Artificial Hydrogenases Based on Cobaloximes and Heme Oxygenase.

The insertion of cobaloxime catalysts in the heme-binding pocket of heme oxygenase (HO) yields artificial hydrogenases active for H2 evolution in neutral aqueous solutions. These novel biohybrids have been purified and characterized by using UV/visible and EPR spectroscopy. These analyses revealed the presence of two distinct binding conformations, thereby providing the cobaloxime with hydrophobic and hydrophilic environments, respectively. Quantum chemical/molecular mechanical docking calculations found open and closed conformations of the binding pocket owing to mobile amino acid residues. HO-based biohybrids incorporating a {Co(dmgH)2 } (dmgH2 =dimethylglyoxime) catalytic center displayed up to threefold increased turnover numbers with respect to the cobaloxime alone or to analogous sperm whale myoglobin adducts. This study thus provides a strong basis for further improvement of such biohybrids, using well-designed modifications of the second and outer coordination spheres, through site-directed mutagenesis of the host protein.

Aqueous light driven hydrogen production by a Ru-ferredoxin-Co biohybrid.

Herein we report the creation of a novel solar fuel biohybrid for light-driven H2 production utilizing the native electron transfer protein ferredoxin (Fd) as a scaffold for binding of a ruthenium photosensitizer (PS) and a molecular cobaloxime catalyst (Co). EPR and transient optical experiments provide direct evidence of a long-lived (>1.5 ms) Ru(III)-Fd-Co(I) charge separated state formed via an electron relay through the Fd [2Fe-2S] cluster, initiating the catalytic cycle for 2H(+) + 2e(-) → H2.

Cu2+ site in photosynthetic bacterial reaction centers from Rhodobacter sphaeroides, Rhodobacter capsulatus, and Rhodopseudomonas viridis.

The interaction of metal ions with isolated photosynthetic reaction centers (RCs) from the purple bacteria Rhodobacter sphaeroides, Rhodobacter capsulatus, and Rhodopseudomonas viridis has been investigated with transient optical and magnetic resonance techniques. In RCs from all species, the electrochromic response of the bacteriopheophytin cofactors associated with Q(A)(-)Q(B) --> Q(A)Q(B)(-) electron transfer is slowed in the presence of Cu(2+). This slowing is similar to the metal ion effect observed for RCs from Rb. sphaeroides where Zn(2+) was bound to a specific site on the surface of the RC [Utschig et al. (1998) Biochemistry 37, 8278]. The coordination environments of the Cu(2+) sites were probed with electron paramagnetic resonance (EPR) spectroscopy, providing the first direct spectroscopic evidence for the existence of a second metal site in RCs from Rb. capsulatus and Rps. viridis. In the dark, RCs with Cu(2+) bound to the surface exhibit axially symmetric EPR spectra. Electron spin echo envelope modulation (ESEEM) spectral results indicate multiple weakly hyperfine coupled (14)N nuclei in close proximity to Cu(2+). These ESEEM spectra resemble those observed for Cu(2+) RCs from Rb. sphaeroides [Utschig et al. (2000) Biochemistry 39, 2961] and indicate that two or more histidines ligate the Cu(2+) at the surface site in each RC. Thus, RCs from Rb. sphaeroides, Rb. capsulatus, and Rps. viridis each have a structurally analogous Cu(2+) binding site that is involved in modulating the Q(A)(-)Q(B) --> Q(A)Q(B)(-) electron-transfer process. Inspection of the Rps. viridis crystal structure reveals four potential histidine ligands from three different subunits (M16, H178, H72, and L211) located beneath the Q(B) binding pocket. The location of these histidines is surprisingly similar to the grouping of four histidine residues (H68, H126, H128, and L211) observed in the Rb. sphaeroides RC crystal structure. Further elucidation of these Cu(2+) sites will provide a means to investigate localized proton entry into the RCs of Rb. capsulatus and Rps. viridis as well as locate a site of protein motions coupled with electron transfer.

EPR investigation of Cu2+-substituted photosynthetic bacterial reaction centers: evidence for histidine ligation at the surface metal site.

The coordination environments of two distinct metal sites on the bacterial photosynthetic reaction center (RC) protein were probed with pulsed electron paramagnetic resonance (EPR) spectroscopy. For these studies, Cu2+ was bound specifically to a surface site on native Fe2+-containing RCs from Rhodobacter sphaeroides R-26 and to the native non-heme Fe site in biochemically Fe-removed RCs. The cw and pulsed EPR results clearly indicate two spectroscopically different Cu2+ environments. In the dark, the RCs with Cu2+ bound to the surface site exhibit an axially symmetric EPR spectrum with g(parallel) = 2.24, A(parallel) = 160 G, g(perpendicular) = 2.06, whereas the values g(parallel) = 2.31, A(parallel) = 143 G, and g(perpendicular) = 2.07 were observed when Cu(2+) was substituted in the Fe site. Examination of the light-induced spectral changes indicate that the surface Cu2+ is at least 23 A removed from the primary donor (P+) and reduced quinone acceptor (QA-). Electron spin-echo envelope modulation (ESEEM) spectra of these Cu-RC proteins have been obtained and provide the first direct solution structural information about the ligands in the surface metal site. From these pulsed EPR experiments, modulations were observed that are consistent with multiple weakly hyperfine coupled 14N nuclei in close proximity to Cu2+, indicating that two or more histidines ligate the Cu2+ at the surface site. Thus, metal and EPR analyses confirm that we have developed reliable methods for stoichiometrically and specifically binding Cu2+ to a surface site that is distinct from the well characterized Fe site and support the view that Cu2+ is bound at or near the Zn site that modulates electron transfer between the quinones QA and QB (QA-QB --> QAQB-) (Utschig, L. M., Ohigashi, Y., Thurnauer, M. C., and Tiede, D. M (1998) Biochemistry 37, 8278-8281) and proton uptake by QB- (Paddock, M. L., Graige, M. S., Feher, G., and Okamura, M. Y. (1999) Proc. Natl. Acad. Sci. U.S.A. 96, 6183-6188). Detailed EPR spectroscopic characterization of these Cu2+-RCs will provide a means to investigate the role of local protein environments in modulating electron and proton transfer.

A new metal-binding site in photosynthetic bacterial reaction centers that modulates QA to QB electron transfer.

Isolated reaction centers (RCs) from Rhodobacter sphaeroides were found to bind Zn(II) stoichiometrically and reversibly in addition to the 1 equiv of non-heme Fe(II). Metal and EPR analyses confirm that Zn(II) is ligated to a binding site that is distinct from the Fe site. When Zn(II) is bound to this site, electron transfer between the quinones QA and QB (QA-QB --> QAQB-) is slowed and the room-temperature kinetics become distributed across the microsecond to millisecond time domain. This effect of metal binding on the kinetics is similar to the more global effect of cooling RCs to 2 degreesC in the absence of Zn(II). This suggests that Zn(II) binding alters localized protein motions that are necessary for rapid QA-QB --> QAQB- electron transfer. Inspection of the RC crystal structure suggests a cluster of histidine ligands located beneath the QB binding pocket as a potential binding site.

Kinetic phases in the electron transfer from P+QA-QB to P+QAQB- and the associated processes in Rhodobacter sphaeroides R-26 reaction centers.

Electron transfer from P+QA-QB to form P+QAQB- was measured in Rhodobacter sphaeroides R-26 reaction centers (RCs) where the native primary quinone, ubiquinone-10 (UQA), was replaced by 2-methyl-3-phytyl-1,4-naphthoquinone (MQA). The native secondary quinone, UQ-10, was retained as UQB. The difference spectrum of the semiquinone MQA- minus UQB- absorption is very similar to that of MQ- minus UQ- in solution (398-480 nm). Thus, the absorption change provides a direct monitor of the electron transfer from MQA- to UQB. In contrast, when both QA and QB are UQ-10 the spectral difference between UQA- and UQB- arises from electrochromic responses of RC chromophores. Three kinetic processes are seen in the near UV (390-480 nm) and near-IR (740-820 nm). Analysis of the time-correlated spectra support the conclusion that the changes at tau1 approximately 3 micros are mostly due to electron transfer, electron transfer and charge compensation are mixed in tau2 approximately 80 micros, while little or no electron transfer occurs at 200-600 micros (tau3) in MQAUQB RCs. The 80-micros rate has been previously observed, while the fast component has not. The fast phase represents 60% of the electron-transfer reaction (398 nm). The activation energy for electron transfer is DeltaG approximately 3.5 kcal/mol for both tau1 and tau2 between 0 and 30 degrees C. In isolated RCs with UQA, if there is any fast component, it appears to be faster and less important than in the MQA reconstituted RCs.

Time-resolved electrochromism associated with the formation of quinone anions in the Rhodobacter sphaeroides R26 reaction center.

The bacterial photosynthetic reaction center contains bacteriochlorophyll (Bchl) and bacteriopheophytin (Bph) cofactors that provide natural probes of electrostatic fields within this protein. We have examined the electrochromic responses of these cofactors, resolved during the lifetimes of the quinone anion states, P+QA-QB and P+QAQB-, and measured as a function of temperature. These measurements provide information on the time-dependent variation in electrostatic field strength on the Bchl and Bph cofactors. Measurements in the near-infrared absorbance bands have revealed the following. First, the QA-QB-->QAQB- electron transfer rate is found to be heterogeneous, consisting of at least two distinct kinetic components. At room temperature, we find a previously unresolved fast kinetic component with a reaction time of 25-40 microseconds, depending upon the preparation, that accounts for approximately 25% of the total reaction yield. The major component was identified with a reaction time of 210-240 microseconds. Below -20 degrees C, QA-QB-->QAQB- electron transfer shows distributed kinetics. The temperature-dependent conversion from biphasic to distributed kinetics suggests that there is a thermal averaging of conformational substates around two reaction center configurations. Interestingly, direct excitation of the Bph with 532 nm light at low temperatures appears to alter the electron transfer kinetics, possibly by inducing a change in the distribution of conformational states. The reaction kinetics were found to be sensitive to the addition of ethylene glycol, which is likely to reflect an osmolarity effect. Second, time-dependent absorption changes of the Bchl and Bph cofactors are found to be kinetically decoupled. The rapid responses of the Bph bands are interpreted to reflect electron transfer, while the slower responses of the Bchl are interpreted to reflect slower relaxation events, possibly including proton uptake. Finally, we find that the electrochromic response and QA-QB-->QAQB- electron transfer to be sensitive to the preparative state of the reaction center, reflecting differences in quinone binding for reaction centers in different states of purification.

Site-specific and compensatory mutations imply unexpected pathways for proton delivery to the QB binding site of the photosynthetic reaction center.

In photosynthetic reaction centers, a quinone molecule, QB, is the terminal acceptor in light-induced electron transfer. The protonatable residues Glu-L212 and Asp-L213 have been implicated in the binding of QB and in proton transfer to QB anions generated by electron transfer from the primary quinone QA. Here we report the details of the construction of the Ala-L212/Ala-L213 double mutant strain by site-specific mutagenesis and show that its photosynthetic incompetence is due to an inability to deliver protons to the QB anions. We also report the isolation and biophysical characterization of a collection of revertant and suppressor strains that have regained the photosynthetic phenotype. The compensatory mutations that restore function are diverse and show that neither Glu-L212 nor Asp-L213 is essential for efficient light-induced electron or proton transfer in Rhodobacter capsulatus. Second-site mutations, located within the QB binding pocket or at more distant sites, can compensate for mutations at L212 and L213 to restore photocompetence. Acquisition of a single negatively charged residue (at position L213, across the binding pocket at position L225, or outside the pocket at M43) or loss of a positively charged residue (at position M231) is sufficient to restore proton transfer activity to the complex. The proton transport pathways in the suppressor strains cannot, in principle, be identical to that of the wild type. The apparent mutability of this pathway suggests that the reaction center can serve as a model system to study the structural basis of protein-mediated proton transport.

Electron-transfer kinetics and electrostatic properties of the Rhodobacter sphaeroides reaction center and soluble c-cytochromes.

The kinetics of electron transfer between the Rhodobacter sphaeroides R-26 reaction center and nine soluble c-cytochromes have been analyzed and compared to the patterns of the surface electrostatic potentials for each of the proteins. Characteristic first-order electron-transfer rates for 1:1 complexes formed at low ionic strength between the reaction center and the different c-cytochromes were identified and found to vary by a factor of almost 100, while second-order rates were found to differ by greater than 10(6). A correlation was found between the location of likely electrostatic interaction domains on each cytochrome and its characteristic rate of electron transfer. The interaction domains were identified by mapping electrostatic potentials, calculated from the Poisson-Boltzmann equation, onto simulated "encounter surfaces" for each of the cytochromes and the reaction center. For the reaction center, the c-cytochrome binding domain was found to have almost exclusively net negative potential (< -3 kT) and to be shifted slightly toward the M-subunit side of the reaction center. The location of interaction domains of complementary, positive potential (> 3 kT) differed for each cytochrome. The correspondence between electrostatic, structural, and kinetic properties of 1:1 reaction center-cytochrome complexes leads to a proposed mechanism for formation of reaction center-cytochrome electron-transfer complexes that is primarily driven by the juxtaposition of regions of delocalized complementary potential. In this mechanism the clustering of charged residues is of primary importance and not the location of specific residues. A consequence of this mechanism is that many different sets of charge distributions are predicted to be capable of stabilizing a specific configuration for a reaction center-cytochrome complex. This mechanism for reaction center association with water-soluble c-cytochromes fits molecular recognition mechanisms proposed for c-cytochromes in nonphotosynthetic systems. In general, the kinetic scheme for reaction center driven cytochrome oxidation was found to vary between a simple two-state model, involving cytochrome in free and reaction center bound states, and a three-state model, that includes cytochrome binding in kinetically competent ("proximal") and incompetent ("distal") modes. The kinetically incompetent mode of cytochrome binding is suggested not to be an intrinsic feature of the reaction center-cytochrome association but is likely to be due to variation in the physical state of the reaction center.

In bacterial reaction centers protons can diffuse to the secondary quinone by alternative pathways.

The mechanisms of proton conduction to the reduced secondary quinone in bacterial reaction centers were studied in wild-type and genetically modified reaction centers from Rhodobacter capsulatus. In the L212-213AA double mutant (L212Glu----Ala, L213Asp----Ala), reaction center function is severely altered. However, a photocompetent revertant of this strain which carries a third 'compensating' mutation, M231Arg----Leu, at about 15 A from the secondary quinone, displays the normal proton binding function of the reaction center. Furthermore, the apparent pK values of group(s) involved in the stabilization of the semiquinone anion are restored by that mutation. We conclude that L212Glu and L213Asp are not obligatory residues for proton donation to QB in Rb. capsulatus. We suggest that protons can be delivered to the QB site from the cytoplasm via a network of proton channels activated by compensatory mutations, possibly involving water molecules bound in the interior of the reaction center.

Measurement of the extent of electron transfer to the bacteriopheophytin in the M-subunit in reaction centers of Rhodopseudomonas viridis.

We have measured the extent of flash-induced electron transfer from the bacteriochlorophyll dimer, P, to the bacteriopheophytin in the M-subunit, HM, in reaction centers of Rhodopseudomonas viridis. This has been done by measuring the transient states produced by excitation of reaction centers trapped in the PHL (-)HM state at 90 K. Under these conditions the normal forward electron transfer to the bacteriopheophytin in the L-subunit, HL, is blocked and the yield of transient P(+)HM (-) can be estimated with respect to the lifetime of P(*). Under these conditions flash induced absorbance decreases of the bacteriochlorophyll dimer 990 nm band suggest that a transient P(+) state is formed with a quantum yield of 0.09±0.06 compared to that formed during normal photochemistry. These transient measurements provide an upper limited on the yield of a transient P(+) HM (-) state. An estimate of 0.09 as the yield of the P(+) HM (-) state is consistent with all current observations. This estimate and the lifetime of P(*) suggest that the electron transfer rate from P(*) to HM, kM, is about 5 × 10(9) sec(-1) (τM = 200ps). These measurements suggest that the a branching ratio kL/kM is on the order of 200. The large value of the branching ratio is remarkable in view of the structural symmetry of the reaction center. This measurement should be useful for electron transfer calculations based upon the reaction center structure.

Geometry for the Primary Electron Donor and the Bacteriopheophytin Acceptor in Rhodopseudomonas viridis Photosynthetic Reaction Centers.

The tetrapyrrole electron donors and acceptors (bacteriochlorophyll, BCh; bacteriopheophytin, BPh) within the bacterial photosynthetic reaction center (RC) are arranged with a specific geometry that permits rapid (picosecond time scale) electron tunneling to occur between them. Here we have measured the angle between the molecular planes of the bacteriochlorophyll dimer (primary donor), B(2), and the acceptor bacteriopheophytin, H, by analyzing the dichroism of the absorption change associated with H reduction, formed by photoselection with RCs of Rhodopseudomonas viridis. This angle between molecular planes is found to be 60 degrees +/- 2. This means that the ultrafast electron tunneling must occur between donors and acceptors that are fixed by the protein to have a noncoplanar alignment. Nearly perpendicular alignments have been determined for other electron tunneling complexes involving RCs. These geometries can be contrasted with models proposed for heme-heme electron transfer complexes, which have emphasized that mutually parallel orientations should permit the most kinetically facile transfers.

Construction of a flash-activated cyclic electron transport system by using bacterial reaction centers and the ubiquinone-cytochrome b-c1/c segment of mitochondria.

Single-turnover electron transfer within the mitochondrial complex III has been studied by combining, in solution, the isolated complex from bovine heart with detergent-solubilized reaction centers of Rhodopseudomonas sphaeroides. Initiation of electron transfer by short flash activation resulted in the prompt oxidation of cytochrome c and reduction of cytochrome b. The subsequent reduction of ferricytochrome c was observed to be concomitant with the oxidation of the ferrocytochrome b, both reactions being inhibited by the addition of actimycin A. The rate of electron transfer through complex III is dependent upon the ambient redox potential poise in a way that is consistent with the presence of a redox component, presumably analogous to the photosynthetic ubiquinone Qz, which is an obligatory intermediate in electron transfer between cytochromes b and c. These results demonstrate cyclic electron transfer in a constructed assembly of mitochondrial complex III, cytochrome c, and photochemical reaction centers.

Studies on the molecular organization of cytochromes P-450 and b5 in the microsomal membrane.

1. The relative orientations of the heme groups of cytochromes P-450 and b5 in the microsomal membrane have been studied by the technique of electron paramagnetic resonance. The results show that the heme plane of cytochrome P-450 lies in the same plane as the membrane surface, whereas the cytochrome b5 heme plane has a random orientation. 2. No significant broadening or change in relaxation properties of the gz component of low spin cytochrome P-450 occurred when cytochrome b5 was reduced by redox poising. It is concluded that there is little or no paramagnetic coupling between the heme groups of the two species. 3. The results favor a model in which no tight complex between cytochromes P-450 and b5 is present, the species being independent and interacting only by random molecular collisions or via other intermediate species.

Structural organization of the Chromatium vinosum reaction center associated c-cytochromes.

Magnetic interactions operating between the Chromatium vinosum reaction center associated c-cytochromes and the electron carriers of the reaction center have been assayed by comparing the magnetic properties of these components alone, and in various combinations with paramagnetic forms of the reaction center electron carriers. These studies have yielded the following results. 1. The oxidized paramagnetic forms of the high potential cytochromes c-555 produce no discernable alteration of the light-induced (BChl)2.+signal. 2. Similarly, analysis of the lineshape of the light-induced (BChl)2.+signal shows that a magnetic interaction with the oxidized low potential cytochromes c-553 is likely to produce less than a 1 gauss splitting of the (BChl)2.+signal, which corresponds to a minimum separation of 25 +/- 3 A between the unpaired spins if the heme and (BChl)2 are orientated in a coplanar arrangement, suggesting a minimum separation of 15+/- 3A between the heme edge and the (BChl)2 edge. 3. a prominent magnetic interaction is observed to operate between the cytochrome c-553 and c-555, which results in a 30-35 gauss splitting of these spectra, and suggests an iron to iron separation of about 8 A.4. Magnetic interactions are not observed between the c-cytochromes and the reaction center "primary acceptor" (the iron . quinone complex) nor with the reaction center intermediate electron carrier (which involves bacteriopheophytin) suggesting separations greater than 10 A. 5. Magnetic interactions are not discerned between the two cytochrome c-553 hemes, nor between the two cytochrome c-555 hemes, implying that the distance between the cytochromes of the same pair is greater than 10 A. 6. EPR studies of oriented chromatophores have demonstrated that the cytochrome c-553 and c-555 hemes are perpendicular to each other, and suggest that the cytochrome c-553 heme plane lies parallel to the plane of the membrane, while the cytochrome c-555 heme plane lies perpendicular to the plane of the membrane surface.

EPR and optical spectroscopic properties of the electron carrier intermediate between the reaction center bacteriochlorophylls and the primary acceptor in Chromatium vinosum.

1. A reaction center-cytochrome c complex has been isolated from Chromatium vinosum which is capable of normal photochemistry and light-activated rapid cytochrome c553 and c555 oxidation, but which has no antenna bacteriochlorophyll. As is found in whole cells, ferrocytochrome c553 is oxidized irreversibly in milliseconds by light at 7 K. 2. Room temperature redox potentiometry in combination with EPR analysis at 7 K, of cytochrome c553 and the reaction center bacteriochlorophyll dimer (BChl)2 absorbing at 883 nm yields identical results to those previously reported using optical analytical techniques at 77 K. It shows directly that two cytochrome c553 hemes are equivalent with respect to the light induced (BChl)2+. At 7 K, only one heme can be rapidly oxidized in the light, commensurate with the electron capacity of the primary acceptor (quinone-iron) being unity. 3. Prior chemical reduction of the quinone-iron followed by illumination at 200K, however, leads to the slow (t1/2 approximately equal to 30 s) oxidation of one cytochrome c553 heme, with what appears to be concommitant reduction of one of the two bacteriophytins (BPh) of the reaction center as shown by bleaching of the 760 nm band, a broad absorbance increase at approx. 650 nm and a bleaching at 543 nm. The 800 nm absorbing bacteriochlorophyll is also involved since there is also bleaching at 595 and 800 nm; at the latter wave-length the remaining unbleached band appears to shift significantly to the blue. No redox changes in the 883 absorbing bacteriochlorophyll dimer are seen during or after illumination under these conditions. The reduced part of the state represents what is considered to be the reduced form of the electron carrier (I) which acts as an intermediate between the bacteriochlorophyll dimer and quinone-iron. The state (oxidized c553/reduced I) relaxes in the dark at 200K in t1/2 approx. 20 min but below 77 K it is trapped on a days time scale. 4. EPR analysis of the state trapped as described above reveals that one heme equivalent of cytochrome becomes oxidized for the generation of the state, a result in agreement with the optical data. Two prominent signals are associated with the trapped state in the g = 2 region, which can be easily resolved with temperature and microwave power saturation: one has a line width of 15 g and is centered at g = 2.003; the other, which is the major signal, is also a radical centered at g = 2.003 but is split by 60 G and behaves as though it were an organic free-radical spin-coupled with another paramagnetic center absorbing at higher magnetic field values; this high field partner could be the iron-quinone of the primary acceptor. The identity of two signals associated with I-. is consistent with the idea that the reduced intermediary carrier is not simply BPh-. but also involves a second radical, perhaps the 800 nm bacteriochlorophylls in the reduced state...