Abstract
The O2 reduction site of cytochrome c oxidase (CcO), comprising iron (Fea3) and copper (CuB) ions, is probed by x-ray structural analyses of CO, NO, and CN- derivatives to investigate the mechanism of the complete reduction of O2. Formation of the
derivative contributes to the trigonal planar coordination of
and displaces one of its three coordinated imidazole groups while a water molecule becomes hydrogen bonded to both the CN- ligand and the hydroxyl group of Tyr244. When O2 is bound to
, it is negatively polarized (
), and expected to induce the same structural change induced by CN-. This structural change allows
to receive three electron equivalents nonsequentially from
,
, and Tyr-OH, providing complete reduction of O2 with minimization of production of active oxygen species. The proton-pumping pathway of bovine CcO comprises a hydrogen-bond network and a water channel which extend to the positive and negative side surfaces, respectively. Protons transferred through the water channel are pumped through the hydrogen-bond network electrostatically with positive charge created at the Fea center by electron donation to the O2 reduction site. Binding of CO or NO to
induces significant narrowing of a section of the water channel near the hydrogen-bond network junction, which prevents access of water molecules to the network. In a similar manner, O2 binding to
is expected to prevent access of water molecules to the hydrogen-bond network. This blocks proton back-leak from the network and provides an efficient gate for proton-pumping.
Keywords: cell respiration, heme copper protein, membrane protein, respiratory inhibitor, x-ray structural analysis
Cytochrome c oxidase (CcO) is the terminal oxidase of cellular respiration, which reductively converts molecular oxygen (O2) to two water molecules coupled to a proton-pumping process. CcO contains four redox active metal sites, heme a, heme a3, CuA, and CuB. Heme a3 and CuB together form the O2 reduction site. Electrons for O2 reduction are transferred from cytochrome c in the positive side space to the O2 reduction site via CuA and heme a (1).
Time-resolved resonance Raman spectroscopy shows that in the process of complete reduction of O2 either by fully reduced CcO or by the mixed valence CcO which contains only two reducing equivalents, the initial intermediate assignable as the O2-bound form (
) is observed at 571 cm-1 (2–4) and the second intermediate assignable as the oxide-bound form (Fea3 = O2-) has a band at 804 cm-1 (5 and 6). A more detailed description of the identification of the second intermediate by resonance Raman spectroscopy is provided in SI Text.
The fact that the O2-bound form is detectable as the initial intermediate is an indication that the bound O2 is unexpectedly stable in the O2 reduction site. The two redox active metals of the O2 reduction sites would readily reduce O2 to the peroxide level. The structure of the second intermediate (which has a resonance Raman band at 804 cm-1) strongly suggests that O2 has received four electron equivalents to form two oxide species (Fea3 = O2- and CuB-OH-) in this intermediate. It has been proposed that one electron equivalent, in addition to the two equivalents from
and one equivalent from
, is supplied from the hydroxyl group of Tyr244 (7). The electron donation to O2 at Fea3 with appropriate timing is needed for O2 reduction without release of active oxygen species. For example, a delay in donation of the electron equivalent for complete O2 reduction would increase production and release of hydroxide radicals (or oxide radical anions) from the O2 reduction site.
The resonance Raman Fe-O2 stretch spectrum of the initial intermediate is closely similar to the Fe-O2 stretch spectrum of hemoglobins and myoglobins (2–4) in which the bound O2 is essentially in the one-electron reduced (
) state (8). Thus, the transition from the initial intermediate to the second intermediate could be called a nonsequential three electron reduction process. Furthermore, the Raman band positions of the intermediate species are not influenced by the overall oxidation state of CcO before the initiation of the O2 reduction as described above. Electron transfer from the CuA/heme a sites is not involved in this O2 reduction process. These observations indicate that upstream electron pressure does not influence the mechanism of complete O2 reduction to the oxide level (2O2-).
An appropriate structural basis has not been established to rationalize the stability of the O2-bound form and to describe how the nonsequential three electron reduction of
at Fea3 is facilitated. The x-ray structural analyses have shown that
has trigonal planar coordination geometry, suggesting that
is a poor electron donor as well as a poor ligand acceptor (9). Tyr244, which is covalently bound to one of the three histidine ligands to CuB, is a possible electron/proton donor to O2 at Fea3. However, the relative location of the Fea3 and the OH group of Tyr244 shown in the high resolution x-ray structure of the fully reduced CcO indicates that the O2 molecule which is bound to Fea3 would not readily form a direct hydrogen bond with the Tyr244-OH group. Furthermore, the imidazole group of His240, which is covalently linked to Tyr244, strongly restricts movement of the phenol group towards the Fea3 site. These x-ray results are consistent with the unexpectedly stable O2-bound form demonstrated by resonance Raman analyses (2–4). Furthermore, x-ray structural analyses of the O2-bound form are desirable in order to test this proposal, since it is impossible to predict the influence of ligand binding on the structure of the O2 reduction site by only examining the x-ray structure of the ligand-free CcO. Unfortunately, the O2-bound form is too unstable to capture by crystallization. Intermediate species between
and Fea3 = O2- are not detectable in the O2 reduction process under normal enzymatic turnover conditions, as described above (5). Therefore, structural and functional analyses of the O2 reduction site using O2 analogues such as CO, NO, and CN- as probes are needed to provide insights into the mechanism of O2 reduction catalyzed by CcO.
It has been well established that the proton-pumping process is not coupled to the O2 reduction process (from [
,
] to [
,
, Tyr244 radical]). Instead, it is coupled to each of the four electron-transfer steps (from [
,
, Tyr244 radical] to [
,
]) from CuA/heme a sites to the O2 reduction site after completion of the O2 reduction (10). These results indicate that the free energy associated with the activation of the metal centers is stored to provide high electron affinity at the metal sites and to preserve the Tyr244 radical species.
CcO has three possible proton-transfer pathways, which are known as the K-, D- and H-pathways (9, 11–13). X-ray structural and mutational analyses of bovine heart CcO strongly suggest that H-pathway pumps protons (14 and 15). The pathway includes a hydrogen-bond network and a water channel which extend to the positive and negative side surfaces, respectively. A peptide bond is located near the positive side end for blocking proton back-leakage from the positive side space (16 and 17). However, an appropriate structure has not yet been identified which would provide unidirectionality to the proton transfer from the negative side up to the peptide bond.
We report on the effects of the respiratory inhibitors, CO, NO, and CN- on the x-ray structure of the fully reduced CcO. These investigations identify critical roles for the O2 reduction site in stabilizing the O2-bound form, facilitating a complete reduction of O2 with minimal production of active oxygen species, and providing a gate for effective proton pumping.
Results
Statistics of the x-ray structural analyses and values of the coordination geometry parameters of the heme a3-CuB site determined in the x-ray structures are summarized in Table S1 and Table S2.
CO and NO Derivatives of the Fully Reduced CcO.
X-ray structural analyses of the fully reduced CO- and NO-bound CcO (with CO and NO bound to
) were performed at 280 K and 100 K, respectively. The essentially perpendicular coordination structure (164 degree in the Fe-C-O angle) and the bent end-on coordination structure (131 degree in the Fe-N-O angle) were obtained for CO at 280 K and for NO at 100 K, respectively (Fig. 1 A and C). The distance between the
-ligands and
are 2.5–2.7 Å. This distance indicates that direct interactions between
and the
-ligands are unlikely. Furthermore, interactions between the Tyr244-OH group and the bound ligands are strongly restricted by the covalent linkage of Tyr244 to the imidazole of His240. These x-ray structures of the O2-analogue derivatives indicate the structural factors which contribute to the stability of the
species inferred from Resonance Raman analyses (2–4).
Fig. 1.
The x-ray structures of the O2 reduction sites of the CO-, NO-, and CN--bound forms of fully reduced CcO. The ligand-binding structures are shown in FO-FC maps. The digits without letters in the FO-FC maps indicate numbering of fixed water molecules. The structural characteristics are indicated in schematic representations. (A) The CO-bound form at 280 K. (B) The CO-bound form at 100 K. (C) The NO-bound form at 100 K. (D) A stereo view of the NO-bound form. (E) The CN--bound form at 100 K. (F) A stereo view of the CN--bound form.
The x-ray structure of the CO derivative determined at 100 K (Fig. 1B) indicates that CO is bound to
in a side-on fashion after being photo-dissociated from
. The metal-to-carbon and metal-to-oxygen atom distances of 2.4 and 2.7 Å indicate that the CuB-CO bond is rather weak. The 3.0 Å distance between CO and
suggests that no significant bonding interaction exists between CO and
. (In SI Text and Fig. S1, the CO-binding function of CuB is discussed in detail with a comparison of the present results with existing spectroscopic data.) The weak CO-binding structure suggests that the CuB site is suitable for reversible O2 binding. The CO molecule bound to
is located 3.0 Å from
: This is an indication that
controls the supply of O2 molecules to
without forming the μ-peroxide intermediate species (Fe3+-O-O-Cu2+), which could produce active oxygen species.
The structure of the O2 reduction site with CO at CuB at100 K indicates that there is enough space for migration of the CO molecule from CuB to Fea3. The stable CuB-CO structure suggests that migration of CO (and thus O2) from
to
is controlled by certain protein movements which would be essentially “frozen” at 100 K. Similar conformational freezing is detectable in the process of reduction of the fully oxidized CcO by hydrated electrons in the experiments conducted at SPring-8 (18).
CN- Derivatives of the Fully Reduced CcO.
The FO-FC map of the O2 reduction site of the fully reduced CN--bound form at 2.05 Å resolution indicates that CN- is located roughly equidistant (2.3 Å) from
and
(Fig. 1E). It has been established by infrared spectroscopy that CN- but not hydrogen cyanide (HCN), binds to the fully reduced enzyme (19 and 20). His290 is one of the three histidine imidazoles which are ligated to
in a trigonal planar coordination structure. Upon binding of CN- to
, His290 is displaced and moves to a location 2.8 Å away from
, a distance which prevents it from coordinating to
. In this arrangement, the cyanide nitrogen atom and the two imidazole nitrogen atoms form a new trigonal planar cuprous structure which is roughly perpendicular to the heme a3 plane (Fig. 1F) in contrast to the trigonal planar structure in the NO-bound form which is parallel to the heme plane (Fig. 1D). In addition, a water molecule (Water 510) enters the site and becomes hydrogen bonded to both CN- and Tyr244 (Fig. 1E). Tyr244, which is covalently bonded to His240, is located at the end of the K-pathway. The bound CN- induces a small but significant translational shift of the plane of heme a3 (∼0.4 Å), which is coupled to shifts of the helices VIII and IX as shown in Fig. S2A. The translational shift is necessary for formation of the hydrogen bond between Water 510 and the bound CN-. Water 510 which appears upon binding of CN- is likely to be supplied from a water storage site located 7.1 Å away from Water 510 site (as shown in Fig. S2B). Thus, Water 510 is unlikely to represent the water molecule produced by the O2 reduction reaction.
The electron density maps in Fig. 1 A, B, C, and E are contoured at 7.5, 5.5, 6.4, and 5.8σ levels, respectively, in which essentially the same cage size of Water 272 is obtained.
Effects of Inhibitor Bindings and Oxidation States on the Conformation of the H-Pathway.
The proton-pumping system of bovine CcO includes the H-pathway. This pathway is composed of a hydrogen-bond network and a water channel which extend to the positive and negative side surfaces respectively, as schematically shown in Fig. 2B. The water channel located near helix X has four cavities in the fully reduced state, each of which is large enough to contain one to three water molecules (Fig. 2B, right illustration). Fig. 2A shows top views (from the positive side) of the portion of helix X extending from Val380 to Met383 in the fully reduced (blue), fully oxidized (red) and fully reduced CO-bound (yellow) forms. These views include heme a, heme a3, and cavities indicated by the dotted surfaces. The portion which is located one turn closer to the positive side surface than the portion shown in Fig. 2A includes His376 and His378 which are the ligands of hemes a3 and a, respectively. A side view and a top stereo view of this region are shown in Fig. S3 A and B, respectively.
Fig. 2.
Conformational change of helix X which occurs upon ligand bindings to
. (A) Atomic models of the segment of helix X from Val380 to Met383 including hemes a and a3. The models show the structures of the fully reduced (blue), the fully reduced CO-bound (yellow), and the fully oxidized (red) forms. The dotted surfaces denote cavities detectable in different oxidation and ligand-binding states near the two helix turns, shown with the same colors as used for the atomic models. The cavities are calculated with a probe radius of 1.3 Å. The green dotted line shows the location of the water channel in this region. (B) Schematic representations of the conformational changes induced by oxidation and ligand-binding states. (Left) the fully oxidized forms, (Center) the fully reduced CO- and NO-bound forms and (Right) the fully reduced ligand-free and CN--bound forms. The red and blue spheres represent the positions of the fixed water molecules, the locations of which are dependent upon the oxidation state of heme a. The black spheres represent the positions of the fixed water molecules independently of the oxidation states of the metal sites.
Upon binding of CO to the fully reduced CcO, a portion of helix X (from Val380 to Ser382) undergoes a significant conformational change which is identical to the change that occurs upon oxidation as previously reported (14). The conformational change driven by the ligand binding eliminates the cavity which is detectable in the fully reduced state (shown with a blue dotted surface in Fig. 2A). The structural details related to the movement of the main chain of helix X are shown in Fig. S3C. A similar conformational change in helix X is also obtained upon binding of NO. On the other hand, binding of CN- to the fully reduced CcO did not induce a significant conformational change in helix X. It has been shown that the carboxyl group of Asp51 and the hydroxyfarnesylethyl group of heme a both undergo redox-coupled conformational changes (9 and 14). However, ligand binding did not induce any significant conformational changes of the two functional groups. The observed conformational changes are schematically summarized in Fig. 2B. These results strongly suggest that the conformation of helix X is controlled by the oxidation and ligand-binding states of heme a3, while those of Asp51 and the hydroxyfarnesylethyl group of heme a are controlled by the oxidation state of heme a.
Discussion
Complete O2 Reduction with Minimization of Active Oxygen Species.
As described above, the O2 molecule bound to
is negatively polarized like CN-. Therefore, it is reasonable to predict that the structural changes which would occur at the O2 reduction site upon O2 binding are similar to those which occur upon CN- binding (Fig. 1E and Fig. S2A). The resulting O2-bound structure suggests that three electron equivalents can be transferred nonsequentially from
, the hydroxyl group of Tyr244, and
to
at
for complete reduction of
. The negative charge of the CN- ligand appears to be critical for inducing these structural changes, since CO and NO do not significantly influence the structure of the O2 reduction site, as shown in Fig. 1 A–C.
Before the introduction of the water molecule, the coordination geometry of O2 must be similar to the geometry of the O2-bound forms of hemoglobins and myoglobins, as has been suggested by data acquired from time-resolved resonance Raman spectroscopy (2–4). The stability of the
form indicates the absence of any direct electron transfer from CuB to the bound O2 before introduction of Water 510. The direct electron transfer would provide a peroxide-bound form,
, which could produce active oxygen species. For example, homolytic cleavage of the O-O bond would give
and release an oxide radical. Thus, the relative location of Fea3 and CuB is also critical for the nonsequential three electron reduction to minimize the production of active oxygen species.
The absence of Water 510 also stabilizes the globin-type O2-bound form, since it is not possible for Tyr244OH to interact with the bound
. Water 510 triggers all of the structural rearrangements observed upon CN- binding. All of these structural rearrangements must be concerted in order for the nonsequential three electron donation to the
bound at
. Water 510 is supplied from the water site located 7.1 Å away from the Tyr244-OH group. The water transfer rate corresponds to the decay rate of the initial intermediate (
). The second intermediate (P form) receives electrons from heme a coupled with proton pumping as described in the introduction section (10).
O2 reduction without release of active oxygen species was previously proposed on the basis of resonance Raman measurements to occur via nonsequential three electron reduction of
at
(21). This proposal was made without the benefit of structural data. The x-ray structural results presented herein provide a possible structural basis for the O2 reduction mechanism shown in Fig. 3 as follows: O2 transferred through the O2 path in subunits III and I (22) is trapped by
(Fig. 3B). With appropriate timing (vide infra), O2 is transferred to
to form the O2-bound species (
) (Fig. 3C). The rearrangement of the O2 reduction site after entry of the water molecule enables the facile nonsequential three electron reduction of
, from
,
, and the Tyr244-OH group (Fig. 3D) to provide the second intermediate, known as the P form (Fig. 3E). Although the OH group of Tyr244 seems most likely to be the donor of one of the three electron equivalents for the complete
reduction as described above, other candidates such as penta-valent iron (
) and trivalent copper (
) cannot be excluded from consideration (23 and 24). Identification of the electron donor site remains a challenging research subject. (An additional discussion of the structure of the P form is provided in SI Text).
Fig. 3.
A schematic representation of the O2 reduction in CcO. The proposed (experimentally undetectable) intermediate is shown in the shadowed area. The formation rate of the intermediate is much slower than the corresponding rate of disappearance.
Since O2 is a strong ligand, the resulting structure is expected to be somewhat different from that of the CN--bound form shown in Fig. 1E. For example, the ligand-Fea3 distance would be shorter and the conformation of helix X would be similar to the conformation observed when strong ligands are bound. However, it is likely that the negatively-charged O2 introduces structural changes which are similar to the structural changes occurring upon CN- binding. It is equally possible that after the water molecule is introduced, a superoxide molecule,
, is released from the
species and trapped by
, Water 510, and
. The
species causes the conformation of helix X to be retained in the same state prior to release of
. In either case,
receives three electrons nonsequentially and helix X eliminates the large water cavity in the water channel.
The x-ray structure of the fully reduced CN--bound form of Rhodobacter sphaeroides CcO (Rs CcO) at 2.2 Å resolution has been published recently (25). The Protein Data Bank (PDB) structure (3FYI) shows that the analogous histidine residue corresponding to bovine His290 is released from CuB, as it is in the case of bovine CcO, although the conformational change of the imidazole group induced by CN- is not mentioned specifically in the report (25). (Further comparisons of the report with the present results are given in SI Text. A possible function of His290 (in bovine number) in bacterial quinol oxidase is discussed in SI Text.)
Conformational Changes in Helix X as a Gate for Proton Pumping in the H-Pathway.
It has been proposed that during the process of proton pumping by bovine CcO, protons are transferred by thermal motion of water molecules throughout the water channel from the negative side space to the Arg38 residue which is located at the negative side end of the hydrogen-bond network in the H-pathway. The proton-pumping (proton-active transport) process of CcO is then driven by electrostatic repulsion between the net positive charge created upon oxidation of heme a and the protons located in the hydrogen-bond network in the H-pathway (14).
The present results show that the conformation of helix X is controlled by heme a3. Upon oxidation or strong ligand binding to heme a3, the water channel is significantly narrowed by elimination of one of the cavities near its junction point to the hydrogen-bond network as shown in Fig. 2. The conformational change would significantly obstruct the proton transfer driven by movement of the water molecules to essentially close the channel within the timescale of enzymatic turnover (closed state). Preliminary results suggest that helix X of the P and F forms, which appear during the enzymatic turnover as the second and third intermediates, adopts the same conformation as that of the fully oxidized form (Fig. S4). The water channel of the O2-bound form is also likely to be in the closed state since O2 is a strong ligand to
. Therefore, under normal enzymatic turnover, water molecules in the negative side space are expected to be readily accessible to Arg38 (or the channel is in the open state) only when heme a3 is in the ligand-free reduced state.
The roles of the water channel in the proton pumping through H-pathway may be summarized as follows: When both metal sites (Fea3 and CuB) are reduced, the channel is in the open state and collects protons from the negative side space. O2 is transferred through the O2 transfer pathway in subunits III and I (22) and trapped at
in a side-on fashion as described above. This arrangement is maintained until the proton collection is completed in the hydrogen-bond network. Upon transfer of O2 from
to
, the channel is closed to block the proton back-leakage from the hydrogen-bond network. These protein movements controlling the CO migration from CuB to Fea3, as described above, are expected to contribute to the coupling between the CuB site and the proton collection area. After the nonsequential three electron reduction of
at the O2 reduction site which generates the P form, four electrons are transferred sequentially from heme a to the O2 reduction site, forming sequentially F (one electron equivalent lower oxidation state than P), O (fully oxidized form under turnover conditions), E (one electron-reduced form of O), and R (both Fea3 and CuB are reduced) forms. Each electron-transfer step is coupled with the process of proton pumping through the H-pathway. During these four electron-transfer steps, the water channel is retained in the closed state to prevent back-leakage of protons. To obtain a measure of overall energy coupling efficiency (pumping proton/electron ratio) of unity (10), approximately four equivalents of protons should be incorporated from the negative side space to the area accessible to Arg38 (and to one of the propionate groups of heme a) during the opening of the water channel.
In the open state of the water channel, Arg38 is in proton equilibrium with the negative space, since protons are transferred by thermal motion of water molecules through the water channel. In the closed state, the portion of the H-pathway between the closed water cavity and the peptide bond near the positive side end of the hydrogen-bond network (which includes Arg38 and the heme a propionate) is isolated from the negative side space. This portion of the H-pathway includes many proton-accepting or donating groups, such as the guanidino group of Arg38, the propionate of heme a, serine OH groups, and water molecules (Fig. 3B). It is well known that the effective proton affinities of these groups are subject to sensitive influence by their local environments within the protein matrix. Therefore, this section of the H-pathway is expected to have the capacity to accept four-proton equivalents or more. When four electron equivalents are transferred to the P-state to form the (
,
) state, in which each electron transfer is coupled with one equivalent of proton pumping, the section of the H-pathway is in a four-proton-depleted state. When the channel is opened, the four-proton-depleted sites are protonated by protons transferred by the thermal motion of water molecules in the water channel to obtain the original protonation state, which is in equilibrium with the negative side space.
The free-energy change required for the active transport of each proton equivalent up to the peptide bond during the catalytic cycle is likely to be dependent on the location and the stability of the protonated group in the section of the H-pathway. On the other hand, the free-energy change in each of the four downhill electron transfers from heme a to the O2 reduction site is dependent on the oxidation state and ligand-binding environment of the O2 reduction site. These factors would cooperate in providing roughly equal proton-pumping efficiency in each of the four steps of the proton transfer coupled electron transfer from heme a to the O2 reduction site. Further structural and functional studies on the O2 reduction site and H-pathway are required for elucidation of the coupling mechanism between the proton-active transport and the electron transfer for the O2 reduction site.
The hydrogen-bond network in the H-pathway is equipped with two functional systems, the peptide bond and the water channel. These systems prevent the back-leakage of protons at each end of the hydrogen-bond network. The coexistence of these two systems is likely to cooperatively contribute to increasing the overall gating efficiency. Thus, the O2 reduction site is not only the site for the O2 reduction (or the electron sink) but also functions to control the proton gating function.
It has been established that the decay of the initial intermediate (
) is one order of magnitude faster in the fully reduced CcO than in the mixed valence CcO, although the Raman band positions of the initial two intermediate species are not influenced by the overall oxidation state (26). These results suggest that oxidation state of heme a influences the decay rate without affecting the O2 reduction mechanism. On the other hand it has been reported that an appropriate overall electron-transfer rate is required in order to obtain the maximal proton-pumping efficiency (H+/e-) (27). Therefore, the fully reduced CcO is unlikely to provide the maximal proton-pumping efficiency upon oxidation. However, the fact that the resonance Raman bands of the initial and second intermediates are not influenced by the oxidation state of the heme a/CuA site suggests that the structures of the O2 reduction site of these intermediate state are not influenced significantly by the oxidation state of the heme a/CuA site. Furthermore, the oxidation state change of heme a induces only small conformational changes near the water channel as shown in Fig. 2, Fig. S3 A and B. Thus, the open/closed transition in the water channel induced by the conformational changes in helix X upon various ligand bindings to
would not be influenced by the oxidation state of the heme a/CuA site. That is, the water channel is closed upon O2 binding to
also in the mixed valence CcO which could exert the maximal proton-pumping efficiency. As described above, the oxidation state of heme a/CuA site affects the kinetics of the O2 reduction. Thus, O2 transfer from
to
is also likely to be faster under higher electron pressure (or in the fully reduced state). Then, the water channel may be closed before completion of the proton collection to the hydrogen-bond network in H-pathway, which lowers the proton-pumping efficiency.
It has been proposed for the bacterial CcO that the D-pathway, one of the proton-transfer pathways from the negative side space to the O2 reduction site, transfers protons for pumping as well as for production of water molecules. Structural evidence for participation of the D-pathway in the proton-pumping process in bovine CcO has not been obtained thus far.
Materials and Methods
All absorption spectral measurements and x-ray diffraction experiments were performed at 100 K, unless otherwise noted, using crystals of bovine heart CcO prepared as previously described (28) including crystallization as the final step. The present purification method provides the oxidized “fast” form characterized by the Soret maximum and the CN--binding rate (28).
Other details related to sample preparation, x-ray structural analyses, and absorption spectral measurements are provided in SI Text and Fig. S1A–I.
Supplementary Material
Acknowledgments.
This work is supported in part by a Grant-in-Aid for Scientific Research on Priority Areas 16087206 (to T.T.) and 16087208 (to S.Y.), the Targeted Protein Research Program (to K.M., K.S,-I, and S.Y.) and the Global Center of Excellence Program (to S.Y.) each provided by the Japanese Ministry of Education, Culture, Sports, Science and Technology. S.Y. is a Senior Visiting Scientist in the RIKEN Harima Institute.
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/cgi/content/full/0910410107/DCSupplemental.
Data deposition: The atomic coordinates and structure factors have been deposited in the Protein Data Bank, www.pdb.org (PDB ID codes 3AG1, 3AG2, 3AG3, and 3AG4).
References
- 1.Ferguson-Miller S, Babcock GT. Heme/copper terminal oxidases. Chem Rev. 1996;96:2889–2908. doi: 10.1021/cr950051s. [DOI] [PubMed] [Google Scholar]
- 2.Varotsis C, Woodruff WH, Babcock T. Time-resolved Raman detection of ν(Fe-O) in an early intermediate in the reduction of O2 by cytochrome oxidase. J Am Chem Soc. 1990;112:1297. [Google Scholar]
- 3.Han S, Ching Y, Rousseau DL. Primary intermediate in the reaction of oxygen with fully reduced cytochrome c oxidase. Proc Natl Acad Sci USA. 1990;87:2491–2495. doi: 10.1073/pnas.87.7.2491. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ogura T, Takahashi S, Shinzawa-Itoh K, Yoshikawa S, Kitagawa T. Observation of the FeII-O2 stretching Raman band for cytochrome oxidase compound A at ambient temperature. J Am Chem Soc. 1990;112:5630–5631. [PubMed] [Google Scholar]
- 5.Ogura T, et al. Time-resolved resonance Raman elucidation of the pathway for dioxygen reduction by cytochrome c oxidase. J Am Chem Soc. 1993;115:8527–8536. [Google Scholar]
- 6.Rousseau D, Han S. Time-resolved resonance Raman spectroscopy of intermediates in cytochrome c oxidase. Methods Enzymol. 2002;354:351–368. doi: 10.1016/s0076-6879(02)54028-3. [DOI] [PubMed] [Google Scholar]
- 7.Proshlyyakov DA, Pressler MA, Babcock GT. Dioxygenactivation and bond cleavage by mixed-valence cytochrome c oxidase. Proc Natl Acad Sci USA. 1998;95:8020–8025. doi: 10.1073/pnas.95.14.8020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Potter WT, Tucker MP, Houtchens RA, Caughey WS. Oxygen infrared spectra of oxyhemoglobins and oxymyoglobins. Evidence of two major O2 structures. Biochemistry. 1987;26:4699–4707. doi: 10.1021/bi00389a016. [DOI] [PubMed] [Google Scholar]
- 9.Yoshikawa S, et al. Redox-coupled crystal structural changes in bovine heart cytochrome c oxidase. Science. 1998;280:1723–1729. doi: 10.1126/science.280.5370.1723. [DOI] [PubMed] [Google Scholar]
- 10.Bloch D, et al. The catalytic cycle of cytochrome c oxidase is not the sum of its two halves. Proc Natl Acad Sci USA. 2004;101:529–533. doi: 10.1073/pnas.0306036101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.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;94:9085–9095. doi: 10.1073/pnas.94.17.9085. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yoshikawa S, et al. Proton pumping mechanism of bovine heart cytochrome c oxidase. Biochim Biophys Acta. 2006;1757:1110–1116. doi: 10.1016/j.bbabio.2006.06.004. [DOI] [PubMed] [Google Scholar]
- 13.Ostermeier C, Harrenga A, Ermler U, Michel H. Structure at 2.7 Å resolution of the Paracoccus denitrificans two-subunit cytochrome c oxidase complexed with an antibody Fv fragment. Proc Natl Acad Sci USA. 1997;94:10547–10553. doi: 10.1073/pnas.94.20.10547. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Tsukihara T, et al. The low-spin heme of cytochrome c oxidase as the driving element of the proton-pumping process. Proc Natl Acad Sci USA. 2003;100:15304–15309. doi: 10.1073/pnas.2635097100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Shimokata K, et al. The proton pumping pathway of bovine heart cytochrome c oxidase. Proc Natl Acad Sci USA. 2007;104:4200–4205. doi: 10.1073/pnas.0611627104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Perrin CL. Proton exchange in amides: surprises from simple systems. Acc Chem Res. 1989;22:268–275. [Google Scholar]
- 17.Kamiya K, Boero M, Tateno M, Shiraishi K, Oshiyama A. Possible mechanism of proton transfer through peptide groups in the H-pathway of the bovine cytochrome c oxidase. J Am Chem Soc. 2007;129:9663–9673. doi: 10.1021/ja070464y. [DOI] [PubMed] [Google Scholar]
- 18.Aoyama H, et al. A peroxide bridge between Fe and Cu ions in the O2 reduction site of fully oxidized cytochrome c oxidase could suppress the proton pump. Proc Natl Acad Sci USA. 2009;106:2165–2169. doi: 10.1073/pnas.0806391106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Yoshikawa S, O’Keeffe DH, Caughey WS. Investigation of cyanide as an infrared probe of hemoprotein ligand binding sites. J Biol Chem. 1985;260:3518–3528. [PubMed] [Google Scholar]
- 20.Yoshikawa S, Mochizuki M, Zhao X-J, Caughey WS. Effects of overall oxidation state on infrared spectra of heme a3 cyanide in bovine heart cytochrome c oxidase. J Biol Chem. 1995;270:4270–4279. doi: 10.1074/jbc.270.9.4270. [DOI] [PubMed] [Google Scholar]
- 21.Ogura T, et al. Time-resolved resonance Raman evidence for tight coupling between electron transfer and proton pumping of cytochrome c oxidase upon the change from the FeV oxidation level to the FeIV oxidation level. J Am Chem Soc. 1996;118:5443–5449. [Google Scholar]
- 22.Shinzawa-Itoh K, et al. Structures and physiological roles of 13 integral lipids of bovine heart cytochrome c oxidase. EMBO J. 2007;26:1713–1725. doi: 10.1038/sj.emboj.7601618. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wagner WD, Nakamoto K. Resonance Raman spectra of nitridoiron (V) porphyrin intermediates produced by laser photolysis. J Am Chem Soc. 1989;111:1590–1598. [Google Scholar]
- 24.Hamilton GA, et al. Trivalent copper, superoxide, and galactose oxidase. J Am Chem Soc. 1978;100:1899–1912. [Google Scholar]
- 25.Qin L, et al. Redox dependent conformational changes in cytochrome c oxidase suggests a gating mechanism for proton uptake. Biochemistry. 2009;48:5121–5130. doi: 10.1021/bi9001387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Han SH, Ching YC, Rousseau DL. Cytochrome c oxidase: decay of the primary oxygen intermediate involves direct electron transfer from cytochrome a. Proc Natl Acad Sci USA. 1990;87:8408–8412. doi: 10.1073/pnas.87.21.8408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Capitanio N, et al. Factors affecting the H+/e- stoichiometry in mitochondrial cytochrome c oxidase: Influence of the rate of electron flow and transmembrane ΔpH. Biochemistry. 1996;35:10800–10806. doi: 10.1021/bi9606509. [DOI] [PubMed] [Google Scholar]
- 28.Mochizuki M, et al. Quantitative reevaluation of the redox active sites of crystalline bovine heart cytochrome c oxidase. J Biol Chem. 1999;274:33403–33411. doi: 10.1074/jbc.274.47.33403. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.



