The function of methyl-menaquinone-6 and polysul®de reductase membrane anchor (PsrC) in polysul®de respiration of Wolinella succinogenes Wiebke Dietrich and Oliver Klimmek Institut fu È r Mikrobiologie, Johann Wolfgang Goethe-Universita È t, Frankfurt am Main, Germany Wolinella suc cinogenes grows by oxidative phosphorylation with polysul®de as terminal electron a cceptor and either H 2 or formate a s electron donor (polysul®de respiration). The function of the respiratory chains catalyzing these reactions was i nvestigated. Proteoliposomes containing polysul®de reductase (Psr) and either hydrogenase or formate dehy- drogenase i solated f rom the membrane fraction of Wolinella succinogenes catalyzed polysul®de respiration, provided that methyl-menaquinone-6 isolated from W. succinogenes was also present. The speci®c activities of electron transport were commensurate with those of the bacterial membrane frac- tion. Using site-directed mutage nesis, certain residues were substituted in P srC, t he membrane a nchor of polysul®de reductase. Replacement of Y23, D76, Y159, D218, E225 or R305 caused nearly full inhibition of polysul®de respiration without aecting the activity of Psr, which was still bound to the membrane. These residues are predicted to be located in hydrophobic helices o f PsrC, or n ext to t hem. Substitution o f 13 other r esidues of PsrC e ither c aused p artial inh ibition of polysul®de respiration or had no eect. The function of methyl-menaquinone-6, which is thought to be bound to PsrC, is discussed. Keywords: methyl-menaquinone; polysul®de respiration; sulfur respiration; hydrogenase; formate dehydrogenase. Wolinella succinogenes grows at the expense of polysul®de ([S]) respiration with H 2 [reaction (a)] or formate [reaction (b)] as electron donor [1±3]. Reactions (a) (DG o ¢  A31 kJámol H À 1 2 )and(b)(DG o ¢A30 kJ mol A1 formate) are coupled to the generation of an electrochemical proton potential (Dp  0.17 V) across the bacterial membrane which drives ATP synthesis [3±6]. H 2 +[S] ® HS ± +H + (a) HCO 2 ± +[S] ® CO 2 +HS ± (b) It was proposed that the electron transport chain catalyz- ing reactions (a) or (b) consisted of the membrane bound components polysul®de reductase (Psr) and either h ydro- genase or formate dehydrogenase [3,6±10]. The catalytic subunits of the three enzymes are orien ted to t he periplasmic side of the m embrane [4,6,10,11]. The three enzymes were isolated from the membrane fraction of W. succinogenes [3,6±8,12,13], and the corresponding genes were sequenced [9,12,14]. Hydrogenase (Hyd) and formate dehydrogenase (Fdh) are identical with the enzymes involved in fumarate respiration with H 2 and f ormate in W. succinogenes [7,8,15]. The cytochrome b subunits of the two enzymes ( HydC and FdhC) which carry the sites of quinone reduction are similar. Their four histidine residues c oordinating the two heme B groups are predicted to be located at similar places on three homologous membrane helices. Proteoliposomes containing Psr and either hydrogenase or formate dehydrogenase were reported to catalyze reaction (a) or (b) [4,6±8]. The electron transport activities amounted to maximally 5 % of those measured in the bacterial membrane fraction. The activities were not higher in proteoliposomes additionally containing vitamin K 1 which is known to substitute for menaquinone in fumarate respiration. The isolated Psr was found to consist of the three subunits predicted by the psrABC operon [3,6,8]. It contained molybdenum (1 mol per mol enzyme), molybdopterine guanine d inucleotide, free i ron and sul®de, and menaqui- none. Heme, ¯avin and other heavy metals were absent. The enzyme catalyzed the reduction of polysul®de by BH 4 ± [reaction (c)] and the oxidation of sul®de by 2,3-dimethyl- 1,4-naphthoquinone (DMN) [reaction (d)]. BH 4 ± +[S] ® BH 3 +HS ± (c) HS ± + DMN + H + ® [S] + DMNH 2 (d) PsrA is similar to the catalytic subunits of several molybdo-oxidoreductases and is probably the catalytic subunit of Psr, which carries molybdenum coordinated by molybdopterin guanine dinucleotide. PsrA and PsrB are predicted to carry one and four iron-sulfur-centers, respec- tively. PsrC is a hydrophobic p rotein which anchors the enzyme in the membrane [10]. Correspondence to O. Klimme k, Institut fu È r Mikrobiologie, Johann Wolfgang Goethe-Universita È t, Marie-Curie-Str. 9, D-60439 Frankfurt am Main, Germany. Fax: + 49 6 9 79829527, Tel. + 49 69 79829509, E-mail: klimmek@em.uni-frankfurt.de Abbreviations: DMN, 2,3-dimethyl-1,4-naphthoquinone; Hyd, hydrogenase; Fdh, formate dehydrogenase; MK, menaquinone; MK 6 , menaquinone with a side chain o f six isoprene units; MM, 5- or 8-methyl-MK 6 ;MM b ,MMboundtoPsrC;MM b H ± , quinol anion of MM bound to PsrC; Ps r, polysul®de reductase; [S], sulfur atom in polysul®de; TTFB, 4,5,6,7-tetrachloro-2-tri¯uoromethyl- benzimidazol; Dp, electrochemical proton potential across a membrane; Dw, e lectric term of Dp; Ttr, tetrathionate reductase. (Received 25 J une 2001, revised 1 November 2001, accepted 7 November 2001) Eur. J. Biochem. 269, 1086±1095 (2002) Ó FEBS 2002 When the membrane f raction o f W. succinogenes was fused to liposomes containin g menaquinone isolated from W. succinogenes, the electron transport a ctivities [reactions (a) and (b)] decreased as a function of the degree of dilution of the m embrane proteins by phospholipid [6,16]. This effect was interpreted to indicate that the electron transfer from the dehydrogenases to Psr requires diffusion and collision of the enzyme molecules within the membrane. Consistent with this interpretation, the electron transport activity with fumarate was not decreased upo n dilution of the membrane p roteins. In this case the elec tron transfer from the dehydrogenases to fumarate reductase is accom- plished by m enaquinone (MK) whose diffusion velocity i s two orders of m agnitude higher than that of the much larger enzymes. When the membrane fraction was fused t o liposomes containing either no quinone or vitamin K 1 instead of menaquinone from W. succinogenes, the activities of poly- sul®de resp iration d ecreased more markedly as t he phos- pholipid content increased [6,16]. This suggested that one of the two menaquinones of W. succinogenes was speci®cally required for polysul®de respiration a nd could not be substituted by vitamin K 1 . Here we report on our attempts to r estore polysul®de respiration in liposomes using one of the menaquinones of W. succinogenes. To elucidate the function of PsrC, w hich is thought to bind quinone, certain amino-acid residues of t his subunit w ere replaced using site-directed mutagenesis. The resulting mutants were characterized by measuring their speci®c a ctivities of polysul®de respiration [reactions (a) and (b)] and of Psr [reactions (c) and (d)]. The mechanism of polysul®de respiration is discussed in the light of the experimental results. MATERIALS AND METHODS Growth of W. succinogenes W. succinogenes was grown with formate as electron donor and either fumarate or nitrate as electron acceptor as described previou sly [17,18]. The medium containing nitrate was supplemented with b rain±heart infusion (1,3% w/v; Gibco BRL). Kanamycin (25 mgáL A1 ) and chloramphen icol (12,5 mgáL A1 ) were added to the medium when indicated. Cell fractionation Cells of W . succinogenes grown with fumarate w ere sus- pended (10 g proteináL A1 )in50m M Tris/HCl, pH 8.0 at 0 °C. The suspension was passed through a French press at 130 Mpa and 10 mLámin A1 ¯ow. The resulting cell homo- genate was centrifuged for 40 min at 100 000 g to yield the soluble cell fraction (supernatant) and the membrane fraction. The membrane fraction was r esuspended in the same buffer (10 g proteináL A1 ). SDS/PAGE, Western blotting and ELISA SDS/PAGE was carried out according t o [19]. Protein was transferred to nitrocellulose sheets b y electro blotting in a discontinuous buffer system [20]. PsrA was visualized by indirect ELISA u sing the corresponding antiserum and goat anti-(rabbit IgG) Ig conju gated to peroxidase [9]. Fusion particles and proteoliposomes Sonic liposomes containing quinone were prepared from a mixture of egg phospholipid (95%, w/w) [21] and phospha- tidylethanolamine (5%, Fluka; cat. no. 60650) as described previously [13,22]. Fusion particles (see below) were pre- pared by freeze±thawing a mixture of sonic liposomes containing 5- or 8-methyl-MK 6 (MM) and bacterial mem- brane f raction [16]. Th e fusion p articles contained e qual amounts of phospholipid from the liposomes and from the membrane fraction. Proteoliposomes were prepared by freeze-thawing sonic liposomes containing the quinone indicated (Table 1; 10 lmolág phospholipid A1 ) with Psr (or fumarate reductase) and either hydrogenase or formate dehydrogenase [8,13,22]. Per g phospholipid, a total of 26 nmol Psr, 31 nmol fumarate reductase, 1 78 nmol hydrogenase and 89 nmol formate dehydrogenase were applied. Quinones MK 6 and MM were extracted from the membrane fraction of W. succinogenes grown with fumarate using a mixture o f petrol ether a nd methanol. The quinones in the e xtract were separated by H PLC according to [23]. The quinones were quanti®ed by HPLC using vitamin K 1 as the s tandard. MK 4 (Sigma; c at. no. V-937 8) and v itamin K 1 (Fluka; cat. no. 95271) are commercially available. Activities of Psr and of polysul®de respiration The activity of Psr was measured at 37 °C by photometric recording of polysul®de reduction with BH 4 ± [24] or by photometric recording of DMN reduction with sul®de [2]. Polysul®de respiration with H 2 or formate (electron trans- port) was recorded photometrically at 37 °C as de scribed previously [2,8,16]. The unit of activity (U) correspo nded to consumption of 1 lmol substrate per min. Protein Protein was determined after precipitation with t richloro- acetic acid using the Biuret method with KCN [25]. Table 1. A ctivities o f p olysul®de r espiration of proteoliposomes con- taining dierent naphthoquinones. Polysul®de respiration with H 2 (H 2 ® [S]) or with formate (HCO 2 ± ® [S]) was m easured in prote- oliposomes containing Psr and either hydrogenase or formate dehydrogenase isolated from W. succinogenes. Fumarate respiration with formate (HCO 2 ® fumarate) was measured in proteoliposomes containing f umarate reductase a nd formate dehydrogenase as described previously [22]. The activities are given as substrate turnovers of Psr or fumarate reductase at 37 °C. All values are in units of s A1 . Quinone H 2 ® [S] HCO 2 ± ® [S] HCO 2 ± ® fumarate ±25517 Methyl-MK 6 (MM) 370 175 35 MK 6 27 5 1490 MK 4 34 7 1455 VitK 1 25 5 1180 Ó FEBS 2002 Polysul®de respiration of W. succinogenes (Eur. J. Biochem. 269) 1087 Genetic techniques Standard genetic procedures were used essentially according to [26]. DNA was isolated from W. succino- genes with the DNeasy Tissue Kit from Qiagen. PCR was carried out using the Expand High Fidelity PCR System (Roche) or the Expand Long Template PCR System (Roche) with standard ampli®cation proto- cols and a Hybaid Omnigene Thermocycler (MWG Biotech). Southern blotting to nylon membranes was performed as described previously [27]. DNA probes were generated with the PCR DIG Probe Synthesis K it (Roche), and hybrids were visualized using the DIG/ Luminescent Detection Kit (Roche). Construction of W. succinogenes KpsrC Mutant KpsrC was constructed by integrating plasmid pKpsrC into the genome of mutant DpsrC [10] (Fig. 1). Plasmid pKpsrC was constructed from pT7-6 [28] by inserting the cat GC gene excised from pDF4a [29] using BamHI restriction. The orientation of cat GC was con®rmed by restriction analysis. Subsequently, a fragment comprising most of psrB (521 bp), psrC, a nd 60 bp of the 3¢ end of rhpR was inserted using HindIII and Xb aI. This fragment was synthesized from genomic DNA by PCR with primers carrying at their 5¢ ends suitable restriction sites for cloning. The identity of the cloned PCR fragment was c on®rmed by sequencing. Cells of W. succinogenes DpsrC were transformed with the plasmid as described previously [30]. Transformants were selected on agar plates containing the nitrate medium, 2.6% (w/v) brain-heart-infusion agar (Gibco BRL) and chloramphenicol (12.5 mgáL A1 ). The genome of several transformants was checked for t he presence of the cat GC and the psrC gene by means of Southern blot analysis using Bgl II restriction (Fig. 1). As expected, only one BglII fragment (9.7 kbp) of mutant KpsrC hybridized to the cat GC and the psrC probe. The in-frame integration of the plasmid was con®rmed by sequencing. Construction of W. succinogenes psrC mutants The psrC mutants of W. succinogenes (see Table 2) were constructed by t ransforming W. succinogenes DpsrC with derivatives of pKpsrC. The derivatives were synthesized using the Quick Change site-directed mutagenesis kit (Stratagene) with pKpsrC as template and speci®cally synthesized oligonucleotides carrying the desired nucleotide mismatches. Modi®ed pKpsrC plasmids were sequenced to con®rm the m utations. Transformation o f W. succinogenes DpsrC with modi®ed plasmids and selection of t ransfor- mants was performed as described above. Computer analysis Database searches made use of the program BLAST [31]. Search for membrane-spanning helices was performed using the TMPRED program [32]. Multiple sequences were aligned using the program CLUSTALW [33]. RESULTS Reconstitution of polysul®de respiration W. succinogenes grown w ith polysul®de or with fumarate catalyzes polysul®de respiratio n [reactions (a) and (b)] as well as fumarate respiration with H 2 or formate [16]. These bacteria contain equal amounts (approximately 3 lmol p er g phospholipid) of MK 6 and o f MM [34]. The methyl group in the a romatic ring o f MM is at position 5 or 8. Like MK, MM appears to be d issolved in the lipid phase of the membrane and is not tightly bound to membrane proteins. Both quinones can be extracted from t he membrane with a mixture of petrol ether and methanol. In the experiment shown in Table 1, Psr and either hydrogenase or formate dehydrogenase were incorporated into liposomes containing one of the quinones indicated (10 lmolág phospholipid A1 ). The activities of polysul®de respiration of p roteoliposomes containing MM were more than an order of magnitude greater than of those p repared with MK 6 ,MK 4 , vitamin K 1 , or w ithout quinone. These Fig. 1. P hysical map of the psr locus of W. succinogenes DpsrC and KpsrC. Mutant KpsrC was obtained by integration of pKpsrC into the genome of the DpsrC mutant. 1088 W. Dietrich and O. Klimmek (Eur. J. Biochem. 269) Ó FEBS 2002 Table 2. P roperties of psrC mutants grown with formate and fumarate. The presence o f PsrA was tested by Western blot and E LISA. The speci®c activities of Psr (BH 4 ± ® [S] a nd HS ± ® DMN) and of polysul®de respiration (H 2 ® [S] and HCO 2 ± ® [S]) refer to total cellular protein (cells) or to the protein of the membrane fraction ( MF). Strain Preparation PsrA present Uámg protein A1 BH 4 ± ® [S] HS ± ® DMN H 2 ® [S] HCO 2 ± ® [S] Wild-type Cells 13 6.7 2.1 1.8 MF + 19 14 1.5 1.1 DpsrC Cells + a 12 7.0 £ 0.01 £ 0.01 MF ± £ 0.1 £ 0.1 £ 0.01 £ 0.01 KpsrC Cells 8.1 5.5 2.2 1.2 MF + 16 9.8 0.9 1.0 W16F Cells + 1.7 1.6 MF 17 9.7 0.6 0.5 Y23F Cells 0.04 0.03 MF + 13 11 0.02 0.03 D76N Cells £ 0.01 £ 0.01 MF + 20 7.2 0.02 0.02 D76H Cells 0.21 0.11 MF + 19 8.3 0.20 0.09 D76L Cells 0.02 £ 0.01 MF + 19 7.9 0.03 0.02 H82A Cells 2.3 1.1 MF + 17 10 1.4 0.6 S94A Cells 1.6 1.1 MF + 13 7.0 0.8 0.8 Y106F Cells 1.8 1.4 MF + 19 9.3 0.8 0.7 E146Q Cells 1.8 1.0 MF + 14 4.6 0.6 0.4 Y159F Cells £ 0.01 £ 0.01 MF + 12 7.0 0.02 0.02 T160V Cells 0.2 0.2 MF + 17 9.3 0.1 0.1 N174D Cells 1.4 1.2 MF + 17 11 0.9 0.8 S185A Cells 0.9 0.8 MF + 12 6.0 0.4 0.4 S188A Cells 0.5 0.5 MF + 11 5.1 0.2 0.2 S192A Cells 2.1 1.7 MF + 19 11 0.8 0.7 E209Q Cells 2.3 2.1 MF + 19 13 1.2 1.1 D218N Cells £ 0.01 £ 0.01 MF + 15 7.9 0.02 0.02 E225Q Cells 0.03 0.03 MF + 11 4.9 0.03 0.03 E225D Cells 1.1 1.2 MF + 14 7.8 0.4 0.7 W261F Cells 2.0 1.3 MF + 21 10 0.7 0.6 R305F Cells 0.04 0.07 MF + 13 7.2 0.02 0.02 R305K Cells £ 0.01 0.02 MF + 19 7.9 0.03 0.04 Y310F Cells 1.6 1.3 MF + 17 14 1.0 1.2 a The cell homogenate was used. Ó FEBS 2002 Polysul®de respiration of W. succinogenes (Eur. J. Biochem. 269) 1089 low a ctivities were probably due to the low level of MM present in the enzyme preparations used for preparing the proteoliposomes. Proteoliposomes prepared with fumarate reductase and formate dehydrogenase c atalyzed fumarate respiration with formate at high activities, i f MK 6 ,MK 4 or vitamin K 1 were also incorporated. However, if MM was incorporated instead, the activity was as low as that o f proteoliposomes p repared w ithout added quinone. I n liposomes containing fumarate reductase and hydrogenase, fumarate respiration with H 2 was restored by MK 6 ,MK 4 or vitamin K 1 but not by MM (data not shown). The turnover number of Psr in polysul®de respiration with formate in the proteoliposomes is close to that measured in the membrane fraction o f w ild-type W. suc- cinogenes (see Table 2) w hich contains approximately 0.1 lmol Psr per g membrane protein [3,8]. The turnover number with H 2 is 50% higher i n the proteoliposomes than in the membrane fraction. This higher a ctivity is p robably due to the higher a mount of hydro genase relative to Psr in the proteoliposomes. In summary, functional electron transport chains catalyzing reactions (a) or (b) at the expected activities can be restored from the isolated enzymes and MM. Hence no further components appear to be required for the electron transport. MM is speci®cally involved in polysul®de respiration and cannot be substituted by MK 6 although this is also present in the membrane of W. succinogenes. In the experiment shown in Fig. 2, t he membrane fraction of W. succinogenes was fused with sonic liposomes contain- ing i ncreasing amounts of MM. The six d ifferent prepara- tions so obtained contained equal amounts of phospholipids from the membrane f raction and from the liposomes. The activity of polysul®de respiration with H 2 increased hyper- bolically with the amount of MM. The activity was 50 a nd 80% of the maximum activity with 2.5 and 10 lmol MM per g phospholipid, respectively. Thus the activity is considerably enhanced by the incorporation of additional MM into the membrane f raction which normally contains approximately 3 lmol MM p er g phospholipid. As noted in the Discussion section, the titration curve (Fig. 2) may re¯ect binding of reduced MM to PsrC. If this assumption is valid, the dissociation constant would be 2.5 lmol MMág phospholipid A1 . The effect of D p on the electron transport activity The speci®c activities of Psr in the membrane fraction are about twice those found in cells of W. succinogenes (see Table 2). This is consistent with the location of Psr in the membrane which contains half of the total cellular protein. In contrast, the speci®c activities of polysul®de respiration are higher in cells than in the membrane fraction (see Table 2). This is probably due to the Dp generated across the membrane of cells by p olysul®de respiration [5]. The Dp is lost upon cell disruption. The addition of a protonophore to cells also caused inhibition of polysul®de respiration (Fig. 3 ). The protono phore 4 ,5,6,7-tetrachloro-2-tri¯uoro- methylbenzimidazol (TTFB) which i s known t o dissipate the Dp [5,35] was found to cause up to 70% inhibition of polysul®de respiration in c ells. Polysul®de respiration in the membrane fraction or fumarate respiration in cells or in the membrane fraction were not inhibited by TTFB (not shown). Thus, the activity of polysul®de respiration appears to be stimulated by Dp. Characterization of psrC mutants PsrC is predicted to form eight membrane-spanning helices and to be s imilar to four hydrophobic subunits of other electron transport enzymes which are likely to react with quinones (Fig. 4). The nrfD genes of E. coli and Haemophilus in¯uenzae are constituents of gene clusters encoding Ôcytochrome c nitrite reductaseÕ.NrfAofE. coli is known to b e the catalytic subunit of this enzyme [37]. The gene product of nrfD was proposed to encode the m embrane anchor of the enzyme and to carry the site of quinol oxidation. Tetrathionate reductase (Ttr) of Salmonella typhimurium is thought to catalyze the reduction of Fig. 2. A ctivity of polysul®de respiration with H 2 as a function of the MM content o f fusion particles. The maximum activity was assumed to be measured with 57 lmo l MMág pho spho lipid -1 . Fig. 3. T he e ect of the protonophore TTFB o n the activities of poly- sul®de respiration in cells of W. succinogenes. TTFB dissolved in demethylsulfoxide was added 3 min before the electron transport was started. The speci®c activity of polysul®de respiration with H 2 and formate was 3.6 and 0.72 Uámg A1 cell protein, respectively at 37°C. 1090 W. Dietrich and O. Klimmek (Eur. J. Biochem. 269) Ó FEBS 2002 tetrathionate to thiosulfate b y a quinol [39]. The catalytic subunit ( TtrA) is p redicted to carry molybdenum and a n iron±sulfur center. TtrB is predicted to harbour four iron± sulfur centers. TtrC was proposed t o anchor the enzyme in the membrane and to carry the quinol site. TtrC is larger than PsrC and probably contains an additional hydropho- bic helix at its C-terminus. A molybdo-iron ±sulfur enzyme which is similar to Psr and tetrathionate reductase is possibly encoded by a gene cluster of Archa eoglobus fulgidus. This cluster inc ludes orf 2386 which is p redicted to code for a hydrophobic protein resembling PsrC. The majority of residues conserved among the ®ve p roteins are located in the hydrophobic stretches or close to them. Conserved residues that may be essential for e lectron or proton t ransfer are presented in highlighted letters in Fig. 4. The corresponding residues of PsrC were replaced b y other residues, and the resulting mutants were characterized (Table 2). The psrC mutants were constructed from a mutant (DpsrC) of W. succinogenes carrying the kan gene instead of psrC (Fig. 1). Cells of this mutant grown with fumarate did not catalyze polysul®de respiration with H 2 or formate, in contrast to the wild-type strain (Table 2). PsrA and the enzymic a ctivities of Psr [reactions (c) and (d)] were found in cells of the mutant, but were missing in the membrane fraction. PsrA and t he activity of reaction (d) w ere previously found to be located in the periplasmic cell fraction of mutant DpsrC [10]. Integration of plasmid pKpsrC into the genome of DpsrC resulted in strain KpsrC which carried the intact psrABC operon including psrC (Fig. 1). This s train had wild-type a ctivities of polysul®de respiration with H 2 and with formate. As with the wild-type strain, P srA and the activities of Psr were f ound in the membrane fraction of strain KpsrC. These results show that PsrC anchors Psr in the membrane a nd is required for polysul®de respiration, but not for t he Psr activities. The other psrC mutants listed in Table 2 were construc- ted by integrating derivatives of pKpsrC with altered codons into the genome of the DpsrC mutant. All the mutants so obtained h ad PsrA bound to the membrane, and the speci®c activities of Psr were similar to those of the wild-type s train. Hence, the i ntegration of Psr into the membrane and its enzymic activities were not affected by the mutations. The mutants differ in their speci®c activities of polysul® de respiration. Eight of the 23 mutan ts either did not catalyze polysul®de respiration or their speci®c activities were less than 5% of those of the wild-type strain. In these mutants a residue was a ltered which is presumably located in one of the eight hydrophobic segments of PsrC (Y23F, Y159F, E225Q, R305F and R305K) or next to their ends (D76N, D76L, and D218N). The replacement of another three residues l ocated in hydrophobic stretches of PsrC had less drastic effects; the a ctivities of T160 V, S185A and S 188A amounted to approximately 10, 50 and 25%, respectively, of those of the wild-type strain. Substitution of Y106, E146, S192 and of W261 which are located in hydrophobic stretches had no effect on the electron transport a ctivities. Wild-type activities were also measured with mutants W16F, H 82A, S94A, N 174D, E209Q, and Y310F. These residues are predicted to be located i n the hydrophilic loops of PsrC. In summary, the properties of the mutants (Table 2) suggest that certain residues of PsrC are essential for electron transfer from the dehydrogenases to polysul®de, indicating that PsrC serves speci®c functions in addition to that of the membrane anchor of Psr. The activity of polysul®de respiration in the membrane fraction of the wild-type strain was found to be considerably enhanced by increasing the amount of MM (Fig. 2). Therefore, it was feasible that higher amounts of MM would restore the electron transport in the mutants lacking this activity. To test this possibility, the membrane fraction Fig. 4. S equence alignment of PsrC of W. succinogenes (W. s.) to four proteins predicted from DNA sequences. Residues which were replaced in PsrC by site-directed mutagenesis are highlighted. These residues are s hown in white on black background, if the corresponding mutants sho wed 5% or less of t he wild-type activities of polysul®de respiration (Table 2). P utative membrane spanning segments are boxed. A. f., Archaeoglobus fulgidus [36]; E. c., Escherich ia coli [37]; H. i., Haemophilus in¯uenzae [38]; S. t ., Salmonella typhimurium [39]. Ó FEBS 2002 Polysul®de respiration of W. succinogenes (Eur. J. Biochem. 269) 1091 of thes e mutants (Y23 F, Y159F, D 218N, E225Q, R305F and R305K) was fused to liposomes containing high concentrations of MM as in the experiment shown in Fig. 2. Electron transport a ctivity was not restored in these fusion particles which contained 20-fold the amount of MM present in the membrane fraction (not shown). This s uggests that the lack of electron transport a ctivity was not caused by a decreased af®nity of the putative MM binding site for MM. However, a stimulation w ould not be seen if this af®nity in the mutants was d ecreased by more t han two orders of magnitude. With mutants showing partial inhibi- tion of the electron transport, this activity was either stimulated (E225D) or was not signi®cantly altered (S185A, S188A) by the increased quinone content. DISCUSSION The function of MM The standard redox potential of MK in organic s olution at pH 7 (E o ¢) was d etermined to be A74 mV [40,41]. A methyl group in the aromatic ring of n apthoquinones was found to lower the E o ¢ by approximately 16 mV [41]. Therefore, the E o ¢ of MM in organic solution is assumed to be A90 mV [reaction (2) in Table 3]. The same value is likely to apply for MM in the bacterial membrane, as the E o ¢ of MK in a bacterial membrane was determin ed to be close to that in organic solution [43]. It will be sho wn below that MMH 2 dissolved i n the membrane is not suf®ciently electro- negative to serve as donor for polysul®d e reduction. Tetrasul®de ( 2À 4 ) an d p entasul®de ( 2À 5 ) a re the only polysul®de species occurring at signi®cant concentrations in the solutions (10 A2 M HS ± , pH 8 ) used for measuring polysul®de respiration [44]. The concentration of pentasul- ®de is about half that of tetrasul®de. The redox potential of tetrasul®de [reaction (4) in Table 3] was evaluated from t hat of elemental sulfur [reaction (3) in Table 3] using the equilibrium constant of reaction (e) ( 3.6 ´ 10 A9 M ;[2,44]). 3/8 S 8 +HS ± ®  2À 4 +H + (e) The standard potential of tetrasul®de at pH 8 [reaction (4) in Table 3] turns out to be nearly equal to that of elemental sulfur; this also holds true for p entasul®de. These potentials a re approximately 150 mV more n egative than that of MM [reaction ( 2) in Table 3]. As a consequence, the reduction of polysul®de by MMH 2 [reaction (f)] is extremely ende rgonic. From the equilibrium constant at pH 8 of reaction ( f) [5 ´ 10 A16 M 3 ,from reactions (2) and (4) in Table 3] it is calculated that t he reaction becomes exergonic when the ratio MM : MMH 2 exceeds 2 ´ 10 A4 with the concentrations of tetrasul®de and sul®de at 10 A4 M and 10 A2 M , respectively. 3MMH 2 +  2À 4 ® 3MM + 4HS ± +2H + (f) The ratio corresponds to a c oncentration of MM w ithin themembrane(0.6´10 A6 M ) which is more than a n order of magnitude below the K m of hydrogenase for DMN (15 ´ 10 A6 M ). The K m for this water soluble menaquinone analogue is thought to re¯ect those f or the quinones within the membrane on the basis t hat the content of 3 lmol qui- noneág phospholipid A1 corresponds to 3 ´ 10 A3 M quinone concentration. If MM/MMH 2 was a component of the electron transport chain c atalyzing polysul®de reduction by H 2 , the steady state concentration of MM would b e below 0.6 ´ 10 A6 M . The corresponding velocity of MM reduction by H 2 would be much lower than that of the overall electron transport f rom H 2 to polysul®de. There- fore, the species of MM involved in polysul®de respiration should h ave a much lower redox potential than that of MM dissolved in the membrane. The redox potential at pH 8 of the species involved in polysul®de respiration can be estimated a ssumin g its equilibrium ratio o xidized/ reduced to be 1 (instead of 2 ´ 10 A4 ) and the equilibrium concentrations of  2À 4 (10 A4 M ) and HS ± (10 A2 M )used above. The value s o obtained [ pH8   A260 mV, reaction (5) in Table 3] is regarded as the upper limit of redox potential of the quinol suited as donor of polysul®de reduction. Hypothetical mechanism of polysul®de respiration It is envisaged that the species of M M serving in polysul®de respiration is bound to PsrC in its o xidized (MM b )andits reduced form (MM b H ± ) which is the q uinol anion. MM b is thought to be reduced to MM b H ± by accepting ele ctrons from the cytochrome b subunit (HydC) of h ydrogenase (or FdhC of formate dehydrogenase) upon collision with PsrC (Fig. 5 ). MM b reduction is assumed to be coupled to proton uptake from the cytoplasmic side of t he membrane, and MM b H ± oxidation to be c oupled to proton release at the periplasmic side. MM b and MM b H ± are thought to be located in the hydrophobic part of PsrC. Therefore, the uptake and release of protons is expected to be gu ided by proton channels. The channel for proton release should be in PsrC and t hat for proton uptake i n HydC. The H + /e ratio of 0.5 predicted by the mechanism is half that determined for fumarate respiration of W. succinogenes,in agreement with the growth yields of polysul®de and fumarate respiration [2,3,6]. The  pH8  of MM b /MM b H ± isassumedtobeA260 mV [reaction (5) in Table 3]. Using this value, the free energy (D pH8  )ofMM b reduction by H 2 and of MM b H ± oxidation by polysul®de is calculated ( Table 4, Dw  0). The free energy values are e xpected to be altered by the Dw  170 mV across the membrane of growing cells. Assuming MM b and MM b H ± to be located in t he center of the membrane, each electron derived from H 2 would become 85 mV more electropositive o n i ts way f rom the periplasmic side to MM b (Fig. 5 ). The simultaneous transfer of a proton from the cytoplasmic side to MM b should a ffect thefreeenergyofMM b reduction by H 2 in the same way. Thus MM b reduction by H 2 should become 24 kJámol A1 more endergonic, as two electrons (derived from H 2 )from Table 3. R edox potentials of compounds involved in polysul®de r espi- ration with H 2 . E o ¢ of reaction (3) was taken from [42]. The values o f reactions (2), ( 4), and (5) are de rived as described in the text. E o ¢ (mV)  pH8  (1) H 2 ® 2H + +2e ± A420 A480 (2) MMH 2 ® MM + 2H + +2e ± A90 A150 (3) HS ± ® 1/8 S 8 +H + +2e ± A275 A305 (4) 4HS ± ® S+4H + +6e ± A260 -300 (5) MM b H ± ® MM b +H + +2e ± A230 A260 1092 W. Dietrich and O. Klimmek (Eur. J. Biochem. 269) Ó FEBS 2002 the periplasmic (positive) side and one proton from the cytoplasmic side are transferred to MM b .MM b H ± oxida- tion should become 8 kJámol A1 more exergonic, as it is coupled to the transfer of two e lectrons and one proton from the center of the membrane t o the periplasmic side. The free energy conserved in the Dp generated by polysul®de respiration is given by the difference i n D pH8  of polysul® de reduction by H 2 in the absence and in the presence of the Dw (16 kJámol  À1 2 ), and is exclusively conserved from MM b reduction by H 2 (Table 4). This value is consistent with the formation of 0 .33 mol ATP per mol H 2 at a phosphoryla- tion potential of 44 kJámol ATP A1 [45]. This ATP yield would b e h alf t hat o f f umarate respiration, in agreement with the growth yields [2,3,6]. The effect of D p on electron transport activity The effect of Dp on the a ctivity of electron t ransport ( Fig. 3) can be explained if it is assumed that the activity of polysul®de resp iration in the membrane fraction is limited by the amount of MM b H ± in the absence of a Dp,andthat MM b H ± dissociates from PsrC (C) according to reaction (g), where    designates a proto n taken up from or released to the cytoplasmic side of the membrane. MM b H ± +    ® MMH 2 +C (g) Assuming that MM b H ± is located i n the center of the membrane, its amount should be increased by a Dpacross the membrane according to reaction (g), whose velocity could be lower than the a ctivity of polysul®d e respiration. The view that polysul®de re spiration is limited by the amount of MM b H ± in the membrane fraction i n the absence of a Dp is supported by t he experiment shown i n F ig. 2. In this experiment, polysul®de respiration was stimulated Fig. 5. H ypothetical mechanism of polysul®de respiration with H 2 . MM b , M M bound to PsrC; MM b H ± , hydroquinone anion of MM boundtoPsrC.ThedottedandthestripedareasdesignateHydand Psr, respectively. Hyd catalyses the reduction of MM b by H 2 , probably at one o f the hemes. T his reaction is coup led to the uptake of a proton from the cytoplasm. Psr catalyses the oxidation of MM b H ± by [S] which is coupled to the release of a p roton to the periplasm. Table 4. S tandard free energies at pH 8 of the reactions thought to make up polysul®de respiration with H 2 .MM b and MM b H ± designate MM bound to PsrC in the oxidized and reduced state (Fig. 4). [S] designates polysul®de sulfur. D pH8  values were calculated from the  pH8  of reactions 1, 4, and 5 given in Table 3. Dw Designates the electrical potential across t he membrane which i s generated by polysul®de respiration [3,5]. D pH8  (kJá mol A1 ) Dw  0mV Dw  170 mV H 2 +MM b ® MM b H ± +H + A43 A19 MM b H ± + [S] ® MM b +HS ± +8 0 Total: H 2 + [S] ® HS A +H + A35 A19 Fig. 6. H ypothetical topology of PsrC. Resi- dues replaced in PsrC by site-directed muta- genesis are indicated. Residues i n bold letters correspond to mutants with 5% or less of the wild-type speci®c activities of poly sul®de res- piration. Ó FEBS 2002 Polysul®de respiration of W. succinogenes (Eur. J. Biochem. 269) 1093 nearly two-fold upon the incorporation of additional MM. As the MM dissolved in the membrane is likely to be fully reduced in the steady s tate of polysul®de respiration, the amount of MM b H ± should increase according to reaction (g) as increasing amounts o f MM are incorporated into the membrane. On the basis of these considerations, the titration curve (Fig. 2) re¯ects the formation of MM b H ± according to reaction ( g). The function of PsrC Like the iron-sulfur subunit of fumarate reductase [46], PsrB is thought to serve as mediator of electron transfer between the prosthetic group (MM b ) of the membrane anchor (PsrC) and the catalytic subunit ( PsrA) of Psr. Therefore, PsrA is probably boun d to PsrB which in turn is bound to PsrC on the periplasmic side of the membrane. Dissociation of PsrA from the membrane was not observed in any of the psrC mutants listed in Table 2 (mutant DpsrC excepted). The activities of Psr are not impaired in the mutants listed in Table 2. This also refers to the mutants lackin g at least 95% of the activities of polysul®de respiration. The residues substitutedinthesemutants(inboldtypeinFig.6)are either charged at neutral p H or are tyrosines, the phenolic hydroxyl groups of which appear to be essential for polysul®de respiration. The function of these residues m ay be explained on the b asis o f t he hypoth etical m echanism depicted in Fig. 5. R305 is well suited for binding and stabilizing the postulate d quinol anion of MM (MM b H ± ) which is thought to be bound to PsrC in a hydrophobic environment. The mere positive c harge of R305 is appar- ently not suf®cient for its function, as the smaller lysine residue d id not substitute for R305. D218 and E225 may serve in guiding the proton formed by the o xidation of MM b H ± to the periplasmic side of the membrane. Consis- tent with this function, the negative charge of E225 appears to be essential, as mutant E225Q was nearly inactive , while E225D showed more t han 25% of the w ild-type a ctivity. D76 a s w ell as Y23 and Y159 may possibly involved in the proton transfer to and from the cytoplasmic side according to reaction (g) w hich describes the binding of reduced MM to PsrC and its dissociation. Consistent with this function, substitution of D76 by asparagine o r leucine r esulted in nearly inactive mutants, whereas D76H exhibited about 10% of the wild-type activity of polysul®de respiration. The hydroxyl groups of Y23 and Y159 may form hydrogen bounds to MM b and MM b H ± . W. succinogenes mutants with one of the four heme ligands of HydC replaced were found to lack the activities of electron transport form H 2 to polysul®de and to fumarate [15]. The heme groups appear to be required for the reduction of MM b and o f MK. However, i t i s not known whether the two quinones are reduced at the same site of Hyd C. Assuming that there is only one site for quinone reduction on HydC (or FdhC), it has to be postulated that MM b protrudes f rom its binding site on PsrC into the lipid phase of the membrane to accept electrons form the heme group(s) together with a proton as depicted in Fig. 5. Alternatively, MM H 2 transiently bound to the c yto- chrome b might transfer a hydride t o MM b . Consistent with the latter mechanism, hydrogenase was found to catalyze MM reduction by H 2 which is coupled to the generation of a Dw across the membrane of liposomes containing hydrogenase (S. Biel, J. Simon & R. Gross, Institu t fu È r Mikrobiologie, Johann Wolfgang Goethe- Universita È t Frankfurt-am-Main, personal c ommunication). In fumarate respiration, MK may be reduced at the same site on HydC. In this case, the resulting MKH 2 would be free to diffuse to fumarate reductase. ACKNOWLEDGEMENTS The authors are indebted to A. Kro È ger for helpful discussion and to O. Schu È rmann for skilful technical assistance. The w ork was supported by the Deutsche Forschungsgemeinschaft (SFB 472). REFERENCES 1. Macy, J.M., Schro È der, I., Thauer, R.K., K ro È ger, A. (1986) Growth of Wolinella succinogenes on H 2 S plus fumarate and on formate plus sulfur as energy sources. Arch. Microbiol. 144, 147±150. 2. Klimmek, O., Kro È ger, A., Steudel, R., Holdt, G. (1991) Growth of Wolinella succinogenes with polysul®de a s terminal acceptor of phosphorylative electron transport. Arch. Microbiol. 155, 177±182. 3. Klimmek, O. (1996) Aufbau und Funktion der Polysul®d- Reduktase von Wolinella succinogenes. PhD Thesis, University of Frankfurt am Main, Germa ny. 4. Schauder, R. & Kro È ger, A. (1993) Bacterial sulphur respiration. Arch. Microbiol. 159, 491±497. 5. Wloczyk, C., Kro È ger, A., Go È bel, T., Holdt, G., Steudel, R. (1989) The electrochemical proton potential generated by the sulphur respiration of Wolinella succinogenes. Arch. Microbiol. 152, 600±605. 6. Hedderich, R., Klimmek, O., Kro È ger, A., Dirmeier, R., Keller, M., Stetter, K.O. (1999) Anaerobic respiration with sulfur and with organic disul®des. FEMS Microbiol. Rev. 22, 353±381. 7. Schro È der, I., Kro È ger, A., Macy, J.M. (1 988) Isolation of the sul- phur reductase and reconstitution of the sulphur resp iration o f Wolinella succinogenes. Arch. Microbiol. 14 9 , 572±579. 8. Fauq ue, G., Klimmek, O., Kro È ger, A. (1994) Sulfur reductases from mesophilic sulfur-reducing e ubacteria. Me thods Enzymol. 243, 367±383. 9. Krat, T., Bokranz, M., Klimm ek, O., Schro È der, I., Fahrenholz, F., Kojro, E., Kro È ger, A. (1992) Cloning and nucleotide sequence of the psrA gene of Wolinella succinogenes polysulphide reductase. Eur. J. Biochem. 206, 503±510. 10. Krat, T., Groû, R., Kro È ger, A. (1995) The function of Wolinella succinogenes psr genes in electron transport with polysulphide as the terminal acceptor. Eur. J. Biochem. 230, 601±606. 11. Gross, R., Simon, J., Theis, F., Kro È ger, A. (1998) Two membrane anchors of Wolinella succinogenes h ydrogen ase and their function in fumarate and polysul®d e r espiration. Arch. M icrobiol. 17 0, 50±58. 12. Dross, F., Geisler, V., Lenger, R., Theis, F., Krat, T., Fahrenholz, F., Kojro, E., Ducheà ne, A., Tripier, D., Juvenal, K., Kro È ger, A. (1992) The q uinone-reactive Ni/Fe-hydrogenase o f Wolinella succinogenes. Eur. J. Biochem. 206, 93±102. 13. Unden, G. & Kro È ger, A. (1982) Reconstitution in liposomes of the electron transport c hain catalyzing fumarate reduction by f or- mate. Biochim. Biophys. Acta 682, 258±263. 14. Bokranz, M., Gutmann, M., Ko È rtner, C., Kojro, E., Fahrenholz, F., Lauterbach, F., Kro È ger, A. (1991) Cloning and nucleotide sequence of the structural genes encoding the f orma te dehydro- genase of Wolinella succinogenes . Arch. Microbiol. 156, 119±128. 15. Gross, R., Simon, J., Lancaster, C.R.D., Kro È ger, A. (1998) Identi®cation of histidine residues in Wolinella succinogenes hydrogenase that are essential for menaquinone r eduction by H 2 . Mol. Microbiol. 30, 639±646. 1094 W. Dietrich and O. Klimmek (Eur. J. Biochem. 269) Ó FEBS 2002 16. Jankielewicz, A., Klimmek, O., K ro È ger, A. (1995) The electron transfer from hydrogen ase and formate d ehydrogenase to poly- sul®de red uctase in the m embrane of Wolinella s uccinogenes. Biochim. Biophys. Acta 1231 , 157±162. 17. Kro È ger, A., Geisler, V., Ducheà ne, A. (1994) Isolation of Wolinella succinogenes hydrogenase, chromatofocusing. In A Practical G uide to Membrane Protein Puri®cation (von Jagow, G. & Scha È gger, H., eds), pp. 141±147. Academic Press, London. 18. Lorenzen, J.P., Kro È ger, A ., Unden, G. (1993) Regulation of anaerobic respiratory pathways in Wolinella succinogenes by the presence of electron acceptors. Arch. Microbiol. 159, 477±483. 19. Laemmli, U.K. ( 1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680±685. 20. K yhse-Andersen, J. (1984) Electroblotting o f multiple gels: a simple apparatus without buer tank for rapid transfer of proteins form polyacrylamide to nitrocellulose. J. Biochem. Biophys. Methods 10, 203±209. 21. Singleton, W.S., Gray, M .S., Brown, M.L., W hite, J.L. (1965) Chromato-graphically ho mogeneou s lecithin fro m egg phosph o- lipids. J. Am. O il Chem. Soc. 42, 53±56. 22. U nden, G. & K ro È ger, A. (1986) Reconstitution of a functional electron transport chain form p uri®ed formate dehydrogenase and fumarate reductase complexes. Methods Enzymol 126, 387±399. 23. U nden, G. (1988) Dierential r oles for m enaquinone and d em- ethyl-menaquinone in anaerobic electron transport of E. coli and their fnr-dependen t expression. Arch. Microbiol. 150, 4 99±503. 24. Klimmek, O., Kreis, V., Klein, C., Simon, J., W ittershagen, A., Kro È ger, A. (1998) The function of the periplasmic Sud prot ein in polysul®de respiration of Wolinella succinogenes. Eur. J. Biochem. 253, 263±269. 25. B ode, Ch, Go eb ell, H., S ta È hler, E. (1968) Zur E liminierung von Tru È bungsfehlern bei der Eiweiûbestimmung mit der Biuretmeth- ode. Z. Klin. C hem. U. Klin. Biochem. 6, 418±422. 26. Sambrook, J., Fritsch, E.F., Maniatis, T. (1989) Molecular Clon- ing. A Laboratory M anual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 27. Lenger, R., Herrmann, U., Gross, R., S imon, J., Kro È ger, A. (1997) Structure and function of a s econd gene cluster encoding the formate d ehydrogenase of Wolinella succinogenes. Eur. J. B io- chem. 246, 646±651. 28. Tabor, S. (1990) Ex pression us ing the T7 RNA p olymerase/pro- moter system. In Current Protocols in Molecular Biology (Ausubel, F.A., Brent, R., Kingst on, R.E., Moore, D.D., S eidman, J.G ., Smith, J.A. & Struhl, K., eds), pp. 1 6.2.1±16.2.11. G reene Pub- lishing and Wiley Interscience, N ew York. 29. Haas, R., Kahrs, A.F., F acius, D., Allmeier, H., Schmitt, R., Meyer, T.F. (1993) TnMax ± a versatile mini-transposon for t he analysis of cloned ge nes and shuttle mutagenesis. Gene 130, 23±31. 30. Simon, J., Gross, R., Ringel, M., S chmidt, E., Kro È ger, A. (1998) Deletion and site-directed mutagenesis of the Wolinella succino- genes fumarate reductase operon. Eur. J. Biochem. 251, 418±426. 31. Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J. (1990) Basic local alignment search tool. J. Mol. Biol. 215, 403±410. 32. Hofmann, K. & Stoel, W. (1993) Tmbase ± a database o f membrane-spanning protein segments. Biol. Chem. Hoppe-Seyler 374, 166. 33. Thompson, J.D., Higgins, D.G., Gibson, T.J. (1994) CLU- STALW: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-speci®c gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673±4680. 34. Collins, M.D. & Fernandez, F. (1984) Menaquinone-6 and thermoplasma-quinone-6 in Wolinella succinogenes. FEMS Microbiol. Lett. 22, 2 73±276. 35. Mell, H., Wellnitz, C., Kro È ger, A. (1986) The e lectrochemical proton p otential and the p ro ton/electron rati o of the electron transport with fumarate in Wolinella succinogenes. Biochim. Bio- phys. Acta 852, 2 12±221. 36. Klenk, H P., Clayton, R.A., Tomb, J F., White, O ., Nelson, K.E., Ketchum, K.A., Dodson, R.J., Gwinn, M., Hickey, E.K., Peterson, J.D. et al. (1997) The c omplet e gen ome seq uence o f the hyperthermophilic, sulphate-reducing archaeon Archaeoglobus fulgidus. Nature 390, 364±370. 37. Hussain, H., Grove, J., Griths, L., Busby, S., Cole, J. (1994) A seven-gene operon e ssential for form ate-dependen t nitrite reduction to ammonia by enteric b acteria. Mol. Microbiol. 12, 153±163. 38. Fleischmann, R.D., Adams, M.D., White, O., Clayton, R .A., Kirkness, E.F., Kerlavage, A.R., Bult, C.J., Tomb, J.F., Dougherty, B.A., Merrick, J.M. (1995) W hole-genome random sequencing and assembly of Haemophilus in¯uenzae Rd. Science 269, 496±512. 39. Hensel, M., Hinsley, A.P., Nikolaus, T., S awers, G., Berks, B.C. (1999) The genetic basis o f tetrathionate respiration in Salmonella typhimurium. Mol. Microbiol. 32, 275±287. 40. S chnorf, U. (1966) D er Ein¯uû von Substituenten auf Redox- potential und Wuchseigenschaften von Chinonen. PhD Disserta- tion no. 3871, ETH, Zu È rich. 41. Clark, W.M. (1960) Oxidation±Reduction Potentials of Organic Systems. Williams & Wilkins, Baltimore. 42. Thauer, R., Jungermann, K., Decker, K. (1977) Energy conser- vation i n c h emotrophic a naero bic bacteria. Bacteriol. Rev. 41 , 100±180. 43. Liebl, U., Pezennec, S ., Riedel, A., Kellner, E., N itschke, W . (1992) The Rieske FeS center fro m the gram-positive bacterium PS3 and its interaction with the menaquinone po ol studied by EPR. J. Biol. Chem. 20, 14068±14072. 44. Giggenbach, W. (1972) Optical spectra and equilibrium distribu- tion of polysul®de ions in aqueous solution at 20°. Inorg. Chem. 11, 1201±1207. 45. Kro È ger, A. & Winkler, E. (1981) Phosphorylative fumarate reduction in Vibrio succinogenes: stoichiometry of ATP synth esis. Arch. Microbiol. 129, 100±104. 46. Lancaster, C.R.D., Kro È ger, A., Auer, M., M ichel, H. (1999) Structure of fumarate reductase from Wolinella succinogenes at 2.2 A Ê resolution. Nature 402, 377 ±387. Ó FEBS 2002 Polysul®de respiration of W. succinogenes (Eur. J. Biochem. 269) 1095 . The function of methyl-menaquinone-6 and polysul®de reductase membrane anchor (PsrC) in polysul®de respiration of Wolinella succinogenes Wiebke. the dehydrogenases to polysul®de, indicating that PsrC serves speci®c functions in addition to that of the membrane anchor of Psr. The activity of polysul®de