Time-resolved thermodynamic profiles for CO photolsysis from the mixed valence form of bovine heart cytochrome c oxidase.

Photoacoustic calorimetry has been utilized to probe the thermodynamics accompanying photodissociation of the CO mixed valence form of bovine heart cytochrome c oxidase (COMV CcO). At pH's below 9 photolysis of the COMV CcO results in three kinetic phases with the first phase occurring faster than the time resolution of the instrument (i.e., < approximately 50 ns), a second phase occurring with a lifetime of approximately 100 ns and a third phase occurring with a lifetime of approximately 2 micros. The corresponding volume and enthalpy changes for these processes are: DeltaH1, DeltaV1 = +79 +/- 10 kcal mol(-1), +9 +/- 1 mL mol(-1); DeltaH2, DeltaV2 = -79 +/- 5 kcal mol(-1), -9 +/- 2 mL mol(-1); DeltaH3, DeltaV3 = +54 +/- 7 kcal mol(-1), +8 +/- 1 mL mol(-1). At pH's above 9 only one phase is observed, a prompt phase occurring in < 50 ns. The overall volume change is negligible above pH 9 and the enthalpy change is +29 +/- 5 kcal mol(-1). The data are consistent with the prompt phase being associated with CO-Fe(a3) bond cleavage, CO-CuB+ bond formation, Fe(a3) low-spin to high-spin transition and fast electron transfer (ET) from heme a3 to heme a followed by proton transfer from Glu242 to Arg38 on an approximately 100 ns timescale. The slow phase is likely a combination of CO thermal dissociation from CuB and additional ET between heme a3 to heme a. Interestingly, this phase is not evident above pH 9 suggesting linkage between CO dissociation/ET and the protonation state of a group or groups near the binuclear center.     

Turnover of cytochrome c oxidase from Paracoccus denitrificans

The heme aa3 type cytochrome oxidase from Paracoccus denitrificans incorporated into vesicles with phospholipid reacts during turnover much as the oxidase from mitochondria does. The spectrophotometric changes observed at various wavelengths are closely similar, and the rate is about one-half of that for beef heart oxidase under the same conditions. The rate of appearance of oxidized cytochrome c on initiation of the reaction is also similar and depends on the previous treatment of the oxidase as described by Antonini, E., Brunori, M., Colosimo, A., Greenwood, C. and Wilson, M. T. (1977) Proc. Natl. Acad. Sci. U.S.A. 74, 3128-3132. In terms of their model the resting Paracoccus enzyme is converted to the pulsed form during turnover. The effect is observed with both cytochrome c and hexamine ruthenium as reductants. With the latter a 60-fold increase in rate is observed.     

FTIR studies of the CO and cyanide adducts of fully reduced bovine cytochrome c oxidase

Photolysis spectra of the CO and cyanide adducts of reduced bovine cytochrome c oxidase have been studied by FTIR difference spectroscopy. Bound CO is predominantly in a single 1963 cm(-1) form whereas cyanide is bound in at least two forms (2058/2045 cm(-1)). These forms are pH-independent between pH 6.5 and 8.5, indicating that there is no titratable protonatable group that influences significantly their binding in this pH range. Photolysis spectra of the cyanide adduct have a positive band around 2090 cm(-1) in H(2)O due at least in part to free HCN and at 1880 cm(-1) in D(2)O due to free DCN. The frequency of the positive band around 2090 cm(-1), and its persistence in D(2)O media, raises the possibility that a transient cyanide-Cu(B) adduct also contributes to this signal, equivalent to the CO-Cu(B) species that is formed when CO is photolyzed. Photolysis produces changes throughout the 1000-1800 cm(-1) region. Reduced minus (reduced + CO) photolysis spectra in H(2)O exhibit a pH-independent and symmetrical peak/trough at 1749/1741 cm(-1). A related feature in homologous oxidases has been suggested to arise from a conserved glutamic acid. However, only around one-third of the feature is shifted to lower frequencies by incubation in D(2)O media, and an additional fraction is shifted if catalytic turnover occurs in D(2)O. Reduced minus (reduced + cyanide) photolysis spectra exhibit multiple features in H(2)O in this region with peaks at 1752, 1725, and 1708 cm(-1) and troughs at 1740, 1715, and 1698 cm(-1). Again, only a part of these features shift in D(2)O, even with catalytic turnover. A variety of additional H/D-sensitive features in the 1700-1000 cm(-1) region of the spectra can be discerned, one of which in cyanide photolysis spectra is tentatively assigned to a conserved tyrosine, Y244. Data are discussed in relation to the structure of the binuclear center and protonatable groups in its vicinity.     

Electronic and vibrational spectroscopy of the cytochrome c:cytochrome c oxidase complexes from bovine and Paracoccus denitrificans.

The 1:1 complex between horse heart cytochrome c and bovine cytochrome c oxidase, and between yeast cytochrome c and Paracoccus denitrificans cytochrome c oxidase have been studied by a combination of second derivative absorption, circular dichroism (CD), and resonance Raman spectroscopy. The second derivative absorption and CD spectra reveal changes in the electronic transitions of cytochrome a upon complex formation. These results could reflect changes in ground state heme structure or changes in the protein environment surrounding the chromophore that affect either the ground or excited electronic states. The resonance Raman spectrum, on the other hand, reflects the heme structure in the ground electronic state only and shows no significant difference between cytochrome a vibrations in the complex or free enzyme. The only major difference between the Raman spectra of the free enzyme and complex is a broadening of the cytochrome a3 formyl band of the complex that is relieved upon complex dissociation at high ionic strength. These data suggest that the differences observed in the second derivative and CD spectra are the result of changes in the protein environment around cytochrome a that affect the electronic excited state. By analogy to other protein-chromophore systems, we suggest that the energy of the Soret pi* state of cytochrome a may be affected by (1) changes in the local dielectric, possibly brought about by movement of a charged amino acid side chain in proximity to the heme group, or (2) pi-pi interactions between the heme and aromatic amino acid residues.     

Reaction of oxygen with cytochrome c oxidase from Paracoccus denitrificans

The reaction of reduced cytochrome c oxidase (EC 1.9.3.1) from Paracoccus denitrificans (American Type Culture Collection 13543) with dioxygen has been followed by laser flash photolysis of the CO derivative. In detergent-stabilized solutions the reaction showed at least two distinct kinetic components, the faster of which was oxygen concentration dependent and had a rate of approximately 60 X 10(6) M-1 s-1. The slower reaction was independent of oxygen concentration and had a rate of 9 X 10(2) s-1. These rates are about 1.5 times greater than comparable rates for ox heart oxidase reported by C. Greenwood and Q. H. Gibson (J. Biol. Chem. (1967) 242, 1782-1787). The kinetic components have markedly different optical spectra which agree precisely in form with those for ox heart enzyme (Greenwood, C., and Gibson, Q. H. (1967) J. Biol. Chem. 242, 1782-1787) but are shifted by 2 nm toward the red. In phospholipid vesicles, the spectral contribution of the faster component was augmented. The dissociation constant for CO at 20 degrees C is 1.6 microM, 6 times greater than for the ox heart enzyme. The bacterial enzyme binds one CO per 2 heme a. The enzyme has an absorption band at 830 nm in the oxidized form similar to that of the ox heart enzyme.     

Copper and manganese electron spin resonance studies of cytochrome c oxidase from Paracoccus denitrificans

The two-subunit cytochrome c oxidase from Paracoccus denitrificans contains two heme a groups and two copper atoms. However, when the enzyme is isolated from cells grown on a commonly employed medium, its electron paramagnetic resonance (EPR) spectrum reveals not only a Cu(II) powder pattern, but also a hyperfine pattern from tightly bound Mn(II). The pure Mn(II) spectrum is observed at -40 degrees C; the pure Cu(II) spectrum can be seen with cytochrome c oxidase from P. denitrificans cells that had been grown in a Mn(II)-depleted medium. This Cu(II) spectrum is very similar to that of cytochrome c oxidase from yeast or bovine heart. Manganese is apparently not an essential component of P. denitrificans cytochrome c oxidase since it is present in substoichometric amounts relative to copper or heme a and since the manganese-free enzyme retains essentially full activity in oxidizing ferrocytochrome c. However, the manganese is not removed by EDTA and its EPR spectrum responds to the oxidation state of the oxidase. In contrast, manganese added to the yeast oxidase or to the manganese-free P. denitrificans enzyme can be removed by EDTA and does not respond to the oxidation state of the enzyme. This suggests that the manganese normally associated with P. denitrificans cytochrome c oxidase is incorporated into one or more internal sites during the biogenesis of the enzyme.     

Subunit III of cytochrome c oxidase is expressed in Paracoccus denitrificans

Polyclonal antibodies have been obtained against a synthetic dodecapeptide identical to the aminoacid sequence 120-131 DSPIKDGVWPPE (inferred from its DNA sequence) of Paracoccus denitrificans cytochrome c oxidase subunit III. The antibodies had a titer higher than 1:10000 when tested against the antigen. These antibodies have been used to produce immunological evidence that, despite the fact that subunit III is not isolated with cytochrome c oxidase, it exists in Paracoccus denitrificans lysates. The antibodies did not show reactivity with bovine heart cytochrome c oxidase either by ELISA or immunoblotting. It was also shown that the antibodies react with a single polypeptide present in Paracoccus denitrificans cell lysates, having an apparent molecular weight close to that of subunit III of bovine heart oxidase.     

The role of tryptophan 272 in the Paracoccus denitrificans cytochrome c oxidase

The mechanism of electron coupled proton transfer in cytochrome c oxidase (CcO) is still poorly understood. The P(M)-intermediate of the catalytic cycle is an oxoferryl state whose generation requires one additional electron, which cannot be provided by the two metal centres. The missing electron has been suggested to be donated to this binuclear site by a tyrosine residue. A tyrosine radical species has been detected in the P(M) and F* intermediates (formed by addition of H2O2) of the Paraccocus denitrificans CcO using electron paramagnetic resonance (EPR) spectroscopy. From the study of conserved variants its origin was determined to be Y167 which is surprising as this residue is not part of the active site. Upon inspection of the active site it becomes evident that W272 could be the actual donor of the missing electron, which can then be replenished from Y167 or from the Y280-H276 cross link in the natural cycle. To address the question, whether such a direct electron transfer pathway to the binuclear centre exists two tryptophan 272 variants in subunit I have been generated. These variants are characterised by their turnover rates as well as using EPR and optical spectroscopy. From these experiments it is concluded, that W272 is an important intermediate in the formation of the radical species appearing in P(M) and F* intermediates produced with hydrogen peroxide. The significance of this finding for the catalytic function of the enzyme is discussed.     

The cytochrome c peroxidase of Paracoccus denitrificans.

The size, visible absorption spectra, nature of haem and haem content suggest that the cytochrome c peroxidase of Paracoccus denitrificans is related to that of Pseudomonas aeruginosa. However, the Paracoccus enzyme shows a preference for cytochrome c donors with a positively charged 'front surface' and in this respect resembles the cytochrome c peroxidase from Saccharomyces cerevisiae. Paracoccus cytochrome c-550 is the best electron donor tested and, in spite of an acidic isoelectric point, has a markedly asymmetric charge distribution with a strongly positive 'front face'. Mitochondrial cytochromes c have a much less pronounced charge asymmetry but are basic overall. This difference between cytochrome c-550 and mitochondrial cytochrome c may reflect subtle differences in their electron transport roles. A dendrogram of cytochrome c1 sequences shows that Rhodopseudomonas viridis is a closer relative of mitochondria than is Pa. denitrificans. Perhaps a mitochondrial-type cytochrome c peroxidase may be found in such an organism.     

Isolation and analysis of the genes for cytochrome c oxidase in Paracoccus denitrificans

Synthetic oligonucleotide probes were used to clone two loci from the chromosomal DNA of Paracoccus denitrificans that contain the genes for cytochrome c oxidase (cytochrome aa₃). One locus seems to contain four or five genes probably forming an operon. Two of these code for the oxidase subunits II and III. Three open reading frames are found between the COII and COIII genes. The other locus codes for the subunit I. A short open reading frame is found upstream of this gene. All three subunits of the Paracoccus enzyme show remarkable homology to the corresponding subunits of the mitochondrial cytochrome oxidase. Possible protein products of the open reading frames have not yet been identified.     

Electrogenic events upon photolysis of CO from fully reduced cytochrome c oxidase

CO photolysis from fully reduced Paracoccus denitrificans aa(3)-type cytochrome c oxidase in the absence of O(2) was studied by time-resolved potential electrometry. Surprisingly, photo dissociation of the uncharged carbon monoxide results in generation of a small-amplitude electric potential with the same sign as the physiological charge separation during activity. The number of electrogenic events after CO photolysis depends on the state of the enzyme. CO photolysis following immediately after activation by an enzymatic turnover, showed a two-component potential development. A fast (~1.5μs) phase was followed by slower potential generation with a time constant varying from 8μs at pH 7 to 250μs at pH 10. The amplitude of the fast phase was independent of the time of incubation after enzyme activation, whereas the slower phase vanished with a time constant of ~25min. CO photolysis from enzyme that had not undergone a prior single turnover showed the fast phase, but the amplitude of the slow phase was reduced to 10-30%. The amplitude of the fast phase corresponds to charge movement of 0.83? perpendicular to the membrane dielectric, and is independent of the time after enzyme activation. Thus it can be used as an internal ruler for normalization of the electrogenic responses of CcO. The slow phase was absent in the K354M mutant with a blocked proton-conducting K channel. We propose that CO photolysis increases the pK of the K354 residue, which results in its partial protonation, and generation of electric potential.     

Time-resolved ATR-FTIR spectroscopy of the oxygen reaction in the D124N mutant of cytochrome c oxidase from Paracoccus denitrificans

Real-time measurements of the cytochrome c oxidase reaction with oxygen were performed by ATR-FTIR spectroscopy, using a mutant with a blocked D-pathway of proton transfer (D124N, Paracoccus denitrificans numbering). The complex spectrum of the ferryl-->oxidized transition together with other bands showed protonation of Glu 278 with a peak position at 1743 cm-1. Since our time resolution was not sufficient to follow the earlier reaction steps, the FTIR spectrum of the CO-inhibited fully reduced-->ferryl transition was obtained as a difference between the spectrum before the laser flash and the first spectrum after it. A trough at 1735 cm-1 due to deprotonation of Glu 278 was detected in this spectrum. These observations confirm the proposal [Smirnova I.A., et al. (1999) Biochemistry 38, 6826-6833] that the proton required for chemistry at the binuclear site is taken from Glu 278 in the perroxy-->ferryl step, and that the rate of the next step (ferryl-->oxidized) is limited by reprotonation of Glu 278 from the N-side of the membrane in the D124N mutant enzyme. The blockage of the D-pathway in this mutant for the first time allowed direct detection of deprotonation of Glu 278 and its reprotonation during oxidation of cytochrome oxidase by O2.     

Kinetics of the interaction of the cytochrome c oxidase of Paracoccus denitrificans with its own and bovine cytochrome c

We have devised a relatively simple method for the purification of cytochrome aa3 of Paracoccus denitrificans with three major subunits similar to those of the larger subunits of the mitochondrial cytochrome oxidase. This preparation has no c-type cytochrome. Studies were made of the oxidation of soluble cytochromes c from bovine heart and Paracoccus. The cytochrome-c oxidase activity was stimulated by low concentrations of either cytochrome c, providing an explanation for the multiphasic nature of plots of v/S versus v. Kinetics of the oxidation of bovine cytochrome c by the Paracoccus oxidase resembled those of bovine oxidase with bovine cytochrome c in every way; the Paracoccus oxidase with bovine cytochrome c can serve as an appropriate model for the mitochondrial system. The kinetics of the oxidation of the soluble Paracoccus cytochrome c by the Paracoccus oxidase were different from those seen with bovine cytochrome c, but resembled the latter if poly(L-lysine) was added to the assays. The important difference between the two species of cytochrome c is the more highly negative hemisphere on the side of the molecule way from the heme crevice in the Paracoccus cytochrome. Thus, the data emphasize the importance of all of the charged groups on cytochrome c in influencing the binding or electron transfer reactions of this oxidation-reduction system. The data also permit some interesting connotations about the possible evolution from the bacterial to the mitochondrial electron transport system.     

Mutations in the Ca2+ binding site of the Paracoccus denitrificans cytochrome c oxidase.

Recent structure determinations suggested a new binding site for a non-redox active metal ion in subunit I of cytochrome c oxidase both of mitochondrial and of bacterial origin. We analyzed the relevant metal composition of the bovine and the Paracoccus denitrificans enzyme and of bacterial site-directed mutants in several residues presumably liganding this ion. Unlike the mitochondrial enzyme where a low, substoichiometric content of Ca2+ was found, the bacterial wild-type (WT) oxidase showed a stoichiometry of one Ca per enzyme monomer. Mutants in Asp-477 (in immediate vicinity of this site) were clearly diminished in their Ca content and the isolated mutant enzyme revealed a spectral shift in the heme a visible absorption upon Ca addition, which was reversed by Na ions. This spectral behavior, largely comparable to that of the mitochondrial enzyme, was not observed for the bacterial WT oxidase. Further structure refinement revealed a tightly bound water molecule as an additional Ca2+ ligand.     

The structure of Paracoccus denitrificans cytochrome c550.

The crystal structure of Paracoccus (formerly Micrococcus) denitrificans cytochrome c550 has been solved by x-ray diffraction to a resolution of 2.45 A. In both amino acid sequence and molecular structure it is evolutionarily homologous with mitochondrial cytochrome c from eukaryotes and photosynthetic cytochrome c2 from purple non-sulfur bacteria. All of these cytochromes c have the same basic folding pattern, with surface insertions of extra amino acids in c550. Various strains of c2 have all, some, or none of the extra insertions observed in c550. The hydrophobic heme environment, position of aromatic rings, and structure and environment of the heme crevice, are virtually identical in cytochromes c55o, c, and c2. Radical changes observed at all regions on the molecular surface except the heme crevice argue for the importance of the crevice and the exposed edge of the heme in the transfer of electrons to and from the cytochrome molecule.     

Proton-coupled electron equilibrium in soluble and membrane-bound cytochrome c oxidase from Paracoccus denitrificans

The pH dependence of electron and proton re-equilibration upon CO photolysis from two-electron-reduced aa3 oxidase was followed by time-resolved electrometry and optical spectroscopy. Optical spectroscopy on soluble Paracoccus denitrificans enzyme at alkaline pH revealed a slow (1 ms) component of electron re-equilibration coupled to the release of protons from the catalytic site. In the work [Br?ndén, M., et al. (2003) Biochemistry 42, 13178-13184], it was proposed that this proton is released from a water molecule in the catalytic site, located deep in the membrane dielectric. Movement of charged particles such as protons across the dielectric should create an electric potential. However, recording of the time course of the potential generation did not show any potential development in the millisecond time domain, but instead, potential generation was found with an apparent time constant of 50-100 micros. This potential was generated upon proton release from the level of the binuclear catalytic site through the K-channel, because mutation in this channel abolishes the potential generation altogether. The apparent inconsistency between results from optical spectroscopy and electrometry was solved by optical experiments on the membrane-incorporated enzyme. Reconstituting the enzyme into proteoliposomes speeds up the slow electron redistribution process by a factor of 10 and shows the same time constant as potential generation. The possible mechanism of such dramatic change in the rate of proton transfer is discussed.     

Tracing the D-pathway in reconstituted site-directed mutants of cytochrome c oxidase from Paracoccus denitrificans

Heme-copper terminal oxidases use the free energy of oxygen reduction to establish a transmembrane proton gradient. While the molecular mechanism of coupling electron transfer to proton pumping is still under debate, recent structure determinations and mutagenesis studies have provided evidence for two pathways for protons within subunit I of this class of enzymes. Here, we probe the D-pathway by mutagenesis of the cytochrome c oxidase of the bacterium Paracoccus denitrificans; amino acid replacements were selected with the rationale of interfering with the hydrophilic lining of the pathway, in particular its assumed chain of water molecules. Proton pumping was assayed in the reconstituted vesicle system by a stopped-flow spectroscopic approach, allowing a reliable assessment of proton translocation efficiency even at low turnover rates. Several mutations at positions above the cytoplasmic pathway entrance (Asn 131, Asn 199) and at the periplasmic exit region (Asp 399) led to complete inhibition of proton pumping; one of these mutants, N131D, exhibited an ideal decoupled phenotype, with a turnover comparable to that of the wild-type enzyme. Since sets of mutations in other positions along the presumed course of the pathway showed normal proton translocation stoichiometries, we conclude that the D-pathway is too wide in most areas above positions 131/199 to be disturbed by single amino acid replacements.     

Interaction of cytochrome c with cytochrome c oxidase: an NMR study on two soluble fragments derived from Paracoccus denitrificans

The functional interactions between the various components of the respiratory chain are relatively short-lived, thus allowing high turnover numbers but at the same time complicating the structural analysis of the complexes. Chemical shift mapping by NMR spectroscopy is a useful tool to investigate such transient contacts, since it can monitor changes in the electron-shielding properties of a protein as the result of temporary contacts with a reaction partner. In this study, we investigated the molecular interaction between two components of the electron-transfer chain from Paracoccus denitrificans: the engineered, water-soluble fragment of cytochrome c(552) and the Cu(A) domain from the cytochrome c oxidase. Comparison of [(15)N,(1)H]-TROSY spectra of the [(15)N]-labeled cytochrome c(552) fragment in the absence and in the presence of the Cu(A) fragment showed chemical shift changes for the backbone amide groups of several, mostly uncharged residues located around the exposed heme edge in cytochrome c(552). The detected contact areas on the cytochrome c(552) surface were comparable under both fully reduced and fully oxidized conditions, suggesting that the respective chemical shift changes represent biologically relevant protein-protein interactions.     

ATR-FTIR spectroscopy and isotope labeling of the PM intermediate of Paracoccus denitrificans cytochrome c oxidase

The structure of the P(M) intermediate of Paracoccus denitrificans cytochrome c oxidase was investigated by perfusion-induced attenuated total reflection-Fourier transform infrared (ATR-FTIR) spectroscopy. Transitions from the oxidized to P(M) state were initiated by perfusion with CO/oxygen buffer, and the extent of conversion was quantitated by simultaneously monitoring visible absorption changes. In prior work, tentative assignments of bands were proposed for heme a(3), a change in the environment of the protonated state of a carboxylic acid, and a covalently linked histidine-tyrosine ligand to Cu(B) that has been found in the catalytic site. In this work, reduced minus oxidized difference spectra at pH 6.5 and 9.0 and P(M) minus oxidized difference spectra at pH 9.0 were compared in unlabeled, universally (15)N-labeled, and tyrosine-ring-d(4)-labeled proteins to improve these assignments. In the reduced minus oxidized difference spectrum, (15)N labeling resulted in large changes in the amide II region and a 9 cm(-1) downshift in a 1105 cm(-1) trough that is attributed to histidine. In contrast, changes induced by tyrosine-ring-d(4) labeling were barely detectable where the isotope-sensitive bands are expected. Both isotope substitutions had large effects on P(M) minus oxidized difference spectra. A prominent trough at 1542 cm(-1) was shifted to 1527 cm(-1) with (15)N labeling, and its magnitude was diminished with the appearance of a 1438 cm(-1) trough with tyrosine-ring-d(4) labeling. Both isotope substitutions also had large effects on a 1314 cm(-1) trough in the same spectra. These shifts indicate that the bands are linked to both a nitrogenous compound and a tyrosine, the most obvious candidate being the covalent histidine-tyrosine ligand of Cu(B). Comparison with model material data suggests that the tyrosine hydroxyl group is protonated when the binuclear center is oxidized but deprotonated in the P(M) intermediate. Positive bands at 1519 and 1570 cm(-1) were replaced with bands at 1504 and 1556 cm(-1), respectively, with tyrosine-ring-d(4) labeling, are characteristic of upsilon(7a)(C-O) and upsilon(C-C) bands of neutral phenolic radicals, and most likely reflect the formation of the neutral radical state of the histidine-tyrosine ligand in P(M).     

Amino acid sequence of Paracoccus denitrificans cytochrome c550.

The amino acid sequence of Paracoccus (formerly Micrococcus) denitrificans cytochrome c550 has been established by a combination of standard chemical techniques and interpretation of a 2.5 A resolution x-ray electron density map. Peptides derived from a trypsin digest were chemically sequenced, and then ordered by fitting them to the density map. The amino acid compositions of chymotryptic peptides confirmed the x-ray map ordering the tryptic peptides. The amino acid sequence of this respiratory, prokaryotic cytochrome with 134 residues is discussed in relation to those of eukaryotic respiratory cytochrome c (103 to 113 amino acids), and prokaryotic, photosynthetic c2 (103 to 124 amino acids). At the primary structure level, c and c550 differ no more from cytochromes c2 than the various cytochromes c2 do from one another. It is suggested that the respiratory electron transport chain in prokaryotes and eukaryotes is a relatively late evolutionary offshoot of the photosynthetic electron transport chain in purple non-sulfur bacteria.