/pnas.95.23.13606. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Siegbahn PEM, Blomberg MRA. Mutations in the D-channel of cytochrome c oxidase causes leakage of the proton pump. FEBS Lett. 2014;588:545–548. doi: 10.1016/j.febslet.2013.12.020. [DOI] [PubMed] [Google Scholar]
  • 39.Lee HJ, et al. Intricate role of water in proton transport through cytochrome c oxidase. J Am Chem Soc. 2010;132:16225–16239. doi: 10.1021/ja107244g. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Sharma V, Ala-Vannesluoma P, Vattulainen I, Wikström M, Róg T. 2015. Role of subunit III and its lipids in the molecular mechanism of cytochrome c oxidase. Biochim Biophys Acta Bioenergetics 1847:690–697. [DOI] [PubMed]
  • 41.Lide DR. CRC Handbook of Chemistry and Physics. CRC Press; Boca Raton, FL: 2004. [Google Scholar]
  • 42.Kaila VR, Verkhovsky MI, Hummer G, Wikström M. Glutamic acid 242 is a valve in the proton pump of cytochrome c oxidase. Proc Natl Acad Sci USA. 2008;105:6255–6259. doi: 10.1073/pnas.0800770105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Yang S, Cui Q. Glu-286 rotation and water wire reorientation are unlikely the gating elements for proton pumping in cytochrome C oxidase. Biophys J. 2011;101:61–69. doi: 10.1016/j.bpj.2011.05.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Henry RM, Yu CH, Rodinger T, Pomès R. Functional hydration and conformational gating of proton uptake in cytochrome c oxidase. J Mol Biol. 2009;387:1165–1185. doi: 10.1016/j.jmb.2009.02.042. [DOI] [PubMed] [Google Scholar]
  • 45.Henry RM, Caplan D, Fadda E, Pomès R. Molecular basis of proton uptake in single and double mutants of cytochrome c oxidase. J Phys Condens Matter. 2011;23:234102. doi: 10.1088/0953-8984/23/23/234102. [DOI] [PubMed] [Google Scholar]
  • 46.Jorgensen WL, Jenson C. Temperature dependence of TIP3P, SPC, and TIP4P water from NPT Monte Carlo simulations: Seeking temperatures of maximum density. J Comput Chem. 1998;19:1179–1186. [Google Scholar]
  • 47.MacKerell AD, et al. All-atom empirical potential for molecular modeling and dynamics studies of proteins. J Phys Chem B. 1998;102:3586–3616. doi: 10.1021/jp973084f. [DOI] [PubMed] [Google Scholar]
  • 48.Klauda JB, et al. Update of the CHARMM all-atom additive force field for lipids: Validation on six lipid types. J Phys Chem B. 2010;114:7830–7843. doi: 10.1021/jp101759q. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Hockney RW, Eastwood JW. Computer Simulation Using Particles. McGraw-Hill; New York: 1981. [Google Scholar]
  • 50.Plimpton S. Fast parallel algorithms for short-range molecular-dynamics. J Comput Phys. 1995;117:1–19. [Google Scholar]
  • 51.Laio A, Parrinello M. Escaping free-energy minima. Proc Natl Acad Sci USA. 2002;99:12562–12566. doi: 10.1073/pnas.202427399. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES