SoxRS-mediated regulation of chemotrophic sulfur oxidation in Paracoccus pantotrophus

Dagmar Rother, Grazyna Orawski, Frank Bardischewsky and Cornelius G. Friedrich

Lehrstuhl für Technische Mikrobiologie, Fachbereich Bio- und Chemieingenieurwesen, Universität Dortmund, Emil-Figge-Strasse 66, D-44221 Dortmund, Germany

Correspondence
Cornelius G. Friedrich
cornelius.friedrich{at}udo.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Paracoccus pantotrophus GB17 requires thiosulfate for induction of the sulfur-oxidizing (Sox) enzyme system. The soxRS genes are divergently oriented to the soxVWXYZA–H genes. soxR predicts a transcriptional regulator of the ArsR family and soxS a periplasmic thioredoxin. The homogenote mutant GB{Omega}S carrying a disruption of soxS by the {Omega}-kanamycin-resistance-encoding interposon expressed a low thiosulfate-oxidizing activity under heterotrophic and mixotrophic growth conditions. This activity was repressed by complementation with soxR, suggesting that SoxR acts as a repressor and SoxS is essential for full expression. Sequence analysis uncovered operator characteristics in the intergenic regions soxS–soxV and soxW–soxX. In each region a transcription start site was identified by primer extension analysis. Both regions were cloned into the vector pRI1 and transferred to P. pantotrophus. Strains harbouring pRI1 with soxS–soxV or soxW–soxX expressed the sox genes under heterotrophic conditions at a low rate, indicating repressor titration. Sequence analysis of SoxR suggested a helix–turn–helix (HTH) motif at position 87–108 and uncovered an invariant Cys-80 and a cysteine residue at the C-terminus. SoxR was overproduced in Escherichia coli with an N-terminal His6-tag and purified to near homogeneity. Electrophoretic gel mobility shift assays with SoxR retarded the soxS–soxV region as a single band while the soxW–soxX region revealed at least two protein–DNA complexes. These data demonstrated binding of SoxR to the relevant DNA. This is believed to be the first report of regulation of chemotrophic sulfur oxidation at the molecular level.


Abbreviations: HTH, helix–turn–helix; Lac, lactose; Sox, sulfur oxidation; TBE, Tris/borate/EDTA


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
The oxidation of hydrogen sulfide or sulfur to sulfuric acid represents the oxidative half of the global sulfur cycle and is mediated primarily by specialized prokaryotes. The ability to oxidize reduced inorganic sulfur compounds (via the sulfur oxidation enzyme system, Sox) is found in aerobic chemotrophic bacteria and anaerobic phototrophic bacteria. Paracoccus pantotrophus is an aerobic, Gram-negative, neutrophilic, facultatively autotrophic bacterium which grows with thiosulfate or molecular hydrogen as energy source and heterotrophically with a large variety of carbon sources (Ludwig et al., 1993; Rainey et al., 1999). This species, a member of the {alpha}-Proteobacteria, was isolated as Thiosphaera pantotropha (Robertson & Kuenen, 1983) and is the closest relative of Paracoccus denitrificans (Rainey et al., 1999). The genus Paracoccus has been of particular interest with respect to its highly flexible energy metabolism, the alternative anaerobic respiratory growth and the electron-transport chain used for aerobic growth, which has long been used as model for mitochondrial electron transport. It shows versatility with respect not only to physiology but also to regulation and promoter structures (reviewed by Baker et al., 1998; Steinrücke & Ludwig, 1993).

P. pantotrophus grows with thiosulfate exclusively under aerobic conditions and not anaerobically with nitrate as electron acceptor (Friedrich, 1998). The formation of proteins required for chemotrophic growth with thiosulfate is induced by thiosulfate. Thiosulfate-induced cells are able to oxidize thiosulfate and hydrogen sulfide at a high rate, while sulfur is slowly oxidized and sulfite is not metabolized by whole cells (Chandra & Friedrich, 1986; Friedrich & Mitrenga, 1981).

The gene region of P. pantotrophus encoding sulfur-oxidizing ability comprises 15 genes, of which seven, soxXYZABCD, encode the periplasmic proteins SoxYZ, SoxB, SoxCD and SoxXA, which catalyse hydrogen sulfide-, sulfur-, thiosulfate- and also sulfite-dependent cytochrome c reduction in vitro (Rother et al., 2001). The first gene of the sox gene region, soxR, previously designated ORF1, predicts a transcriptional regulator of the ArsR family, and soxS (formerly ORF2) a periplasmic thioredoxin (Friedrich et al., 2001). Both genes are oriented divergently to the other genes of the sox cluster (Fig. 1). The Sox proteins required for sulfur oxidation are expressed in the presence of thiosulfate in chemotrophic bacteria (Friedrich et al., 2001; Kelly et al., 1997) as are the respective proteins in phototrophic purple bacteria, purple non-sulfur and green bacteria (reviewed by Friedrich, 1998). Apart from marginal physiological studies no information is available on the regulation of the proteins involved in sulfur oxidation at the molecular biological level, or on the mechanisms that lead to the formation of the Sox proteins in chemotrophic or phototrophic bacteria.



View larger version (52K):
[in this window]
[in a new window]
 
Fig. 1. Diagram of the sox gene cluster of P. pantotrophus showing the intergenic regions soxS–soxV and soxW–soxX in detail. The nucleotides of the soxS–soxV and soxW–soxX regions are specified according to GenBank accession number X79242. Palindromic sequences are indicated by boxes and starts or stops of genes by arrows. Transcription starts identified by primer extension experiments are marked +1. Brackets demarcate the intergenic regions which were used for the experiments.

 
In P. pantotrophus the thioredoxin-encoding soxW gene (formerly shxW) is expressed in the presence of thiosulfate (Bardischewsky & Friedrich, 2001), as are the structural genes soxXYZABCD and subsequent genes soxEFGH of unknown function (Friedrich et al., 2000; Rother et al., 2001). In the mutant strain P. pantotrophus GB{Omega}V the soxV gene (formerly shxV) is disrupted by the {Omega}-kanamycin interposon. In the presence of thiosulfate, strain GB{Omega}V does not express soxW due to the polarity of the interposon but fully expresses the subsequent genes soxX–H (Bardischewsky & Friedrich, 2001). This result suggested a regulatory region upstream of soxX.

In this study we report the construction of the homogenote mutant GB{Omega}S, which expresses the sox genes under heterotrophic growth conditions, and the identification of two sox promoter regions by primer extension analysis. We have expressed soxR of P. pantotrophus in Escherichia coli, isolated the His-tagged SoxR, demonstrated SoxR binding to the two promoter regions of the sox gene cluster and identified SoxR as specific repressor protein.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Bacterial strains and plasmids.
Strains and plasmids used and constructed in this study are listed in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains and plasmids

 
Media and growth conditions.
E. coli was cultivated in Luria–Bertani (LB) medium (Sambrook et al., 1989) at 37 °C. P. pantotrophus was cultivated mixotrophically in mineral medium, pH 7·2, with 20 mM thiosulfate plus 20 mM disodium succinate (Bardischewsky & Friedrich, 2001) unless otherwise stated, and heterotrophically in mineral medium pH 7·2 with 10 mM disodium succinate at 30 °C. Antibiotics were added where appropriate (for E. coli 100 µg ampicillin ml–1, 25 µg chloramphenicol ml–1 or 25 µg kanamycin ml–1, and for P. pantotrophus 5 µg chloramphenicol ml–1).

DNA and gene transfer techniques.
Standard DNA techniques were applied (Sambrook et al., 1989). Plasmid DNA was isolated according to Kieser (1984). Restriction enzymes and T4 DNA ligase were obtained from Promega, Roche or Fermentas (OliI) and used as recommended by the manufacturer. PCR reactions were performed with DyNAzyme EXT polymerase (Finnzymes) essentially as described elsewhere (McPherson et al., 1992). For DNA sequencing, plasmid DNA was prepared with the Invisorb Spin Plasmid Mini Kit (Invitec). DNA was sequenced as described previously (Rother et al., 2001). E. coli was transformed as described by Chung et al. (1989). E. coli S17-1 was used to mobilize plasmids into P. pantotrophus GB17 (Simon et al., 1983). P. pantotrophus and E. coli S17-1 were conjugated as described by Rother et al. (2001).

Construction of the soxS : : {Omega}-kanamycin insertion.
soxS was disrupted by inserting the {Omega}-kanamycin interposon of pHP45{Omega}-Km by gene replacement using the suicide plasmid pOG10A. From plasmid pEG9 a 9533 bp EcoRI fragment containing soxR–soxC' was isolated and inserted into pUC19 (Yanisch-Perron et al., 1985), resulting in the vector pOG5A. pOG5A was cut with XhoI and SacI and the generated 5673 bp soxR–soxZ' fragment was cloned into pBluescript SK, resulting in the vector pOG8. The {Omega}-kanamycin interposon was isolated from pHP45{Omega}-Km at the BamHI site, treated with Klenow polymerase and inserted into pOG8 at the OliI site of soxS, leading to the plasmid pOG9A. From pOG9A a Klenow-polymerase-treated 7·8 kb XhoI–SacI fragment was cloned into the EcoRV site of the mobilizable plasmid pSUP202, resulting in pOG10A. Selection of heterogenote transconjugants and homogenote mutants was done as described for strain GB{Omega}V (Bardischewsky & Friedrich, 2001).

Cloning of soxR.
For complementation of the mutant P. pantotrophus GB{Omega}S with SoxR the plasmid pOG7 was cut with SmaI and the generated 1942 bp fragment containing soxR was cloned into the shuttle vector pRI1, resulting in the plasmid pRIsoxR.

Cloning of the intergenic regions soxW–soxX and soxS–soxV.
The soxW–soxX region was amplified by PCR using the primers S30 (5'-AAAACCCGGGGCATGGCTATGAAAATG-3') and S31 (5'-TTTTGAATTCGCAATGGCCATGGCCAC-3'), with plasmid pJOEB9 as template. S30 is compatible to bp 2473 and S31 to bp 2700 according to the DNA sequence of the sox gene region, which is accessible at the EMBL database under accession number X79242 PDSOX. The 252 bp DNA fragment generated was restricted with EcoRI and SmaI and cloned into pUC19, yielding the vector pRD148.3. The soxS–soxV region was amplified by PCR using primer S32 (5'-AAAACCCGGGTGTTCGAACATCAAGAG-3') and S33 (5'-TTTTGAATTCAAAGGACAGCAGCCCCG-3'), with plasmid pJOEB9 as template. Primer S32 is compatible to bp 1037 and S33 to bp 1278. The resulting 268 bp fragment was restricted with EcoRI and SmaI and cloned into pUC19, yielding the vector pRD149.2. The DNA fragments were verified by sequence analyis.

From pRD148.3 the putative binding region soxW–soxX was isolated as a 260 bp BamHI–EcoRI fragment and cloned into the EcoRV site of the shuttle vector pRI1, resulting in plasmid pRD154.7. The intergenic region soxS–soxV was isolated as a 246 bp BamHI–EcoRI fragment from pRD149.2 and transferred to pRI1, resulting in plasmid pRD156.5.

Cloning of soxR-his.
A truncated soxR gene (soxR') was generated by PCR with the primers S36 (5'-AAAAGGATCCGCGATAAATTGGCCCAG-3'), matching bp 753 of the sox gene region, and S37 (5'-TTTTAAGCTTTCAGCAATCCGTCTCGA-3'), which is compatible to bp 320. Plasmid pEG9 was used as template and a 446 bp fragment was produced. The primers introduced restriction sites for BamHI and HindIII into the ends of the fragments. The PCR fragment was digested with BamHI and HindIII and the resulting 434 bp fragment was cloned between the BamHI and HindIII sites of pUC19, yielding plasmid pRD150.6. After verification of the inserted DNA by DNA sequence analysis the BamHI–HindIII soxR' fragment was transferrred to the expression vector pQE30 (Qiagen), which adds a His6-tag to the N-terminus of a protein. The resulting plasmid was pRD151.2. The truncated soxR' gene was completed with the hybridized oligonucleotides S34 (5'-GATCCGATGACGATGACAAAATGATCCCGGCCCCGTG-3') and S35 (5'-GATCCACGGGGCCGGGATCATTTTGTCATCGTCATCG-3'), which delivered the codons for the first 6 aa of SoxR and which contained the sequence for the enterokinase cleavage site. This protease cleavage site was inserted in order to maintain the option to remove the His6-tag from the purified protein if required. This short double-stranded DNA was compatible with BamHI restriction sites on the 5'-end as well as the 3'-end. The correct orientation of the inserted DNA was verified by sequence analysis; the final expression vector was named pRD152.4 (Table 1).

Isolation of RNA and primer extension.
Total RNA was isolated from P. pantotrophus cells after mixotrophic growth with succinate plus thiosulfate with the High Pure RNA isolation kit (Roche). 5'-Fluorescently labelled primers (MWG Biotech) were used for primer extension analyses. The transcription start upstream of soxV was determined with primers pEG_7.5 (PV1; bp 1199–1221, 5'-CCCCCGAGTGCTGCCGGGCTGCC-3') and pEG_7.2 (PV2; bp 1286–1308, 5'-CGGCACCATCGGCAGGATGCAGG-3') and that upstream of soxX with primers pEG_7.3 (PX1; bp 2675-2696, 5'-TGGCCATGGCCACTACCGACGC-3') and pEG_7.4 (PX2; 2731–2753, 5'-CGCCTTCGACATAGACCACTTCC-3'). One picomole of primer was annealed to 1 µg total RNA and extended for 1 h at 42 °C by using 200 U Moloney murine leukaemia virus reverse transcriptase according to the manufacturer's protocol.

Expression of soxR-his and purification of His6-SoxR.
Cells of E. coli M15(pREP4, pRD152.4) were grown at 37 °C in 1 l LB medium supplemented with 100 µg ampicillin ml–1 and 25 µg kanamycin ml–1. Plasmid pREP4 supplied additional lactose-repressor (Lac) to allow soxR-his expression exclusively upon induction. At an OD600 of 0·6, IPTG was added to a final concentration of 0·4 mM and the culture was incubated for an additional 4 h. Cells were harvested by centrifugation (Sorvall GS3 rotor, 6000 r.p.m., 10 min, 4 °C), washed twice with lysis buffer (50 mM Tris/HCl pH 7·5, 500 mM NaCl, 10 mM MgCl2, 2 mM CaCl2, 1 mM imidazole, 5 mM 2-mercaptoethanol, 1 µM PMSF, 5 %, w/v, glycerol), resuspended in 25 ml of this buffer and passed four times through a French press at 150 MPa. Cell debris was removed by centrifugation (Sorvall SS34 rotor, 20 000 r.p.m., 30 min, 4 °C), and the supernatant was referred to as crude extract. This extract was subjected to Ni-NTA affinity chromatography. A column filled with 4 ml Ni-NTA agarose (Qiagen) was equilibrated with lysis buffer. An aliquot of crude extract (5 ml) was applied to the column and rinsed with 20 ml lysis buffer, followed by two washing steps, each with 16 ml wash buffer (50 mM Tris/HCl pH 7·5, 500 mM NaCl, 10 mM MgCl2, 2 mM CaCl2, 20 mM imidazole, 5 mM 2-mercaptoethanol, 1 µM PMSF, 5 %, w/v, glycerol). SoxR was eluted with elution buffer (50 mM Tris/HCl pH 7·5, 500 mM NaCl, 10 mM MgCl2, 2 mM CaCl2, 250 mM imidazole, 5 mM 2-mercaptoethanol, 1 µM PMSF, 5 %, w/v, glycerol). Fractions (2 ml) containing His-tagged protein were pooled and concentrated with a Vivaspin concentrator (Vivascience). To remove NaCl and imidazole during concentration the protein was washed with buffer A (50 mM Tris/HCl pH 7·5, 10 mM MgCl2, 2 mM CaCl2, 1 µM PMSF, 5 %, w/v, glycerol). This extract was applied to a 6 ml Resource Q-column equilibrated against buffer A. Protein was eluted with a linear gradient of 0–1 M NaCl. His6-SoxR-containing fractions were pooled (40 ml), washed with SoxR buffer (100 mM Tris/HCl pH 7·5, 1 mM EDTA, 10 mM MgCl2, 2 mM CaCl2, 5 %, w/v, glycerol, 2 mM 2-mercaptoethanol) and concentrated to 1 ml.

DNA labelling.
For gel mobility shift assays the binding regions were isolated from the plasmids pRD148.3 and pRD149.2 after restriction with EcoRI and PstI. From pRD148.3 a 260 bp fragment with the binding region soxW–soxX and from pRD149.2 a 274 bp fragment with the binding region soxS–soxV were isolated by gel elution. These DNAs were labelled using the Biotin 3' End DNA Labelling Kit (Perbio). Deoxynucleotidyl transferase (TdT) catalyses the incorporation of biotin-N4-CTP onto the 3'-OH end of DNA (Roychoudhury & Wu, 1980; Roychoudhury et al., 1976). As TdT exhibits a substrate preference for single-stranded DNA, the isolated fragments were denatured at 95 °C for 5 min prior to the labelling reaction, which was performed according to the manufacturer's instructions. Subsequently the DNA single strands were renatured by hybridization.

Gel mobility shift assays.
These assays were performed with the LightShift chemiluminescent EMSA Kit (Perbio). Biotin-labelled DNA was incubated with varying amounts of protein for 20 min in gel shift assay buffer (10 mM Tris pH 7·5, 50 mM KCl, 1 mM DTT, 2·5 %, v/v, glycerol, 50 ng poly(dIdC) µl–1, 0·05 % NP-40) at room temperature. DNA and DNA–protein complexes were separated in a 8 % polyacrylamide gel. The gels were prerun at 100 V in Tris/borate/EDTA (TBE) buffer (Sambrook et al., 1989) for 30 min, and after samples (20 µl) had been loaded into the wells, gels were run at 100 V for 90 min. The gel was then transferred to Hybond-N+ nylon membrane (Amersham). Blotting was done by a semi-dry procedure using the Multiphor electrophoresis system (Pharmacia) in 0·5x TBE buffer at 380 mA and 10 °C for 45 min. Biotin-labelled DNA was detected according to the manufacturer's instructions. The membrane was exposed to Kodak BioMax light film for 2–5 min.

Analytical procedures.
Denatured proteins were separated by SDS-PAGE according to Laemmli (1970). Proteins were stained with Coomassie blue as described by Weber et al. (1972). Protein from cell-free extracts was determined by the method of Bradford (1976). A ligand blotting procedure (Towbin et al., 1979) was performed by the semi-dry procedure using the Multiphor electrophoretic system (Pharmacia) with nickel-nitrilotriacetate (Ni-NTA) conjugate (Qiagen). The transfer buffer was 39 mM glycine, 48 mM Tris, 0·0375% (w/v) SDS, 20% (v/v) methanol. The transfer took place at room temperature for 60 min. Ni-NTA conjugate consisting of Ni-NTA coupled to calf intestinal alkaline phosphatase was used to detect the His6-tag of SoxR.

Enzyme assays.
The thiosulfate oxidation rate of whole cells was determined with an oxygen electrode (Rank Brothers). The assay (3 ml) contained 50 µl cell suspension of OD436 30, equivalent to 150 µg protein, and 100 µmol sodium/potassium phosphate buffer (pH 8·0). Reactions were started with 30 µl 0·2 M sodium thiosulfate. One unit (U) of thiosulfate oxidation activity was defined as 1 µmol O2 consumed (mg protein)–1.

Sequence analyses.
Database searches were performed with BLASTN and BLASTP (http://www.ncbi.nlm.nih.gov/blast/) (Altschul et al., 1997), including a CD search (Marchler-Bauer et al, 2003). Pairwise alignments were carried out with BLAST 2 (http://www.ncbi.nlm.nih.gov/blast/bl2seq/bl2.html) (Tatusova & Madden, 1999) or with LALIGN (http://www.fasta.bioch.virginia.edu/fasta_www/lalign.htm) (Huang & Miller, 1991) and multiple alignments with CLUSTALW (http://www.ebi.ac.uk/clustalw/) (Thompson et al., 1994). HTH motifs were predicted with Network Protein Sequence Analysis (http://www.npsa-pbil.ibcp.fr/cgi-bin/npsa_automat.pl?page=/NPSA/npsa_hth.html) (Dodd & Egan, 1990). DNA sequences were analysed with the Clone Manager 6 program (Scientific & Educational Software).


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Inactivation of soxS
The transposon Tn5 acts in a polar manner on the expression of sox genes located downstream of the insertion, as demonstrated for soxV : : {Omega}-Km (Bardischewsky & Friedrich, 2001) and soxX : : {Omega}-Km (F. Bardischewsky, unpublished data). The homogenote mutant GB{Omega}S carried the soxS : : {Omega}-kanamycin interposon, which inactivated soxS and very likely acted in a polar manner on the expression of soxR. In contrast to the wild-type, this strain expressed thiosulfate-oxidizing activity under heterotrophic conditions, albeit at a low rate. However, also under mixotrophic growth with succinate plus thiosulfate a low thiosulfate-oxidizing activity of 18 % of the wild-type was expressed (Table 2). Complementation in trans with soxR restored the repression of the formation of thiosulfate-oxidizing activity under heterotrophic growth with succinate but did not restore the full expression when thiosulfate was included in the medium (Table 2). These data suggested SoxR as repressor protein for expression of the sox genes and SoxS as an essential periplasmic component for their optimal expression.


View this table:
[in this window]
[in a new window]
 
Table 2. Trans complementation of P. pantotrophus GB{Omega}S with pRIsoxR

 
Analysis of promoter/operator regions of the sox cluster
The mutant GB{Omega}V suggested a regulatory site upstream of soxV and of soxX (Bardischewsky & Friedrich, 2001). Sequence characteristics for operator regions were detected in the 69 bp intergenic region between soxS and soxV and in a second 114 bp intergenic region between soxW and soxX. Both regions, as well as parts of adjacent genes, contain characteristics for binding of dimeric regulatory proteins. For soxS–soxV the perfect inverted repeat CAAGCATCGGC and the perfect direct repeat TGGCGGGGC were detected. For soxW–soxX the inverted repeat TGAAAATG and the direct repeats CAGGGAG and TGGCCAT were identified. Complete palindomic sequences TGTTCGAACA, TGCCGCGGCA, GAATATTC and CGCCGGCG were present for the soxS–soxV region, and TGGCCATGGCCA was located downstream of soxX (Fig. 1). Such sequence features are typical recognition elements of DNA-binding proteins possessing an HTH motif. Similar repeats are observed in the respective regions of Rhodovulum sulfidophilum, Rhodopseudomonas palustris, Silicibacter pomeroyi and Bradyrhizobium japonicum. One of the sequences (TGTTCGAACA) is also present as a perfect palindrome in soxS of S. pomeroyi and Rv. sulfidophilum with an identity of 80 % (data not shown). The significance of these observations is presently unknown.

Further intergenic regions exist between soxA and soxB (119 bp) and between soxE and soxF (53 bp). No promoter/operator sites were considered for the soxA–soxB and soxE–soxF intergenic regions of P. pantotrophus since the homogenote mutant strain GB{Omega}X did not express any of the sox genes downstream of the soxX : : Tn5 insertion (F. Bardischewsky, unpublished data). For Rv. sulfidophilum, however, it was concluded that soxF can be transcribed from a promoter lying between soxC and soxF (Appia-Ayme et al., 2001).

Determination of sox transcription start sites
Two transcription starts were determined by primer extension for the sox gene cluster of P. pantotrophus GB17. The existence of a transcription start site upstream of soxX was evident from strain GB{Omega}V, which expressed the sox structural genes. To identify the site upstream of soxV the 5' ends of the sox mRNAs were determined by primer extension analysis. For the primer extension reactions two different primers were used for each transcription start site. For examination of the soxS–soxV intergenic region the primers used were complementary to a sequence upstream of the ATG codon of soxV and a region internal to soxV, and for soxW–soxX two different primers were used internal to soxX. Primer extension analysis using primers PV1 and PV2 determined the transcription start at thymine-1124 (Fig. 2a), which is 104 nt upstream of the translation start of soxV. Using the primers PX1 and PX2 a second transcription site was determined at thymine-2568 or thymine-2569 (Fig. 2b), which are 85 or 86 nt upstream of the translation start codon of soxX (Fig. 1). Thus, the nucleotide sequences suggesting operator sites for binding of regulatory proteins were located downstream of the RNA polymerase binding sites for both intergenic regions.



View larger version (51K):
[in this window]
[in a new window]
 
Fig. 2. Determination of the 5'-end of the P. pantotrophus soxVW and soxXYZABCDEFGH transcripts by primer extension. The same labelled oligonucleotides as those used for the primer extensions were used to generate the sequencing ladders in lanes A, C, G and T. The bases representing the starts of transcription of the genes soxVW (a) and soxX–soxH (b) are marked by asterisks.

 
Expression of soxR in E. coli
To study the role of SoxR in regulation of the sox genes by in vitro gel mobility shift assays, SoxR was required in significant amounts; therefore, soxR–his was expressed in E. coli. The complementary oligonucleotides S34 and S35 completed soxR' and introduced sequence variations at the third position of the third and the fifth codon without changing the amino acid sequence, and led to plasmid pRD152.4. In both cases the rarely used codon CCC was changed to CCG, most frequently used for proline in E. coli. This nucleotide exchange was performed since the efficiency of translation depends on the RNA sequence having optimal codons for the first five N-terminal amino acids downstream of the translational start codon as well as on the RNA sequence immediately upstream of the translation start (Degryse, 1990).

After transformation with plasmid pRD152.4, E. coli M15(pREP4, pRD152.4) cells were analysed for SoxR production by SDS-PAGE. A strong IPTG-inducible protein band of 20 kDa was observed, which was in agreement with the predicted size of 18·46 kDa of the 166 aa comprising His6-SoxR (Fig. 3a, lane 3). The protein was present in the soluble fraction of the cell-free extracts and absent from cell extracts from E. coli M15(pREP5, pQE30) not expressing soxR-his (Fig. 3a, lane 2). Ni-NTA-AP conjugate (Qiagen) identified the His-tag and demonstrated that the additional band represented the His6-tagged SoxR (Fig. 3a, lanes 7 and 8). The calculated pI of the His-tagged SoxR is 6·04, whereas for SoxR a pI of 5·71, a size of 149 aa and a molecular mass of 16·47 kDa were predicted.



View larger version (79K):
[in this window]
[in a new window]
 
Fig. 3. SDS-PAGE analysis of His6-SoxR expressed in E. coli M15. (a) Coomassie stain; (b) ligand blotting of His6-SoxR with Ni-NTA conjugate. Lanes: 1, molecular mass markers; 2, crude extract from E. coli M15(pREP4, pQE30), induced, not expressing His6-SoxR, 15 µg; 3, crude extract from E. coli M15(pREP4, pRD152.4), induced, 15 µg; 4, eluate from affinity chromatography with Ni-NTA-agarose, 3 µg; 5, eluate from anionic-exchange chromatography with a Resource Q column, 3 µg; 6, crude extract from E. coli M15(pREP4, pQE30), induced, 25 µg, not expressing His6-SoxR; 7, crude extract from E. coli M15(pREP4, pRD152.4), induced, 25 µg; 8, purified His6-SoxR, 3 µg.

 
SoxR binds to the intergenic regions soxS–soxV and soxW–soxX
Gel retardation assays were used to examine binding of SoxR to the intergenic regions soxS–soxV and soxW–soxX. The soxS–soxV DNA fragment formed a protein–DNA complex with His6-SoxR-containing cell extracts from E. coli M15(pREP4, pRD152·4), increasing in intensity with increasing protein concentration (Fig. 4a, lanes 4–6). The soxW–soxX DNA fragment formed three protein–DNA complexes when various amounts of cell extract were used (Fig. 4b, lanes 4–6), indicating multiple binding sites on the fragment used in the shift experiment.



View larger version (43K):
[in this window]
[in a new window]
 
Fig. 4. Gel retardation analysis of the soxS–soxV– and the soxW–soxX–His6-SoxR complex. The amount of labelled DNA was 0·5 ng for each lane. (a, c) soxS–soxV DNA, (b) soxW–soxX DNA. For binding studies of SoxR, crude extracts from E. coli M15(pREP4, pRD152.4) (SoxR+) and E. coli M15(pREP4, pQE30) (SoxR) (a, b) and purified His6-SoxR (c) were used. (a, b) Lanes: 1, DNA only; 2, DNA plus 100 ng extract (SoxR+); 3, DNA plus 200 ng extract (SoxR+); 4, DNA plus 500 ng extract (SoxR+); 5, DNA plus 1000 ng extract (SoxR+); 6, DNA plus 2000 ng extract (SoxR+); 7, DNA plus 1000 ng extract (SoxR); 8, DNA plus 2000 ng extract (SoxR); 9, DNA plus 1000 ng extract (SoxR+) and 20 ng of unlabelled competitor DNA; 10, DNA plus 2000 ng extract (SoxR+) and 20 ng unlabelled competitor DNA. (c) Lanes: 1, DNA only; 2, DNA plus 200 ng crude extract (SoxR+); 3, DNA plus 500 ng crude extract (SoxR+); 4–10, DNA plus decreasing amounts of purified His6-SoxR (4, 500 ng; 5, 200 ng; 6, 100 ng; 7, 50 ng, 8, 10 ng; 9, 5 ng; 10, 1 ng). Due to the ageing of the protein extracts the signal intensity decreases with time.

 
The specificity of the binding was evident from gels using different concentrations of cell extract of E. coli M15(pREP4, pQE30), which is specifically devoid of SoxR. No protein–DNA complexes were formed with these extracts lacking His6-SoxR (Fig. 4a, lanes 7 and 8; Fig. 4b, lanes 7 and 8). The specificity of the binding was further examined by addition of 40-fold excess of unlabelled fragments to the assays, which abolished the retardation of the labelled DNA (Fig. 4a, lanes 9 and 10; Fig. 4b, lanes 9 and 10). These findings demonstrated that His6-SoxR bound specifically to the intergenic regions soxS–soxV and soxW–soxX. Further evidence for the specificity of His6-SoxR regarding the DNA-binding regions was achieved by using purified His6-SoxR in the gel mobility shift assay. When 0·5 ng of DNA carrying the intergenic region soxS–soxV was used with decreasing amounts of protein the typical protein–DNA complex was seen. With crude extract, 200–500 ng of the protein was required to observe a band-shift whereas with purified SoxR protein a faint band appeared with 10 ng and a more intense band with 50 ng of His6-SoxR, demonstrating that binding of the purified SoxR protein was specific, although a slight loss of activity occurred during the purification process (Fig. 4c). As expected from the binding assays using cell-free extract of E. coli, the purified His6-SoxR also bound to the soxW–soxX promoter region (data not shown).

When thiosulfate was added to the in vitro binding assays at concentrations of 10 µM to 10 mM no notable release of His6-SoxR was observed from the binding regions soxS–soxV and soxW–soxX (data not shown). This result indicated that thiosulfate per se was not involved in derepression of the sox genes in P. pantotrophus and suggested a so far unknown sulfur intermediate as the active principle in regulation of chemotrophic sulfur oxidation of P. pantotrophus.

SoxR, a transcriptional repressor
The function of SoxR as a repressor of sox gene expression would require an interaction of the protein with operator sequences. To confirm the findings from in vitro binding studies, thiosulfate-dependent oxygen consumption rates were examined from P. pantotrophus GB17 containing plasmid pRI1, pRD156.5 or pRD154.7 and compared to the wild-type cultivated under mixotrophic and heterotrophic growth conditions. Under mixotrophic conditions similar thiosulfate oxidation rates of 2·29–2·83 U were obtained for all four strains: the wild-type P. pantotrophus GB17 (2·85 U), P. pantotrophus GB17(pRI1) harbouring the vector without any insert (2·29 U), P. pantotrophus GB17(pRD156.5) carrying the intergenic region soxS–soxV (2·63 U), and P. pantotrophus GB17(pRD154.7) carrying the intergenic region soxW–soxX (2·75 U). As the inducer thiosulfate was not supplied under heterotrophic growth conditions absolutely no thiosulfate oxidation rates were measured for the wild-type and the control strain GB17(pRI1). In strain GB17(pRD156.5) and GB17(pRD154.7) carrying soxS–soxV and soxW–soxX low but distinct thiosulfate oxidation rates of 0·10 U and 0·19 U, respectively, were determined. These findings suggested that SoxR was titrated from the genomic binding regions, leading to a constitutive albeit low expression of the sox genes.

Sequence analysis of SoxR and the sox regulatory regions
The gel shift assays identified SoxR as a protein binding to the intergenic regions soxS–soxV and soxW–soxX, and the constitutive synthesis of Sox proteins upon introduction of the regions suggested the function of SoxR as repressor protein of the sox gene cluster of P. pantotrophus. The DNA-binding property of SoxR was in accordance with the HTH motif at position 87–108 of the deduced primary structure (Network Protein Sequence Analysis), as HTH motifs are diagnostic for DNA-binding proteins (Fig. 5). An invariant Cys-80 is located 7 aa in front of the HTH motif present exclusively in SoxR homologues (Fig. 6) but not in other ArsR transcriptional regulators (data not shown). Cys-149 is the C-terminus of SoxR of P. pantotrophus, and cysteine residues are generally present at the C-termini of SoxR and its ArsR homologues from various sources (data not shown). Streptomyces lividans 1326 harbours an inducible mercury resistance for which MerR, belonging to the ArsR family, is the transcriptional regulator. The C-terminal cysteine of MerR binds the inducer mercury(II) ions (Rother, 1998). Since thiosulfate did not affect binding of SoxR to the relevant sox intergenic regions thiosulfate-dependent formation of a protein disulfide in SoxR may be involved in the regulation of sox expression.



View larger version (65K):
[in this window]
[in a new window]
 
Fig. 5. Alignment of SoxR of P. pantotrophus and other bacteria. The box indicates the HTH motif. The invariant cysteine in front of the HTH motif and the cysteine in the C-terminal region are highlighted in bold. Pp, P. pantotrophus; Bj, B. japonicum; Me, M. extorquens; Ps, P. salicylatoxidans; Rp, R. palustris; Rs, R. sulfidophilum; Rsp, R. sphaeroides; Sp, S. pomeroyi.

 
Database comparisons of the primary structure of SoxR showed high amino acid sequence similarities to other putative proteins of other chemo- and phototrophic bacteria. soxR of Rv. sulfidophilum predicting SoxR of 140 aa (accession no. AAO11780 is 70 % identical in a 74 aa overlap. A 118 aa SoxR homologue (IGwit database RMQ02638 of Methylobacterium extorquens shows an identity of 54·3 %. A SoxR homologue of 116 aa of Bradyrhizobium japonicum (accession no. BAC48771 (Kaneko et al., 2002) and the 124 aa homologue of Rps. palustris (accession no. CAE29915 (Larimer et al., 2004) have identities of 52 %. Pseudaminobacter salicylatoxidans (accession no. AJ404005.4) has a SoxR (121 aa) with an identity of 48 % of the entire primary structure to that of P. pantotrophus. The same degree of identity is given for SoxR (128 aa) of Silicibacter pomeroyi (accession no. YP 166241) (Moran et al., 2004). SoxR of Rhodobacter sphaeroides (127 aa) (accession no. NZ_AAAE0100156.1) (Matsuzaki et al., 2003) has a 61 % identity in a 91 aa overlap (data not shown). These chemo- and phototrophic bacteria all contain essential sox genes homologous to those of P. pantotrophus, and the SoxR homologues may represent repressor proteins of sox expression in these strains.

The essentially invariant HTH motif of SoxR homologues led us to examine the identity of the nucleotide sequences of the two binding regions. With LALIGN a 59·4 % identity in a 64 nt overlap was observed between the soxS–V and soxW–X regions of P. pantotrophus. Moreover, a 64·2 % identity in a 81 nt overlap with the soxW–X region of the phototroph Rv. sulfidophilum was found. From the highly conserved HTH motif in SoxR (Fig. 5) similarly conserved DNA sequences were expected. However, although similar structural features like repeats are present in the intergenic regions of these bacteria there is no sequence similarity in any of them.


   ACKNOWLEDGEMENTS
 
We thank Josefina Ringk and Bettina Höller for expert technical assistance. This study was supported by grant Fr 318/9-1 of the Deutsche Forschungsgemeinschaft.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Altenbuchner, J., Viell, P. & Pelletier, I. (1992). Positive selection vector based on palindromic sequences. Methods Enzymol 216, 457–566.[Medline]

Altschul, S. F., Madden, T. L., Schäffer, A. A., Zhang, J., Zhang, Z., Miller, W. & Lipman, D. J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25, 3389–3402.[Abstract/Free Full Text]

Appia-Ayme, C., Little, P. J., Matsumoto, Y., Leech, A. P. & Berks, B. C. (2001). Cytochrome complex essential for photosynthetic oxidation of both thiosulfate and sulfide in Rhodovulum sulfidophilum. J Bacteriol 183, 6107–6118.[Abstract/Free Full Text]

Baker, S. C., Ferguson, S. J., Ludwig, B., Page, M. D., Richter, O.-M. H. & van Spanning, R. J. M. (1998). Molecular genetics of the genus Paracoccus: metabolically versatile bacteria with bioenergetic flexibility. Microbiol Mol Biol Rev 62, 1046–1078.[Abstract/Free Full Text]

Bardischewsky, F. & Friedrich, C. G. (2001). The shxVW locus is essential for oxidation of inorganic sulfur and molecular hydrogen by Paracoccus pantotrophus GB17: a novel function in lithotrophy. FEMS Microbiol Lett 202, 215–220.[CrossRef][Medline]

Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248–254.[CrossRef][Medline]

Chandra, T. S. & Friedrich, C. G. (1986). Tn5-induced mutations affecting sulfur-oxidizing ability (Sox) of Thiosphaera pantotropha. J Bacteriol 166, 446–452.[Medline]

Chung, C. T., Niemela, S. L. & Miller, R. H. (1989). One-step preparation of competent Escherichia coli: transformation and storage of bacterial cells in the same solution. Proc Natl Acad Sci U S A 86, 2172–2175.[Abstract/Free Full Text]

Degryse, E. (1990). Influence of the second and third codon of the expression of recombinant hirudin in E. coli. FEBS Lett 269, 244–246.[CrossRef][Medline]

Dodd, I. B. & Egan, J. B. (1990). Improved detection of helix-turn-helix DNA-binding motifs in protein sequences. Nucleic Acids Res 18, 5019–5026.[Abstract]

Friedrich, C. G. (1998). Physiology and genetics of sulfur-oxidizing bacteria. Adv Microb Physiol 39, 235–289.[Medline]

Friedrich, C. G. & Mitrenga, G. (1981). Oxidation of thiosulfate by Paracoccus denitrificans and other hydrogen bacteria. FEMS Microbiol Lett 10, 209–212.[CrossRef]

Friedrich, C. G., Quentmeier, A., Bardischewsky, F., Rother, D., Kraft, R., Kostka, S. & Prinz, H. (2000). Novel genes coding for lithotrophic sulfur oxidation of Paracoccus pantotrophus GB17. J Bacteriol 182, 4677–4687.[Abstract/Free Full Text]

Friedrich, C. G., Rother, D., Bardischewsky, F., Quentmeier, A. & Fischer, J. (2001). Oxidation of reduced inorganic sulfur compounds by bacteria: emergence of a common mechanism? Appl Environ Microbiol 67, 2873–2882.[Free Full Text]

Huang, X. & Miller, W. (1991). A time-efficient, linear-space local similarity algorithm. Adv Appl Math 12, 337–357.[CrossRef]

Kaneko, T., Nakamura, Y., Sato, S. & 14 other authors (2002). Complete genomic sequence of nitrogen-fixing symbiotic bacterium Bradyrhizobium japonicum USDA110. DNA Res 9, 189–197.[Medline]

Kelly, D. P., Shergill, J. K., Lu, W.-P. & Wood, A. P. (1997). Oxidative metabolism of inorganic sulfur compounds by bacteria. Antonie van Leeuwenhoek 71, 95–107.[CrossRef][Medline]

Kieser, T. (1984). Factors affecting the isolation of ccc DNA from Streptomyces lividans and Escherichia coli. Plasmid 12, 19–36.[CrossRef][Medline]

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685.[Medline]

Larimer, F. W., Chain, P., Hauser, L. & 16 other authors (2004). Complete genome sequence of the metabolically versatile photosynthetic bacterium Rhodopseudomonas palustris. Nat Biotechnol 22, 55–61.[CrossRef][Medline]

Ludwig, W., Mittenhuber, G. & Friedrich, C. G. (1993). Transfer of Thiosphaera pantotropha to Paracoccus denitrificans. Int J Syst Bacteriol 43, 363–367.[Abstract]

Marchler-Bauer, A., Anderson, J. B., DeWeese-Scott, C. & 24 other authors (2003). CDD: a curated Entrez database of conserved domain alignments. Nucleic Acids Res 31, 383–387.[Abstract/Free Full Text]

Matsuzaki, M., Abe, M., Hara, S., Iwasaki, Y., Yamamoto, I. & Satoh, T. (2003). An abundant periplasmic protein of the denitrifying phototroph Rhodobacter sphaeroides f. sp. denitrificans is PstS, a component of an ABC phosphate transport system. Plant Cell Physiol 44, 212–216.[Abstract/Free Full Text]

McPherson, M. J., Quirke, P. & Taylor, G. R. (1992). PCR – a Practical Approach. New York: Oxford University Press.

Mittenhuber, G., Sonomoto, K., Egert, M. & Friedrich, C. G. (1991). Identification of the DNA region responsible for sulfur-oxidizing ability of Thiosphaera pantotropha. J Bacteriol 173, 7340–7344.[Medline]

Moran, M. A., Buchan, A., Gonzalez, J. M. & 32 other authors (2004). Genome sequence of Silicibacter pomeroyi reveals adaptations to the marine environment. Nature 432, 910–913.[CrossRef][Medline]

Pfitzner, U., Odenwald, A., Ostermann, T., Weingard, L., Ludwig, B. & Richter, O. M. (1998). Cytochrome c oxidase (heme aa3) from Paracoccus denitrificans: analysis of mutations in putative proton channels of subunit I. Bioenerg Biomembr 30, 89–97.[CrossRef][Medline]

Rainey, F. A., Kelly, D. P., Stackebrandt, E., Burghardt, J., Hiraishi, A., Katayama, Y. & Wood, A. P. (1999). A re-evaluation of the taxonomy of Paracoccus denitrificans and a proposal for the combination Paracoccus pantotrophus comb. nov. Int J Syst Bacteriol 49, 645–651.[Abstract]

Rémy, F., Frey, J. & Krisch, H. (1987). Interposon mutagenesis of soil and water bacteria: a family of DNA fragments designed for in vitro insertional mutagenesis of Gram-negative bacteria. Gene 52, 147–154.[CrossRef][Medline]

Robertson, L. A. & Kuenen, J. G. (1983). Thiosphaera pantotropha gen. nov. sp. nov., a facultatively anaerobic, facultative autotrophic sulphur bacterium. J Gen Microbiol 129, 2847–2855.

Rother, D. (1998). MerR der Regulator der Quecksilberresistenz-Gene aus Streptomyces lividans 1326 – Expression, Reinigung und Charakterisierung. PhD thesis, Universität Stuttgart: Stuttgart: Verlag Ulrich E. Grauer.

Rother, D., Henrich, H.-J., Quentmeier, A., Bardischewsky, F. & Friedrich, C. G. (2001). Novel genes of the sox gene cluster, mutagenesis of the flavoprotein SoxF, and evidence for a general sulfur oxidizing system in Paracoccus pantotrophus GB17. J Bacteriol 183, 4499–4508.[Abstract/Free Full Text]

Roychoudhury, R. & Wu, R. (1980). Terminal transferase-catalyzed addition of nucleotides to the 3' termini of DNA. Methods Enzymol 65, 43–62.[Medline]

Roychoudhury, R., Jay, E. & Wu, R. (1976). Terminal labeling and addition of homopolymer tracts to duplex DNA fragments by terminal deoxynucleotidyl transferase. Nucleic Acids Res 3, 863–877.[Abstract]

Sambrook, J., Maniatis, T. & Fritsch, E. F. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Simon, R., Priefer, U. & Pühler, A. (1983). A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria. Bio/Technology 1, 784–790.[CrossRef]

Steinrücke, P. & Ludwig, B. (1993). Genetics of Paracoccus denitrificans. FEMS Microbiol Rev 104, 83–118.[CrossRef]

Tatusova, T. A. & Madden, T. L. (1999). BLAST 2 sequences, a new tool for comparing protein and nucleotide sequences. FEMS Microbiol Lett 174, 247–250.[CrossRef][Medline]

Thompson, J. D., Higgins, D. G. & Gibson, T. J. (1994). CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22, 4673–4680.[Abstract]

Towbin, H., Staehelin, T. & Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci U S A 76, 6762–6766.

Weber, K., Pringle, J. R. & Osborn, M. (1972). Measurement of molecular weights by electrophoresis on SDS polyacrylamide gel. Methods Enzymol 26, 3–27.[CrossRef][Medline]

Wodara, C., Kostka, S., Egert, M., Kelly, D. P. & Friedrich, C. G. (1994). Identification and sequence analysis of the soxB gene essential for sulfur oxidation of Paracoccus denitrificans GB17. J Bacteriol 176, 6188–6191.[Abstract]

Yanisch-Perron, C., Vieira, J. & Messing, J. (1985). Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103–119.[CrossRef][Medline]

Received 22 October 2004; revised 22 December 2004; accepted 24 January 2005.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Rother, D.
Articles by Friedrich, C. G.
Articles citing this Article
PubMed
PubMed Citation
Articles by Rother, D.
Articles by Friedrich, C. G.
Agricola
Articles by Rother, D.
Articles by Friedrich, C. G.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2005 Society for General Microbiology.