Charge translocation coupled to electron injection into oxidized cytochrome c oxidase from Paracoccus denitrificans.

Electrons were discretely injected into oxidized cytochrome c oxidase in liposomes by laser flash excitation of bound ruthenium [II] bispyridyl, and the membrane potential was recorded by time-resolved electrometry. Membrane potential is generated in a fast phase when an electron is transferred from the excited dye, via the CuA center, to heme a at a relative dielectric depth d inside the membrane [Zaslavsky, D., Kaulen, A. D., Smirnova, I. A., Vygodina, T., and Konstantinov, A. A. (1993) FEBS Lett. 336, 389-393]. Subsequently, membrane potential may develop further in a slower event, which is due to proton transfer into the enzyme from the opposite side of the membrane [Ruitenberg, M., Kannt, A., Bamberg, E., Ludwig, B., Michel, H., and Fendler, K. (2000) Proc. Natl. Acad. Sci. U.S.A. 97, 4632-4636]. Here, we confirm that injection of the first electron into the fully oxidized cytochrome c oxidase from Paracoccus denitrificans is associated with a fast electrogenic 11 micros phase, but there is no further electrogenic phase up to 100 milliseconds when special care is taken to ensure that only fully oxidized enzyme is present initially. A slower electrogenic 135 micros phase only becomes apparent and grows in amplitude upon increasing the number of light flashes. This occurs in parallel with a decrease in amplitude of the 11 micros phase and correlates with the number of enzyme molecules that are already reduced by one electron before the flash. The electrogenic 135 micros phase does not appear with increasing flash number in the K354M mutant enzyme, where electron and proton transfer into the binuclear center is delayed. We conclude that the 135 micros phase, and its associated proton uptake, take place on electron injection into enzyme molecules where the binuclear heme a3-CuB site is already reduced by one electron, and that it is accompanied by oxidation of heme a with a similar time constant. Reduction of heme a is not associated with electrogenic proton uptake into the enzyme, neither in the fully oxidized nor in the one-electron-reduced enzyme. The extent of the electrogenic 135 micrcos phase also rules out the possibility that reduction of the binuclear center by the second electron would be coupled to proton translocation in addition to the electrogenic uptake of a proton.

Heme-copper oxidases with modified D- and K-pathways are yet efficient proton pumps.

The cytochrome aa(3)-type quinol oxidase from the archaeon Acidianus ambivalens and the ba(3)-type cytochrome c oxidase from Thermus thermophilus are divergent members of the heme-copper oxidase superfamily of enzymes. In particular they lack most of the key residues involved in the proposed proton transfer pathways. The pumping capability of the A. ambivalens enzyme was investigated and found to occur with the same efficiency as the canonical enzymes. This is the first demonstration of pumping of 1 H(+)/electron in a heme-copper oxidase that lacks most residues of the K- and D-channels. Also, the structure of the ba(3) oxidase from T. thermophilus was simulated by mutating Phe274 to threonine and Glu278 to isoleucine in the D-pathway of the Paracoccus denitrificans cytochrome c oxidase. This modification resulted in full efficiency of proton translocation albeit with a substantially lowered turnover. Together, these findings show that multiple structural solutions for efficient proton conduction arose during evolution of the respiratory oxidases, and that very few residues remain invariant among these enzymes to function in a common proton-pumping mechanism.

Electron and proton transfer in the arginine-54-methionine mutant of cytochrome c oxidase from Paracoccus denitrificans.

Arginine 54 in subunit I of cytochrome c oxidase from Paracoccus denitrificans interacts with the formyl group of heme a. Mutation of this arginine to methionine (R54M) dramatically changes the spectral properties of heme a and lowers its midpoint redox potential [Kannt et al. (1999) J. Biol. Chem. 274, 37974-37981; Lee et al. (2000) Biochemistry 39, 2989-2996; Riistama et al. (2000) Biochim. Biophys. Acta 1456, 1-4]. During anaerobic reduction of the mutant enzyme, a small fraction of heme a is reduced first along with heme a(3), while most of heme a is reduced later. This suggests that electron transfer is impaired thermodynamically due to the low redox potential of heme a but that it still takes place from Cu(A) via heme a to the binuclear site as in wild-type enzyme, with no detectable bypass from Cu(A) directly to the binuclear site. Consistent with this, the proton translocation efficiency is unaffected at 1 H(+)/e(-) in the mutant enzyme, although turnover is strongly inhibited. Time-resolved electrometry shows that when the fully reduced enzyme reacts with O(2), the fast phase of membrane potential generation during the P(R )()--> F transition is unaffected by the mutation, whereas the slow phase (F --> O transition) is strongly decelerated. In the 3e(-)-reduced mutant enzyme heme a remains oxidized due to its lowered midpoint potential, whereas Cu(A) and the binuclear site are reduced. In this case the reaction with O(2) proceeds via the P(M) state because transfer of the electron from Cu(A) to the binuclear site is delayed. The single phase of membrane potential generation in the 3e(-)-reduced mutant enzyme, which thus corresponds to the P(M)--> F transition, is decelerated, but its amplitude is comparable to that of the P(R)--> F transition. From this we conclude that the completely (4e(-)) reduced enzyme is fully capable of proton translocation.

The role of the D- and K-pathways of proton transfer in the function of the haem-copper oxidases.

The X-ray structures of several haem-copper oxidases now at hand have given important constraints on how these enzymes function. Yet, dynamic data are required to elucidate the mechanisms of electron and proton transfer, the activation of O(2) and its reduction to water, as well as the still enigmatic mechanism by which these enzymes couple the redox reaction to proton translocation. Here, some recent observations will be briefly reviewed with special emphasis on the functioning of the so-called D- and K-pathways of proton transfer. It turns out that only one of the eight protons taken up by the enzyme during its catalytic cycle is transferred via the K-pathway. The D-pathway is probably responsible for the transfer of all other protons, including the four that are pumped across the membrane. The unique K-pathway proton may be specifically required to aid O-O bond scission by the haem-copper oxidases.

Proton translocation by cytochrome c oxidase can take place without the conserved glutamic acid in subunit I.

A glutamic acid residue in subunit I of the heme-copper oxidases is highly conserved and has been directly implicated in the O(2) reduction and proton-pumping mechanisms of these respiratory enzymes. Its mutation to residues other than aspartic acid dramatically inhibits activity, and proton translocation is lost. However, this glutamic acid is replaced by a nonacidic residue in some structurally distant members of the heme-copper oxidases, which have a tyrosine residue in the vicinity. Here, using cytochrome c oxidase from Paracoccus denitrificans, we show that replacement of the glutamic acid and a conserved glycine nearby lowers the catalytic activity to <0.1% of the wild-type value. But if, in addition, a phenylalanine that lies close in the structure is changed to tyrosine, the activity rises more than 100-fold and proton translocation is restored. Molecular dynamics simulations suggest that the tyrosine can support a transient array of water molecules that may be essential for proton transfer in the heme-copper oxidases. Surprisingly, the glutamic acid is thus not indispensable, which puts important constraints on the catalytic mechanism of these enzymes.