Oxygen-linked equilibrium CuB-CO species in cytochrome ba3 oxidase from thermus thermophilus. Implications for an oxygen channel ar the CuB site.

We report the first study of O(2) migration in the putative O(2) channel of cytochrome ba(3) and its effect to the properties of the binuclear heme a(3)-Cu(B) center of cytochrome ba(3) from Thermus thermophilus. The Fourier transform infrared spectra of the ba(3)-CO complex demonstrate that in the presence of 60-80 micro m O(2), the nu(C-O) of Cu(B)1+-C-O at 2053 cm(-1) (complex A) shifts to 2045 cm(-1) and remains unchanged in H(2)O/D(2)O exchanges and in the pH 6.5-9.0 range. The frequencies but not the intensities of the C-O stretching modes of heme a(3)-CO (complex B), however, remain unchanged. The change in the nu(C-O) of complex A results in an increase of k(-2), and thus in a higher affinity of Cu(B) for exogenous ligands. The time-resolved step-scan Fourier transform infrared difference spectra indicate that the rate of decay of the transient Cu(B)1+-CO complex at pH 6.5 is 30.4 s(-1) and 28.3 s(-1) in the presence of O(2). Similarly, the rebinding to heme a(3) is slightly affected and occurs with k(2) = 26.3 s(-1) and 24.6 s(-1) in the presence of O(2). These results provide solid evidence that in cytochrome ba(3), the ligand delivery channel is located at the Cu(B) site, which is the ligand entry to the heme a(3) pocket. We suggest that the properties of the O(2) channel are not limited to facilitating ligand diffusion to the active site but are extended in controlling the dynamics and reactivity of the reactions of ba(3) with O(2) and NO.

Cytochrome ba 3 from Thermus thermophilus is a member of the large family of structurally related heme-copper oxidases (1,2). It catalyzes both the four-electron reduction of O 2 to H 2 O, converting the energy of this reaction to a transmembrane proton motive force, and the two-electron reduction of NO to N 2 O (1-4). Based on the crystal structure, the enzyme contains a homodinuclear copper (Cu A ), a low-spin heme b, and a heme a 3 -Cu B binuclear center (1). The ba 3 -oxidase retains the electron transport chain functional under low oxygen concentration in the medium. In both structurally characterized aa 3 -type heme-copper oxidases, three possible O 2 channels have been suggested (5,6). One of them leads from a trapped lipid pocket in subunit III to the active site and contains Glu-278 (Paracoccus denitrificans numbering), a residue near Cu B (5,6). In the presence of three O 2 channels in aa 3 -type oxidases, it is not known yet whether a facilitated oxygen channel is necessary, because the O 2 concentrations normally far exceed the O 2 affinity of the enzyme; thus, excess O 2 would not be rate-limiting. The proposed oxygen input channel in ba 3 contains Ile-235 instead of Glu-278 (conserved to aa 3 oxidases), which optimizes the formation of a hydrophobic pore (1,2). The evolutionary development of an optimized oxygen channel is appropriate for organisms such as T. thermophilus ba 3 oxidase, the archaeal Sulfolobus acidocaldarius aa 3 -quinol oxidase, and Natronomonas pharaonics, which grow under low oxygen tension and at high temperatures (2). However, no data exist in the literature to demonstrate the nature of the conformational changes that occur in the binuclear center in the presence of O 2 in the channel. Because of the unusual ligand-binding and kinetic properties of the binuclear center, cytochrome ba 3 oxidase is unique among the heme-copper oxidases in that it is susceptible to a detailed kinetic analysis of its ligand dynamics (4,7). The binding of CO to the binuclear center of ba 3 follows that found in all heme-copper oxidases and proceeds according to the Scheme 1 (7)(8)(9)(10)(11)(12).
In our previous work (7), we identified the C-O stretching mode of the equilibrium Cu B 1ϩ -CO species (complex A) at 2053 cm Ϫ1 and concluded that the environment in the binuclear center does not alter the protonation state of the Cu B histidine ligands. Understanding the conformational transitions that are associated with protonation/deprotonation of labile residues is essential because ionizable groups whose pK a values are near physiological pH are involved in proton uptake or release. A hydrogen-bonded connectivity between the propionates of heme a 3 , Asp-372, and H 2 O was also reported. Accordingly, plausible mechanisms of proton pathway(s) directly associated with the propionates of the heme a 3 redox center and the proton-labile side chain of Asp-372 were suggested.
The nature of heme-copper oxidases is to bring in O 2 through ligand-entry channels to the binuclear center and then remove protons and H 2 O from the active site. Because it has been proposed that 1) Cu B is a way-stop for ligand entry to heme a 3 (8) and 2) the O 2 channel is located at the Cu B site, we sought to determine the properties of the binuclear center by applying our FTIR approach (7,9,11) to study the CO bound cytochrome ba 3 at room temperature in the presence of low-oxygen concentrations in the medium (ϳ70 M), preventing spontaneous replacement of CO by O 2 . In cytochrome ba 3 , the exceptionally high affinity for CO binding to Cu B (K 1 Ͼ 10 4 ) has allowed us to perform such experiments. We have also used time-resolved * This work was supported in part by the Greek Ministry of Education. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. step-scan Fourier transform infrared spectroscopy (FTIR) 1 to investigate the ligand dynamics subsequent to CO photolysis at room temperature in the presence of ϳ70 M O 2 and compare the results with those obtained in the absence of O 2 , which is essential for elucidating the unique chemical mechanisms of the redox processes catalyzed by the enzyme and the dynamics of the binuclear center.

EXPERIMENTAL PROCEDURES
Cytochrome ba 3 was isolated from T. thermophilus HB8 cells according to previously published procedures (3). The samples used for the FTIR measurements had an enzyme concentration of ϳ1 mM and were placed in a desired buffer (pD 5.5-6.5, MES; pD 7.5, HEPES; pD 8.5-9.5 CHES). The pD solutions prepared in D 2 O buffers were measured by using a pH meter and assuming pD ϭ pH (observed) ϩ 0.4. Dithionite reduced samples were exposed to 1 atmosphere CO (1 mM) in an anaerobic cell to prepare the CO adduct and transferred to a tightly sealed FTIR cell under anaerobic conditions (pathlength l ϭ 15 and 25 m for the experiments in H 2 O and D 2 O, respectively). CO gas was obtained from Messer (Frankfurt, Germany) and isotopic CO ( 13 CO) was purchased from Isotec (Miamisburg, OH). The ba 3 carbonmonoxy/O 2 adduct was prepared by addition of aliquots of 5 l of an oxygen-saturated buffer solution (1.2 mM O 2 ) to make the final O 2 concentration of 60 -80 M. The 532 nm pulse from a Continuum neodymium-yttrium aluminum garnet (Nd-YAG) laser (7-ns width, 3 Hz) was used as a pump light (10 mJ/pulse) to photolyze the ba 3 -CO complex. The time-resolved stepscan FTIR spectra were obtained with spectral resolution of 8 cm Ϫ1 and 100-s time resolution for the 0 -75-ms measurements. A total of 10 co-additions per retardation data point was collected. Changes in intensity were recorded with a mercury cadmium telluride detector, amplified (dc-coupled), and digitized with a 200-kHz, 16-bit, analog-todigital converter. Blackman-Harris three-term apodization function with 32 cm Ϫ1 phase resolution and the Mertz phase correction algorithm were used. Difference spectra were calculated as ⌬A ϭ Ϫlog(intensity of sample/intensity of reference). The detailed experimental set-up for the time-resolved step-scan FTIR has been described previously (7,9,11). The rate constants for each phase of the decay of the Cu B 1ϩ -CO complex and for CO rebinding to heme a 3 were calculated, assuming first-order kinetics, with three-parameter exponential fits to the experimental data. Optical absorbance spectra were recorded before and after FTIR measurements to assess sample stability with a PerkinElmer Lamda 20 UV-visible spectrometer.

RESULTS AND DISCUSSION
The FTIR spectrum of the CO-bound cytochrome ba 3 complex at neutral pH exhibits peaks at 1967, 1973, 1982, and 2053 cm Ϫ1 (Fig. 1, trace A). In the 13 C 16 O derivative, these peaks shift to 1923,1928,1937, and 2007 cm Ϫ1 , respectively (Fig. 1,  trace B). The peaks at 1967, 1973, and 1982 cm Ϫ1 have been assigned to the C-O stretching modes of heme a 3 -CO (complex B), and the peak at 2053 cm Ϫ1 to the C-O stretching mode of Cu B 1ϩ -CO (complex A). As shown in Fig. 1, trace C, addition of 70 M buffered O 2 to the ba 3 -CO complex leads to a shift of the CO mode of Cu B 1ϩ to 2045 cm Ϫ1 . Addition of oxygen-saturated buffer (Ͼ300 M) causes the spontaneous replacement of CO by O 2 , as evidenced by the full disappearance of both the heme a 3 -CO modes at 1967, 1973, and 1982 cm Ϫ1 and the Cu B -CO mode at 2053 cm Ϫ1 . Confirmation that the newly developed 2045 cm Ϫ1 mode is CO-sensitive is shown in the 13 CO spectrum (Fig. 1, trace D), where it shifts to 1999 cm Ϫ1 . These data also demonstrate a significant growth of the 2045 cm Ϫ1 conforma-  and thus on the frequency of (C-O) (7). Because we see no change in the frequency of the 2045 cm Ϫ1 mode, we conclude that the Cu B -His environment is very rigid and not subject to conformational transitions that are associated with protonation/deprotonation events of the Cu B -His ligands. We suggest that low concentrations of O 2 (60 -80 M) in the putative O 2 channel play a role in modulating the structure of the Cu B -CO complex that facilitates the transition from 2053 to 2045 cm Ϫ1 . We attribute the transition to an increased electron density in the CO antibonding orbitals that results in weakening the C-O bond strength (lower C-O ). It remains to be determined whether the O 2 is hydrogen-bonded to one of the Cu B -His ligands, causing the weakening of the C-O bond, or the newly developed conformer is the result of a dynamic effect in the delivery channel generating a conformational change in the Cu B environment. It becomes intriguing to speculate that the properties of the O 2 channel are not limited to facilitate ligand diffusion to the active site but are extended in controlling the dynamics and reactivity of the reactions of ba 3 with O 2 and nitric oxide and other ligands, including cyanide. The increased value of k Ϫ2 indicates that Cu B undergoes structural changes to behave as an efficient trap for CO. We suggest that this behavior is extended to the physiological function of the enzyme. This way, conformational changes associated with Cu B occur, facilitating the entry and the later coordination of O 2 to the binuclear center. Fig. 3A shows the time-resolved step-scan FTIR difference spectra (t d ϭ 0 -75 ms, 8 cm Ϫ1 spectral resolution) of fully reduced ba 3 -CO subsequent to CO photolysis by a nanosecond laser pulse (532 nm). Upon photolysis, CO is transferred from heme a 3 to Cu B . It should be noted that the Cu B -CO complex (complex A) is not photolabile and thus remains a spectator in the photodynamic events occurring to complex B. Under our 8 cm Ϫ1 spectral resolution, the heme a 3 Fe-CO peaks at 1967, 1973, and 1982 cm Ϫ1 are not resolved; thus, a single negative peak at 1976 cm Ϫ1 indicates the photolyzed heme a 3 2ϩ -C-O complex. The positive peak that appears at 2053 cm Ϫ1 is attributed to the C-O stretch ( C-O ) of the transient Cu B 1ϩ -CO complex, and its frequency is the same as that obtained at pD 8.5 (7) and that of the equilibrium Cu B 1ϩ -CO complex. At early times (0 -3000 s), the intensity of the 1976 and 2053 cm Ϫ1 modes in the transient difference spectra remains unchanged, suggesting that dissociation of CO from Cu B does not occur on this time scale. At later times (3-75 ms), there is a decrease in intensity of the 2053 cm Ϫ1 mode that is accompanied by an increased intensity of the 1976 cm Ϫ1 mode. At 75 ms after CO photolysis, the intensities of both the 1976 and 2053 cm Ϫ1 modes are almost diminished. The intensity ratio (ϳ2) of the Fe-CO/Cu B -CO remains constant for all data points, indicating that no significant fraction of CO escapes the binuclear center. This is also consistent with both the low-temperature experiments (21 K) and those obtained at pD 8.5 at room temperature (7,8). The high signal-to-noise ratio in the time-resolved FTIR difference spectra has allowed us to monitor the decay and reappearance of the 2053 and 1976 cm Ϫ1 modes, respectively.  1, 2, 4, 5, 6, 8, 10, 12, 15, 18, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, and 75 ms after CO photolysis, in the absence (A) and in the presence (B) of O 2 . The spectral resolution was 8 cm Ϫ1 , the time resolution was 100 s, and 10 co-additions were collected per data point. The excitation wavelength was 532 nm (10 mJ/pulse), and each data set is the average of three measurements. Insets, kinetic analysis of the 2053 cm Ϫ1 (Cu B -CO) (squares) and 1976 cm Ϫ1 (Fe-CO) (circles) modes versus time after CO photolysis in the absence (A) and in the presence (B) of O 2 . ⌬A was measured from the intensity of the corresponding modes at times between 0 and 75 ms after the photolysis of CO from heme a 3 . The curves are three-parameter exponential fits to the experimental data, according to first-order kinetics.
ing to heme a 3 is 26.3 s Ϫ1 . Both rates are 10% lower than those observed at pD 8.5 (7). The pH/pD dependence of the rates of decay of the Cu B complex and the rebinding to heme a 3 will be presented elsewhere. 2 Fig . 3B shows the time-resolved step-scan FTIR difference spectra (t d ϭ 0 -75 ms, 8 cm Ϫ1 spectral resolution) of fully reduced ba 3 -CO in the presence of 70 M O 2 after CO photolysis by a nanosecond laser pulse (532 nm). The results are very similar to those obtained in the absence of O 2 , including the intensity ratio (ϳ2) of the Fe-CO/Cu B -CO modes. This observation indicates that O 2 is not coordinated to Cu B in complex B, before photolysis. Importantly, the observed rates for the decay of the transient Cu B 1ϩ -CO complex (28.3 s Ϫ1 ) and the heme a 3 2ϩ recombination (k 2 ϭ 24.6 s Ϫ1 ) that we have determined, shown in Fig. 3B, inset,  This work presents a very peculiar and unexpected observation. However, the ba 3 oxidase is expressed under limited amounts of oxygen; thus, an O 2 -concentration-dependent behavior, such that presented in this work, is feasible. Several control experiments with Mb-CO (data not shown) indicate that under the same experimental conditions (O 2 concentration, pH-range, temperature), O 2 spontaneously replaces CO. The experiments reported here have been repeated with different enzyme preparations to avoid some sort of artifact and clearly demonstrate O 2 migration in the delivery channel that is located at the Cu B site. The presence, but not coordination, of O 2 at the Cu B site results in a structural reorientation of the Cu B environment and concomitantly to an increase of k Ϫ2 . This effect can be either steric or electrostatic involving either one of the Cu B -His ligands (polar) or the Cu B atom directly (electrostatic), producing an increased electron density in the CO antibonding orbitals that results in weakening the C-O bond strength (lower C-O ). Our data also support the developing consensus that Cu B is a way stop for O 2 en route to its heme a 3 binding site (8). Whether the presence of Glu-278 in other heme-copper oxidases (5,6) interrupts the putative oxygen channel provided by Ile-235 in ba 3 , and thus the efficient diffusion of low O 2 concentrations to the binuclear center, remains to be determined. Moreover, it is important to establish whether the conformational changes induced to Cu B environment by O 2 , which is present in low concentrations at physiological conditions because of the reduced gas solubility at higher temperatures, in the putative channel of ba 3 , are extended to the superfamily of cytochrome oxidases. Experiments toward this goal are in progress in our laboratory.