Docking Site Dynamics of ba 3 -Cytochrome c Oxidase from Thermus thermophilus

Ligand trajectories trapped within a docking site or within an internal cavity near the active site of proteins are important issues towards the elucidation of the mechanism of reaction of such complex systems, whose activity requires the shuttling of oriented ligands to and from their active site. The ligand motion within ba 3 -cytochrome c oxidase from Thermus thermophilus has been investigated by measuring time-resolved step-scan Fourier transform infrared difference spectra of photodissociated-CO from heme a 3 at ambient temperature. Upon photodissociation, 15-20% of the CO is not covalently attached to Cu B , but is trapped within a docking site near the ring A of heme a 3 propionate. Two trajectories of CO that are distinguished spectroscopically and kinetically ( n CO =2131 cm -1 /t d =10-35 m s and n CO =2146 cm -1 / t d =85 m s) are observed. At later times (t d =110 m s) the docking site reorganizes about the CO and quickly establishes an energetic barrier that facilitates equilibration of the ligand with the protein solvent. The time-dependent shift of the CO trajectories we observe is attributed to a conformational motion of the docking site surrounding the ligand. The implications of these results with respect to the ability of the docking site to constrain ligand orientation, and the reaction dynamics of the docking site are discussed.


Introduction
The docking sites and internal cavities of proteins and enzymes play a significant role in controlling the pathway(s) for diffusion of substrates/ligands to the active site. Time-resolved x-ray and time-resolved IR experiments of photolyzed CO-bound myoglobin (Mb) have demonstrated the existence of docking sites and their relevance in physiological ligand binding (1)(2)(3)(4)(5)(6)(7)(8). However, the different trajectories of the funneled photodissociated ligand in the docking site can be distinguished only through time-resolved IR spectroscopy (2)(3)(4)(5).
There is now a relatively good understanding in the ligand dynamics of photolyzed Mb-CO.
Up to date Mb has played the role of a model system for determining docking sites and channels through which, ligands enter and escape proteins, establishing a foundation for their chemical dynamics (9)(10)(11). Similar measurements in enzymes, whose activity may require the shuttling of oriented ligands to and from their active site, however, have never been reported either under physiological conditions or at cryogenic temperatures. In addition to activating O 2 and conserving the energy of the O 2 reduction for subsequent ATP synthesis, ba 3 -cytochrome c oxidase is able to catalyze the reduction of nitric oxide (NO) to nitrous oxide (N 2 O) under reducing anaerobic conditions (12). To understand the mechanism of reaction in such complex systems, it is crucial to probe in detail the intermediates along the reaction pathway. The time evolution and the trajectories of ligand binding intermediates can be monitored by time-resolved FTIR spectroscopy providing profound information on whether ligands gain access to the binding sites through specific channels and docking sites, or by random diffusion through the protein matrix. Such measurements would enable us to address some new issues regarding the properties of the docking sites that may function to discriminate between ligands such as O 2 , NO and CO.
The crystal structure of cytochrome ba 3 indicates that subunit I consists of a low-spin heme b and a high-spin heme a 3 /Cu B binuclear center, where the dioxygen and nitric oxide by guest on March 24, 2020 http://www.jbc.org/ Downloaded from 4 k 2 reactions take place (13). Subunit II contains a homodinuclear copper complex. The COligation/release mechanism in cytochrome ba 3 follows that found in other heme-copper oxidases (14)(15)(16) and proceeds according to the following scheme:

A B
Cytochrome ba 3 has a relative high affinity for CO (K 1 >10 4 ); the transfer of CO to heme a 3 2+ is characterized by a small k 2 = 8 s -1 , and by a k -2 = 0.8 s -1 (15), that is 30-fold greater than that of the bovine aa 3 (17,18).
Time-resolved step scan FTIR (TRS 2 ) spectra can now be unambiguously interpreted to yield specific information concerning the docking sites and their respective dynamic behavior in heme-copper oxidases. These observations may be made at room temperature under physiologically relevant conditions. In the work presented here, we have continued our TRS 2 -FTIR approach to heme-copper oxidases (18)(19)(20)(21) and report the dynamics of the docking site that shelters the photodissociated CO. In analogy to Mb studies, the results demonstrate the formation of the B 1 and B 0 states that are associated with the photolyzed CO within the docking site of ba 3 -cytochrome c oxidase.

Materials and Methods
Cytochrome ba 3 was isolated from Th. thermophilus HB8 cells according to previously published procedures (15). The samples were stored in liquid nitrogen until further use. The Ar. Dithionite reduced samples were exposed to 1 atm CO (1 mM) in an anaerobic cell to prepare the carbonmonoxy adduct and transferred to a tightly sealed FTIR cell composed of two 3 mm-thick CaF 2 windows, under anaerobic conditions. The pathlength was sufficiently small (15 µm) to avoid the strong absorbance of the water around 1650 cm -1 and keep the response of the MCT detector linear. The TRS 2 experiments were performed on a system described elsewhere (17). 12  Optical absorption spectra were recorded with a Perkin-Elmer Lamda 20 UV-visible spectrometer before and after the FTIR measurements to ensure the formation and stability of the CO adducts. No sample degradation occurred during the time course of the experiments.

Results and Discussion
For the sake of comparison with earlier work on Mb, we ascribe the 2131 and 2146 cm -1 modes we have detected in the TRS 2 -FTIR difference spectra of fully reduced ba 3 -CO to the states referred to in the literature as B 1 and B 0 , respectively. of fully reduced ba 3 -CO subsequent to CO photolysis by a 7-nanosecond laser pulse (532 nm). The negative peak at 1976 cm -1 arises from the photolyzed heme a 3 -CO complex. The positive peak that appears at 2053 cm -1 is the C-O stretch (ν CO ) of Cu B , as previously reported (17). Concurrently with the formation of the Cu B

1+
-CO complex, a positive peak appears at 2131 cm -1 that persists for 35 µs subsequent to CO photolysis, and is displaced by a peak at 6 conjunction with the reported extinction coefficients for heme a 3 -CO and Cu B -CO (16) another mode located at 2146 cm -1 gains intensity. At t d =10-50 µs, the 2146 cm -1 mode is below noise level, but at 85 µs, it is clearly observed in the spectrum. This feature decays at t d =110 µs, becoming "buried" to the background as free "solvated" CO. Importantly, the decay of the 2146 cm -1 mode occurs prior to rebinding of CO to heme a 3 Fe. Fig. 2A presents the TRS 2 -FTIR difference spectra of fully reduced ba 3 -13 CO subsequent to 13  The center frequency of the B 2 state (ν CO = 2119 cm -1 ) appeared static from 0.2 to 10 ps and thus, it was concluded that the spectral shifting appeared to come only from the trajectory leading to the B 1 state. In addition, from the surprisingly narrow band shapes of B 1 (∆ν 1/2 =9.1 8 cm -1 ) and B 2 (∆ν 1/2 =6.0 cm -1 ), it was suggested that the orientation of the docked CO is constrained by a static potential (7). It was finally concluded that the two B -states correspond to opposite orientations of CO within the same docking site and that interconversion between B 1 and B 2 requires end-to-end rotation of CO (2). The same docking site that shelters the photodissociated CO from heme a 3 is expected to trap the thermally dissociated CO from Cu B (k 2 = 29.5 s -1 ), as well as the thermally dissociated CO from heme a 3 under non-photolytic conditions. The latter process occurs very slowly with a time constant of k -2 =0.8 s -1 . However, we found no spectroscopic evidence that the thermally dissociated CO from Cu B is trapped within the docking site prior to its rebinding to the Fe of heme a 3 . We also expect the same docking site to trap any ligand having similar size to CO such as NO and O 2 , and the residues constituting the docking site to be responsible for the kinetic control of ligand binding and escape. The short lifetime of the B 0 and B 1 states in conjunction with the slow recombination of CO to heme a 3 is not a consequence of a diffusional barrier created by the docking site, but it is rather due to steric restrictions directly adjacent to the binuclear center.

10
In Fig. 3  potential. This way, the formation of the Cu B -C bond is rapid and concurrent with the Fe-C rupture (23)(24)(25). The