Photosynthesis Research23: 29-38, 1990. © 1990 KluwerAcademic Publishers. Printed in the Netherlands. Regular paper

Isolation and characterization of the membrane-bound cytochrome c-554 from the thermophilic green photosynthetic bacterium Chloroflexus aurantiacus John C. Freeman ~ & Robert E. Blankenship

Department of Chemistry, Arizona State University, and Center for Study of Early Events in Photosynthesis, Tempe, AZ. 85287-1604, USA 1Present address: Department of Chemistry, The Pennsylvania State University, University Park, PA 16802, USA. Received 11 October 1988; accepted in revised form 21 November 1988

Key words: cytochrome c, photosynthesis, photosynthetic bacteria, electron transport, Chloroflexus aurantiacus, green bacteria Abstract

The membrane-bound photooxidizable cytochrome c-554 from Chloroflexus aurant&eus has been purified. The purified protein runs as a single heme staining band on SDS-PAGE with an apparent molecular mass of 43 000 daltons. An extinction coefficient of 28 + 1 mM -I cm -t per heme at 554nm was found for the dithionite-reduced protein. The potentiometric titration of the hemes takes place over an extended range, showing clearly that the protein does not contain a single heme in a well-defined site. The titration can be fit to a Nernst curve with midpoint potentials at 0, + 120, + 220 and + 300 mV vs the standard hydrogen electrode. Pyridine hemochrome analysis combined with a Lowry protein assay and the SDS-PAGE molecular weight indicates that there are a minimum of three, and probably four hemes per peptide. Amino acid analysis shows 5 histidine residues and 29% hydrophobic residues in the protein. This cytochrome appears to be functionally similar to the bound cytochrome from Rhodopseudomonas viridis. Both cytochrome c-554 from C. aurantiacus and the four-heine cytochrome c-558-553 from R. viridis appear to act as direct electron donors to the special bacteriochlorophyll pair of the photosynthetic reaction center. They have a similar content of hydrophobic amino acids, but differ in isoelectric point, thermodynamic characteristics, spectral properties, and in their ability to be photooxidized at low temperature.

Abbreviations; L D A O - lauryl dimethyl amine-N-oxide, SDS-PAGE - sodium dodecyl sulfate polyacrylamide gel electrophoresis, m V - millivolt, Era,8 - m i d p o i n t potential at pH 8.0, O D V - optical density × volume in mi Introduction

The central involvement ofcytochromes in electron transport processes has long been recognized. They operate in a variety of systems including the respiratory systems of mitochondria and bacteria, and in the photosynthetic apparatus of plants, algae, cyanobacteria, and the anoxygenic bacteria. They are involved as water soluble components, such as mitochondrial cytochrome c, and bacterial cytoch-

rome c2, or as membrane-bound components such as the cytochrome b/cl complexes and cytochrome oxidase. The membrane-bound cytochromes can be further separated into those which reside primarily in the membrane, such as cytochrome oxidase (Pettigrew and Moore 1987) and the cytochrome b portion of the cytochrome b/ct or b6/fcomplexes (Widger et al. 1984), and those that have a hydrophobic anchor in the membrane and function primarily in

30 the adjacent aqueous environment. Examples of the latter class are cytochrome c~, cytochrome f (Wakabayashi et al. 1982, Willey et al. 1984, Pettigrew and Moore 1987), and cytochrome c-558-553 from Rhodopseudomonas viridis (Weyer et al. 1987a). The membrane-bound multiheme cytochrome c-554 from the thermophilic green photosynthetic bacterium Chloroflexus aurantiacus serves as the electron donor to the reaction center (Bruce et al. 1982, Blankenship et al. 1983, Blankenship 1985, Wynn et al. 1987, Zannoni and Venturoli 1988). This cytochrome is present only in photosynthetically grown cells, and follows an induction pattern similar to that of bacteriochlorophylls a and c and the reaction center upon shifting from aerobic to photosynthetic growth conditions (Pierson 1985, Foster et al. 1986). C. aurantiacus appears to have diverged from other photosynthetic organisms at a very early point (Woese 1987) and thus has a unique evolutionary history. It contains a reaction center similar to that of the purple photosynthetic bacteria (Pierson and Castenholtz 1974, Bruce et al. 1982, Pierson and Thornber 1983, Blankenship 1985, Amesz 1987, Kirmaier and Holten 1987, Blankenship et al. 1988b, Ovchinnikov et al. 1988a, b), but an antenna system similar to the green sulfur bacteria (Olson 1980, Fieck and Fuller 1984, Blankenship et al. 1988a). C. aurantiacus also appears to lack detectable water-soluble cytochromes, which are normally involved in bacterial photosynthesis (Bartsch 1978, Wynn et al. 1987). Instead, a blue copper protein may serve as a mobile electron carrier (Trost et al. 1988, McManus et al. 1988). The organism contains a very high heme c to protoheme ratio (Wynn et al. 1987). It has also been reported to have antimycin A and myxothiazol cytochrome b/cj inhibition patterns similar to the green plant cytochrome b6/fcomplex (Zannoni 1987). This latter finding suggests that this complex is significantly different from the bacterial cytochrome b/c~ complexes studied so far. This places C. aurantiacus in a position that could contribute greatly to our understanding of the evolution of cytochromes in photosynthesis. Cytochrome c-554 has been purified and partially characterized in an effort to better understand photosynthetic electron transport in this unusual organism.

Methods

Sources All chemicals used were reagent grade unless specified otherwise. Ultrapure grade SDS was obtained from Boehringer Mannheim, Indianapolis, Indiana. Acrylamide and bis acrylamide were from Bio-Rad, Richmond, California. Pyocyanin was prepared as described by Prince et al. (1981).

Cell growth C. aurantiacus strain J10-fl was grown under high light conditions in the modified medium D of Pierson and Castenholtz (1974) at 55°C in a 16 liter fermenter. Cultures were routinely checked for purity by light microscopy. Cells were harvested by centrifugation at 12,000 x g for 10 rain.

Purification All steps were carried out at 4°C unless stated otherwise. Wet-packed cell paste (150-200 g), was suspended in a total volume of 300 ml of 50mM Tris pH 8.0 buffer containing 1 mM phenylmethylsufonyl fluoride as a protease inhibitor, cooled to 0°C, then disrupted with a Branson model 350 sonifier at a power setting of 8. Centrifugation at 12,000 × g for 10 min removed unbroken cells and cellular debris. The membrane fragments were pelleted by centrifugation at 200,000 × g for two hours in a Beckman 50.2 Ti rotor. In some instances the membranes were salt-washed to remove the blue copper protein auracyanin (Trost et al. 1988) and then repelleted as above. The pellets were suspended and adjusted to a final optical density of 16 at 865nm in 50mM Tris pH 8.0, 1.5% lauryl dimethyl amine-N-oxide (LDAO), and incubated with gentle shaking at 37 °C for 1 h. After centrifugation at 200,000 × g for two hours, the supernatant liquid, containing solubilized cytochrome, reaction centers, and free pigments, was dialyzed against 10 volumes of 50mM Tris pH 9.0 for two days with a buffer change at 24 h. The dialyzed material was loaded on D E A E Sephacel anion exchange media (Phamacia) in a radial flow column

31 (Sepragen 500) equilibrated with 50 m M Tris pH 9.0 containing 0.2% L D A O (buffer 1). The column was eluted with 40 liters of buffer 1 to remove free chlorophylls and carotenoids. Four liters each of 25 mM, 50 mM, and 200 m M NaC1 in buffer 1 were used to elute remaining chlorophylls, reaction centers, and crude cytochrome respectively. After overnight dialysis against 10 volumes of 50mM Tris pH 8 (buffer 2), the cytochrome fraction was placed on a 2.5cm x 40cm D E A E Sephacel anion exchange column equilibrated with buffer 2 containing 0.1% LADO, and eluted with 4 liters each of 0 m M , 25mM, 50mM, and 100mM NaC1, 0.1% L D A O in buffer 2. Cytochrome fractions eluting with A280/A420ratios of 0.7 or less were concentrated using an Amicon PM-30 membrane to an optical density of 2 at 554 nm, loaded in 5 ml fractions on a 2.5cm x 40cm column packed with Sephacryl S-300 gel permeation media (Pharmacia), and eluted with buffer 2 containing 0.05% LDAO. Fractions with A280/m420 ratios less than 0.5 were further chromatographed at a on a 9.2mm x 250ram SOTA Chromatography D E A E analog anion exchange H P L C column. A 45 min gradient from 100-700mM NaC1 in buffer 2 containing 0.05% L D A O was used to elute the cytochrome at a flow rate of 6 ml per minute. Spectral measurements were performed on either a Varian Instruments Cary 219 spectrophotometer interfaced with an Apple II + computer or a Shimadzu UV 160 UV/VIS spectrophotometer. Protein concentrations were determined in triplicate on parallel samples using a modified Lowry assay (Peterson 1977) and the SDS-PAGE molecular weight of 43,000 D. Pyridine hemochrome analysis was performed in triplicate on several samples of different concentrations and purity using the procedure described by Berry and Trumpower (1987). The heme concentration was determined by the dithionite-reduced absorption at 550 nm using an extinction coefficient of 30.27mM-~cm -~ (Berry and Trumpower 1987). "Hemes per peptide" in Table 1 refers to the heme concentration divided by the protein concentration. Values represent the mean of each of the samples. Horse heart cytochrome c (Sigma) was used as the standard. Potentiometric titrations were carried out in a homemade, three-neck cuvette with stirring on the side and bottom. The potential was followed with

a Radiometer P101 platinum working electrode using a Radiometer K401 calomel electrode as the reference. All potentials are stated relative to the standard hydrogen electrode. The titrants used were 50 m M potassium ferricyanide and 50 mM sodium dithionite in buffer 2 added through a rubber septum using a microliter syringe. Spectral measurements were taken when the drift was less than one millivolt per minute. This typically took from 3 to 5 min after the addition of titrant. The total titration time was 5-7h. Mediators were: 10#M, N,N,N,N-tetramethyl- 1,4-phenylenediamine, Em8 --- d- 260 mV; 10/~M, 2,3,5,6-tetramethylp-phenylenediamine, Em8 + 200 mV; 10/tM, 2-hydroxy-l,4-napthoquinone, Era8 = - 200 mV; 50 #M, phenazinemethosulfate, Em8 = q- 70 mV; 50 #M, phenazineethosulfate, Em8 --- d- 30 mV; 50 ~tM, pyocyanin, Era8 = - 9 0 m V (Prince et al. 1981); 50#M, duroquinone Era8 = - 5 5 m V ; 50/~M, 1,4-napthoquinone, Em8 = - 80mV; 50 #M, 1,2-napthoquinone, Era8 -- 24 mV. Midpoint potentials for the last three quinones at pH 7.0 were obtained from Loach (1970) and the midpoint potentials at pH 8 were calculated by Em8 --- Era7 - 60mV (Dutton 1978). Samples for titrations were prepared by exhaustive dialysis against detergent free buffer 2, followed by the addition of sodium dithionite to destroy the remaining LDAO. The cytochrome was filtered and then concentrated by centrifugation in a PM 30 centricon (Amicon) to 40 pl. Buffer 2, containing 0.02% Triton-X-100, was then used to dilute the reduced cytochrome to an optical density of approximately 0.5 at 554 nm. The difference in absorbance between the alpha peak at 554 nm and the isosbestic point at 560 nm was used to follow the titration and account for any shift in the baseline. The titrations were carried out until no change was observed with the addition of more titrant. The difference A554 - - A560 was normalized by division by the greatest difference in absorbance to allow several data sets to be graphed together. Discontinuous SDS-PAGE was carried out using gels with 12.5% total acrylamide and 2.7% crosslinker using the buffer system of Laemmli (1970). The sample buffer contained 4% SDS and 6 M urea in pH 7.0, 250 mM Tris. Samples were diluted with an equal volume of sample buffer and then heated at 85 °C for I min before being placed

32 Table 1. Determination of number of hemes per peptide. A,80/A420 ratio I

Heme concentration 2 /~mol

Protein concentration 3 /t mol

Hemes per peptide 4

0.28 0.31 0.43 0.50

ll.l 7.0 71.7 5.9

2.71 1.80 25.9 3.04

4.1 3.9 2.8 1.9

This value is an indication of the purity of the sample. A lower ratio indicates lower levels of contaminating protein. 2 Heme concentrations have a standard error of ___0.1/~mol. 3 Protein concentrations have a standard error of + 0.05 gtmol. 4 This is an apparent number ofhemes, and is influenced by the amount of protein impurity in the sample. The presence of contaminating proteins will lower the apparent number of hemes per peptide.

on gels. Gels were stained with Coomassie blue to visualize proteins. Heme staining followed the procedure of Thomas et al. (1976), and quantitative heine staining used the technique described by Goodhew et al. (1985). Horse heart cytochrome c (Sigma) was used as a standard. Wide range isoelectric focusing was performed in a Bio-Rad Rotofor preparative isoelectric focusing unit with Bio-Rad 3-10 ampholytes. The isoelectric point was determined by direct measurement using a Ross combination pH electrode (Orion) on recovered samples. Narrow range isoelectric focusing used a 4% polyacrylamide slab gel with 5% crosslinker containing 2% Nonidet P40, 9 M urea and 4% 4-6 ampholytes from Serva. Gels were run at a constant 200 V for four to six hours and maintained at 15 °C. Standards used for calibration were rabbit erythrocyte carbonic anhydrase, pI 6.1; fl lactoglobin, pI 5.31 and 5.14; bovine serum albumin, pI 4.9; and chicken egg albumin, pI 4.2. Amino acid analysis was carried out by the Biotechnology Instrumentation Facility at the University of California-Riverside on both native and performic acid-oxidized samples. Twenty-four hour hydrolysis products were analyzed as their phenylthiohydantoin derivatives. Hydrophobic amino acids are assumed to be those that have a relative hydrophobicity value greater than 1.0 kcal/ mole when compared to glycine. These include Met, Leu, Ile, Val, Phe, Tyr, and Trp (Creighton 1984). Results

Yields of crude cytochrome from 150-180 g of cells ranged from 130 to 180 optical density units at

554 nm times volume in ml (ODV) in the ascorbatereduced form. This corresponds to 70 to 110 mg of protein. Final yields of pure cytochrome ranged from 20 to 30 ODV. Use of the radial flow column allows flow rates of 100ml/min and shortens the time needed for the first separation to a single day. During DEAE anion exchange chromatography at pH 8.0, the cytochrome elutes as a very broad band with 100mM NaC1. The final step of purification yields cytochrome with an A28o/A42o (ascorbate-reduced) ratio of 0.28, which runs as a single heme-staining band at 4 3 k D a on 12.5% SDS-PAGE (Fig. 1). A faint Coomassie-staining band is present at 65kDa, which amounts to a 2% impurity as determined by integrated peak areas on densitometer scans. A high molecular weight heme-staining band was present on SDS-PAGE at approximately 9 0 k D a when a glycerol/Tris sample buffer was used. The band was present in both membranes and purified cytochrome samples, and its intensity was significantly reduced with the 6 M urea sample buffer. Foster et al. (1986) have also observed this band and have described cross reactivity with an affinitypurified antibody to cytochrome c-554. This band is presumably a dimeric aggregate of cytochrome c-554. Optical spectra of the purified protein (Fig. 2a) show an alpha band at 554 nm in the reduced form. The Soret band appears at 420 nm in the reduced form and shifts to 415 nm in the oxidized form. The beta band is present in both forms at 525 nm. There are isosbestic points at 339, 512 and 560nm. The

Fig. 1. SDS-PAGE of purified cytochrome c-554. Lanes 1, 2 and 3 were stained for protein, and lanes 4 and 5 are stained for heine. Lane 1, whole membranes; Lane 2, purified c-554; Lane 3, molecular weight standards; Lane 4, purified c-554; Lane 5, whole membranes.

33

0.50

Table 2. Amino acid composition of cytochrome c-554 from Chloroflexus aurantiacus.

~1.

Amino acid

0.37 o• o~

~ 0.25 O.12 ~ r

:~1/

p~ O. O0 250

'x/i,!

.'1

350

450 550 WavQlen9th (nm)

650

750

b.

1.20

25O

Asx Thr Ser Glx Pro Gly Ala Cys2 Val Met Ile Leu Tyr Phe Trp Lys His Arg

47.78 26.81 25.48 45.54 33.96 38.82 41.91 6.37 28.69 6.40 26.14 35.51 19.08 13.27 NA3 15.40 5.07 25.60

Total

443.61

mol% 10.77 6.04 5.74 10.27 7.65 8.75 9.45 1.56 6.47 1.71 5.89 8.00 4.30 2.99 NA 3.47 1.14 5.77 100

Integer residues per 43,000MW ~ 43 24 23 41 30 35 37 6 26 7 23 32 17 12 NA 14 5 23 398

Integer values were calculated using the gel molecular weight of 43,000 Da. 2 The amount of cysteine in the protein is likely to be underestimated due to the thioether linkage to the heine moiety. 3 NA indicates not analysed.

2 o.eo

i"

nmol in sample

350

450 550 Wavelength (nm)

650

750

FLg. 2. Panel a: UV/Vis Spectra of purified cytochrome c-554. Reductions were accomplished by sequentially adding sodium borohydride to a fully oxidized sample. Absorbance maxima are at 420 nm and 554 nm and are 1.6052 and 0.2582 respectively. isosbestic points are at 560nm, 512nm, and 339nm. Panel b: Difference spectrum of fully-reduced minus fully-oxidizedprotein. Minima are at 292, 369, 41 I, 457,543 and 571 nm. Maxima are at 275, 317, 376, 423, 525 and 554nm. r e d u c e d - m i n u s - o x i d i z e d difference spectrum is shown in Fig. 2b. P o t e n t i o m e t r i c titrations o f purified c y t o c h r o m e c - 5 5 4 (Fig. 3) can be fit to a N e r n s t curve with four one-electron m i d p o i n t potentials at 0mV, + 120 mV, + 220 mV, a n d + 300 mV (solid curve), all values are ___10 mV. T h e relative a m p l i t u d e s o f the curves are 27%, 27%, 23% a n d 23% respectively. The alpha b a n d of the purified c y t o c h r o m e bleaches completely at a p p r o x i m a t e l y + 360 mV. A

N e r n s t curve with three potentials at + 10mV, + 130 mV, a n d + 280 m V a n d relative amplitudes of 29%, 31%, a n d 4 0 % (dashed curve) fits the d a t a of Fig. 3 m o d e r a t e l y well, b u t exhibits deviations from the data in the high potential region. In addition, a three-potential fit requires that the high potential heme be one third m o r e a b s o r b a n t than either of the r e m a i n i n g two. D u r i n g the p o t e n t i o m e t r i c titrations, it was noticed that the alpha b a n d position of the c y t o c h r o m e exhibits a slight red shift o f a p p r o x i m a t e l y 1 n m as it is reduced. The peak starts to the blue of 554 n m a n d moves progressively towards the red as the reduction takes place. T h e pyridine h e m o c h r o m e analysis a n d the dithionite-reduced spectra o f the purified p r o t e i n gave an extinction coefficient of 28 + 1 m M - ~c m - ' per heme. By c o m b i n i n g this i n f o r m a t i o n with a Lowry p r o t e i n assay a n d the gel m o l e c u l a r mass o f 43 kD, an extinction coefficient at 554 n m of 112 + 5 m M - 1cm ~ was o b t a i n e d for the dithionite-reduced protein. This indicates there are p r o b ably four hemes per peptide, as indicated in T a b l e 1 (see Discussion).

34 1 . 0 0 ' iI I I

•

•

iinii~i

0.90 '

0.80 •

O (D

0.70"

t~

0,60 " t~

050

"

¢n O

0.40 "

"0

0.30 -

120

0= O

0.20

*

Z

220 0.10

300 0.00 -200

-I00

0

100

200

300

400

P o t e n t i a l vs H (mV)

Fig. 3. Potentiometric titration of cytochrome c-554 as described in text. The titration can be fit with a Nernst curve with midpoint potentials o f 0 m V , 120mV, 220mV, and 300mV, (marked by arrows), which account for 27%, 27%, 23% and 23% of the height of the curve respectively. Potentials are stated relative to the standard hydrogen electrode. The difference in A554-A560 was normalized by division by the greatest difference in absorbance and then plotted against the potential. Open squares are the reductive titration and the filled squares are the oxidative titration. The solid line is the theoretical curve for the four potentials given above. The dashed line represents the best three potential fit with midpoints at 10 mV, 140 mV, and 280 mV.

Preparative isoelectric focusing indicates a pI between 4.9 and 5.3. In narrow range (pH 4-6) polyacrylamide isoelectric focussing gels pH, cytochrome c-554 focussed between the two bands of fl lactoglobin, indicating a pI of 5.2. Amino acid analysis, shown in Table 2, indicates that cytochrome c-554 contains 29% hydrophobic amino acids and most likely has 5 histidine residues. Assuming the Cys-x-x-Cys-His consensus sequence for c-type heme-binding sites, four of the histidines must be involved in heme ligation (see Discussion). The number of cysteines is probably underestimated in this analysis because of their involvement in heme ligation. Discussion

There is some disagreement in the literature concerning the amounts and types of cytochromes

found in Chloroflexus aurantiacus. Membranes of photosynthetically-grown C. aurantiacus isolated for our experiments and for those of Bruce et al. (1982) showed a single symmetric alpha band at 554nm in reduced-minus-oxidized spectra (data not shown). This has been attributed to membranebound cytochrome c-554 (Bruce et al. 1982, Foster et ai. 1986). From both the spectroscopic data and heme-stained membranes on SDS-PAGE (Fig. 1), it appears that cytochrome c-554 is the major membrane-bound cytochrome in cells grown photosynthetically under high-light conditions. We have partially purified a second membrane-bound c-type cytochrome (data not shown) that corresponds to a 23.5 kilodalton band seen in membranes from both dark and light grown cells (Foster et al. 1986). This cytochrome also has an alpha band near 554nm and would contribute to the membrane absorption spectrum.

35 Zannoni and Ingledew (1985) and Pierson (1985) both describe b-type cytochromes not seen in our preparations. Wynn et al. (1987) found that the heme c to protoheme ratio was 28 to 1 in cells grown under conditions similar to ours. The b-type cytochromes are thus a minor component and would be very difficult to detect in the spectrum of whole membranes. Pierson's (1985) preparations had the carotenoids and chlorophylls removed, thus reducing the level of background absorbance. This procedure would allow for a better characterization of minor heme components in the membranes. Zannoni and Ingledew's (1985) ability to see the b-type cytochromes in the membranes may be due to the extremely high light used during growth, which appears to increase the amount of b-type cytochrome relative to the c-type. The potentiometric titration of the purified cytochrome can be fit with a Nernst equation with four waves of similar amplitude (solid line Fig. 3). Zannoni and Ingledew (1985) have assigned three midpoint potentials to titrations of membranes as belonging to c-type cytochromes. It is possible to assign three potentials to our data, but the fit to the Nernst curve is visibly poorer than the four-potential fit (dashed line Fig. 3). In the course of purification it is possible to modify the tertiary structure of the protein in subtle ways, thus affecting the potentials of the heme sites. This increases the difficulty of assigning both potentials and the number of sites from a potentiometric titration. The highest potential heme shows a midpoint potential of + 300 mV at pH 8. This differs from the + 260 mV at pH 8.1 value found by Bruce et al. 0982). In a membrane flash titration a midpoint potential of 4-295 mV was found by Zannoni and Venturoli (1988). This is comparable to the highest potential found in the dark potentiometric titration of purified cytochrome c-554. This seems to lend support for the higher value, although a shift of this magnitude could easily be induced by subtle structural changes taking place in purification. The Bruce et al. (1982) data might be better fit by Nernst curves with a sum of midpoint potentials of 4- 220 and 4- 300mV. Naively one expects to see only the midpoint potential of the highest potential heme in single turnover flash titrations. However Zannoni and Venturoli (1988) show a two wave titration with the second potential at 4- 140 mV. In the same figure midpoint potentials of 4- 220 and

+ 20 are also shown as time delayed components of cytochrome activity. All of these midpoint potentials are close to those found in the purified cytochrome and would appear to be related. It is clear that this will remain a fertile area of study. The pyridine hemochrome data along with the potentiometric titration indicates that there are probably four hemes per peptide. Lowry protein assays and gel molecular weights have inherent inaccuracies, and thus any heme assignment also has some uncertainty. The presence of contaminating protein can cause additional problems in the assignment of hemes per peptide. The present assignment of four hemes was made after an initial assignment of two (Blankenship et al. 1985) and a second assignment of three hemes (Freeman and Blankenship 1988). The increase in the number of hemes per peptide follows the improvement of the purification (see Table 1). This pattern is reminiscent of the problems that occurred in assigning the number of heme sites to the c3 cytochromes (Postgate 1956, Drucker et al. 1970, Meyer et al. 1971, Meyer and Kamen 1982). Current work on sequencing should confirm the number of heme sites. The amino acid composition of the protein indicates the presence of 5 histidines. Thus the consensus sequence for the heme binding region of c-type cytochromes would allow a maximum of five heme groups. It is likely that two of the histidines are involved in the ligation of the lowest-potential heme, although they are not required (Dracheva et al. 1986, Dracheva et al. 1988). This and the presence of four hemes in the analogous cytochrome c-558-553 from Rhodopseudomonas viridis (Weyer et al. 1987b) lend support to the assignment of four hemes. The function of cytochrome c-554 appears to be similar to that of the c-558-553 cytochrome that is bound to the reaction center of Rhodopseudomonas viridis (Prince et al. 1976, Trosper et al. 1977, Holten et al. 1978, Dracheva et al. 1986, Dracheva et al. 1988) and to the reaction center cytochromes of Chromatium vinosum (Case and Parson 1971, Halsey and Byers 1975, Lin and Thornber 1975, Tiede et al. 1976) and Chromatium tepidum (Nozawa et al. 1987). Because cytochrome c-554 requires detergent to release it from the membrane and is not removed by salt washings, it is clearly a membrane-bound protein. The percentage of

36 hydrophobic residues in cytochrome c-554 (29%) is similar to that of the corresponding cytochrome from R. viridis 27% (Weyer et al. 1987b), and similar to an average value of 28% for selected water soluble proteins (Reeck 1970). This suggests a position in a hydrophilic environment rather than primarily in the lipid portion of the membrane. Cytochrome c-554 could possibily have a fatty acid tail anchoring it to the membrane similar to that of cytochrome c-558-553 from R. viridis (Weyer et al. 1987a), or a single transmembrane helix as in cytochrome c~ o r f ( W a k a b a y a s h i et al. 1982, Wiiley et al. 1984, Pettigrew and Moore 1987). In contrast to cytochrome c-558-553 from R. viridis, C. aurantiacus cytochrome c-554 fails to co-purify with the reaction center. This might be caused by the absence of an H subunit in C. aurantiacus (Shiozawa et al. 1987, Blankenship et al. 1988b, Ovchinnikov et al. 1988a, b). In R. viridis the H subunit appears to interact with the cytochrome and may help bind it to the reaction center (Deisenhofer et al. 1985). The spacing of the midpoint potentials in C. aurantiacus is also unique. In R. viridis, Rhodocyclus gelatinosus, and the Chromatiaceae, the reaction center cytochrome midpoint potentials appear as closely spaced p a i r s - o n e high-potential pair and one low-potential pair (Case and Parson 1971, Halsey and Byers 1975, Lin and Thornber 1975, Prince et al. 1976, Tiede et al. 1976, Trosper et al. 1977, Holten et al. 1978, Dracheva et al. 1986, Alegria and Dutton 1987, Nozawa et al. 1987, Dracheva et al. 1988, Fukushima et al. 1988). C. aurantiacus cytochrome c-554 seems to have a pair of closely spaced high potential hemes, but the lower potential hemes are thermodynamically distinct. The possible functional significance of this difference is not yet clear. Cytochrome c-554 also differs from the R. viridis cytochrome in that it does not participate in low temperature electron transfer in isolated membranes (Wynn et al. 1987). This could be explained by the lack of the paired low potential hemes which seem to be involved in that phenomenon (Prince et al. 1976, Tiede et al. 1976, Holten et al. 1978, Dracheva et al. 1986, Alegria and Dutton 1987, Dracheva et al. 1988). Preliminary evidence indicates that the small blue copper protein auracyanin (McManus et al. 1988, Trost et al. 1988) might reduce cytochrome

c-554 in a manner similar to the water-soluble cytochrome c2 in R. viridis (Shill and Wood, 1984). Kinetic experiments to better determine the roles of c-554 and auracyanin in photosynthetic electron transport are currently being undertaken.

Acknowledgements

This work was supported by grant #88-372623480 from the Competitive Research Grants office of the U.S.D.A., and an Arizona State University Graduate College research assistantship. This is publication # 6 from the Arizona State University Center for the Study of Early Events in Photosynthesis. The Center is funded by U.S. Department of Energy grant # D E - F G 0 2 - 8 8 E R I 3 9 6 9 as part of the U S D A / D O E / N S F Plant Science Centers program. We thank Phan Huynh, Lori Mancino and Chris Bender for carrying out some of the early experiments on cytochrome c-554.

References Alegria G and Dutton PL (1987) Construction and characterization of monolayer films of the reaction center cytochrome-c protein from Rhodopseudomonas viridis. In: Papa S, Chance B, Ernster L (eds) International Symposium on Cytochrome Systems: Molecular Biology and Bioenergetics, pp 601-608. New York: Plenum Press AmeszJ (1987) Primary electron transport and related processes in green photosynthetic bacteria. Photosynthetica 21: 225-235 Bartsch RG (1978) Cytochromes. In: Clayton RK and Sistrom WR (eds) The Photosynthetic Bacteria, pp 249-279. New York: Plenum Press Berry EA and Trumpower BL (1987) Simultaneous determination of hemes a, b and c from pyridine hemochrome spectra. Anal Biochem 161:1-15 Blankenship RE, Fieck R, Bruce BD, Kimaicr C, Holtcn D and Fuller RC (1983) Primary photochemistry in the facultative green photosynthetic bacterium Chloroflexus aurantiacus. J Cellular Biochem 22:251-261 Blankenship RE (1985) Electron transport in green photosynthetic bacteria. Photosynth Res 6:317-333 Blankenship RE, Huynh P, Gabrielson H and Mancino LJ 0985) Purification, physical properties and kinetic behavior of cytochrome c-554 from Chloroflexus aurantiacus. Biophys J 47: 2a Blankenship RE, Brune DC, Freeman JM, King GH, McManus JD, Nozawa T, Trost J T and Wittmershaus BP (1988a) Energy trapping and electrotransfer in Chloroflexus auran-

37 tiacus. In: Olson JM, Ormerod JG, Amesz J, Stackebrandt E and Tr/iper HG (eds) Green Photosynthetic Bacteria, pp. 57-68. New York: Plenum Press Blankenship RE, Trost JT and Mancino LJ (1988b) Properties of reaction centers from the green photosynthetic bacterium Chloroflexus aurantiacus. In: Breton J and Vermeglio A (eds) Structure of Bacterial Reaction Centers, pp 119-127. New York: Plenum Press Bruce BD, Fuller RC and Blankenship RE (1982) Primary photochemistry in the facultatively aerobic green photosynthetic bacterium Chloroflexus aurantiacus. Proc Natl Acad Sci USA 79:6532-6536 Case GD and Parson WW (1971) Thermodynamics of the primary and secondary photochemical reactions in Chromatium. Biochim Biophys Acta 253:187-202 Creighton TE (1984) Proteins: Structure and Molecular Properties. p. 142. New York: W.H. Freeman and Co. Deisenhofer J, Epp O, Miki K, Huber R and Michel H (1985) Structure of the protein subunits in the photosynthetic reaction center of Rhodopseudomonas viridis at 3 A resolution. Nature 318:618-624 Dracheva SM, Drachev LA, Zaberezhnaya SM, Konstantinov AA, Semenov AY and Skulachev VP (1986) Spectral, redox and kinetic characteristics of high-potential cytochrome c hemes in Rhodopseudomonas viridis Reaction Center. FEBS Lett 205:41-46 Dracheva SM, Drachev LA, Konstantinov AA, Semenov AY, Skulachev VP, Arutjunjan AM, Shuvalov VA and Zaberezhnaya SM (1988) Electrogenic steps in the redox reactions catalyzed by photosynthetic reaction-centre complex from Rhodopseudomonas viridis. Eur J Biochem 171:253-264 Drucker H, Trousil EB, Campbell LL, Barlow GH and Margoliash E (1970) Amino acid composition heme content and molecular weight of cytochrome c3 of Desulfovibrio desulficans and Desulfovibrio vulgaris. Biochem 9:1515-1518 Dutton PL (1978) Redox potentiometry: determination of midpoint potentials of oxidation-reduction components of biological electron-transfer systems. Meth Enzymol 54:411-434 Feick RG and Fuller RC (1984) Topography of the photosynthetic apparatus of Chloroflexus aurantiacus. Biochemistry 23: 3693-3700 Foster JM, Redlinger TE, Blankenship RE and Fuller RC (1986) Oxygen regulation of development of the photosynthetic membrane system in Chloroflexus aurantiacus. J Bacteriol 167:655-677 Freeman JC and Blankenship RE (1988) Purification and physical properties of membrane bound cytochrome c-554 from Chloroflexus aurantiacus. Photochem Photobiol 47: 41S Fukushima A, Matsuura K, Shimada K and Satoh T (1988) Reaction center-B870 pigment protein complexes with bound cytochrome c-555 and c-551 from Rhodocyclus gelatinosus. Biochim Biophys Acta 933:399-405 Goodhew CF, Brown KR and Pettigrew GW (1986) Haem staining in gels, a useful tool in the study of bacterial c-type cytochromes. Biochim Biophys Acta 852:288-294 Halsey YG and Byers B (1975) A large photoreactive particle from Chromatium vinosum chromatophores. Biochim Biophys Acta 387:349-367 Holten D, Windsor MW, Parson WW and Thornber JP (1978) Primary photochemical processes in isolated reaction centers

of Rhodopseudomonas viridis. Biochim Biophys Acta 501: 112-126 Kirmaier C and Holten D (1987) Primary photochemistry of reaction centers from the photosynthetic purple bacteria. Photosynth Res 13:225-260 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685 Lin L and Thornber JP (1975) Isolation and partial characterization of the photochemical reaction center of Chromatium vinosum (strain D). Photochem Photobiol 22:37-40 Loach PA (1970) Oxidation-reduction potentials, absorbance bands and molar absorbance of compounds used in biochemical studies. In: Sober HA (ed) Handbook of Biochemistry: Selected Data for Molecular Biology, pp C-281-C-287. Cleveland: The Chemical Rubber Co. McManus JD, Trost JT and Blankenship RE (1988) Kinetic behavior and N-terminal amino acid sequence of Auracyanin. Biophys J 53: 268a Meyer TE, Bartsch RG and Kamen MD (1971) Cytochrome c3: A class of electron transfer heme proteins found in both photosynthetic and sulfate-reducing bacteria. Biochim Biophys Acta 245:453-464 Meyer TE and Kamen MD (1982) New perspectives on c-type cytochromes. Adv Protein Chem 35:105-212 Nozawa T, Trost JT, Fukada T, Hatano M, McManus JD and Blankenship RE (1987) Properties of the reaction center of the thermophilic purple photosynthetic bacterium Chromatium tepidum. Biochim Biophys Acta 894:468-476 Olson JM (1980) Chlorophyll organization in green photosynthetic bacteria. Biochim Biophys Acta 594:33-51 Ovchinnikov YA, Abdulaev NG, Zolotarev AS, Shmukler BE, Zargarov AA, Kutuzov MA, Telezhinskaya IN and Levina NB (1988a) Photosynthetic reaction centre of Chloroflexus aurantiacus. I. Primary structure of the L-subunit. FEBS Lett 231:237-242 Ovchinnikov YA, Abdulaev NG, Shmukler BE, Zargarov AA, Kutuzov MA, Telezhinskaya IN, Levina NB and Zolotarev AS (1988b) Photosynthetic reaction centre of Chloroflexus aurantiacus. Primary structure of the M-subunit. FEBS Lett 232:364-368 Peterson GL (1977) A simplification of the protein assay method of Lowry et al. which is more generally applicable. Anal Biochem 83:346-356 Pettigrew GW and Moore GR (1987) Cytochromes c Biological Aspects. Berlin: Springer-Verlag Pierson BK (1985) Cytochromes in Chloroflexus aurantiacus grown with and without oxygen. Arch Microbiol 143: 260-265 Pierson BK and Thornber JP (1983) Isolation and spectral characterization of photochemical reaction centers from the thermophilic green bacterium Chloroflexus aurantiacus strain J-10-fl. Proc Natl Acad Sci USA 80:80-94 Pierson BK and Castenholz RW (1974) A phototrophic gliding filamentous bacterium of hot springs, Chloroflexus aurantiacus, gen. and sp. nov. Arch Microbiol 100:5-24 Postgate JR (1956) Cytochrome c3 and Desulphoviridin: pigments of the anaerobe Desulphovibrio desulphuricans. J Gen Microbiol 14:545-572 Prince RC, Leigh JS and Dutton PL (1976) Thermodynamic

38 properties of the reaction center of Rhodopseudomonas viridis, in vivo measurements of the reaction center bacteriochlorophyll primary acceptor intermediary electron carrier. Biochim Biophys Acta 440:622-636 Prince RC, Linkletter SJG and Dutton PL (1981) The thermodynamic properties of some commonly used oxidationreduction mediators, inhibitors and dyes as determined by polarography. Biochim Biophys Acta 635:132-148 Reeck G (1970) Amino acid compositions of selected proteins. In: Sober HA (ed) Handbook of Biochemistry Selected Data for Molecular Biology, pp C281-C-287. Cleveland: The Chemical Rubber Co. Shill DA and Wood PM (1984) A role for cytochrome c~ in Rhodopseudomonas viridis. Biochim Biophys Acta 764:1-7 Shiozawa JA, Lottspeich F and Feick R (1987) The photochemical reaction center of Chloroflexus aurantiacus is composed of two structurally similar polypeptides. Eur J Biochem 167: 595-600 Thomas PE, Ryan D and Levin W (1976) An improved staining procedure for the detection of the peroxidase activity of cytochrome P-450 on sodium dodecyl sulfate polyacrylamide gels. Anal Biochem 75:168-176 Tiede DM, Prince RC and Dutton PL (1976) EPR and optical spectroscopic properties of the electron carrier intermediate between the reaction center bacteriochlorophylls and the primary acceptor in Chromatium vinosum. Biochim Biophys Acta 449:447-467 Trosper TL, Benson DL and Thornber JP (1977) Isolation and spectral characteristics of the photochemical reaction center of Rhodopseudomonas viridis. Biochim Biophys Acta 460: 318-330 Trost JT, McManus JD, Freeman JC, Ramakrishna BL and Blankenship RE (1988) Auracyanin: A blue copper protein from the green photosynthetic bacterium Chloroflexus aurantiacus. Biochemistry 27:7858-7863 Wakabayashi S, Matsubara H, Kim CH and King TE (1982) Structural studies of bovine heart cytochrome c~. J Biol Chem 257:9335-9344 Weyer KA, Sch~ifer W. Lottspeich F and Michel H (1987a) The

cytochrome subunit of the photosynthetic reaction center from Rhodopseudomonas viridis is a lipoprotein. Biochemistry 26:2909-2914 Weyer KA, Lottspeich F, Gruenberg H, Land F, Osterhelt D and Michel H (1987b) Amino acid sequence of the cytochrome subunit of the photosynthetic reaction centre from the purple bacterium Rhodopseudomonas viridis. EMBO J 6: 2197-2202 Widger WR, Cramer WA, Herrmann RA and Trebst A (1984) Sequence homology and structural similarity between cytochrome b of mitochondrial complex III and the chloroplast boc complex: position of the cytochrome b hemes in the membrane. Proc Natl Acad Sci USA 81:674-678 Willey DL, Auffret AD and Gray JC (1984) Structure and topology of cytochrome f i n pea chloroplast membranes. Cell 36:555-562 Woese CR (1987) Bacterial evolution. Microbiol Rev 51: 221-271 Wynn RM, Redlinger TE, Foster JM, Blankenship RE, Fuller RC, Shaw RW and Knaff DB (1987) Electron transport chains of phototrophically and chemotrophically grown Chloroflexus aurantiacus. Biochim Biophys Acta 891:216-226 Zannoni D (1987) The interplay between photosynthesis and respiration in facultative anoxygenic phototrophic bacteria. In: Papa S, Chance B and Ernster L (eds) International Symposium on Cytochrome Systems: Molecular Biology and Bioenergetics, pp 575-583. New York: Plenum Press Zannoni D and Ingledew WJ (1985) A thermodynamic analysis of the plasma membrane electron transport components in photoheterotrophically grown cells of Chloroflexus aurantiacus: an optical and electron paramagnetic resonance study. FEBS Lett 193:93-98 Zannoni D and Venturoli G (1988) The mechanism of photosynthetic electron transport and energy transduction by membrane fragments from Chloroflexus aurantiacus. In: Olson JM, Ormerod JG, Amesz J, Stackebrandt E and Triiper HG (eds) Green Photosynthetic Bacteria, pp 135-143. New York: Plenum Press