Exploring the proton pump and exit pathway for pumped protons in cytochrome ba3 from Thermus thermophilus

Edited* by Harry B. Gray, California Institute of Technology, Pasadena, CA, and approved February 21, 2012 (received for review May 8, 2011)
March 19, 2012
109 (14) 5259-5264


The heme-copper oxygen reductases are redox-driven proton pumps. In the current work, the effects of mutations in a proposed exit pathway for pumped protons are examined in the ba3-type oxygen reductase from Thermus thermophilus, leading from the propionates of heme a3 to the interface between subunits I and II. Recent studies have proposed important roles for His376 and Asp372, both of which are hydrogen-bonded to propionate-A of heme a3, and for Glu126II (subunit II), which is hydrogen-bonded to His376. Based on the current results, His376, Glu126II, and Asp372 are not essential for either oxidase activity or proton pumping. In addition, Tyr133, which is hydrogen-bonded to propionate-D of heme a3, was also shown not to be essential for function. However, two mutations of the residues hydrogen-bonded to propionate-A, Asp372Ile and His376Asn, retain high electron transfer activity and normal spectral features but, in different preparations, either do not pump protons or exhibit substantially diminished proton pumping. It is concluded that either propionate-A of heme a3 or possibly the cluster of groups centered about the conserved water molecule that hydrogen-bonds to both propionates-A and -D of heme a3 is a good candidate to be the proton loading site.
The oxygen reductase members of the heme-copper superfamily are critical redox-driven proton pumps in the aerobic respiratory chains of most aerobic bacteria and archaea as well as in the mitochondria of eukaryotes (14). These membrane-bound proteins catalyze the reduction of O2 to water and use the free energy of this reaction to generate a proton motive force that is used for a number of essential biological functions, including the synthesis of ATP.
The heme-copper oxygen reductases constitute a diverse group of phylogenetically related enzymes, the vast majority of which can be classified as members of the A-, B-, or C-family of heme-copper oxygen reductases (1, 5, 6). Enzymes from each of these oxygen reductase families have been shown to couple the redox chemistry to proton pumping (711). Proton pumping is one of the mechanisms by which the heme-copper oxygen reductases generate the proton motive force, and it is reasonable to assume that all the heme-copper oxygen reductases use the same fundamental mechanism to couple the redox chemistry to proton pumping. A number of models of proton pumping have been proposed (8, 1222), but much remains to be elucidated about the proton pump mechanism.
All models of proton pumping include at least one site in the enzyme whose proton affinity (pKa) changes during the catalytic cycle such that the residue picks up a proton from the N-side (negative or inside) of the membrane and releases the proton to the P-side (positive or outside) of the membrane. Each time this occurs, a proton is pumped from inside to outside and a full charge crosses the membrane, generating a voltage. For the A-family heme-copper oxygen reductases, such as the bovine heart mitochondrial enzyme, this occurs four times for each O2 converted to water (i.e., 4 protons are pumped per O2). For the B-family and C-family enzymes, this occurs twice per catalytic cycle (i.e., 2 protons are pumped per O2) (7, 10, 11). The identity of the so-called “proton loading site” and the exit pathway for pumped protons remain unknown, and are addressed in the current work.
If one makes the reasonable assumption that the proton loading site is common to all the heme-copper oxygen reductases, the number of candidates is limited because of the sequence diversity between the A-, B-, and C-family enzymes. Essentially, one is left with the six histidine ligands to the metals, the four propionate side chains of the hemes, and conserved internal water molecules (23). Several models of proton pumping have focused on one or more of the histidine ligands to CuB (2429), but no compelling evidence has emerged. Propionate-D of the active-site heme in the A-family oxygen reductases has also been suggested as a possible proton loading site, or at least as a transient site for proton transfer to the proton loading site (30). This propionate forms a salt bridge with an arginine in the A- and B-family oxygen reductases but is ligated to a Ca2+ in the C-family enzymes (31), making this an unlikely site for protonation, at least in the C-family enzymes. Site-directed mutagenesis experiments in the aa3-type oxygen reductase from Rhodobacter sphaeroides also strongly suggest that propionate-D of the active-site heme is an unlikely candidate for the proton loading site, although this site may be important for the transfer of protons to the proton loading site (32).
Propionate-A of the active-site heme remains a viable candidate, and has been suggested based on several indirect lines of evidence (8, 23, 28, 33). From measurements of the time-resolved generation of the transmembrane potential of the aa3-type oxygen reductase from Paracoccus denitrificans, Wikström and Verkhovsky (8) concluded that the most likely candidate for the proton loading site is propionate-A of heme a3. Recently, we reported (23) a comparison of the structural properties conserved between the heme-copper oxygen reductase families, and concluded that propionate-A of the active-site heme is a reasonable candidate as the proton loading site. Computational studies have also supported this conclusion for the aa3-type heme-copper oxygen reductase from P. denitrificans, indicating that the pKa of this propionate should be sensitive to reduction of heme a (33) and to the redox state of the heme-copper binuclear center (34).
There have been several suggestions specifically concerning the proton loading site of cytochrome ba3 from Thermus thermophilus. These have all involved either propionate-A of heme a3 or hydrogen bond partners of this propionate. Varotsis and coworkers (35) postulated a critical role in proton pumping of cytochrome ba3 from T. thermophilus for the hydrogen-bonded carboxylate pair of propionate-A of heme a3 and D372 (Fig. 1). This suggestion is based on perturbations observed by room temperature FTIR difference spectroscopy on photolysis of CO bound to heme a3 of the fully reduced enzyme. H376, which is also hydrogen-bonded to propionate-A of heme a3, has also been suggested to be the proton loading site based on computational results (28). The proton is proposed (28) to be passed from H376 to E126II, which is hydrogen-bonded to H376 (Fig. 1), and then into the bulk aqueous phase. Finally, the structurally conserved water bridging propionate-A and propionate-D of the active-site heme have also been proposed to be the proton loading site (22, 23), which is the model most favored by the results presented here.
Fig. 1.
Two key subunits I (cyan) and II (yellow) are depicted in the cell membrane along with low-spin heme b and a heme a3-CuB binuclear center. In the B-family, there is only one functional pathway, the K-channel analog, also referred to as the KB channel in the main text (dashed blue arrow). The pumped protons are released to the P-side of the membrane. The region near propionate-A, illustrating the residues of interest in this work, is shown in more detail. The figure was prepared with the program VMD (University of Illinois of Urbana-Champaign Biotechnology Center) and PyMOL (Delano Scientific) from the crystal structures (PDB ID codes 3S8F and 1XME) reported by Luna et al. (36), Hunsicker-Wang et al. (37), and Tiefenbrunn et al. (38).
In the current work, site-directed mutations were made in each of the residues hydrogen-bonded to either propionate-A or propionate-D of the active-site heme a3 of cytochrome ba3, as well as in several nearby residues postulated to be functionally important. Non–hydrogen-bonding residues can be substituted for Y133 (H-bond to propionate-D), H376 (H-bond to propionate-A), or D372 (H-bond to propionate-A), without significant loss of either oxidase activity or proton pumping. In addition, E126II and N377, both adjacent to H376, can be mutated without loss of function. None of these residues is essential. Two particular mutations, D372I and H376N, however, have high activity (50–70%), but most preparations of these mutants fail to pump protons; some preparations are observed to pump but at a lower stoichiometry compared with the WT. This implicates propionate-A or groups close to propionate-A as being important for proton pumping. We suggest that the proton loading site may be centered around the structurally conserved water molecule, which is hydrogen-bonded to both propionate-A and propionate-D of heme a3 as well as to H283, which is one of the ligands to CuB.


Conservation of Residues Hydrogen-Bonded to Propionate-A of the Active-Site Heme.

In cytochrome ba3 from T. thermophilus (36, 37), propionate-A of heme a3 is within hydrogen bond distance to D372 and H376, as well as to two water molecules (PDB ID code 3S8F) (38) (Fig. 1). If either of the two amino acid residues is critical to the proton-pumping mechanism, one would expect the residue to be totally or very highly conserved in all proton-pumping heme-copper oxygen reductases. Within the B-family, which includes the T. thermophilus cytochrome ba3, H376 is totally conserved. However, in the majority of B-family enzymes, asparagine is present in the location of D372, and one also finds glutamine, serine, and threonine at this site.
In the C-family (or cbb3), heme-copper oxygen reductase from Pseudomonas stutzeri, propionate-A of the active-site heme b3 is hydrogen-bonded to a histidine and asparagine (31). Throughout the C-family, the histidine is found in nearly all enzymes, with the exception of a few enzymes (from Chlamydia) that have serine in the same location. As with the B-family, the propionate is also hydrogen-bonded to asparagine in most enzymes, but glutamine, serine, and threonine are also found in the same position.
In the A-family, the histidine is found in a number of enzymes, but there are also many enzymes that have glutamine, serine, or threonine at this location. There are also several halophilic archaea that have an arginine at this position and a group of cyanobacterial enzymes that have a glycine. The second position near propionate-A of the active-site heme is predominantly aspartate in the A-family enzymes, but one also finds asparagine, serine, and threonine.
In summary, with the exception of a few A-family enzymes, all the heme-copper oxygen reductases have two residues that can hydrogen-bond to propionate-A of the active-site heme. In the exceptions, which have glycine, it is reasonable to postulate that water could be present to hydrogen-bond to the propionate. Although histidine is always present in the B-family and (nearly all) C-family enzymes, it is not conserved in the A-family heme-copper oxygen reductases. In the second position, the aspartate in cytochrome ba3 from T. thermophilus (D372) is not found even in most other B-family enzymes, where asparagine is more frequently found in this location. The bioinformatics data indicate that proton pumping does not require either a histidine or an aspartate hydrogen-bonded to propionate-A.

Effects of D372 and H376 Mutants on Activity and Proton Pumping.

Table 1 summarizes the effects of a set of mutants in which either D372 or H376 has been replaced in cytochrome ba3 from T. thermophilus. D372N has high activity and pumps protons, similar to the WT (Fig. 2 and Table 1). The reported turnover number was measured using purified enzyme in detergent solution. The enzymes were also reconstituted in phospholipid vesicles for the purpose of measuring proton pumping, and the oxidase activity of the reconstituted enzymes was measured, both in the absence and presence of ionophores (valinomycin plus CCCP). The ionophores eliminate the proton motive force that would otherwise be generated by the enzyme and that slows the enzyme velocity. The respiratory control ratio (RCR) is the activity in the presence of the ionophores divided by the activity in the absence of the ionophores. The WT has an RCR of about 8, and the D372N mutant has an RCR of 7 (Table 1).
Table 1.
Comparison of the properties of the WT and mutant ba3 oxidases
Mutant ba3UV-visible spectrum (normal or perturbed)Turnover, % (e/s ba3)RCRProton pumping*
WTNormal100 (500 e/s)8+
ND, not determined.
*The “+” indicates proton-pumping stoichiometry equal to that observed with the WT oxidase. The “−” implies that experimental evidence shows severely reduced or no proton pumping. In many cases, the activity was too low to obtain a reliable RCR and proton-pumping data.
The rates of oxygen reduction were measured under several different conditions and are presented relative to WT enzyme determined under the same conditions. The measured enzyme velocity obtained at 25 °C for the WT enzyme is near 500 e−1/s with 30 μM cytochrome c552. Activity was measured at 25 °C because of thermal instability of some mutants.
In some assays, steady-state values as high as 150% of WT were obtained.
Fig. 2.
Proton pumping of the WT, D372V, and D372I mutants of cytochrome ba3 from T. thermophilus. The enzymes were reconstituted into proteoliposomes (7), and a suspension of these vesicles was placed in one syringe of a stopped-flow spectrophotometer [50 μM Hepes⋅KOH, 55 mM KCl, 55 mM sucrose (pH 7.3)]. This solution was mixed with a solution of prereduced cytochrome c552 in the same buffer at a concentration sufficient for about 20 turnovers of the enzyme present (10 μM cytochrome c552 and 0.5 μM enzyme). The change in pH of the bulk solution was measured by monitoring the absorbance of 55 μM phenol red (final concentration) at 558.7 nm. Three traces are shown for each mutant examined: black, no additions to the solution; blue, 5 μM valinomycin plus 5 μM CCCP (final concentrations); and red, 5 μM valinomycin (final concentration). Further details are provided by Kannt et al. (7).
Among the D372 mutations, the most damaging is D372E, which has very low activity (7%) and an RCR of about 1 (i.e., the activity is not influenced by the ionophores).
In contrast, the D372V mutant has relatively high activity (65%) and pumps protons as well as the WT (Fig. 2). Unexpectedly, the D372I mutant exhibits little or no proton pumping, measured using a stopped-flow spectrophotometer (Fig. 2) or with a pH meter (Fig. S1), but retains high oxygen reductase activity (50%) and normal spectral features (Fig. S2). Because of the unique properties of this mutant, several independent enzyme preparations were examined, most of which had negligible proton pumping, although in some cases, pumping was observed but with a very low stoichiometry. The entire sequence of the operon in the expression plasmid was verified.
Among the H376 mutants, the most damaging are H376D and H376A (Table 1). However, replacing H376 by either tyrosine or phenylalanine (H376Y or H376F) yields enzyme that is essentially WT, with high activity, a high RCR, and normal proton pumping (Table 1 and Fig. S3). The H376N mutant retains high activity (70%) but is produced in very low amounts. In all measurements, using several independent preparations, proton pumping was either negligible or much less than in the WT (Table 1 and Fig. S3).

Mutagenesis of Other Residues That Might Be Important for Internal Proton Translocation in Cytochrome ba3: E126II, N377, D287, T302, and Y133.

One plausible pathway for pumped protons is from the K-analog (KB, where the superscript indicates the B-family) channel to H282 (one of the histidine ligands to CuB), followed by subsequent proton transfer from H282 to H376 and then to E126II (28) and out (Fig. 1). H282 is totally conserved as a CuB ligand and appears to be hydrogen-bonded to either a serine or threonine in all the heme-copper oxygen reductases (T302 in cytochrome ba3). The position occupied by E126II is a methionine in most B-family enzymes. There is another acidic residue in the vicinity, D287 (Fig. 1), which is always either a glutamate or aspartate (with one exception) in the B-family enzymes.
Two additional residues of interest are N377 and Y133. N377 is adjacent to H376 and is totally conserved in all the B-family oxygen reductases. Furthermore, the planes of the side chains of H376 and N377 are stacked within van der Waals distance. Y133 is hydrogen-bonded to propionate-D of the active-site heme a3, although it is conserved in only about 75% of the B-family sequences. In most cases where a tyrosine is not present at this position, it is replaced by a phenylalanine.
Several mutants were made to test the functional importance of the above-mentioned residues (Table 1 and Fig. S3):
i)E126II was replaced by glutamine (E126QII) without any adverse affects. This residue is clearly not functionally important. The double mutant E126IIM/D372N corresponds to the most prevalent sequence found in the B-family oxygen reductases, as deduced from amino acid sequence comparisons. It has the same oxidase activity and proton pumping as the WT. Ongoing structural analysis of this form indicates that the methionine residue is in close contact with R225.
ii)D287 is hydrogen-bonded to R255 and a water molecule (Fig. 1) as well as to Q284. D287 was replaced by E, L, I, and N. D287E is not expressed and is apparently unstable. D287I and D287L have very low activity (<10%). Most interesting is D287N, which has high activity (77%), but the RCR is only 2. Proton-pumping results indicate that the stoichiometry of proton pumping is substantially reduced. D287 forms a salt bridge with R225 and is located at the interface between subunits I and II. Hence, a structural perturbation attributable to mutations at this location might be anticipated. For this reason, the R225M mutant was constructed. The R225M mutant is fully active. Proton pumping was not measured.
iii)H282 is essential, and no effort was made to replace it. The T302V mutant was designed to put a hydrophobic group in place of T302, which is hydrogen-bonded to H282. The mutant enzyme has low activity (10%) and is expressed at low levels, implying some instability or assembly problems.
iv)N377 is adjacent to H376, and the side chains are in contact. Replacement of N377 by alanine does not alter the oxidase activity or proton pumping of the enzyme.
v)Y133F and Y133W were both examined. Y133W is about 40% active and pumps protons. Y133F is fully active.

Testing for Proton Backflow.

If the pumped protons are delivered through the KB channel to propionate-A of heme a3, it is plausible that mutations to the hydrogen-bonding partners of propionate-A may result in opening a pathway for protons to be driven in the reverse direction from the periplasm-facing side of the protein, through the putative exit pathway for pumped protons past propionate-A, and then to the active site. To examine this, several double mutants were examined in which T312V was used to eliminate enzyme activity completely by blocking the KB channel, accompanied by a mutation in either D372 or H376. The results are shown in Table S1. T312V has virtually no enzyme activity (<1%). Two of the double mutants, T312V/D372V and T312V/H376F, have between 3% and 4% oxygen reductase activity. Both the D372V and H376F mutants have substantial oxygen reductase activity (65% and 130%), in the absence of the T312V mutation. It is possible, although not compelling, that these data imply a gating system to prevent proton backflow that has been compromised by the D372V and H376F mutants.

FTIR Difference Spectroscopy.

Reduced-minus-oxidized FTIR difference spectroscopy can be informative about changes in the protonation state and hydrogen bonding of carboxyl groups (39), and this was used to characterize changes induced by the nonpumping D372I mutation. Assignments of many of the features of the redox difference spectrum have been reported previously (40), although not with great certainty. The two spectroscopic regions of interest are 1,710–1,770 cm−1, where the (C=O) stretching band of protonated Asp/Glu residues are located, and 1,670–1,710 cm−1, where the protonated heme propionates absorb. Previously, absorption bands associated with protonated heme propionates have been assigned at 1,698 cm−1 in the reduced enzyme and at 1,692 and 1,708 cm−1 in the oxidized enzyme (40). Fig. 3 shows the double-difference spectrum between the reduced-minus-oxidized FTIR difference spectra of the WT and D372I mutant. There is a prominent trough/peak at 1,698/1,692 cm−1. A reasonable interpretation is that the increased magnitude of this feature may result from a higher population of a protonated heme propionate whose absorption shifts on reduction of the enzyme from 1,698 to 1,692 cm−1.
Fig. 3.
Reduced-minus-oxidized double-difference spectrum of WT-minus-D372I. The reduced-minus-oxidized spectrum of the D372I mutant is subtracted from that of the WT cytochrome ba3. The trough/peak feature at 1,698/1,692 cm−1 is attributable to changes to the one or more protonated heme propionates, and this is more prominent in the D372I mutant. Δa.u. is the difference in absorption units.
Fig. 3 also shows a peak/trough at 1,745/1,736 cm−1, which reports a perturbation by the D372I mutation in the spectral region where protonated acidic residues (Asp and Glu) absorb. There are redox-induced perturbations in this region of the spectrum for both the WT and a number of mutants (Fig. S4), and the alterations appear to be attributable to changes in the environments of several residues. Efforts to identify the origins of these spectral features were not successful (SI Materials and Methods).


Four amino acid residues make direct hydrogen-bonding contact with the propionate groups of heme a3 (Fig. 1): Y133 and R449 to propionate-D and H376 and D372 to propionate-A. His376 has additional interactions with both E126II and N377. In addition, a water molecule bridges between propionates-A and -D, and it is also hydrogen-bonded to His283, one of the ligands to CuB (Fig. 1). The data presented here show that Y133, H376, D372, E126II, and N377 are not essential for either oxidase activity or proton pumping. Point mutations in each of these residues, examined individually, maintain function. Indeed, the high activity and proton pumping exhibited by D372A, D372V, and H376F indicate that having either an ionizable group or one that can form a hydrogen bond to propionate-A is not essential at these positions. Similarly, the Y133F and Y133W mutants show that a hydrogen bond to propionate-D is not important at this position.
The unexpected finding in the current work is that two mutants, D372I and H376N, do not pump protons or do so with a substantially diminished stoichiometry, despite exhibiting high catalytic turnover (i.e., 50% and 70% of the WT). It is noted that either the isolation procedure or the procedure to reconstitute these enzymes into liposomes for the purpose of measuring proton pumping may perturb mutants that are less stable than the WT. This may explain some of the variability in the proton-pumping results (i.e., no pumping vs. some pumping) observed with these two mutants. The important point is that of all the mutants examined, only preparations of D372I and H376N consistently exhibit compromised proton pumping (with most showing no pumping), and both of these sites are adjacent to propionate-A of heme a3. Although neither D372 nor H376 is essential, these specific mutations (D372I and H376N) perturb the proton pump. This is particularly striking because both D372V and H376F pump protons as well as the WT.
The reduced-minus-oxidized UV-visible difference spectrum of D372I (Fig. S2) is not perturbed, but the FTIR difference spectra (Fig. 3 and Fig. S4) indicate some perturbations attributable to D372I. The most salient change in the double-difference spectrum (Fig. 3) attributable to the D372I mutation is the enhanced trough/peak at 1,698/1,692 cm−1, which could result from an alteration to the environment around propionate-A of heme a3 (40). However, further studies and definitive band assignments are needed to test this hypothesis.

Previous Results from Mutating the Equivalent of D372 in A-Family Oxygen Reductases.

Although an aspartate (or aspartyl) residue near propionate-A of the active-site heme is not a conserved feature of heme-copper oxygen reductases, it is commonly found in A-family enzymes, and thus has attracted attention. Mutagenesis experiments have been performed on the equivalent residues: D399 in the P. denitrificans aa3-type oxygen reductase, D407 in the R. sphaeroides aa3-type oxygen reductase, and D407 in the Escherichia coli bo3-type oxygen reductase (14, 27, 4346). The D399N mutation in the P. denitrificans oxidase does not significantly change the properties of the oxidase, whereas the D399L mutant has only ∼7% activity and does not pump protons (41). These data are similar to those reported in the current work with the T. thermophilus enzyme. Qian et al. (42) concluded that D407 in the R. sphaeroides enzyme does not play an essential role based on a number of mutations that did not significantly influence the enzyme properties compared with the WT. Similar results were also obtained within the cytochrome bo3 from E. coli (43). The results of the current work with the D372 mutations also show a variety of phenotypes, ranging from a lack of proper assembly to only minor changes compared with the WT, as previously observed. However, only D372I in the T. thermophilus enzyme decouples the proton pump.

Efforts to Test for a Backleak of Protons to the Active-Site.

During the normal operation of the WT enzyme, there must be a gating mechanism to prevent protons from slipping through the exit channel, to propionate-A, and then to the enzyme active-site. One question addressed in the current work is whether any of the D372 or H376 mutations result in the perturbation of the gating mechanism preventing protons from the outside from being consumed in making water at the active site. To examine this, the normal flux through the KB channel was blocked by the T312V mutation, and second mutations were introduced at D372 or H376. The data, shown in Table S1, are suggestive but not compelling that some backward flux of protons is observed. The activity of T312V is nil (<1%), whereas two of the double mutants have 3–4% activity. Whether this means that protons are delivered through the exit channel to the active site remains unproven and will require further study.

Importance of H282, T302, and D287 for Proton Pumping.

Whether H282, a ligand to CuB, helps convey pumped protons from the KB channel to the “exit pathway” cannot be determined from the current work. Because H282 is always hydrogen-bonded to either a serine or threonine, the T302V mutant was examined to eliminate this hydrogen bond. The severe loss of activity and spectroscopic perturbation caused by the T302V mutation reflects previous reports of placing a hydrophobic residue in the equivalent location in several A-family heme-copper oxygen reductases (T352 in E. coli cytochrome bo3 and in R. sphaeroides cytochrome aa3 as well as T344 in P. denitrificans cytochrome aa3) (4447).
The role of D287 requires further examination. The low respiratory control and apparent low proton pumping of D287N are suggestive of an important role for D287 in the proton exit pathway. However, the location of D287 at the interface between subunits I and II and the salt bridge with R225 leave open the possibility that less specific structural changes beyond the site of the mutation may be involved. It is noted that the optical spectra of the hemes are not perturbed by the D287N mutation (Fig. S2). Further work will be needed to determine the role of D287.

Implications About the Identity of the Proton Loading Site.

The current experiments show that a structural perturbation adjacent to propionate-A (i.e., D372I, H376N) can result in decoupling the proton pump. The decoupling is very mutation-specific. The lack of proton pumping by D372I, for example, seems to be attributable to a packing problem unique to isoleucine (but not valine) at this position. Although the lack of proton pumping could be attributable to perturbations elsewhere in the protein, it is reasonable that the elimination of proton pumping is likely attributable to a perturbation to propionate-A of heme a3. This leaves us with the puzzle of why other mutations, such as D372V and H376F, which alter the hydrogen bonding to propionate-A, have little effect. If propionate-A were the proton loading site of the pump, one might think that the pump would be sensitive to mutating the residues that are hydrogen-bonded to the carboxyl group.
One intriguing possibility (22, 23) is that the proton loading site is not propionate-A itself but a cluster of residues centered about the water molecule that is hydrogen-bonded to both propionate-A and propionate-D of heme a3 as well as to His283, one of the CuB ligands (Fig. 1). This arrangement appears to be universal to all the proton-pumping heme-copper oxygen reductases [38 (and supporting information therein)]. This might explain why the proton pump is robust and not easily altered by any single mutation. Presumably, the unique structural perturbations by the D372I or H376N mutation alter the hydrogen bond pattern of this water molecule, which is also hydrogen-bonded to propionate-A. This model is also consistent with the decoupling caused by some R481 mutations adjacent to propionate-D of the active site heme o3 of cytochrome bo3 from E. coli (30). Further experiments, including determining the structure of the D372I mutant enzyme, as well as computational work, will be required to test the role of this conserved hydrogen-bonded cluster centered about a water molecule.

Materials and Methods

Protein expression, purification, and steady-state activity measurements were as previously described (48). Proton pumping was measured following reconstitution of the purified oxidase into proteoliposomes (7) and measurement of the acidification of the external solution concomitant with a known number of turnovers using cytochrome c552 as the electron donor. The acidification was monitored in most cases using phenol red as a pH indicator (7); however, in some cases, a sensitive pH electrode was used (7). Perfusion-induced FTIR difference spectroscopy was done with a sample attached to an attenuated total reflectance diamond prism, as described previously (49, 50). Details are provided in SI Materials and Methods.

Supporting Information

Supporting Information (PDF)
Supporting Information


MM Pereira, FL Sousa, AF Veríssimo, M Teixeira, Looking for the minimum common denominator in haem-copper oxygen reductases: Towards a unified catalytic mechanism. Biochim Biophys Acta 1777, 929–934 (2008).
J Hemp, RB Gennis, Diversity of the heme-copper superfamily in archaea: Insights from genomics and structural modeling. Results Probl Cell Differ 45, 1–31 (2008).
I Belevich, DA Bloch, N Belevich, M Wikström, MI Verkhovsky, Exploring the proton pump mechanism of cytochrome c oxidase in real time. Proc Natl Acad Sci USA 104, 2685–2690 (2007).
P Brzezinski, RB Gennis, Cytochrome c oxidase: Exciting progress and remaining mysteries. J Bioenerg Biomembr 40, 521–531 (2008).
MM Pereira, M Santana, M Teixeira, A novel scenario for the evolution of haem-copper oxygen reductases. Biochim Biophys Acta 1505, 185–208 (2001).
J Hemp, et al., Comparative genomics and site-directed mutagenesis support the existence of only one input channel for protons in the C-family (cbb3 oxidase) of heme-copper oxygen reductases. Biochemistry 46, 9963–9972 (2007).
A Kannt, et al., Electrical current generation and proton pumping catalyzed by the ba3-type cytochrome c oxidase from Thermus thermophilus. FEBS Lett 434, 17–22 (1998).
M Wikström, MI Verkhovsky, Mechanism and energetics of proton translocation by the respiratory heme-copper oxidases. Biochim Biophys Acta 1767, 1200–1214 (2007).
M Toledo-Cuevas, B Barquera, RB Gennis, M Wikström, JA García-Horsman, The cbb3-type cytochrome c oxidase from Rhodobacter sphaeroides, a proton-pumping heme-copper oxidase. Biochim Biophys Acta 1365, 421–434 (1998).
E Arslan, A Kannt, L Thöny-Meyer, H Hennecke, The symbiotically essential cbb(3)-type oxidase of Bradyrhizobium japonicum is a proton pump. FEBS Lett 470, 7–10 (2000).
H Han, et al., Adaptation of aerobic respiration to low O2 environments. Proc Natl Acad Sci USA 108, 14109–14114 (2011).
I Belevich, et al., Initiation of the proton pump of cytochrome c oxidase. Proc Natl Acad Sci USA 107, 18469–18474 (2010).
VR Kaila, M Verkhovsky, G Hummer, M Wikström, Prevention of leak in the proton pump of cytochrome c oxidase. Biochim Biophys Acta 1777, 890–892 (2008).
I Belevich, MI Verkhovsky, Molecular mechanism of proton translocation by cytochrome c oxidase. Antioxid Redox Signal 10, 1–29 (2008).
S Yoshikawa, et al., Proton pumping mechanism of bovine heart cytochrome c oxidase. Biochim Biophys Acta 1757, 1110–1116 (2006).
H Michel, The mechanism of proton pumping by cytochrome c oxidasex127e [comments]. Proc Natl Acad Sci USA 95, 12819–12824 (1998).
PEM Siegbahn, MRA Blomberg, Proton pumping mechanism in cytochrome c oxidase. J Phys Chem A 112, 12772–12780 (2008).
PE Siegbahn, MR Blomberg, Energy diagrams and mechanism for proton pumping in cytochrome c oxidase. Biochim Biophys Acta 1767, 1143–1156 (2007).
MH Olsson, PE Siegbahn, MR Blomberg, A Warshel, Exploring pathways and barriers for coupled ET/PT in cytochrome c oxidase: A general framework for examining energetics and mechanistic alternatives. Biochim Biophys Acta 1767, 244–260 (2007).
R Sugitani, ES Medvedev, AA Stuchebrukhov, Theoretical and computational analysis of the membrane potential generated by cytochrome c oxidase upon single electron injection into the enzyme. Biochim Biophys Acta 1777, 1129–1139 (2008).
VR Kaila, MI Verkhovsky, G Hummer, M Wikström, Glutamic acid 242 is a valve in the proton pump of cytochrome c oxidase. Proc Natl Acad Sci USA 105, 6255–6259 (2008).
V Daskalakis, SC Farantos, V Guallar, C Varotsis, Regulation of electron and proton transfer by the protein matrix of cytochrome c oxidase. J Phys Chem B 115, 3648–3655 (2011).
HY Chang, J Hemp, Y Chen, JA Fee, RB Gennis, The cytochrome ba3 oxygen reductase from Thermus thermophilus uses a single input channel for proton delivery to the active site and for proton pumping. Proc Natl Acad Sci USA 106, 16169–16173 (2009).
MA Sharpe, S Ferguson-Miller, A chemically explicit model for the mechanism of proton pumping in heme-copper oxidases. J Bioenerg Biomembr 40, 541–549 (2008).
DM Popovic, IV Leontyev, DG Beech, AA Stuchebrukhov, Similarity of cytochrome c oxidases in different organisms. Proteins 78, 2691–2698 (2010).
DV Makhov, DM Popović, AA Stuchebrukhov, Improved density functional theory/electrostatic calculation of the His291 protonation state in cytochrome C oxidase: Self-consistent charges for solvation energy calculation. J Phys Chem B 110, 12162–12166 (2006).
M Wikström, Mechanism of proton translocation by cytochrome c oxidase: A new four-stroke histidine cycle. Biochim Biophys Acta 1458, 188–198 (2000).
JA Fee, DA Case, L Noodleman, Toward a chemical mechanism of proton pumping by the B-type cytochrome c oxidases: Application of density functional theory to cytochrome ba3 of Thermus thermophilus. J Am Chem Soc 130, 15002–15021 (2008).
TK Das, et al., Active site structure of the aa3 quinol oxidase of Acidianus ambivalens. Biochim Biophys Acta 1655, 306–320 (2004).
A Puustinen, M Wikström, Proton exit from the heme-copper oxidase of Escherichia coli. Proc Natl Acad Sci USA 96, 35–37 (1999).
S Buschmann, et al., The structure of cbb3 cytochrome oxidase provides insights into proton pumping. Science 329, 327–330 (2010).
HJ Lee, L Ojemyr, A Vakkasoglu, P Brzezinski, RB Gennis, Properties of Arg481 mutants of the aa3-type cytochrome c oxidase from Rhodobacter sphaeroides suggest that neither R481 nor the nearby D-propionate of heme a3 is likely to be the proton loading site of the proton pump. Biochemistry 48, 7123–7131 (2009).
VR Kaila, V Sharma, M Wikström, The identity of the transient proton loading site of the proton-pumping mechanism of cytochrome c oxidase. Biochim Biophys Acta 1807, 80–84 (2011).
PEM Siegbahn, MRA Blomberg, ML Blomberg, A theoretical study of the energetics of proton pumping and oxygen reduction in cytochrome oxidase. J Phys Chem B 107, 10946–10955 (2003).
C Koutsoupakis, T Soulimane, C Varotsis, Probing the Q-proton pathway of ba3-cytochrome c oxidase by time-resolved Fourier transform infrared spectroscopy. Biophys J 86, 2438–2444 (2004).
VM Luna, Y Chen, JA Fee, CD Stout, Crystallographic studies of Xe and Kr binding within the large internal cavity of cytochrome ba3 from Thermus thermophilus: Structural analysis and role of oxygen transport channels in the heme-Cu oxidases. Biochemistry 47, 4657–4665 (2008).
LM Hunsicker-Wang, RL Pacoma, Y Chen, JA Fee, CD Stout, A novel cryoprotection scheme for enhancing the diffraction of crystals of recombinant cytochrome ba3 oxidase from Thermus thermophilus. Acta Crystallogr D Biol Crystallogr 61, 340–343 (2005).
T Tiefenbrunn, et al., High resolution structure of the ba3 cytochrome c oxidase from Thermus thermophilus in a lipidic environment. PLoS ONE 6, e22348 (2011).
PR Rich, M Iwaki, Methods to probe protein transitions with ATR infrared spectroscopy. Mol Biosyst 3, 398–407 (2007).
P Hellwig, T Soulimane, G Buse, W Mäntele, Electrochemical, FTIR, and UV/VIS spectroscopic properties of the ba(3) oxidase from Thermus thermophilus. Biochemistry 38, 9648–9658 (1999).
U Pfitzner, et al., Tracing the D-pathway in reconstituted site-directed mutants of cytochrome c oxidase from Paracoccus denitrificans. Biochemistry 39, 6756–6762 (2000).
J Qian, et al., Aspartate-407 in Rhodobacter sphaeroides cytochrome c oxidase is not required for proton pumping or manganese binding. Biochemistry 36, 2539–2543 (1997).
JW Thomas, A Puustinen, JO Alben, RB Gennis, M Wikström, Substitution of asparagine for aspartate-135 in subunit I of the cytochrome bo ubiquinol oxidase of Escherichia coli eliminates proton-pumping activity. Biochemistry 32, 10923–10928 (1993).
JW Thomas, LJ Lemieux, JO Alben, RB Gennis, Site-directed mutagenesis of highly conserved residues in helix VIII of subunit I of the cytochrome bo ubiquinol oxidase from Escherichia coli: An amphipathic transmembrane helix that may be important in conveying protons to the binuclear center. Biochemistry 32, 11173–11180 (1993).
JR Fetter, et al., Possible proton relay pathways in cytochrome c oxidase. Proc Natl Acad Sci USA 92, 1604–1608 (1995).
JP Hosler, et al., Polar residues in helix VIII of subunit I of cytochrome c oxidase influence the activity and the structure of the active site. Biochemistry 35, 10776–10783 (1996).
U Pfitzner, et al., Cytochrome c oxidase (heme aa3) from Paracoccus denitrificians: Analysis of mutations in putative proton channels of subunit I. J Bioenerg Biomembr. 30, 89–97 (1998).
Y Chen, L Hunsicker-Wang, RL Pacoma, E Luna, JA Fee, A homologous expression system for obtaining engineered cytochrome ba3 from Thermus thermophilus HB8. Protein Expr Purif 40, 299–318 (2005).
AS Vakkasoglu, JE Morgan, D Han, AS Pawate, RB Gennis, Mutations which decouple the proton pump of the cytochrome c oxidase from Rhodobacter sphaeroides perturb the environment of glutamate 286. FEBS Lett 580, 4613–4617 (2006).
K Yang, et al., Glutamate 107 in subunit I of the cytochrome bd quinol oxidase from Escherichia coli is protonated and near the heme d/heme b595 binuclear center. Biochemistry 46, 3270–3278 (2007).

Information & Authors


Published in

Go to Proceedings of the National Academy of Sciences
Proceedings of the National Academy of Sciences
Vol. 109 | No. 14
April 3, 2012
PubMed: 22431640


Submission history

Published online: March 19, 2012
Published in issue: April 3, 2012


  1. cytochrome c oxidase
  2. exit channel
  3. bioenergetics
  4. respiratory chain


*This Direct Submission article had a prearranged editor.



Hsin-Yang Chang
Department of Biochemistry and
Sylvia K. Choi
Center for Biophysics and Computational Biology, University of Illinois, Urbana, IL 61801; and
Ahmet Selim Vakkasoglu
Center for Biophysics and Computational Biology, University of Illinois, Urbana, IL 61801; and
Ying Chen
Department of Molecular Biology, The Scripps Research Institute, La Jolla, CA 92037
James Hemp
Department of Biochemistry and
James A. Fee1 [email protected]
Department of Molecular Biology, The Scripps Research Institute, La Jolla, CA 92037
Robert B. Gennis1 [email protected]
Department of Biochemistry and
Center for Biophysics and Computational Biology, University of Illinois, Urbana, IL 61801; and


To whom correspondence may be addressed. E-mail: [email protected] or [email protected].
Author contributions: H.-Y.C. and R.B.G. designed research; H.-Y.C., S.K.C., A.S.V., Y.C., and J.H. performed research; H.-Y.C., S.K.C., A.S.V., J.A.F., and R.B.G. analyzed data; and H.-Y.C., A.S.V., J.H., and R.B.G. wrote the paper.

Competing Interests

The authors declare no conflict of interest.

Metrics & Citations


Note: The article usage is presented with a three- to four-day delay and will update daily once available. Due to ths delay, usage data will not appear immediately following publication. Citation information is sourced from Crossref Cited-by service.

Citation statements



If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

Cited by


    View Options

    View options

    PDF format

    Download this article as a PDF file


    Get Access

    Login options

    Check if you have access through your login credentials or your institution to get full access on this article.

    Personal login Institutional Login

    Recommend to a librarian

    Recommend PNAS to a Librarian

    Purchase options

    Purchase this article to access the full text.

    Single Article Purchase

    Exploring the proton pump and exit pathway for pumped protons in cytochrome ba3 from Thermus thermophilus
    Proceedings of the National Academy of Sciences
    • Vol. 109
    • No. 14
    • pp. 5135-5548







    Share article link

    Share on social media