Scientific Report 2006

where the subscripts in and out refer, respectively, to matrix and the intermembrane space of the mitochondrion or, in bacteria, to the cytoplasm and the periplasmic space. The free energy of dioxygen reduction is thus captured as a proton gradient; the out side is positive and the in side is negative.

During the past 5 decades, enormous progress has been realized in understanding the chemical properties of the 3 redox centers, and the enzyme from several different sources has been crystalized and its structure determined at resolutions ranging from about 3 to 1.8 Å. Moreover, much has been learned about the pathways of electron transfer within the enzyme and the detailed mechanisms whereby the oxygen molecule is reduced to 2 water molecules. The outstanding questions pertain to the flow of protons into the enzyme from its in side, across the hydrophobic core of the membrane, to exit on its out side. The mystery lies in how all this scalar chemistry comes together to “pump” 4 protons across the membrane. Although the enzyme has been examined by using virtually every available spectroscopic technique, no one has addressed directly the pathways of those protons becoming either water or part of the proton gradient.

Much of the past work was done with enzymes in a single clade typified by the mitochondrial enzyme (derived from an ancient bacterium) and with enzymes isolated from common bacteria, notably Rhodobacter sphaeroides, Paraccocus denitrificans, and Escherichia coli (the quinol oxidase). The enzymes from these sources are highly similar in amino acid sequence, 3-dimensional structure, electron-transfer paths, mechanism of oxygen reduction, and, most likely, mechanisms of proton pumping. Our research is based on a 1988 description of a highly sequence-divergent form of the enzyme from Thermus thermophilus, cytochrome ba₃, that represents a distinct clade of enzymes widely distributed among archaebacteria. Respectively, these clades represent A- and B-type oxidases.

We recently developed a homologous expression system for cytochrome ba₃. This system allows easy purification of the enzyme in amounts of 2–3 mg/L of culture medium by using an N-terminal heptahistidine tag on subunit I. The recombinant enzyme is equally active with native, wild-type protein, and the results of a 2.3-Å x-ray structure determination, done in collaboration with C.D. Stout, Department of Molecular Biology, revealed the expected details at full occupancy and at least one notable surprise.

All the A-type oxidases have a glutamic acid residue within the hydrophobic interior of the enzyme that is thought to be at the “end” of the D pathway of proton transport and close to the Fe_a3-Cu_B site of dioxygen reduction (Fig. 1A). Mutation of this residue to, for example, glutamine blocks all but a small percentage of the enzyme’s electron-transfer activity, and infrared studies indicate that this residue “senses” changes in the chemical structure of the dioxygen reduction site. Indeed, scientists think that glutamate 286 actually donates the pumped proton to the exit part of the molecule, becoming deprotonated with a pK_a of about 9.4. However, no direct evidence exists for this notion, and because glutamate 286 resides in a highly hydrophobic region of the structure, its pK_a most likely is much higher.

Fig. 1. Cross-eyed stereo view of the Fea3-CuB binuclear active-site structures (leftmost Fe and Cu) of the cytochrome aa3 from R sphaeroides (A) and the cytochrome ba3 from T thermophilus (B), emphasizing the overlapping position of the glutamate 286 (Glu-238) in R sphaeroides and the isoleucine 235 (Ile-235) in T thermophilus. Under the influence of oxygen reduction, protons most likely enter the protein from the upper left of the figure and exit at the lower right.

The surprise in the structure of cytochrome ba₃ is that an isoleucine residue is isopositional with glutamate 286 as isoleucine 235, as shown in Fig. 1B. We mutated this residue to both a glutamine and a glutamate residue in the cytochrome with no apparent loss of electron-transfer activity. To determine if the substituted glutamate residue can be used to monitor changes in the Fe_a3-Cu_B pair, as it does in the A-type oxidases, we have initiated an infrared study of these mutant proteins in collaboration with R. Gennis, University of Illinois, Urbana-Champaign, Illinois, and J. Heberle, Jülich Research Center, Jülich, Germany. How these studies will advance our understanding of proton-pumping mechanisms remains unclear. What is clear is that cytochrome ba₃ provides new openings to explore the mechanism of the cytochrome oxidases.

Publications

Chen, Y., Hunsicker-Wang, L.M., Pacoma, R.L., Luna, E., Fee, J.A. A homologous expression system for obtaining engineered cytochrome ba₃ from Thermus thermophilus HB8. Protein Expr. Purif. 40:299, 2005.

Farver, O., Chem, Y., Fee, J.A., Pecht, I. Electron transfer among the Cu_A-, heme b- and a₃-centers of Thermus thermophilus cytochrome ba₃. FEBS Lett. 580:3417, 2006.

Hunsicker-Wang, L.M., Pacoma, R.L., Chen, Y., Fee, J.A., Stout, C.D. A novel cryoprotection scheme for enhancing the diffraction of crystals of recombinant cytochrome ba₃ oxidase from Thermus thermophilus. Acta Crystallogr. D Biol. Crystallogr. 61(Pt. 3):340, 2005.