Sulfite oxidation in Sinorhizobium meliloti

Biochimica et Biophysica Acta 1787 (2009) 1516–1525 Contents lists available at ScienceDirect Biochimica et Biophysica Acta j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b b a b i o Sulfite oxidation in Sinorhizobium meliloti Jeremy J. Wilson, Ulrike Kappler ⁎ Centre for Metals in Biology, School of Chemistry and Molecular Biosciences, The University of Queensland, St. Lucia, Qld 4072, Australia a r t i c l e i n f o Article history: Received 26 May 2009 Received in revised form 16 July 2009 Accepted 16 July 2009 Available online 24 July 2009 Keywords: Sulfite oxidation Metalloenzyme Taurine metabolism Energy generation Gene expression Enzyme a b s t r a c t Sulfite-oxidizing enzymes (SOEs) are crucial for the metabolism of many cells and are particularly important in bacteria oxidizing inorganic or organic sulfur compounds. However, little is known about SOE diversity and metabolic roles. Sinorhizobium meliloti contains four candidate genes encoding SOEs of three different types, and in this work we have investigated the role of SOEs in S. meliloti and their possible link to the metabolism of the organosulfonate taurine. Low level SOE activity (∼ 1.4 U/mg) was present under all conditions tested while growth on taurine and thiosulfate induced high activities (5.5–8.8 U/mg) although S. meliloti cannot metabolize thiosulfate. Protein purification showed that although expression of two candidate genes matched SOE activity patterns, only a single group 2 SOE, SorT (SMc04049), is responsible for this activity. SorT is a heme-free, periplasmic homodimer (78 kDa) that has low homology to other bacterial SOEs. SorT has an apparent kcat of 343 s− 1 and high affinities for both sulfite (KMapp_pH8 15.5 μM) and ferricyanide (KMapp_pH8 3.44 μM), but not cytochrome c, suggesting a need for a high redox potential natural electron acceptor. KMapp_sulfite was nearly invariant with pH which is in contrast to all other well characterized SOEs. SorT is part of an operon (SMc04049-04047) also containing a gene for a cytochrome c and an azurin, and these might be the natural electron acceptors for the enzyme. Phylogenetic analysis of SorT-related SOEs and enzymes of taurine degradation indicate that there is no link between the two processes. © 2009 Elsevier B.V. All rights reserved. 1. Introduction Bacteria have a major role in both oxidative and reductive reactions of the biological sulfur cycle, and the bacterial sulfur conversions can be roughly classified into reactions with a role in energy metabolism and reactions involved in assimilation of sulfur into cell biomass. In soil environments most of the available sulfur occurs as organosulfonates or organosulfate esters [1,2]. It is thus not surprising that many soil bacteria including plant symbionts such as Sinorhizobium meliloti are not only able to utilize these organically bound forms of sulfur in order to gain access to sulfur for cell biomass, but can use the carbon skeleton of these compounds as a growth substrate [2–4]. A particularly abundant organosulfonate that can be used by many bacteria as a source of cell carbon, sulfur or nitrogen is taurine (2-aminoethanesulfonate) [2–4]. A common pathway for taurine degradation involves taurine deamination by a taurine dehydrogenase (TauXY) to sulfoacetaldehyde [5,6] which is then desulfonated by the action of sulfoacetaldehyde acetyltransferase (Xsc), yielding acetyl-phosphate and sulfite. Acetyl-phosphate can either be assimilated into cell biomass via acetyl-CoA or used for substrate level phosphorylation with subsequent excretion of acetate [3]. Most steps of this pathway have been characterized in some form, however, ⁎ Corresponding author. Tel.: +61 7 3365 2978; fax: +61 7 3365 4620. E-mail address: u.kappler@uq.edu.au (U. Kappler). 0005-2728/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.bbabio.2009.07.005 unclear is the fate of the sulfonate sulfur, which is initially present as sulfite but is usually recovered as sulfate, indicating the presence of an as yet incompletely characterized enzymatic oxidation step [4,7,8]. Information on possible links between the occurrence of certain types of sulfite-oxidizing enzymes (SOEs) and the presence of organosulfonate degradation pathways is also scarce. Sulfite can be enzymatically converted to sulfate by molybdenumcontaining enzymes of the sulfite oxidase family [EC1.8.3.1, sulfite oxidase (SO) and EC1.8.2.1, sulfite dehydrogenase (SDH)] [9,10] and here we have systematically investigated the enzymology of sulfite oxidation in S. meliloti grown in the presence of a variety of organic or inorganic sulfur compounds. Bacterial sulfite oxidation is generally linked to the respiratory chain and is therefore regarded as part of the cell's energy generating systems. The SO family contains three distinct groups of enzymes based on the size and structure of the Mo-domain (Fig. 1) [9], a classification that corresponds well with the conserved domains identified within the SO enzyme family (cd_00321) [11]. Interestingly, SOE group 2 is the only group that contains sequences of eukaryotic origin, while both group 1 and 3 SOEs seem to be exclusive to prokaryotes. The first bacterial enzyme of the SO family to be fully characterized was the SorAB sulfite dehydrogenase from Starkeya novella, a periplasmic group 2 enzyme consisting of a large, Mo-binding subunit (SorA) and a small cytochrome c-binding subunit (SorB) that form an integral complex [12,13]. Since then a number of sulfite-oxidizing enzymes from other bacteria have been reported [14–16], but all of J.J. Wilson, U. Kappler / Biochimica et Biophysica Acta 1787 (2009) 1516–1525 1517 Fig. 1. The sulfite oxidase enzyme family and putative SOEs from S. meliloti: relationship between conserved motifs (COG, cd) and phylogenetic/functional groupings. these enzymes are only distantly related [9]. In addition, other bacterial SO-like enzymes such as the SoxCD sulfur dehydrogenase (group 2) [17,18] and the YedY oxidoreductase (group 1) [19,20] are known that do not oxidize sulfite. In this study we have used the α-Proteobacterium S. meliloti as a model organism for elucidating the nature and functions of SOEs in the metabolism of this bacterium. While S. meliloti is well known for its capacity to act as a nitrogen-fixing symbiont of plants, little is known about its ability to grow with inorganic and organic sulfur compounds, which is likely, however, to aid the survival of free-living S. meliloti in soils in the absence of suitable host plants. A complete operon for taurine catabolism via a taurine dehydrogenase (TauXY) has been identified in the genome of S. meliloti RM1021 [5]. However, as is the case for other sulfonate-metabolizing bacteria, the nature of the SOE that is part of this process in S. meliloti has not been established, and here we have investigated the molecular details of this process as well as the phylogenetic relationship of SOEs and taurine degradation pathways. 2. Materials and methods 2.1. Bacterial strains, media and growth conditions S. meliloti strain 1021 [21] was routinely grown aerobically at 30 °C using liquid or solid TYS medium [22] containing 25 μg/mL Streptomycin. Sulfur compound utilization was assessed using a modified version of DSMZ (German type culture collection) medium 69 containing twice the original concentration of phosphate buffer. The medium was prepared as a basal salt medium without the addition of thiosulfate, and different carbon or sulfur sources (methanol, ethanol, sodium thiosulfate, potassium tetrathionate, taurine, methanesulfonic acid, ethanesulfonic acid, dimethylsulfoxide, sodium fumarate, sodium hydrogen carbonate) were added at a final concentration of 20 mM, glucose was used at a final concentration of 10 mM. Solutions of volatile or instable compounds were prepared immediately prior to use. Where necessary, substrate stock solutions were adjusted to near neutral pH before use. Cultures containing inorganic sulfur compounds as primary electron donor were grown under autotrophic conditions in the absence of added carbon other than the 0.3 g/L yeast extract contained in the medium base. Cultures supplemented with sodium carbonate are equivalent to basal salt controls as addition of sodium carbonate only alters the level of carbon dioxide available for fixation, not the level of reduced carbon compounds present in the medium. 2.2. Growth experiments Cultures (10 mL medium in sterile 50 mL tubes) were set up in triplicate from precultures grown with the same sulfur or carbon compound and incubated with shaking at 30 °C for 48 h. OD600 readings were taken after inoculation, 24 h and 48 h. Cultures were harvested by centrifugation and pellets stored at − 20 °C. Growth rates of S. meliloti on taurine, thiosulfate, methanesulfonic acid and glucose were determined using 250 mL flasks containing 35 mL medium. Cultures were inoculated at OD600 0.1 (glucose and taurine) or OD600 0.01 (methanesulfonic acid and thiosulfate), OD600 readings were taken every hour. Growth characteristics determined here were used in the preparation of cultures for RNA isolation. 2.3. Biochemical methods The activity of sulfite-oxidizing enzymes was routinely determined using 20 mM Tris–Cl buffer pH 8.0, 1 mM ferricyanide, 2 mM freshly prepared sodium sulfite and varying amounts of cell extracts. Assays were performed at 25 °C by monitoring the absorbance change at 420 nm (ɛferricyanide = 1.09 mM− 1 cm− 1). One unit of activity corresponds to the amount of enzyme required to oxidize 1 μmol of sulfite per minute. For easy comparison with data reported by others, activities are also reported in mkat/kg where applicable. Cytochrome c based assays for SOEs [23] were used where specified. Kinetic parameters of SorT were determined using the ferricyanide assay and 20 mM Tris-acetate buffers. Kinetic parameters were derived by direct non-linear fitting of the data to the Michaelis–Menten equation using SigmaPlot 9 (SYSTAT Software). The pH dependence of activity was determined in 50 mM buffers (bis-Tris, Tris, glycine, CAPS) between pH 6 and 12. The activity of malate dehydrogenase was determined as in [24]. Assays for sulfite oxidase activity were carried out using a Hansatech Oxygen electrode essentially as in [25]. Small scale cell-free extracts for use in enzyme assays were prepared from cell pellets of 10 mL cultures using BugBuster Mastermix™ (Novagen) according to the manufacturer's instructions. Cellular fractionation was performed on cultures in late exponential growth phase using the osmotic shock method [26]. After removal of the periplasmic fraction, cell pellets were resuspended in 20 mM Tris– 1518 J.J. Wilson, U. Kappler / Biochimica et Biophysica Acta 1787 (2009) 1516–1525 Cl pH 8, passed three times through a French Pressure cell (Aminco, 12 000 psi) followed by separation of the membrane and soluble protein fractions by ultracentrifugation (90 min, 4 °C, 145 000 ×g) in a Beckman L8-80 ultracentrifuge. Membrane fractions were resuspended in 20 mM Tris–Cl pH 8 using a handheld glass homogenizer. Cellular fractionation by osmotic shock uses very mild conditions, and as no enzymatic digestion of the cell wall is involved, release of periplasmic proteins is often incomplete. However, the osmotic shock method avoids accidental release of cytoplasmic proteins due to cell lysis during isolation of periplasmic proteins using spheroplasts and was chosen for this reason. Protein gel electrophoresis used the method of [27]. Protein determinations were performed using the BCA-1 kit (Sigma Aldrich) or the 2DQuant kit (GE Healthcare Biosciences). Mass fingerprints of proteins separated on SDS-PAGE gels were prepared as in [28] and analyzed using a VoyagerSTR MALDI-Tof mass spectrometer (Applied Biosystems). Electrospray mass spectrometry was performed on a Q-Star mass spectrometer (Applied Biosystems) essentially as in [28]. Molecular mass determination of native, purified proteins by MALLS (Multi Angle Laser Light Scattering) was carried out as in [29] using a miniDAWN Tristar laser light scattering photometer and Optilab DSP interferometric refractometer (both Wyatt technology) equipped with a Superdex 75 HR (10/30) gel filtration column and using 20 mM Tris–Cl pH 7.8, 150 mM NaCl as the buffer. A sample volume of 50 μL of an approximately 200 μM protein solution was used per injection, experimental errors are reported as standard deviations of the molecular mass estimate. Mo-content of protein samples was determined by ICP-MS at the ENTOX centre at the University of Queensland. 2.4. Purification of the S. meliloti sulfite-oxidizing enzyme 35–40 g (wet weight) of S. meliloti cell material was resuspended in 20 mM Tris–Cl pH 8.8 and passed three times through a French Press (see above). Cell debris was removed by centrifugation (30 000 ×g, 30 min, 4 °C), and the resultant crude extract applied to a DEAESepharose column (2.6 × 24.5 cm, GE Healthcare Biosciences) equilibrated in 20 mM Tris–Cl pH 8.8. Proteins were eluted using a linear gradient from 0 to 300 mM NaCl and fractions testing positive for SDH activity pooled. Ammonium sulfate (15% w/v) was added to the pool which was then applied to a Phenyl-Sepharose FF column (1.6 × 20 cm, GE Healthcare Biosciences) equilibrated in 20 mM Tris–Cl pH 7.8, 15% (w/v) ammonium sulfate. SDH activity eluted in the flow-through and was concentrated (AmiconUltra 30 kDa MWCO, Millipore) before being applied to a Superdex 75 (16/60, GE Healthcare Biosciences) gel filtration column (running buffer: 20 mM Tris–Cl pH 7.8, 150 mM NaCl). Where further purification was required, pooled samples were desalted by dialysis against 20 mM Tris–Cl pH 8.8 and applied to a MonoQ column (5/5, GE Healthcare Biosciences) (gradient: 0–250 mM NaCl). Protein purity and enrichment were monitored by SDS-PAGE and SDH activity assays throughout the purification. 2.5. Molecular biological methods Standard methods were used throughout [26]. All oligonucleotide primers were from Invitrogen. Genomic DNA was isolated using the DNAZOL reagent (Invitrogen), standard PCR reactions used the GoTaq green Mastermix (Promega) according to the manufacturer's instructions. RNA was isolated from S. meliloti cultures grown to midexponential phase and preserved using the Bacteria Protect Reagent (Qiagen) using either the Qiagen RNeasy Kit or the Illustra RNAspin Mini kit (GE Healthcare Biosciences) according to the manufacturer's instructions. DNA contamination of RNA samples was assessed by PCR using generic primers 27F and 1492R [30] that target the bacterial 16S rRNA gene. If a PCR product was detectable after 34 cycles of amplification, the RNA sample was subjected to further DNAse treatment and PCR-testing until no PCR product was detectable. Only DNA- free RNA samples were used for RT-PCR and quantitative PCR. RNA concentrations were determined using the Quant-it RNA kit (Invitrogen). cDNA was synthesized from 500 μg RNA using the Superscript III reverse transcriptase (Invitrogen) and the RNAsin RNAse Inhibitor (Promega). Diluted cDNA (1:10 to 1:100 000) was used as the template for real-time and standard PCR. Quantitative real-time PCR used PCR products of 200 bp and the SYBR green Mastermix (Applied Biosystems). Real-time PCR reactions were set up using an epMotion robot (Eppendorf) and performed on an Applied Biosystems 7900T cycler in 384-well plates. Primer concentrations were optimized for each target gene. Data analysis was carried out as in [31]. 2.6. Bioinformatic analysis of the S. meliloti RM1021 genome A number of known genes involved in the metabolism of sulfur compounds were translated into amino acid sequences and used in BLAST_P searches [32] of the S. meliloti RM1021 genome and the megaplasmids pSymA and pSymB. These included the tau operon from Paracoccus pantotrophus (taurine metabolism, genes tauR, tauA, tauB, tauC, tauX, tauY, xsc, tauZ, acc. no AY498615.3), the SO family-related genes yedY (Escherichia coli, acc. no NC_000913.2, proteins NP_416480 and NP_416481), sorA and soxC (both from St. novella, acc. no AF139113 and AF154565), the DMSO reductase genes dorA (Rhodobacter capsulatus, acc. no U49506) and dmsABC (E. coli, acc. no. J03412) as well as the sox gene cluster from St. novella (genes soxC, soxB, soxD, soxY, soxZ, soxA, soxX, acc. no AF154565). Candidate genes identified in the S. meliloti genome were analyzed using the programs ProtParam [33], TMHMM [34,35], DAS [36], TMPRed [37], TatP [38] and Signal P [39,40] available through the Expasy homepage (www. expasy.org; accessed May 2009) and the Vector NTi Advance 10 program suite (Invitrogen). Phylogenetic analyses were performed on sequence homologues of the Xsc (SMb21530), TauY (SMb21529) and SorT (SMc04049) proteins from S. meliloti. Homologous sequences were aligned using AlignX (Vector NTi Advance11, Invitrogen). The MEGA4.0 software package [41] was used for phylogenetic and bootstrap analyses. Evolutionary distances were computed using the Poisson correction method and are in the units of the number of amino acid substitutions per site. All positions containing gaps and missing data were eliminated from the datasets. 3. Results 3.1. Identification of genes encoding putative sulfite-oxidizing enzymes in the S. meliloti genome Using the sorA and soxC genes from St. novella and the yedY gene from E. coli [19] as the search models, four candidate genes that encode proteins related to the molybdenum subunit of enzymes of the SO family (Table 1, Fig. 2) were identified in S. meliloti [9]. No homologues of dissimilatory adenylylphosphosulfate reductases were found, and therefore, indirect oxidation of sulfite [10] was not considered any further in this context. All four putative S. meliloti SOEs belong to the COG2041 group (sulfite oxidase and related enzymes), and to the conserved domain cd_00321 ‘SO_family_Moco’ [11]. Cd_00321 is a superfamily that contains 4 major subcategories (Fig. 1), cd_02107 to cd_02110 (YedY-like Moco; bact_SO_family_ Moco; arch_bact_SO_family_Moco and SO_family Moco_dimer), with the latter, cd_02110, containing another four subgroups, cd_02111 to cd_02114 that are based on different, well characterized pro- and eukaryotic SOEs (Fig. 1). In the classification of SOEs based on the structure of the catalytic Mo-binding domain [9] cd_02107 and cd_02108 form group 1 ‘enzymes from pathogenic bacteria’, cd_02110 corresponds to group 2 ‘sulfite-oxidizing enzymes and plant nitrate reductases’ and cd_02109 is equivalent to group 3 ‘enzymes from archaea and soil bacteria’ (Fig. 1). 1519 J.J. Wilson, U. Kappler / Biochimica et Biophysica Acta 1787 (2009) 1516–1525 Table 1 Properties of SDH-related genes identified in the S. meliloti genome. Gene name SMc01281 SMb20584 SMc04049 SMa02103 CDD domainsa SO family groupb Location in S. meliloti genome Cellular location No. of aa MW (kDa) pI Transmembrane domains Signal peptide SP cleavage site MW-processed protein pI processed protein cd02107, COG2041 Group 1A Chromosome Periplasmic 313 34.69 6.78 No Yes (TatP), 49.9% (Signal P) res. 46–47, AAA-LE 29.82 5.43 cd02109, COG2041 Group 3 pSYMb Cytoplasmic 239 27.38 5.65 No No cd02110, COG2041 Group 2 Chromosome Periplasmic 399 42.47 5.88 No Yes res. 31–32, AEA-KE 39.42 5.7 cd02110, COG2041 Group 2 pSYMa Cytoplasmic 468 51.13 5.55 No No a b Based on [11]. Based on [9]. Of the four putative SOEs present in S. meliloti, two are group 2 SOEs (cd_02110 SO_family_Moco_dimer) while one each belongs to the group 1 (cd_02107) and group 3 (cd_02109) SO family proteins (Table 1, Fig. 1). Although the group 2 SOEs (cd_02110) contain four well established subgroups, neither of the group 2 S. meliloti SOEs was strongly associated any of these subgroups (Fig. 1), indicating that further differentiation of cd_02110 may be necessary in the future. The S. meliloti group 1 SOE SMc01281 is related to YedY from E. coli (59% identity/69% conserved amino acids). This periplasmically located protein is encoded by the chromosomally located yedY gene (SMc01281) which, like the majority of yedY genes [9], is found in conjunction with a gene (yedZ) encoding a heme b-binding protein with 6 transmembrane helices (Fig. 2, Table 1). The two group 2 SOElike proteins (cd_02110) are encoded by genes SMa02103 and SMc04049. The SMa02103 gene is located on megaplasmid pSymA and encodes a cytoplasmic protein (Fig. 2), while the chromosomally located SMc04049 gene encodes a periplasmic protein. The fourth SOE-candidate gene, SMb20584, is located on megaplasmid pSYMb and encodes a cytoplasmic group 3 SOE with no characterized relatives to date. 3.2. When does S. meliloti require sulfite-oxidizing enzymes? To establish conditions where sulfite oxidation is required by S. meliloti, cells were grown in the presence of a number of carbon com- pounds (glucose, formate, methanol, ethanol, hydrogencarbonate) as well as inorganic (thiosulfate, tetrathionate) and organic sulfur compounds (methanesulfonic acid (MSA), ethanesulfonic acid, taurine, dimethylsulfoxide). Significant biomass production was observed only when glucose or taurine were present in the growth medium (final OD600 of ∼2.4 and ∼1.2, respectively), while formate supported some growth (final OD600 ∼ 0.46) (Table 2). All other substrates led to final OD600 values of 0.2–0.22. As S. meliloti can grow to an OD600 of ∼0.2 on the base medium without added carbon sources it is likely that this level of growth is supported by the small amount of yeast extract (0.3 g/L) present in the basal medium. It is thus unclear whether any of the additional substrates tested actually contributed to the growth of S. meliloti. Cell extracts prepared for all of the above growth conditions (Table 2) contained mostly low levels of SOE activity (between 0.5 and 1.9 U/mg, average 1.4 ± 0.4 U/mg), while high levels of activity were only detected in cell extracts of cultures grown on taurine (5.54 ± 0.42 U/mg) or thiosulfate (8.8± 0.63 U/mg). This is a surprising finding as S. meliloti lacks a pathway for thiosulfate utilization. Cell extracts from MSA grown cultures had a slightly enhanced SOE activity of 2.43 ± 0.13 U/mg (Table 2). Thus, the substrates taurine, thiosulfate, MSA and glucose were chosen for further investigation as they are representative of the three different types of growth substrates used in this study, and distinct SOE-like enzymes might be associated with different prevailing growth modes. Fig. 2. S. meliloti operons encoding proteins related to known sulfite-oxidizing enzymes. Panel A: SMc01281 gene region encoding a YedY related SOE group 1A protein. Panel B: SMc04049 gene region (SOE group 2). Panel C: SMa02103 gene region (SOE group 2). Panel D: SMb20584 gene region (SOE group 3). Colour coding for genes shown: light grey: Molybdenum subunits, grey: heme-containing proteins, dark grey: pseudoazurin-related proteins, hatched: repeat regions, open arrows: unrelated genes present in the gene region. 1520 J.J. Wilson, U. Kappler / Biochimica et Biophysica Acta 1787 (2009) 1516–1525 Table 2 Growth characteristics and sulfite-oxidizing enzyme activity observed for S. meliloti grown with different carbon or sulfur sources. Condition Average U/mg SDH activity OD600 after 48 h Methanol (20 mM) Ethanol (20 mM) Glucose (10 mM) Potassium tetrathionate (10 mM) Sodium thiosulfate (20 mM) Taurine (20 mM) MSA (20 mM) ESA (20 mM) DMSO (20 mM) Sodium carbonate (20 mM) Sodium formate (20 mM) 1.31 ± 0.08 1.89 ± 0.19 1.12 ± 0.06 0.52 ± 0.04 8.85 ± 0.63 5.54 ± 0.42 2.43 ± 0.13 1.45 ± 0.09 1.72 ± 0.15 1.74 ± 0.11 1.57 ± 0.07 0.203 0.241 2.41 0.205 0.22 1.15 0.201 0.199 0.206 0.209 0.464 Enzyme activity averages shown are the mean of at least 14 determinations carried out on crude extracts from three separate cultures for each condition. Errors are given as 95% confidence intervals. Shown in bold and italics: growth substrates chosen for further study. ESA = ethanesulfonic acid, MSA = methanesulfonic acid, DMSO = dimethylsulfoxide. The cellular location of SOE activity was determined for the four model substrates: in all cases, the fractionation by osmotic shock was successful as indicated by the absence of the cytoplasmic marker enzyme, malate dehydrogenase, in the periplasmic protein fraction (data not shown). Periplasmic fractions contained significant amounts of SOE activity (1.67–2.04 U/mg for taurine and thiosulfate), indicating a periplasmic location of the responsible enzyme in all four cases and thus involvement of either SMc04049 (SorT) or the YedY-like SMc01281. 3.3. Expression of SOE genes in S. meliloti Expression of the genes encoding the four putative S. meliloti SOEs was determined, initially using two-step RT-PCR (Fig. 3A). As a control the xsc gene encoding a sulfoacetaldehyde acetyltransferase that is specific to taurine metabolism was used, and the control gene was only expressed in cells that had been grown on taurine (Fig. 3A). Similar results were obtained for the tauY gene that encodes a subunit of taurine dehydrogenase (data not shown). In contrast, three of the four SDH-like genes were expressed in varying amounts under all conditions tested, while no expression of the gene encoding the cytoplasmic group 2 SMa02103 protein was detected in any of the samples, ruling out its involvement in sulfite oxidation (Fig. 3A). Expression of the SMb20584 gene encoding the cytoplasmic group 3 SOE-like protein was very low throughout, confirming that the two periplasmic enzymes SMc01281 (YedY-like, group 1) and SMc04049 (SorT, group 2) were most likely causing the observed SDH activity. Quantification of gene expression by real-time PCR revealed that the SMc04049/sorT gene (group 2 SOE) had the highest expression levels under all conditions tested as well as an expression pattern matching the observed SOE activity and was thus likely to be the major S. meliloti SOE (Fig. 3B). In fact, during growth on taurine and thiosulfate expression of this gene was similar to that of the xsc gene that is induced to high levels in the presence of taurine and is central to taurine catabolism. The SMc01281 gene encoding the second periplasmic SOE (YedY-like, group 1) also showed increased expression in the presence of taurine and thiosulfate, the two growth conditions associated with the highest observed SOE activity in S. meliloti cell extracts. However, expression of SMc01281 on thiosulfate and taurine reached only 7% and 8.8% of the transcript levels detected for the SMc04049/sorT gene under the respective growth conditions. In contrast, the SMb20584 gene was expressed at low levels (0.7% and 1.7% of SMc04049/sorT expression on thiosulfate and taurine), with the highest expression for this gene being observed in the presence of glucose or thiosulfate. This differing expression pattern and the cytoplasmic location of the encoded protein indicate that the group 3 protein encoded by SMb20584 is not likely to be involved in sulfite oxidation in S. meliloti. Fig. 3. Expression of SOE-like genes from S. meliloti in cells grown in the presence of glucose (G), thiosulfate (TS), taurine (T) or methanesulfonic acid (MSA). Panel A: Reverse Transcriptase PCR. RNA was isolated from culture in early to mid-exponential phase, for glucose (G) and taurine (T) additional RNA samples (G2, T2) were isolated from cells in mid to late exponential growth phase. The latter samples were only used in RT-PCR not qPCR experiments, N = no template control. The cDNA was diluted 1:100 for RT-PCR except for SMb20584 where a dilution of 1:10 was used. Panel B: Average normalized gene expression derived from quantitative real-time PCR experiments carried out using the SYBR green technology. Gene expression was normalized relative to the expression of the 16S rRNA gene in each sample. Experimental errors are given as relative standard deviation. Co-transcription with neighboring genes was tested for the SMc04049/sorT and SMc01281/yedY genes that are found in conserved genetic environments. As shown in Fig. 4, the putative SMc01281/yedY and SMc01282/yedZ genes form a transcriptional unit, while SMc04049/sorT is co-transcribed with both the cytochrome c encoding SMc04048 gene and the azu2 gene that encodes a pseudoazurin (Figs. 1 and 3). This suggests that the two most highly expressed SOE-like proteins in S. meliloti are either multisubunit proteins or require accessory proteins for function. 3.4. Purification and characterization of the sulfite-oxidizing enzyme from taurine-grown S. meliloti cells Purification of the S. meliloti SOE was undertaken using cells grown on taurine. Throughout the purification SOE activity was always associated with a single protein peak, making the involvement of more than one enzyme unlikely. As reported for the SOEs from Comamonas/ Delftia acidovorans and Cupriavidus necator [7,8], several unknown proteins tended to co-purify with SMc04049 (SorT). Using peptide mass fingerprints the major co-purifying proteins were identified as a putative dipeptide binding periplasmic protein (SMc02634), a putative oligopeptide transporter protein (SMb21196), a probable fructose-bisphosphate aldolase class I protein (SMc03983) and a putative uncharacterized protein (SMc02156). No functional link 1521 J.J. Wilson, U. Kappler / Biochimica et Biophysica Acta 1787 (2009) 1516–1525 Fig. 4. Co-transcription of SOE-related genes in S. meliloti. Black bars indicate the position and expected sizes of the PCR products generated. Panel A: yedYZ genes (SMc01281 and 01282), panel B: SMc04047-04049 genes. Lanes: R = cDNA template generated from RNA isolated from cells grown with taurine, N = no template control, P = positive control with genomic DNA as template. Amplification conditions were the same for all samples. between these proteins and SMc04049 (SorT) is apparent. The amount of co-purifying proteins could be reduced by changes in the purification parameters. The final enzyme preparation contained a protein with an apparent molecular mass of ∼40 kDa (Fig. 5, top, inset). The preparation was essentially free of other proteins, although minute amounts of a contaminating cytochrome c were present in some fractions. The purified protein was tryptically digested and the resulting mass fingerprint clearly identified it as SMc04049 (SorT) Fig. 5. Kinetic data for purified SorT (SMc04049) S. meliloti sulfite dehydrogenase. Top: Direct non-linear fit of data for KMapp_sulfite at pH 8; inset: SDS-PAGE of purified S. meliloti SorT (SMc04049), left lane: molecular mass standard, right lane: purified SorT (SMc04049) (1.2 μg). Bottom: pH dependence of sulfite-oxidizing activity fitted to a bell-shaped curve. (data not shown). The molybdenum content of the purified enzyme was 59% as determined by ICP-MS. 3.5. Characterization of SMc04049 (SorT) The SMc04049 (SorT) protein eluted from a calibrated Superdex75 (16/60) gelfiltration column with an apparent molecular mass of ∼68 kDa which is larger than the mass of a monomer, but too small for a dimer. In the case of the SorAB sulfite dehydrogenase a similar discrepancy had been due to the presence of an additional subunit with unusual electrophoretic properties [13,42]. As SorT appeared to consist of a single, ∼40 kDa subunit, we used a combination of electrospray mass spectrometry and MALLS to determine the molecular masses of the subunit and the native protein. The mature SMc04049 (SorT) protein has a predicted molecular mass of 39.42 kDa, and using electrospray mass spectrometry a subunit molecular mass of 39.424 ± 0.005 kDa was determined for SorT. In contrast, MALLS analysis of the size of the native protein resulted in a mass of 78.04 ± 0.8 kDa (peak polydispersity: 1.001), clearly showing that the native protein is a homodimer (α2) without additional subunits. The purified SorT (SMc04049) SOE was kinetically characterized using the three established electron acceptors for SOEs, ferricyanide, cytochrome c (horse heart) and oxygen. SorT activity with cytochrome c was only ∼13% (65 ± 6 U/mg or 1081 ± 106 mkat/kg) of the activity observed with ferricyanide (522± 17 U/mg or 8701 ± 284 mkat/kg). No oxygen dependent sulfite oxidation was detected although S. meliloti cells show respiratory activity with sulfite (Voreck and Kappler, unpublished observation), indicating that the reaction catalysed by SorT is linked to the respiratory chain. All further characterization was therefore carried out using ferricyanide as the electron acceptor. At pH 8.0 in 20 mM Tris-acetate buffer, an apparent KMsulfite of 15.5± 2.0 μM was determined (Fig. 5, top), the corresponding apparent KMferricyanide was 3.4± 0.8 μM and the apparent kcat was 343 ± 11 s− 1 (Table 3). The pH dependence of SorT activity was determined between pH 6 and pH 12. The data could be fitted to a bell-shaped curve with apparent pKa values of pH 5.5 and pH 11.1 and maximal sulfite-oxidizing activity between pH 8 and 9 (Fig. 5, bottom). Table 3 Variation of catalytic parameters of the SorT (SMc04049) sulfite dehydrogenase with pH. pH 6 7 8 8.8 9.6 KMapp_sulfite (μM) kcat (s− 1) 22.1 ± 2.7 309 ± 8 10.6 ± 1.3 341 ± 9 15.5 ± 1.9 343 ± 11 10.8 ± 1.2 385 ± 9 7.9 ± 1.2 387 ± 11 1522 J.J. Wilson, U. Kappler / Biochimica et Biophysica Acta 1787 (2009) 1516–1525 Apparent KMsulfite and kcat values of SorT were also determined at pH 6, 7, 8.8 and 9.6 (Table 3). Both parameters were remarkably insensitive to pH, with apparent KMsulfite values varying between 22.1 ± 2.7 μM at pH 6 and 7.9 ± 1.2 μM at pH 9.6. Turnover numbers increased slightly at higher pH values with values varying between 309 ± 8 and 387 ± 11 s− 1 at pH 6 and pH 9.6 respectively. This behaviour of the catalytic parameters is clearly different from similar data reported for the vertebrate sulfite oxidases or the bacterial SorAB sulfite dehydrogenase [43,44], where both apparent KMsulfite and kcat values are strongly pH dependent (Fig. S1). Unlike many of the well characterized SOs and SDHs [13,44,45] SorT was found to be relatively insensitive to inhibition by changes in ionic strength, with 200 mM Tris-acetate buffer causing ∼ 20% inhibition of activity, while sodium chloride at 40 mM and 200 mM caused 11% and 32% inhibition, respectively. Sodium sulfate was a slightly more potent inhibitor of activity with addition of 10 mM and 50 mM causing 9% and 23% inhibition of activity. In contrast, the presence of 15 mM sodium sulfate causes 50% inhibition of enzyme activity in the St. novella SorAB sulfite dehydrogenase [43]. 4. Discussion This is the first study of sulfite-oxidizing enzymes in S. meliloti and the relationship of these enzymes to pathways for the degradation of sulfur compounds. S. meliloti contains four genes that encode SOEs and these include representatives of all three known groups of sulfiteoxidizing enzymes, making S. meliloti an ideal model organism for studying bacterial sulfite oxidation. The ability of S. meliloti to oxidize sulfur compounds has also not been investigated in detail so far, although a complete operon for the degradation of taurine had been previously identified in the S. meliloti genome [5]. Here we have shown that while taurine is readily metabolized neither related sulfonates such as methane- and ethanesulfonic acid nor a number of other organic and inorganic sulfur compounds support growth. The final uncharacterized step in the degradation of organosulfonates is the conversion of sulfite, which is highly reactive and therefore cytotoxic [10,46], to sulfate and our work has unambiguously identified the group 2 SOE SorT (SMc04049) as the enzyme responsible for SOE activity observed in S. meliloti grown in the presence of thiosulfate and the alkanesulfonate taurine (Table 2, Fig. 3A). We also detected expression of the genes encoding two other SOEs, the group 3 enzyme SMb20584 and a group 1, YedY-like SOE (SMc01281). While expression levels for the former were low and its possible function is unknown, expression of the yedY-related gene SMc01281 was high during growth of S. meliloti on taurine and thiosulfate, indicating a possible role for this enzyme in S. meliloti sulfur metabolism although the nature of this role is unclear and does not appear to involve sulfite oxidation. To the best of our knowledge this is the first time that growth conditions under which a YedY-like SOE is expressed have been identified for any bacterium, thereby opening up new possibilities for future studies of this class of SOE. The S. meliloti SorT SOE belongs, like all other pro- and eukaryotic SOEs studied to date, to the group 2 SOEs (cd_02110) within the sulfite oxidase enzyme family, however, it is not closely related to any of the bacterial SOEs studied to date. SorT shares 59% sequence identity and 76% sequence similarity with the largely uncharacterized enzyme from Delftia acidovorans [7,8], while sequence identities with the other bacterial SOEs that have been studied in varying levels of detail range between 27 and 35% (sequence similarity values: 44–50%) (Table S1). In view of the low similarity of the SorT sequence to other bacterial SOEs and the obvious structural and kinetic differences discussed below we have decided to refer to this enzyme as SorT rather than naming it ‘SorA’ as has been done in some cases [7]. The designation ‘SorA’ was coined for the heterodimeric, hemecontaining SOE from St. novella which has only 32% sequence identity to SorT and also has a clearly differing structure. This classification is also supported by the position of the two enzymes in question within the cd_02110 SOE group: SorA from St. novella is a representative of subgroup cd_02114 ‘SorA-Moco_dimer’, while the S. meliloti SorT enzyme does not fall into any of the four existing subcategories of cd_02110 (Fig. 1). Unlike the well studied SOE from St. novella, SorT is a homodimeric enzyme containing only a molybdenum redox centre, a structure that is reminiscent of that of the plant sulfite oxidase (PSO), however, SorT is not a sulfite oxidase as it cannot transfer electrons to molecular oxygen. Heme groups, that are a typical feature of vertebrate sulfite oxidases and some bacterial sulfite deydrogenases [43,44], are absent from SorT. As for most bacterial SOEs [7,8,14,15], the natural electron acceptor for SorT is unknown at present and all these enzymes show much larger activities with the artificial electron acceptor ferricyanide than with mitochondrial cytochrome c which is preferred by the heme-containing bacterial SOEs [13,16] found in St. novella and Campylobacter jejuni. The high SorT activities observed with ferricyanide may reflect a need for an electron acceptor with a higher redox potential than that of the vertebrate cytochrome and/or a differing primary structure that enables more efficient interactions between SorT and the acceptor protein. Small bacterial electron transfer proteins with high redox potentials include copper proteins such as azurin [47], ferredoxins and certain cytochromes c [48], and we have shown here that genes encoding both a cytochrome c and an azurin are co-transcribed with the sorT gene. The need for a different electron acceptor may also be related to the absence of a heme domain/ subunit in SorT as these domains are thought to mediate electron transfer to cytochrome c in SorAB and the vertebrate SOs [12]. In addition to its unusual structural features, the catalytic mechanism of SorT also shows novel features that differ from those of all SOEs that have been studied in detail so far. SorT has high affinities for its substrate, sulfite, and fast turnover rates, however, while in vertebrate SOs and bacterial SorAB SOEs KMapp_sulfite is clearly pH dependent with very strong increases seen above pH 8 (Fig. S1) [43,44], in SorT KMapp_sulfite is nearly invariant with pH. In fact, in contrast to the other SOEs the highest substrate affinities were observed at pH values above 8. Similarly, kcat_app was very insensitive to pH changes, with only a small increase seen towards the high pH range (Table 3). In vertebrate SOs and the SorAB SOE kcat_app is clearly pH dependent and shows either one or two pKa values between pH 6 and 10. For vertebrate SOs and SorAB SDHs it has been suggested that the higher substrate affinity at low pH is indicative of a higher affinity of the enzyme for the protonated form of the substrate, HSO− 3 [43,49,50]. It is then possible that the differing behaviour of KMapp_sulfite seen in SorT could indicate that this enzyme 2− works equally well with either HSO− 3 or SO3 . There is some possibility that the use of an artificial electron acceptor such as ferricyanide, could have some influence on the pH dependence of kcat_app [43,44]. However, the altered behaviour of KMapp_sulfite is highly unlikely to be due to the use of ferricyanide as an electron acceptor in SorT steady-state assays: It has been shown for several other SOEs including those found in vertebrates that the strong increase in KMapp_sulfite above pH 8 is a property of the Mo centre alone and it has been seen in steady-state assays, and is also apparent when only the reductive half reaction of the enzyme is studied, i.e. when no electron acceptor is present [43,49,50]. The altered properties of the SorT catalytic parameters are an exciting finding and make SorT a prime candidate for elucidating how changes in the redox properties of the Mo centre and in the active site environment alter the kinetics of sulfite oxidation. Is there a link between the pathway for taurine oxidation and the presence of a particular type of SOE? Sulfite-oxidizing enzymes appear to be associated with the metabolism of taurine in many bacteria [4], and we have therefore investigated whether SorT represents an SOEtype that is generally associated with taurine degradation pathways (represented by the TauY and Xsc proteins from S. meliloti). In both cases sequences originating from α- and β-Proteobacteria form distinct but related clusters (Fig. S2), however, while there is a good J.J. Wilson, U. Kappler / Biochimica et Biophysica Acta 1787 (2009) 1516–1525 correlation between the presence of TauY and the presence of Xsc, Xsc-like sequences exist in many organisms that do not appear to contain a TauY-like taurine dehydrogenase. For both TauY and Xsc from S. meliloti the most closely related sequences were from Ochrobacterium, Roseobacter and Rhodobacter species. However, there was no obvious correlation between the occurrence of TauY/Xsc-like sequences and SorT-related sequences. SorT groups with sequences from both α- and β-Proteobacteria such as Sulfitobacter, Acidovorax, Comamonas and Delftia acidovorans. SOErelated sequences from Roseobacter and Rhodobacter species that contain sequences closely related to TauY and Xsc from S. meliloti, fall into a different clade that also contains sequences from Nitrobacter (Fig. 6). The majority (N90%) of the SOE-encoding genes in this group, including the SOEs from D. acidovorans, C. necator or T. thermophilus [7,15] are found upstream of genes encoding soluble c-type cyto- 1523 chromes with between 63 and 90% amino acid similarity to the cytochrome encoded by the SMc04048 gene (Fig. 2), i.e. these enzymes have an operon structure that on the surface resembles that of the heme-containing, heterodimeric SorAB SOE from St. novella [9,12,13]. However, despite these similarities, neither SorT nor any of the other characterized SOEs contain a heme group as an integral part of the enzyme. This proves that it is not possible at present to predict the structure of bacterial SOEs based on an inspection of the genetic environment of SOE-encoding genes alone. As there is no direct link between taurine metabolism and the occurrence of a particular type of SOEs, the question as to what the main function of bacterial SOEs in cellular metabolism is remains to be answered. As a result of our analyses it seems possible that contrary to earlier assumptions [4,7,51], SOEs are not linked to particular metabolic pathways but may be induced in response to exposure of the bacteria to sulfite. This is supported by the expression pattern of sorT that differed noticeably Fig. 6. Phylogenetic relationship of S. meliloti SorT to other bacterial sulfite-oxidizing enzymes. The tree was generated using the neighbor-joining algorithm integrated into MEGA4.0 [41], bootstrap values (500 replicates) above 50% are shown. SorT and other bacterial SOEs that have been described in the literature are shown in bold. 1524 J.J. Wilson, U. Kappler / Biochimica et Biophysica Acta 1787 (2009) 1516–1525 from those of xsc and tauY which encode key enzymes of taurine degradation. Clearly further work is needed to uncover the factors governing the formation of bacterial SOEs and the functional differences inherent in their structure in order to understand how these enzyme are integrated into cellular metabolism. [20] [21] Acknowledgements This work was supported by the Australian Research Council by an Australian Research Fellowship (ARF) and an ARC discovery project (DP0878525) to UK and by the University of Queensland. We would like to thank Prof. G. King and Dr. S. Rowland for making their MALLS analysis facility available for use in this project and for assisting with the analysis of the resulting data. We thank Dr. C.J. Jones for critically reviewing this manuscript. [22] [23] Appendix A. Supplementary data [24] Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbabio.2009.07.005. [25] References [26] [1] M.A. Kertesz, P. Mirleau, The role of soil microbes in plant sulphur nutrition, J. Exp. Bot. 55 (2004) 1939–1945. [2] M.A. Kertesz, Riding the sulfur cycle — metabolism of sulfonates and sulfate esters in Gram-negative bacteria, FEMS Microbiol. Rev. 24 (1999) 135–175. [3] A.M. Cook, K. Denger, T.H.M. Smits, Dissimilation of C3-sulfonates Arch, Microbiol. 185 (2006) 83–90. [4] A.M. Cook, K. Denger, S.S. Oja, P. Saransaari, Metabolism of taurine in microorganisms — a primer in molecular biodiversity? Advances in Experimental Medicine and Biology, Springer, Berlin, 2006, pp. 3–13. [5] C. Bruggeman, K. Denger, A.M. Cook, J. Ruff, Enzymes and genes of taurine and isethionate dissimilation in Paracoccus denitrificans, Microbiology 150 (2004) 805–816. [6] J. Ruff, K. Denger, A.M. Cook, Sulphoacetaldehyde acetyltransferase yields acetyl phosphate: purification from Alcaligenes defragrans and gene clusters in taurine degradation, Biochem. J. 369 (2003) 275–285. [7] K. Denger, S. Weinitschke, T.H.M. Smits, D. Schleheck, A.M. Cook, Bacterial sulfite dehydrogenases in organotrophic metabolism: separation and identification in Cupriavidus necator H16 and in Delftia acidovorans SPH-1, Microbiology 154 (2008) 256–263. [8] W. Reichenbecher, D.P. Kelly, J.C. Murrell, Desulfonation of propanesulfonic acid by Comamonas acidovorans strain P53: evidence for an alkanesulfonate sulfonatase and an atypical sulfite dehydrogenase, Arch. Microbiol. 172 (1999) 387–392. [9] U. Kappler, C.G. Friedrich, C. Dahl, Bacterial sulfite dehydrogenases — enzymes for chemolithtrophs only?, Microbial Sulfur Metabolism, Springer, Berlin, 2008, pp. 151–169. [10] U. Kappler, C. Dahl, Enzymology and molecular biology of prokaryotic sulfite oxidation, FEMS Microbiol. Lett. 203 (2001) 1–9. [11] A. Marchler-Bauer, J.B. Anderson, P.F. Cherukuri, C. DeWeese-Scott, L.Y. Geer, M. Gwadz, S. He, D.I. Hurwitz, J.D. Jackson, Z. Ke, C. Lanczycki, C.A. Liebert, C. Liu, F. Lu, G.H. Marchler, M. Mullokandov, B.A. Shoemaker, V. Simonyan, J.S. Song, P.A. Thiessen, R.A. Yamashita, J.J. Yin, D. Zhang, S.H. Bryant, CDD: a Conserved Domain Database for protein classification, Nucleic Acids Res. 33 (2005) D192–196. [12] U. Kappler, S. Bailey, Molecular basis of intramolecular electron transfer in sulfiteoxidizing enzymes is revealed by high resolution structure of a heterodimeric complex of the catalytic molybdopterin subunit and a c-type cytochrome subunit, J. Biol. Chem. 280 (2005) 24999–25007. [13] U. Kappler, B. Bennett, J. Rethmeier, G. Schwarz, R. Deutzmann, A.G. McEwan, C. Dahl, Sulfite: cytochrome c oxidoreductase from Thiobacillus novellus — purification, characterization and molecular biology of a heterodimeric member of the sulfite oxidase family, J. Biol. Chem. 275 (2000) 13202–13212. [14] G. D'Errico, A. Di Salle, F. La Cara, M. Rossi, R. Cannio, Identification and characterization of a novel bacterial sulfite oxidase with no heme binding domain from Deinococcus radiodurans, J. Bact. 188 (2006) 694–701. [15] A. Di Salle, G. D'Errico, F. La Cara, R. Cannio, M. Rossi, A novel thermostable sulfite oxidase from Thermus thermophilus: characterization of the enzyme, gene cloning and expression in Escherichia coli, Extremophiles 10 (2006) 587–598. [16] J.D. Myers, D.J. Kelly, A sulphite respiration system in the chemoheterotrophic human pathogen Campylobacter jejuni, Microbiology 151 (2005) 233–242. [17] F. Bardischewsky, A. Quentmeier, D. Rother, P. Hellwig, S. Kostka, C.G. Friedrich, Sulfur dehydrogenase of Paracoccus pantotrophus: the heme-2 domain of the molybdoprotein cytochrome c complex is dispensable for catalytic activity, Biochemistry 44 (2005) 7024–7034. [18] A. Quentmeier, R. Kraft, S. Kostka, R. Klockenkamper, C.G. Friedrich, Characterization of a new type of sulfite dehydrogenase from Paracoccus pantotrophus GB17, Arch. Microbiol. 173 (2000) 117–125. [19] S.J. Brokx, R.A. Rothery, G.J. Zhang, D.P. Ng, J.H. Weiner, Characterization of an [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] Escherichia coli sulfite oxidase homologue reveals the role of a conserved active site cysteine in assembly and function, Biochemistry 44 (2005) 10339–10348. L. Loschi, S.J. Brokx, T.L. Hills, G. Zhang, M.G. Bertero, A.L. Lovering, J.H. Weiner, N.C.J. Strynadka, Structural and biochemical identification of a novel bacterial oxidoreductase, J. Biol. Chem. 279 (2004) 50391–50400. F. Galibert, T.M. Finan, S.R. Long, A. Puhler, P. Abola, F. Ampe, F. Barloy-Hubler, M.J. Barnett, A. Becker, P. Boistard, G. Bothe, M. Boutry, L. Bowser, J. Buhrmester, E. Cadieu, D. Capela, P. Chain, A. Cowie, R.W. Davis, S. Dreano, N.A. Federspiel, R.F. Fisher, S. Gloux, T. Godrie, A. Goffeau, B. Golding, J. Gouzy, M. Gurjal, I. HernandezLucas, A. Hong, L. Huizar, R.W. Hyman, T. Jones, D. Kahn, M.L. Kahn, S. Kalman, D.H. Keating, E. Kiss, C. Komp, V. Lalaure, D. Masuy, C. Palm, M.C. Peck, T.M. Pohl, D. Portetelle, B. Purnelle, U. Ramsperger, R. Surzycki, P. Thebault, M. Vandenbol, F.J. Vorholter, S. Weidner, D.H. Wells, K. Wong, K.C. Yeh, J. Batut, The composite genome of the legume symbiont Sinorhizobium meliloti, Science 293 (2001) 668–672. J.E. Beringer, R factor transfer in Rhizobium leguminosarum, J. Gen. Microbiol. 84 (1974) 188–198. U. Kappler, S. Bailey, C.J. Feng, M.J. Honeychurch, G.R. Hanson, P.V. Bernhardt, G. Tollin, J.H. Enemark, Kinetic and structural evidence for the importance of Tyr236 for the integrity of the Mo active site in a bacterial sulfite dehydrogenase, Biochemistry 45 (2006) 9696–9705. J.P. Markwell, J. Lascelles, Membrane-bound, pyridine nucleotide-independent L-lactate dehydrogenase of Rhodopseudomonas sphaeroides, J. Bact. 133 (1978) 593–600. R. Hansch, C. Lang, E. Riebeseel, R. Lindigkeit, A. Gessler, H. Rennenberg, R.R. Mendel, Plant sulfite oxidase as novel producer of hydrogen peroxide — combination of enzyme catalysis with a subsequent non-enzymatic reaction step, J. Biol. Chem. 281 (2006) 6884–6888. F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, K. Struhl, in: K. Janssen (Ed.), Current Protocols in Molecular Biology, vol. 1, John Wiley and Sons Inc., Hoboken, NJ, 2005. U.K. Laemmli, Cleavage of structural protein during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. U. Kappler, K.F. Aguey-Zinsou, G.R. Hanson, P.V. Bernhardt, A.G. McEwan, Cytochrome c551 from Starkeya novella: characterization, spectroscopic properties, and phylogeny of a diheme protein of the SoxAX family, J. Biol. Chem. 279 (2004) 6252–6260. S.L. Rowland, W.F. Burkholder, K.A. Cunningham, M.W. Maciejewski, A.D. Grossman, G.F. King, Structure and mechanism of action of Sda, an inhibitor of the histidine kinases that regulate initiation of sporulation in Bacillus subtilis, Mol. Cell 13 (2004) 689–701. D.J. Lane, E. Stackebrandt, M. Goodfellow, 16S–23S rRNA sequencing, Nucleic Acid Techniques in Bacterial Systematics, John Wiley and Sons, Chichester, 1991, pp. 115–175. U. Kappler, L.I. Sly, A.G. McEwan, Respiratory gene clusters of Metallosphaera sedula — differential expression and transcriptional organization, Microbiology 151 (2005) 35–43. S.F. Altschul, T.L. Madden, A.A. SchÑffer, J. Zhang, Z. Zhang, W. Miller, D.J. Lipman, Gapped BLAST and PSI-BLAST: a new generation of protein database search programs, Nucleic Acids Res. 25 (1997) 3389–3402. E. Gasteiger, C. Hoogland, A. Gattiker, S. Duvaud, M.R. Wilkins, R.D. Appel, A. Bairoch, J.M. Walker, Protein identification and analysis tools on the ExPASy Server, The Proteomics Protocols Handbook, Humana Press, 2005, pp. 571–607. A. Krogh, B. Larsson, G. VonHeijne, E.L.L. Sonnhammer, Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes, J. Mol. Biol. 305 (2001) 567–580. S. Moller, M.D.R. Croning, R. Apweiler, Evaluation of methods for the prediction of membrane spanning regions, Bioinformatics 17 (2001) 633–646. M. Cserzo, E. Wallin, I. Simon, G. VonHeijne, A. Elofsson, Prediction of transmembrane alpha-helices in prokaryotic membrane proteins: the Dense Alignment Surface method, Prot. Eng. 10 (1997) 673–676. K. Hofmann, W. Stoffel, TMbase — a database of membrane spanning proteins segments, Biol. Chem. 374 (1993) 166. J.D. Bendtsen, H. Nielsen, D. Widdick, T. Palmer, S. Brunak, Prediction of twinarginine signal peptides, BMC Bioinformatics 6 (2005) 167–176. O. Emanuelsson, S. Brunak, G. von Heijne, H. Nielsen, Locating proteins in the cell using TargetP, SignalP and related tools, Nat. Protocols 2 (2007) 953–971. H. Nielsen, S. Brunak, G. VonHeijne, Machine learning approaches to the prediction of signal peptides and other protein sorting signals, Prot. Eng. 12 (1999) 3–9. K. Tamura, J. Dudley, M. Nei, S. Kumar, MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24 (2007) 1596–1599. F. Toghrol, W.M. Southerland, Purification of Thiobacillus novellus sulfite oxidase. Evidence for the presence of heme and molybdenum, J. Biol. Chem. 258 (1983) 6762–6766. S. Bailey, T. Rapson, K. Winters-Johnson, A.V. Astashkin, J.H. Enemark, U. Kappler, Molecular basis for enzymatic sulfite oxidation — how three conserved active site residues shape enzyme activity, J. Biol. Chem. 284 (2009) 2053–2063. M.S. Brody, R. Hille, The kinetic behavior of chicken liver sulfite oxidase, Biochemistry 38 (1999) 6668–6677. C.J. Feng, R.V. Kedia, J.T. Hazzard, J.K. Hurley, G. Tollin, J.H. Enemark, Effect of solution viscosity on intramolecular electron transfer in sulfite oxidase, Biochemistry 41 (2002) 5816–5821. H. Niknahad, P.J. O'Brien, Mechanism of sulfite cytotoxicity in isolated rat hepatocytes, Chem. Biol. Interact. 174 (2008) 147–154. M. Kukimoto, M. Nishiyama, T. Ohnuki, S. Turley, E.T. Adman, S. Horinouchi, T. Beppu, Identification of interaction site of pseudoazurin with its redox partner, J.J. Wilson, U. Kappler / Biochimica et Biophysica Acta 1787 (2009) 1516–1525 copper-containing nitrite reductase from Alcaligenes faecalis S-6, Prot. Eng. 8 (1995) 153–158. [48] C.A. McDevitt, P. Hugenholtz, G.R. Hanson, A.G. McEwan, Molecular analysis of dimethylsulfide dehydrogenase from Rhodovulum sulfidophilum; its place in the DMSO reductase family of microbial molybdopterin-containing enzymes, Mol. Microbiol. 44 (2002) 1576–1587. [49] T.D. Rapson, U. Kappler, P.V. Bernhardt, Direct catalytic electrochemistry of sulfite 1525 dehydrogenase: mechanistic insights and contrasts with related Mo enzymes, Biochim. Biophys. Acta-Bioenerg. 1777 (2008) 1319–1325. [50] H.L. Wilson, K.V. Rajagopalan, The role of tyrosine 343 in substrate binding and catalysis by human sulfite oxidase, J. Biol. Chem. 279 (2004) 15105–15113. [51] U. Kappler, C.G. Friedrich, H.G. Truper, C. Dahl, Evidence for two pathways of thiosulfate oxidation in Starkeya novella (formerly Thiobacillus novellus), Arch. Microbiol. 175 (2001) 102–111.