ARCHIVES

OF BIOCHEMISTRY

AND BIOPHYSICS

Vol. 287, No. 1, May 15, pp. 128-134, 1991

Redox Potentials of Flavocytochromes c from the Phototrophic Bacteria, Chromatium vinosum a nd Chlorobium thiosulfa tophilum T. E. Meyer,l

R. G. Bartsch,

M. S. Caffrey,

and M. A. Cusanovich

Department of Biochemistry, University of Arizona, Tucson, Arizona 85721

Received November 26, 1990, and in revised form January 14, 1991

The redox potentials of flavocytochromes c (FC) from Chromatium vinosum and Chlorobium thiosulfatophilum have been studied as a function of pH. Chlorobium FC has a single heme which has a redox potential of f98 mV at pH 7 (N = 1) that is independent of pH between 6 and 8. The average two-electron redox potential of the flavin extrapolated to pH 7 is $28 mV and decreases 35 mV/ pH between pH 6 and 7. The anionic form of the flavin semiquinone is stabilized above pH 6. The redox potential of Chromatium FC is markedly lower than for Chlorobium. The two hemes in Chromatium FC appear to have a redox potential of 15 mV at pH 7 (N = l), although they reside in very different structural environments. The hemes of Chromatium FC have a pa-dependent redox potential, which can be fit in the simplest case by a single ionization with pK = 7.05. The flavin in Chromatium FC has an average two-electron redox potential

of -26 mV at pH 7 and decreases 30 mV/pH between pH 6 and 8. As with Chlorobium, the anionic form of the flavin semiquinone of Chromatium FC is stabilized above pH 6. The unusually high redox potential of the flavin, a stabilized anion radical, and sulfite binding to the flavin in both Chlorobium and Chromatium FCs are characteristics shared by the flavoprotein oxidases. By analogy with glycolate oxidase and lactate dehydrogenase for which there are three-dimensional structures, the properties of the FCs are likely to be due to a positively charged amino acid side chain in the vicinity of the Nl nitrogen of the fiavin. %,1991 Academic Press, Inc.

Flavocytochromes c (FC)” are generally found in those purple and green phototrophic bacterial species utilizing reduced sulfur compounds and in particular those which utilize both sulfide and thiosulfate (1). FC appears to 1To whom correspondence should be addressed. ’ Abbreviation used: FC, flavocytochrome c. 128

function as a sulfide dehydrogenase (a), although there are species which utilize sulfide, but which do not have detectable FC. The molecular weight of FC is 50-72,000, consisting of both flavoprotein and cytochrome c subunits (containing one or two hemes depending on source), which are tightly bound and which appear to dissociate only under denaturing conditions. Moreover, the FAD is covalently bound to the flavoprotein subunit via a thioether bridge between a cysteine sulfur and the 8-01 position of the flavin ring (3). Inhibitors of enzyme activity such as cyanide (2) form adducts with the flavin, which have charge transfer absorbance in the region of 650 nm (4). Other nucleophiles, such as the physiologically significant substrates thiosulfate, sulfite, and mercaptans also form adducts with the flavin. Free flavins and flavoproteins also form adducts with nucleophiles like sulfite, but these do not have charge transfer bands (5, 6). Reactivity with sulfite can be correlated with increasing flavin redox potential (5), which suggests a rationale for reactivity of FC with nucleophiles. Electron transfer in FC is from donor to the bound flavin followed by rapid intramolecular electron transfer (>106 s-l) to the heme and finally to an electron acceptor (7). Thus, the FCs transfer electrons along a defined path and have positioned the heme and flavin groups to permit rapid intramolecular electron transfer. In view of the unusual properties of the FCs, it is of considerable interest to understand the oxidation-reduction potentials of the prosthetic groups in order to obtain insight into the chemistry of adduct formation as well as Aavin-heme interaction. To this end we have determined the redox potentials of FC from Chromatium vinosum (72 kDa, two hemes, one flavin) and from Chlorobium thiosulfatophilum (50 kDa, one heme, one flavin). The redox potential of the heme in Chromatium FC has been reported to range between +lO and +45 mV at pH 7 (S-11) and of Chlorobium to be +98 mV (12), but the flavin potential has not been studied to date. 0003-9%1/91 Copyright All rights

0

1991

of reproduction

by Academic in any

form

$3.00

Press,

Inc.

reserved.

REDOX

POTENTIALS

OF

Chromatium

and

Chlorobium

FLAVOCYTOCHROMES

c

129

platinum electrode in an anaerobic cuvet. The buffer contained 10 mM EDTA, 1 mM FeCI,, 50 mM phosphate, 50 mM acetate, 50 mM Tris, and 33 fiM FC, adjusted to the appropriate pH. The solution potential was changed by titration with an anaerobic solution of 0.1 M Fe(NH&(SO& in 10 mM HCl. The pK for the loss of 690.nm absorbance was determined at a protein concentration of 100 pM. The buffer contained 1 mM phosphate adjusted to the desired pH by addition of 0.3 M NaOH.

RESULTS

Wovelength FIG.

1.

The heme heme and conditions

MATERIALS

Redox titration has maximal Aavin absorb of titration.

AND

(nm)

of Chromatium flavocytochrome c at pH 7.0. absorbance at 552 and 522 nm, whereas both light between 440 and 500 nm. See text for Full scale absorbance is 1.0.

METHODS

C. uinosum strain D FC was prepared as previously described (13). The ratio of 280. to 410.nm absorbance was 0.55 and the ratio of 475. to 525.nm absorbance was 1.3. The former is the ratio to be expected for pure protein. The latter shows that the flavin is not complexed with any of the naturally occurring ligands and is believed to be indicative of the native flavin-to-heme stoichiometry. The extinction coefficient is 54 mM-’ cm- I for reduced protein at 552 nm (14). C. thiosulfatophilum strain Tassajara FC was prepared as described by Meyer et al. (12). The purity index was 0.94 and the 475. to 525.nm absorbance ratio was 1.5, both of which are to be expected for pure, uncomplexed protein. The extinction coefficient is 30 mM~ ’ cm ’ for reduced protein at 553 nm. The redox potential of Chromatium FC was measured using a stirred spectroelectrochemical cell which had a volume of about 3 ml and a lcm optical path similar to that described by Dutton (15). The buffer contained 20 mM acetate, 20 mM phosphate, 20 mM Tris, 10 mM EDTA, and 18 pM FC, adjusted to the appropriate pH. The mediator was 1 mM iron-EDTA, the reductant was 1 mM methyl viologen, and the oxidant was 1 mM potassium ferricyanide. The redox potential of Fe-EDTA is 100 mV at pH 7 (16). Ag/AgCl reference and auxiliary electrodes and a platinum working electrode were fitted to the reaction vessel through tapered glass joints. The solution potential applied between auxiliary and Pt electrodes was provided by a Princeton Applied Research Model 363 potentiostat. The potential between the reference and the Pt electrodes was measured using a Keithley Model 191 digital multimeter. Absorption spectra were measured with a Cary 118 spectrophotometer fitted with a magnetic stirrer. Titrations were performed at room temperature, 23°C. The redox potential of Chlorobium FC was measured prior to acquisition of the spectroelectrochemical cell by using a combination calomel/

A representative redox titration of Chromatium FC at pH 7 is shown in Fig. 1. The apparent isosbestic points at 435, 507, 533, 540, and 560 nm indicate that no significant concentration of neutral flavin semiquinone is produced during the course of the titration, and therefore no correction to heme absorbance changes at 552 nm is required. A Nernst plot of the 552-nm data of Fig. 1 is shown in Fig. 2. The data points define a straight line with a slope of 60 mV (N = 1) and a midpoint potential of +15 mV at pH 7. The redox reaction is fully reversible with superposition of both forward and reverse titrations, and there is no significant curvature in the data extending to 90% oxidation and reduction. Errors in slope are estimated at about lo%, which indicates that the two hemes have redox potentials, which differ less than about 60 mV. If the two heme potentials differed by more, there would be obvious curvature at the extremities of the Nernst plot or there would be significant deviation from N= 1. The Chromatium FC heme gave linear Nernst plots with N = 1 at all six pH values examined and the titrations were all fully reversible. This establishes that the two hemes have nearly equal midpoint potentials at all pH values studied. The effect of pH on the redox potential of the heme is shown in Fig. 3. The dotted line was calculated assuming a single ionization of the oxidized protein with a pKO of 7.05, using the equation E = E + 59

-60-1

4

-1

FIG.

0 log Ox/Red

2. Nernst plot of the 552-nm data for the heme of Chromatium Havocytochrome c shown in Fig. 1. Open circles are from the reductive titration and tilled circles are from the oxidative titration. Least squares tit of the data give a line with a slope of 60 mV (N = 1).

1

130

MEYER

ET

AL.

-604 5

6

7

6

PH FIG. 3. Dependence of Chromatium flavocytochrome c on pH. Heme data (552 nm, filled circles) were fit with 7.05 (dotted line) or with two pK’s (solid line). Flavin open circles) were connected with a straight line, which -30 mV/pH.

redox potential a single pK at data (475 nm, has a slope of

log (H+)/(K,, + (H+)) (17). A better fit to the data was obtained by assuming two ionizations (solid line), using the equation E = E + 59 log (K, + (H+))/(K, + (H+)). The two pK values were found to be 6.7 and 7.5 for the oxidized and reduced forms, respectively. However, it is not possible to distinguish between these two models without consideration of additional data and the subsequent discussion will focus on the single pK model. We considered the possibility that the ionization observed in the redox titration might be correlated with loss of a 695nm absorbance band. Chromatium FC does have a weak absorbance band at 690 nm (extinction coefficient about 0.7 rnM--’ cm-’ hemeel, compared to 1.1 mM-’ cm-’ for horse cytochrome c), which titrated with a pK of 10.3 as shown in Fig. 4. To determine the midpoint potential of the flavin, we first corrected the absorbance changes at 475 nm for the contribution of the heme. The absorbance change at 475 nm due to the heme is estimated to be 18% of the absorbance change at 552 nm as determined with the isolated heme subunit. A Nernst plot was then constructed as shown in Fig. 5. The average two-electron potential for the flavin at pH 7 is -26 mV. The titration was completely reversible upon oxidation and reduction in the sense that the forward and reverse plots were congruent. However, the flavin absorbance was not completely recovered on reoxidation (about 5% loss), indicating a small amount of denaturation. Brown (9) found that in his protein sample, only about 50% of the flavin absorbance was recovered on reoxidation. The data, plotted in Fig. 5, fall on a straight line which has a slope of 77 mV (N = 0.76). A nonintegral N-value was also observed by Brown (9), who obtained a value of 0.63. The large slope (77 mV instead of the normal 30 mV for an N = 2 system) indicates that the first one-electron step of reduction of the flavin has

FIG. 4. pH titration of the 690.nm chrome c. The pK value was calculated the slope was found to be 1.0 + 0.1.

2 PH peak of Chromatium flavocytofrom a Hill plot of the data, and

a significantly higher potential than does the second, and thus the semiquinone is stabilized (17). This conclusion implies a potential source of error in the two-electron potential since it was assumed that the absorbance of half-reduced flavin was half that of the oxidized form. For example, in lactate oxidase, the semiquinone anion has only 30% of the absorbance of oxidized flavin at 450 nm rather than 50% (18). The semiquinone formation constant was estimated from analysis of the Nernst plot according to Clark (17). The formation constant was calculated using the formula Kf = (A - 3/A)‘, where A = antilog (Ey5% - E,)/59, and ETsB is the potential at 75% oxidized flavin (17). The difference in one-electron redox potentials is Ed = 59 log Kf. The formation constant Kf was found to be 12.5 and Ed about 65 mV at pH 7 for Chromatium FC. If the semi-

“V

1

40 -20-0 -5

-2o--40.. -60.. -801 -1

0 log Ox/Red

-I 1

FIG. 5. Nernst plot of the Chromatium flavin 475 nm data from Fig. 1. The heme contribution at 475 nm was taken to be 18% of the absorbance change at 552 nm and substracted prior to plotting the data. Least squares fit of the data gave a line with a slope of 77 mV or N = 0.76.

REDOX

POTENTIALS

OF

Chromatium

and

FLAVOCYTOCHROMES

c

131

160

160

P .5 El

Chlorobium

100 80 60 40 20 -1 log

0 Ox/Red

1 log

FIG. 6. Nernst plot of the 553.nm heme data from the Ch2orobium flavocytochrome c reductive redox titration at pH 6.76 (data not shown). See text for conditions.

quinone is stabilized as suggested by the large slope in the Nernst plot, then it must be in the anionic form at pH 7, because there was no sign of the neutral radical in the visible spectrum between 500 and 600 nm. The average two-electron potential for the flavin plotted vs pH is shown in Fig. 3. The titration of the flavin resulted in denaturation at pH 5 (partial precipitation of the flavin subunit) and was not fully reversible at pH 6 and 7.76 (approximately lo-20 mV hysteresis), therefore the semiquinone formation constant was not calculated other than at pH 7. A 50-mV hysteresis at pH 8.76 was also observed by Brown (9). Another reason we did not generally calculate the one-electron redox potentials is that the slope of the Nernst plot and thus the formation constant is sensitive to the size of the correction for the heme contribution. Thus, it is clear that the semiquinone is stabilized, but we do not know the magnitude except within large error limits.

1201 100

6

7

8

PH FIG. 7. tential on pK of 8.6. a straight

Dependence of the Chlorobium Aavocytochrome c redox popH. The heme data in filled circles were fitted with a single The two flavin data points (open circles) were connected with line, which has a slope of -35 mV/pH.

1 Ox/Red

FIG. 8. Nernst plot of the 475.nm flavin data from the Chlorobium flavocytochrome c redox titration at pH 6.76 after the heme contribution was subtracted out. The change due to heme at 475 nm was taken to be 18% of that at 553 nm. Least squares fit of the data gave a straight line with a slope of 52 mV or an apparent N value of 1.13.

The slope of the E, vs pH curve for the flavin of Chromatium FC is -30 mV/pH between pH 6 and 8, which is typical of free flavins such as FMN which has a -30 mV/ pH slope between pKa and pKo (19). This result indicates that the pKR for fully reduced flavin in FC is below 6 and the pKo for fully oxidized flavin in FC is above 8 (they are 6.7 and 10.3 in free FMN, respectively). The pKs or ionization constant for flavin semiquinone in FC must also be slightly below pH 6 because a small amount of neutral radical was visible at that pH (through loss of isosbestic points). In a separate study, the pKs was estimated as 5.7 from kinetics of FC flavin reduction (7). In contrast, the pKs for free FMN is 8.6 (19). The downward shift of nearly 3 pH units for pKs suggests that there is a positive charge near the FAD in FC which both stabilizes the anionic form of the semiquinone and lowers the pK for the fully reduced flavin. This is analyzed in more detail in the discussion. A representative Nernst plot for the heme in the Chlorobium FC redox titration is shown in Fig. 6. The midpoint redox potential is nearly independent of pH between pH 5 and 8 as shown in Fig. 7, with the redox potential 98 mV at pH 7. If the 5-mV drop at pH 8 is real, then there could be a pK in the oxidized form at 8.6. This behavior is in marked contrast to that of Chromatium FC, which, as noted above, has at least one redox potential-dependent ionization near pH 7. The absorbance changes at 475 nm were corrected for the changes due to heme at this wavelength (18%) and a Nernst plot of the flavin data is shown in Fig. 8. The slope of the Nernst plot is 52 mV instead of the expected 30 mV for an N = 2 reaction, which indicates that the semiquinone is stabilized in Chlorobium FC as well as in Chromatium. However, the apparent difference between oneelectron potentials (Ed) for the flavin in Chlorobium FC

132

MEYER

is only 21 mV (Kf = 2.27). The redox potential is pH dependent as shown in Fig. 7, with the average two-electron potential extrapolated to pH 7 equal to +28 mV. The flavin potential decreases 35 mV/pH between pH 6 and 7, consistent with a pKs value below pH 6, just as with Chromatium FC. The flavin potential was too low at pH 8 to be reached by the iron-EDTA couple. DISCUSSION The redox potentials of the hemes in the two phototrophic bacterial FCs are among the lowest reported for c-type cytochromes having histidine and methionine ligands (15 and 98 mV). The average potential for 17 cytochromes ca is 343 * 48 mV (20), which is more or less typical of the class I cytochromes c. The lowest potential class I c-type cytochromes previously reported are from Rhodocyclus gelatinosus and Ectothiorhodospira hulophila, which have potentials of 28 and 58 mV, respectively (21) and from Desulfovibrio species, which are 20 and 40 mV (22). Possible explanations for the unusually low redox potentials for the cytochrome subunits of FC are that the heme is more exposed to solvent or there is a buried negative charge which stabilizes the oxidized state of the heme. However, the three-dimensional structure is necessary to determine those factors which contribute to the unusually low redox potentials. Chromatium FC is unusual among diheme cytochromes in that both hemes appear to have the same potential. This would not be surprising if the two halves of the protein had nearly identical sequences, but the amino acid sequences of the two halves of the protein are actually as different as any two cytochromes c from different species (Van Beeumen, unpublished results). By way of contrast, the Pseudomonas and Azotobacter cytochromes cq are diheme proteins homologous to FC, and have measurably different potentials for the two hemes (23). All four hemes in the Desulfovibrio cytochromes c3 (24) and in the Rhodopseudomonas viridis reaction center cytochrome (25) also have markedly different redox potentials. The low temperature EPR spectrum of Chromatium FC indicates that the two hemes are distinguishable and that one of the two hemes undergoes a pH-dependent conformational change with a pK about 7.5 (26). The redox potential was not measured at low temperature, thus it is not possible to relate the low temperature pK of 7.5 with the room temperature pK for the oxidized heme (pK 6.7-7.05). Furthermore, only one of the two hemes binds carbon monoxide (14, 27). Chromatium FC is also unusual in that there is a strong heme-heme interaction which is enhanced by CO binding as indicated by the circular dichroism spectrum (14, 28). It may be that the strong heme-heme interaction is responsible for the observation of only one midpoint potential for the two hemes. Most c-type cytochromes whose redox potentials have been studied as a function of pH show a redox-linked pK

ET

AL.

in the oxidized form above neutrality. For example, fifteen cytochromes c2 have an average pK of 8.5 + 0.5 although some are as low as 7.2 (20). We considered the possibility that the pK of 7 observed with the Chromatium FC heme could be due to the same causes as seen with the cytochromes c2. This redox-linked pK is generally associated with loss of a 695-nm absorbance band in the near infrared and to changes in the chemical shifts of protons of the heme and its ligands in NMR experiments. These combined phenomena have most often been explained as substitution of a lysine for the methionine sixth ligand (29). However, we determined the pK for loss of 690-nm absorbance in Chromatium FC and found that it is 10.3. Thus, the redox-linked ionization for the oxidized heme with pK 7 must be different. An explanation for additional heme-linked ionizations not connected with the 695-nm absorbance change in some of the cytochromes c2 is that a buried heme propionate has no positively charged residue nearby and thus has a higher than normal pK (20). In these examples, the pK’s in the oxidized and reduced cytochrome are separated by about 1 pH unit and are lower than that for the 695-nm band titration. If we assume that there are two pK’s in the Chromatium FC at 6.7 for oxidized protein and 7.5 for reduced protein, then a buried heme propionate would be a reasonable explanation because the pK’s are separated by about the same amount as in other cytochromes (0.8 unit). The redox potentials of the hemes in the FCs have been established with greater certainty (+5 mV) than have those of the flavin (~15 mV) because there is a larger absorbance change due to heme without significant interference by the flavin. For the flavin titration, it is necessary to subtract the contribution of the heme from the smaller absorbance change at 475 nm, with the correction 25-40s of the total absorbance change at 475 nm, depending on species. Nevertheless, the slope of the Nernst plots for the flavin were found to be significantly different from what is expected for free flavin, and consistent with stabilization of the semiquinone intermediate by the protein. However, the slope of the Nernst plots for the flavin titration were found to vary with the size of the correction due to heme, which made it difficult to determine oneelectron flavin redox potentials with confidence. Furthermore, there was no clear break in the flavin titration which would signal the halfway point. Therefore, it appears that the two one-electron potentials for the flavin in FC are relatively close together. Our best estimate at pH 7 is 65 and 21 mV for the Chromatium and Chlorobium FCs, respectively. Based on this analysis, we can estimate one-electron potentials of +6 and +38 mV for the oxidized to semiquinone transition for Chromatium and Chlorobium FC, respectively. Similarly, the semiquinone-to-fully reduced transition would have potentials of -58 and $18 mV. In order to obtain more reliable one-electron potentials, it would be desirable to use a technique which is not

REDOX

POTENTIALS

OF

Chromatium

and

Chlorobiun

FLAVOCYTOCHROMES

c

133

dependent on observations of optical changes, such as addition, both have exceptionally high two-electron redox potentials for the flavin and have a stabilized anionic EPR, where the semiquinone can be monitored directly. semiquinone. Based on the analogy to flavin-containing In contrast to the heme subunits of FC, which have class II dehydrogenases/oxidases, we can conclude that a some of the lowest potentials in their class, the flavin basic amino acid side chain (presumably from lysine) is subunits have some of the highest known two-electron positioned near the N-l position of the flavin. Moreover, potentials of any flavoprotein (-26 mV and f28 mV for Chromatium and Chlorobium, respectively). This is in studies on kinetics of flavin-sulfite adduct formation suggest the presence of histidine and cysteine (liberated by sharp contrast to free FMN, which has a two-electron reaction of sulfite with a cystine disulfide) near the flavin. potential of -205 mV (19) and to flavodoxin with a twoelectron potential of -293 mV and a stabilized neutral The observation of charge transfer absorbance for the adducts also demands an acceptor, which may be an arradical (30). Moreover, the flavin in FC has a stabilized red or anionic semiquinone at pH 6-8 and forms a strong omatic amino acid side chain such as phenylalanine. complex with sulfite. These characteristics are typical of Taken together, the evidence supports the existence of the class 2 dehydrogenase/oxidase enzymes of Massey and four amino acid residues (lysine, histidine, cystine, and phenylalanine) which are sufficiently close to the flavin Hemmerich (31). It has been proposed that the stabilized red radical, the high potential, and the ligand binding in FC to influence its properties. Although structural incharacteristics of class 2 flavoprotein oxidases may be formation is required to confirm the proposals made here, due to the presence of a basic residue near the N-l position it is clear that the FC are unique among both heme and of the flavin (31). The three-dimensional structures of flavin-containing proteins and will yield new insights into glycolate oxidase and yeast flavocytochrome b show that biological electron transfer. there is in fact a lysine which is near the N-l position of the flavin (32,33). Both proteins have relatively high two- ACKNOWLEDGMENT electron redox potentials (-68 mV and -64 mV, respecThis work was supported in part by a grant from the National Science tively) (34, 35) and are reactive with sulfite. Thus, it is Foundation, DMB 88-05513. reasonable to conclude that the flavin in FC also has a positively charged residue near the N-l position which is largely responsible for the high redox potential, for the REFERENCES stabilized red radical, and for the strong reactivity with 1. Cusanovich, M. A., Meyer, T. E., and Bartsch, R. G. (1991) in sulfite as well as other nucleophiles. Irrespective of these Chemistry and Biochemistry of Flavoenzymes Vol. II (Muller, F., Ed.), CRC Press, Boca Raton, FL, in press. similarities, the FC sulfite adducts form charge transfer 2. Yamanaka, T., and Kusai, A. (1976) in Flavins and Flavoproteins complexes, which have not been observed with any other (Singer, T. P., Ed.), pp. 292-302, Elsevier, Amsterdam. flavoproteins, and suggest the presence of a unique charge 3. Kenney, W. C., McIntire, W., and Yamanaka, T. (1977) Biochim. transfer acceptor in FC. Riophys. Acta 483, 467-474. It is possible that covalent binding of FAD in FC is 4. Meyer, T. E., and Bartsch, R. G. (1976) in Flavins and Flavoproteins partly responsible for the high redox potentials and low(Singer, T. P., Ed.), pp. 312-317, Elsevier, Amsterdam. ered pK’s. It has been found that the redox potentials of 5. Muller, F., and Massey, V. (1969) J. Biol. Chem. 244, 4007-4016. 8-a-N-imidazole-substituted flavins are about 50 mV F., Feldberg, R., Schuman, M., Sullivan, P. A., 6. Massey, V., Muller, higher than for free flavins and their pK’s have been lowHowell, L. G., Mayhew, S. G., Matthews, R. G., and Foust, G. P. ered by about 1 unit (36). The redox properties of B-(u(1969) J. Biol. Chem. 244, 3999-4006. cysteinyl-substituted flavins, such as found in FC, have I. Cusanovich, M. A., Meyer, T. E., and Tollin, G. (1985) Biochemistry not been studied. However, the lack of charge and reduced 24, 1281-1287. inductive effect of cysteinyl sulfur suggeststhat covalent 8. Bartsch, R. G., and Kamen, M. D. (1960) J. Biol. Chem. 235,825831. binding has only a small effect on the redox properties of Arizona. of FC. 9. Brown, S. L. (1981) M.S. Thesis, University None of the three flavin ionizations (pK,., pK,, and pK,) 10. Meyer, T. E., Vorkink, W. P., Tollin, G., and Cusanovich, M. A. (1985) Arch. Biochem. Biophys. 236, 52-58. estimated from the redox titrations of Chromatium FC 11. Guo, I,. H., Hill, H. A. O., Hopper, D. J., Lawrence, G. A., and are similar to the pK values observed in flavin ligandSanghera, G. S. (1990) J. Biol. Chem. 265, 1958-1963. binding studies (6.6 and 8.4) (37, 38). Thus, the ligandM. A., and Mathewson, binding pK values likely result from the interaction of 12. Meyer, T. E., Bartsch, R. G., Cusanovich, J. H. (1968) Biochim. Biophys. Acta 153, 854-861. the flavin with amino acid side chains, such as histidine 13. Bartsch, R. G. (1971) in Methods in Enzymology (San Pietro, A., (the pK at 6.6) and cysteine (the pKat 8.4) which influence Ed.), Vol. 23, pp. 3444363, Academic Press, San Diego. adduct formation, but not flavin redox potentials. 14. Bartsch, R. G., Meyer, T. E., and Robinson, A. B. (1968) in Structure In summary, the two FCs, although structurally distinct and Function of Cytochromes (Okunuki, K., Kamen, M. D., and in terms of size and number of hemes, have very similar Sekuzu, I., Eds.), pp. 443-451, University Park Press, Baltimore. redox properties. Both have relatively low heme redox 15. Dutton, P. L. (1978) in Methods in Enzymology (Fleischer, S., and potentials as compared to class I c-type cytochromes. In Packer, L., Eds.), Vol. 54, pp. 411-435, Academic Press, San Diego.

134

MEYER

16. Kolthoff, I. M., and Auerbach, C. (1952) J. Am. Chem. Sot. 74, 1452-1456. 17. Clark, W. M. (1960) Oxidation Reduction Potentials of Organic Systems, Williams and Wilkins, Baltimore. 18. Stankovich, M., and Fox, B. (1983) Biochemistry 22, 4466-4472. 19. Draper, R. D., and Ingraham, L. L. (1968) Arch. Biochem. Biophys. 125,802-808. 20. Pettigrew, G. W., Bartsch, R. G., Meyer, T. E., and Kamen, M. D. (1978) Biochim. Biophys. Acta 503, 509-523. T. E., Przysiecki, C. T., Watkins, J. A., Bhattacharyya, A., 21. Meyer, Simondsen, R. P., Cusanovich, M. A., and Tollin, G. (1983) Proc. Nutl. Acad. Sci. USA 80,6740-6744. 22. Bianco, P., Haladjian, J., Lout& M., and Bruschi, M. (1983) Biochem. Biophys. Res. Commun. 113, 526-530. 23. Leitch, F. A., Brown, K. R., and Pettigrew, G. W. (1985) Biochim. Biophys. Acta 808, 213-218. 24. Fan, K., Akutsu, H., Kyogoku, Y., and Niki, K. (1990) Biochemistry 29, 2257-2263. 25. Fritzsch, G., Buchanan, S., and Michel, H. (1990) Biochim. Biophys. Acta 977, 1577162. 26. Strekas, T. C. (1976) Biochim. Biophys. Acta 446, 179-191. 27. Bartsch, R. G. (1963) in Bacteria1 Photosynthesis (Gest, H., San Pietro, A., and Vernon, L., Eds.), pp. 475-494, Antioch Press, Yellow Springs, Ohio.

ET

AL.

28. Yong,

F. C., and King,

29. Davis, L. A., Schejter, 249, 262442632.

T. E. (1970)

J. Biol.

A., and Hess,

Chem.

245,1331-1335.

G. P. (1974)

J. Biol.

30. Fillat, M. F., Edmondson, D. E., and Gomez-Moreno, Biochim. Biophys. Acta 1040, 301-307. 31. Massey, 257. 32. Lindqvist, 3628. 33. Xia,

V., and Hemmerich, Y., and Branden,

Z., and Mathews,

34. Pace, C., and Stankovich,

P. (1980)

Biochem.

C. I. (1989)

F. S. (1990) M. (1986)

J. Biol.

J. Mol.

Biol.

Biochemistry

35. Capeillere-Blandin, C., Bray, R. C., Iwatsubo, (1975) Eur. J. Biochem. 54, 549-566. 36. Williamson, 7790-7797.

G., and Edmondson,

D. E. (1985)

C&m.

C. (1990)

Sot. Trans. Chem. 212, 25,

13,246~

264,36248377863. 2516-2522.

M., and Labeyrie, Biochemistry

F. 24,

37. Cusanovich, M. A., and Meyer, T. E. (1982) in Flavins and Flavoproteins (Massey, V., and Williams, C. H., Eds.), pp. 844-848, Elsevier, Amsterdam. 38. Meyer, T. E., Holden, H. M., Rayment, I., Bartsch, R. G., and Cusanovich, M. A. (1987) in Flavins and Flavoproteins (Edmondson, D. E., and McCormick, D. B., Eds.), pp. 365-369, Walter de Gruyter, New York.

Redox potentials of flavocytochromes c from the phototrophic bacteria, Chromatium vinosum and Chlorobium thiosulfatophilum.

The redox potentials of flavocytochromes c (FC) from Chromatium vinosum and Chlorobium thiosulfatophilum have been studied as a function of pH. Chloro...
766KB Sizes 0 Downloads 0 Views