The novel extracellular Streptomyces reticuli haem-binding protein HbpS influences the production of the catalase-peroxidase CpeB

Darío Ortiz de Orué Lucana, Tanja Schaa and Hildgund Schrempf

Universität Osnabrück, FB Biologie/Chemie, Barbarastraße 11, D-49069 Osnabrück, Germany

Correspondence
Darío Ortiz de Orué Lucana
ortiz{at}biologie.uni-osnabrueck.de
Hildgund Schrempf
schrempf{at}biologie.uni-osnabrueck.de


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The Gram-positive soil bacterium and cellulose degrader Streptomyces reticuli synthesizes the mycelium-associated enzyme CpeB, which displays haem-dependent catalase and peroxidase activity, as well as haem-independent manganese-peroxidase activity. Downstream of the cpeB gene, a so far unknown gene was identified. The new gene and its mutated derivatives were cloned in Escherichia coli as well as in Streptomyces lividans and a gene-disruption mutant within the chromosome of the original S. reticuli host was constructed, comparative physiological, biochemical and immunological studies then allowed the deduction of the following characteristics of the novel gene product. (i) The protein was found extracellularly; the substitution of twin arginines within the signal peptide abolished its secretion. (ii) The highly purified protein interacted specifically with haem and hence was designated HbpS (haem-binding protein of Streptomyces). (iii) HbpS contained three histidine residues surrounded by hydrophobic amino acids; one of them was located within the motif LX3THLX10AA, which is related to the motif within the yeast cytochrome c peroxidase LX2THLX10AA whose histidine residue interacts with haem. (iv) The addition of haemin (Fe3+ oxidized form of haem) to the Streptomyces cultures led to enhanced levels of HbpS which correlated with increased haemin-resistance. (v) The presence of HbpS increased synthesis of the highly active catalase-peroxidase CpeB containing haem. In this process HbpS could act as a chaperone that binds haem and then delivers it to the mycelium-associated CpeB; HbpS could also interact with membrane-associated proteins involved in a signal transduction cascade regulating the expression of cpeB. (vi) HbpS shared varying degrees of amino acid identities with bacterial proteins of so far unknown function. This report contributes to the elucidation of the biological function of these proteins.


The GenBank/EMBL/DDBJ accession number for the sequence of hbpS from Streptomyces reticuli reported in this paper is Y14336.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Haem is synthesized in a multistep biosynthetic pathway with well-defined intermediates. The haems are tetrapyrroles. The tetrapyrrole precursor, {delta}-aminolaevulinic acid (ALA), is formed either from glutamate by the C5 pathway or from glycine and succinyl coenzyme A via ALA synthase. The subsequent conversion of ALA to the cyclic intermediate uroporphyrinogen III (UroIII) requires three successive enzymic steps which are catalysed by ALA dehydratase, porphobilinogen deaminase and uroporphyrinogen III synthase. UroIII is the final common precursor for all tetrapyrroles and thus presents the major branch point in haem biosynthesis. The later steps of haem biosynthesis require uroporphyrinogen III decarboxylase, coproporphyrinogen III oxidase, followed by ferrochelatase, which catalyses the protoporphyrin chelation of ferrous iron to form protohaem. This is the prosthetic group of numerous haem proteins, and can be modified further to other types of haem (Panek & O'Brian, 2002; Thony-Meyer, 1997). In some Gram-positive bacteria investigated (Bacillus subtilis, Clostridium josui, Mycobacterium tuberculosis and Staphylococcus aureus), the genes for enzymes catalysing the biosynthesis of UroIII are organized in operons (hemAXCDBL) (Johansson & Hederstedt, 1999). The compact organization of hem genes in Gram-positive bacteria is probably important for concerted regulation and may also be of evolutionary significance. Several Gram-negative bacteria have specific receptors recognizing free haem or haem-containing proteins. Free haem or haem from carrier proteins is actively transported through the outer membrane to the periplasm via an energy-dependent mechanism (Braun & Braun, 2002).

Haem is the prosthetic group of numerous proteins involved in a wide variety of biological processes, including oxygen carriers, redox enzymes and regulatory proteins. Haem-containing enzymes are abundant in many micro-organisms and include cytochrome c oxidase, catalases, different types of peroxidases and catalase-peroxidases (Woloszczuk et al., 1980; Zou et al., 1999). During early stages of growth, the Gram-positive bacterium Streptomyces reticuli was found to produce a mycelia-associated, haem-containing enzyme (CpeB), which exhibits a catalase-peroxidase activity with broad substrate specificity and manganese-peroxidase activity (Zou & Schrempf, 2000). The cpeB gene and the regulator gene furS form an operon which is transcribed under the control of a promoter(s) located upstream of the furS gene. Thus FurS also acts as autoregulator in a redox-dependent fashion (Ortiz de Orué Lucana & Schrempf, 2000). The thiol form of FurS contains one zinc ion per monomer and binds in this state to its cognate operator upstream of the furS gene. Oxidation of -SH groups within FurS induces Zn2+ release (Ortiz de Orué Lucana et al., 2003).

CpeB can use H2O2 to oxidize a number of substrates in dependence on an attached haem group (ferric-protophorpyrin) or in a haem-independent reaction which is coupled to Mn(II)/(III) peroxidation (Zou & Schrempf, 2000). The additional haem-dependent catalase activity of the enzyme leads to a disproportionation of H2O2 to O2. Thus the mycelia-associated enzyme also plays an important part in detoxifying H2O2 and in minimizing reactions caused by highly reactive oxygen species arising from the interaction of H2O2 with certain divalent metal ions (Fe2+ and others). Therefore S. reticuli is well equipped to minimize the Fenton reaction, the reaction products of which are hazardous to every organism (Fridovich, 1986).

In this report we describe the identification of a gene whose product has been characterized as a novel haem-binding protein from Streptomyces (HbpS). Using biochemical studies, we showed that the secretion of HbpS depends on the presence of twin arginine residues within its signal peptide. Chromosomal mutants carrying a disruption of the hbpS gene were designed and analysed to gain insights as to the physiological role of HbpS.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and plasmids.
Streptomyces reticuli Tü45 (H. Zähner, Tübingen, Germany) and Streptomyces lividans 66 (D. A. Hopwood, John Innes Institute, Norwich, UK) were used. The plasmid pUC18 (Sambrook et al., 1989) was a gift from J. Messing (State University of New Jersey, Piscataway, USA). Escherichia coli strains DH5{alpha} and M15 (Villarejo et al., 1972) and plasmids pWHM3 (Vara et al., 1989; Table 1) and pQE32 (Qiagen; Table 1) were used for routine cloning purposes. The constructs pUKS10 (a pUC18 derivative; Table 1) and pWKS10 (a pWHM3 derivative; Table 1), both of which contain the furS–cpeB operon, have been described earlier (Zou et al., 1999; Ortiz de Orué Lucana & Schrempf, 2000).


View this table:
[in this window]
[in a new window]
 
Table 1. Plasmids

 
Media and culture conditions.
For cultivation of S. reticuli and S. lividans complete or minimal media (Schlochtermeier et al., 1992a) supplemented with the indicated carbon source were used. Depending on the purpose of the experiments, cultures were grown in baffled Erlenmeyer flasks containing 5–200 ml on a rotary shaker for 1–4 days. For cultivation of S. reticuli strains in solid media, R2 medium (without sucrose) (Hopwood et al., 1985) containing 0·25 g K2SO4 l–1, 10·12 g MgCl2.6H2O l–1, 10 g glucose l–1, 0·10 g Casamino acids l–1, 5 g yeast extract l–1, 3·03 g Tris-base l–1 (adjusted with HCl to pH 7·5) and 14 g agar l–1 was used. E. coli strains (DH5{alpha} or M15) were grown in LB medium at 37 °C (Sambrook et al., 1989).

Chemicals and enzymes.
Chemicals for SDS-gel electrophoresis were obtained from Serva. Molecular mass markers, nitrophenyl, o-dianisidine, 4-chloro-1-naphthol, haemin and 3,3',5,5'-tetramethylbenzidine were supplied by Sigma. Hydrogen peroxide (30 %, w/v) was bought from Merck.

Test for peroxidase activity.
CpeB was released from the mycelium using acetate buffer (pH 5·5) containing 0·1 % Triton X-100 (Zou et al., 1999), and aliquots (30 µl) were tested for activity. Samples were loaded onto a native 10 % polyacrylamide gel. After the run, the gel was washed twice with acetate buffer (20 mM, pH 5·5) and activity staining was carried out with 4-chloro-1-naphthol and 5 mM H2O2 (Conyers & Kidwell, 1991).

Cleavage of DNA, ligation and agarose gel electrophoresis.
DNA was cleaved with various restriction enzymes according to the suppliers' instructions. Ligation (Sambrook et al., 1989) was performed with T4 ligase (New England Biolabs). Gel electrophoresis was carried out in 0·8–2 % agarose gels using TBE buffer.

Transformation and isolation of plasmids.
E. coli was transformed with plasmid DNA by electroporation (Dower et al., 1988). Plasmids were isolated from E. coli with the aid of a mini plasmid kit (Qiagen).

Hybridization experiments.
The transfer of DNA fragments of the restricted genomic S. reticuli DNA onto nylon membranes was performed as described by Sambrook et al. (1989). The hybridization probes were labelled using Klenow enzyme and digoxigenin-11-dUTP (Roche). Hybridization and immunological detection were carried out according to standard procedures (Sambrook et al., 1989).

RNA isolation.
To obtain well-grown mycelia, S. reticuli (wild-type and hbpS mutant) spores [20 µl in 80 %, (v/v) glycerol] were inoculated in 10 ml complete medium (Schlochtermeier et al., 1992b) and grown as a standing culture at 30 °C for 16 h. The culture was diluted (1 : 10) in the same medium, and cultivation was continued on a rotary shaker at 30 °C for 12 h. For further scaling up, the culture was diluted to 1 l, and cultivation was continued (16 h). The cultures were washed twice in minimal medium (MM) without supplement, and kept shaking after resuspension in the same medium. The mycelia were suspended in 1 l MM, divided into 100 ml portions and supplemented with Avicel (1 % final concentration). Samples (100 ml) were taken for enzyme activity tests, immunoblotting and RNA isolation. Genomic RNA was isolated as previously described (Ortiz de Orué Lucana & Schrempf, 2000).

Analysis of transcripts.
Total RNA was electrophoretically separated on a 2 % agarose gel containing 2 % formamide and 1x MOPS buffer and was transferred over 3 h to a positively charged nylon membrane under vacuum using 20x SSC. RNA size marker I (0·3–6·9 kb; Roche) was used for size determination. The membrane was dried at 80 °C for 30 min and subsequently exposed to UV radiation for 3 min to fix the RNA. Hybridization was performed in a solution containing 5x SSC, 0·1 % SDS, 100 µg salmon sperm DNA ml–1 and 5x Denhardt's reagent (Sambrook et al., 1989) at 64 °C for 2 h. The 32P-labelled probe was added and hybridization was continued for another 20 h.

The cpeB probe was radioactively labelled with the Rediprime DNA labelling system (Amersham Biosciences) using Klenow polymerase and [32P]dCTP. Firstly, the DNA region of cpeB was amplified by PCR. The following primers were used: CD, 5'-GAGTTCCGCACAGTTCGGAAGG-3', and CE, 5'-CGTTGGCACCGCCGCGCTTGTCGC-3'.

The membrane was washed twice with 2x SSC containing 0·1 % SDS at room temperature for 20 min and subsequently with 0·1x SSC supplemented with 0·1 % SDS at 64 °C for 45 min, and subjected to autoradiography at –70 °C.

Cloning of the hbpS gene in E. coli.
The hbpS-encoding region of the previously described construct pWKS10 (Zou et al., 1999; Table 1), harbouring S. reticuli furS, cpeB and the newly identified hbpS genes, was amplified by PCR using the following primers: HBP1, 5'-CCTGAGCATGCCCAGCCGCAAGAAGCCGTCCC-3' consisting of an SphI restriction site, followed by the sequence encoding N-terminal amino acids of HbpS; and HBP2, 5'-GCCGGATCCTGGCCGAGCACGGCCGCGCC-3', determining the C-terminal amino acids of HbpS, followed by a BamHI restriction site. The PCR product was digested with SphI and BamHI. After digestion with BamHI the DNA was treated with the Klenow polymerase to get a blunt-ended fragment which was ligated with SphI/SmaI-digested pQE32 (Table 1). The resulting plasmid pQH1 (Table 1) was transformed into E. coli M15(pREP4). The plasmid was sequenced and the correctness of the designed hbpS gene and its in-frame fusion with the His-tag codons were confirmed.

Cloning of a hbpS fusion gene in S. lividans.
The downstream region of the hbpS gene was amplified using the oligonucleotides HBP3, 5'-GTCGGTGGCCGTCGTCGACCGCAACGGCAACACCCTGGTCACCCTG-3', annealing upstream of the BsiWI restriction site in the middle of hbpS, and HBP4, 5'-CTCGGCCCGGGCGGGTCAGTGGTGGTGGTGGTGGTGGCCGAGCACGG-3', containing an SrfI restriction site, a stop codon and six histidine codons. Following the PCRs, the generated product was cleaved with BsiWI and SrfI and ligated with the large BsiWI–SrfI fragment of the plasmid pUKS10 (Zou et al., 1999; Table 1). The 4·6 kb EcoRI–HindIII fragment of the resulting plasmid pUKS15 (Table 1) was ligated with EcoRI/HindIII-digested pWHM3 vector. The resulting plasmid pWKS15 (Table 1) carried a hbpS gene with six histidine codons in front of its stop codon. The correctness of the in-frame insertion was confirmed by restriction and sequencing. The plasmid pWKS15 (Table 1) was transformed into S. lividans protoplasts, which were generated as described by Hopwood et al. (1985). Transformants were selected using an overlay of 0·4 % agarose containing 200 µg thiostrepton ml–1 (Hopwood et al., 1985).

Site-directed mutagenesis of hbpS.
The codons for arginines (R9 and R10) from the identified twin arginines within the signal sequence of HbpS (Fig. 1b) were replaced by two lysine-encoding codons. Lysines and arginines are highly basic with a long side-chain and possess similar hydrophobicity coefficients; thus, it is less likely that the substitutions induce unfavourable steric interactions in the resulting protein. For this purpose the upstream region of hbpS was amplified using the following oligonucleotides: HBP5, 5'-CTTCCTCAGCATGTCCAGCCLGCAAGAAGCCGTCCAAGAAGACCCGCGTCCTCGT-3', containing a BbvCI restriction site and two lysine codons replacing the two arginine codons of the hbpS wild-type gene, and HBP4 (see above). For the PCR and further ligation the plasmid pUKS10 (Zou et al., 1999; Table 1) was used. The PCR product was digested with BbvCI and BsiWI, resulting in a 0·328 kb mutagenized hbpS fragment which was ligated with the large BbvCI–BsiWI fragment of plasmid pUKS10 (Table 1). The 4·6 kb EcoRI–HindIII fragment of the resulting plasmid pUKS16 (Table 1) was ligated with the EcoRI/HindIII-digested pWHM3. The resulting plasmid pWKS16 (Table 1) carried an hbpS gene containing mutations in the codons encoding arginines 9 and 10 within HbpS. The correctness of the in-frame replacement was checked by restriction and sequencing. The plasmid pWKS16 was transformed into S. lividans protoplasts as described above.



View larger version (59K):
[in this window]
[in a new window]
 
Fig. 1. Relative location of the hbpS gene and alignments. (a) Gene locations. In the wild-type, the hbpS gene, and one other ORF located within the SrfI–SphHI fragment of pUKS10 (Zou et al., 1999) are presented downstream of the furScpeB operon (GenBank/EMBL accession no. Y14336). In the hbpS mutant, the hbpS gene within the chromosomal DNA has been disrupted by the insertion of the hygromycin resistance cassette (hyg). (b) The HbpS protein from S. reticuli (HbpS_reti) was aligned to an ORF from S. coelicolor A3(2) (Orf_coel), Atu1117 from A. tumefaciens C58 (Atu1117_tume), CpmX from Sphingomonas aromaticivorans (CpmX_sphin) and NahX from P. putida G7 (NahX_pseud). The arginine residues at positions 9 and 10 as well as the three histidine residues in HbpS are in bold, white letters and are underlined. (c) The region surrounding the histidine residue at position 60 (indicated by *) of HbpS was aligned to the region containing the histidine residue involved in haem binding by the yeast cytochrome c peroxidase (CCP).

 
Purification of the fusion protein and generation of antibodies.
The complete hbpS-encoding region from the construct pUKS10 described previously (Zou et al., 1999) was amplified using PCR, and subsequently cloned into pQE32 (Table 1). The resulting construct pQH1 (Table 1) was transformed into E. coli M15(pREP4). The corresponding transformant containing the pQH1 plasmid was inoculated in LB medium with ampicillin (100 µg ml–1) and kanamycin (50 µg ml–1) and during its exponential growth phase (OD600 0·6–0·8) was induced with 1 mM IPTG for 3 h. The cells were washed with chilled solution A (10 mM HEPES, 60 mM KCl, pH 8·0) and disrupted in the same solution by ultrasonication (Branson sonifier, 5x10 s, with 10 s intervals). After centrifugation (16 000 g), the debris and the inclusion bodies (constituting about 50 % of the insoluble fusion protein) were separated from the cytoplasm. The inclusion bodies were denatured with solution A containing 6 M urea and supplemented with 30 mM imidazole. After centrifugation (16 000 g) the supernatant containing the fusion protein was mixed with Ni2+-NTA (Qiagen) on a wheel at 4 °C for 2 h. The agarose column containing the bound proteins was washed with solution A supplemented with 30 mM imidazole (using about 10-fold of the column volume), and eluted with the same solution containing 200–500 mM imidazole. The whole procedure was repeated. To remove imidazole, the sample was dialysed against solution A and the protein was concomitantly concentrated (to about 0·45 mg ml–1) by using an Amicon device. The protein concentrations of the samples were determined by the methods of Lowry, and Bradford (1976). The isolated protein (150 µg) was used to generate antibodies in a guinea pig (Eurogentec). The antisera were stored in aliquots at –20 °C.

Localization of HbpS in Streptomyces.
S. reticuli (wild-type), S. reticuli hbpS disruption mutant, S. lividans pWKS10 (Zou et al., 1999), S. lividans pWKS17 (derivative of pWKS10 lacking hbpS, see below and Table 1) and S. lividans pWHM3 (control strain with vector) were grown in complete medium. The pWKS17 (Table 1) plasmid was generated by self-ligation of the longer SrfI–BbvCI (Fig. 1a) restriction fragment of pWKS10. Thus, pWKS17 lacks the hbpS gene. The cultures were transferred to a minimal medium containing 1 % Avicel (for S. reticuli) or 0·5 % yeast extract (for S. lividans) and cultivated on a rotary shaker at 30 °C. S. reticuli was incubated for 6 h, S. lividans for 14 h. Mycelia were centrifuged (4000 g). The extracellular proteins from the culture filtrate were precipitated by adding (NH4)2SO4 (90 % saturation). The precipitated proteins were tested for immuno-reactivity with anti-HbpS antibodies.

Purification of HbpS–His-tag from Streptomyces.
S. lividans transformants carrying the plasmid pWKS15 (Table 1) or the construct pWKS17 (derivative of pWKS10 lacking hbpS) were grown as described above. After precipitation [using (NH4)2SO4, 90 % saturation] of proteins from the culture filtrate, the protein solutions were dialysed against solution A (see above). Afterwards, the HbpS–His-tag protein was purified using Ni–NTA-affinity chromatography as described previously (Ortiz de Orué Lucana & Schrempf, 2000).

Detection of haem binding by HbpS.
Haem binding in HbpS was identified in SDS-polyacrylamide gels as described by Moore et al. (1978) and Smalley et al. (2001) using tetramethylbenzidine/H2O2, which detects the presence of haem in proteins. For the preparation of gels, loading buffer and electrophoresis buffer with 0·1 % SDS was used. The HbpS–His-tag-containing sample was free from DTT and was not boiled prior to electrophoresis, which was carried out at 4 °C in the dark. Then, the gel was incubated for 30 min in 20 mM Tris/HCl (pH 7·3) buffer containing 50 % methanol to fix the protein and to lower the concentration of SDS within the gel. This was consecutively stained for 45 min using the following solutions: 0·25 mM sodium acetate, pH 5·3; 0·25 % (w/v) 3,3',5,5'-tetramethylbenzidine, 25 % (v/v) methanol, 0·75 % H2O2 [2·5 % (v/v) of a commercial 30 % solution]. H2O2 was added immediately prior to use. After staining, the gel was incubated for 10 min in 25 % (v/v) methanol to remove excess of tetramethylbenzidine. The gel was then stored in 0·1 M Tris/HCl, pH 7·3.

Generation of disruption mutant.
The plasmid pUKS10 (Zou et al., 1999) was digested with EcoRI. The linearized DNA was then partially digested with SalI to obtain a 4·6 kb DNA fragment containing part of cpeB, the hbpS gene and its downstream region. After treatment with the Klenow enzyme, the DNA was circularized by self-ligation. The resulting construct was named pUCH1 (Table 1). Using a pBR322 derivative with a hygromycin-resistance gene containing terminator sequences (hyg) (Blondelet-Rouault et al., 1997), a HindIII fragment (2·3 kb) containing this hyg was generated. This fragment was blunt-ended using the Klenow enzyme and ligated with the BsiWI-digested pUCH1 which was also blunt-ended. The resulting plasmid pUCH2 (Table 1) contains hyg, which is flanked on the left by 0·87 kb and on the right by 1·15 kb of the above-described plasmid pUCH1. The ligation mixture was added to electrocompetent E. coli DH5{alpha}. Hygromycin-resistant E. coli transformants were selected, and the correctness of their plasmid constructs was analysed by restriction. One of the correct constructs was isolated. Ten micrograms thereof was denatured (0·2 M NaOH, 10 min, 37 °C), chilled on ice and neutralized by rapid addition of HCl. Then the DNA was used to transform 50 µl protoplasts (~109 ml–1) generated from S. reticuli, which were spread onto osmotically stabilized medium, as described by Hopwood et al. (1985) and incubated at 30 °C for 19 h. The plates were overlaid with 2 ml molten agarose (40 °C) containing hygromycin (1 mg ml–1). Hygromycin-resistant colonies were restreaked several times, and their genomic DNA was analysed as to the size of fragments carrying the hygromycin gene.

Growth assays.
The sensitivity of S. reticuli wild-type and S. reticuli hbpS mutant to haemin was determined using a disc inhibition assay. A sample of 100 µl spores (5x108) was added to 3 ml soft agar (Sambrook et al., 1989) poured onto the respective R2 plates and allowed to solidify. Sterile 6 mm-diameter blank papers discs (Schleicher & Schuell) were added to the bacteria-overlaid plates and saturated with 20 µl of different concentrations (100, 200, 300 and 400 µM) of haemin. Plates were incubated overnight at 30 °C before zones of inhibition were measured.

SDS-PAGE and Western blotting.
SDS-PAGE was performed in the presence of 0·1 % SDS (Laemmli, 1970). Proteins were separated by 12·5 % SDS-PAGE and transferred to a fluorotrans membrane (Sambrook et al., 1989). The membrane was blocked for 1 h at room temperature with PBS (40 mM Na2HPO4, 8 mM NaH2PO4, 150 mM NaCl, pH 7·4) containing 5 % skimmed milk powder, and subsequently incubated overnight at 4 °C with the generated anti-CpeB or anti-HbpS antibodies. After treatment with secondary anti-rabbit or anti-guinea pig antibodies, respectively, conjugated with alkaline phosphatase, the membrane was stained with 5-bromo-4-chloro-3-indolyl-phosphate and nitro blue tetrazolium (Blake et al., 1984).

PCR, DNA sequencing and computer analysis.
PCR was performed using Pfu DNA polymerase (Invitrogen). To test the correctness of cloned genes, sequencing was done using the Ready Reaction mix and ABI PRISM equipment (PE Biosystems) by the departmental sequence service (U. Coja, FB Biologie, University of Osnabrück). Sequence entry, primary analysis and ORF searches were performed using Clone Manager 5.0. Database searches using the PAM120 scoring matrix were carried out with BLAST algorithms (BLASTX, BLASTP and TBLASTN) on the NCBI file server (BLAST{at}ncbi.nlm.nih.gov) (Altschul et al., 1997). Multiple sequence alignments were generated by means of the CLUSTAL W (1.74) program (Higgins et al., 1992). Putative Shine–Dalgarno (ribosome-binding) sites (Gold et al., 1981; Strohl, 1992) and signal peptide cleavage sites were predicted as described by Nielsen et al. (1997).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Analysis of the hbpS gene and the deduced protein
S. reticuli produces the mycelia-associated, haem-containing catalase-peroxidase CpeB; the corresponding cpeB gene is regulated by the redox regulator FurS encoded by furS, which is located upstream of cpeB (Zou et al., 1999). The genomic DNA from S. reticuli downstream of cpeB had been cloned in plasmid pUBBI previously (Zou et al., 1999), and was subsequently sequenced (see Fig. 1a and GenBank/EMBL accession no. Y14336). The cpeB gene was found to be followed by an intergenic region and by an ORF comprising 566 bp. The deduced protein has 188 aa with a predicted molecular mass of 18·5 kDa and theoretical pI of 9·6. Fig. 1(a) (wild-type) displays the physical map of the furS–cpeB operon and the newly identified ORF, which in the course of additional studies was subsequently named hbpS (for nomenclature, see results of functional tests and Discussion). The hbpS gene is separated by 117 bp from the cpeB gene, which is located in the opposite orientation. The deduced HbpS protein contains an N-terminally located signal peptide which includes twin arginines (at positions 9 and 10; Fig. 1b).

Comparative analysis revealed that the amino acid sequence of HbpS shows the highest identity (69·9 %) to an ORF from Streptomyces coelicolor A3(2). HbpS also shares a limited number of identical amino acids with a few other deduced proteins with so far uncharacterized features, including Atu1117 from Agrobacterium tumefaciens C58 (38·3 %) (Wood et al., 2001), CpmX from Sphingomonas aromaticivorans (36·6 %) (Yrjala et al., 1997) and NahX from Pseudomonas putida G7 (33 %) (Grimm & Harwood, 1999) (Fig. 1b). The genes cpmX and nahX are each located in an operon, which is responsible for degradation of chlorinated or methylated aromatic compounds via the meta-cleavage pathway. The N-terminal region of Atu1117 shows a predictable signal peptide cleavage site, however without twin arginines.

The following investigations were focused on analysing the function of the protein encoded by the hbpS gene.

Secretion of HbpS requires a twin arginine in its signal peptide
A S. lividans transformant was generated previously which has the plasmid pWKS10 (Table 1); this containing the furS–cpeB operon and hbpS (Zou et al., 1999). As outlined in Methods, the hbpS fusion gene with six histidine codons was cloned and overexpressed in E. coli. The subsequently purified His-tag–HbpS protein molecules consist predominantly of the monomeric form, including its signal peptide (25 kDa), and smaller quantities of the dimeric form (50 kDa) (Fig. 2). After generation of antibodies, immunological studies revealed that the culture filtrate of S. lividans pWKS10 contained the mature monomeric (apparent molecular mass 16·5 kDa) and dimeric (apparent molecular mass 33 kDa) forms of HbpS (Fig. 3b, lanes 3 and 6). The control strains S. lividans pWKS17 (without hbpS) and S. lividans pWHM3 (vector alone) lacked any extracellular protein immuno-reacting with anti-HbpS antibodies (Fig. 3b, lanes 4 and 5). N-terminal sequencing of HbpS (secreted by S. lividans pWKS15) showed that the cleavage site in the signal peptide could be located between the residues Asp37 and Thr38 and not between the alanine residues (positions 35 and 36) which would have been predicted for a signal peptidase cleavage site (Nielsen et al., 1997). Thus, it has to be assumed that the N-terminus of the mature HbpS was further modified via action of an extracellular peptidase. The predicted mature secreted protein has a molecular mass of 15·1 kDa and a pI of 5·9.



View larger version (119K):
[in this window]
[in a new window]
 
Fig. 2. Purification of the HbpS fusion protein from E. coli. E. coli M15(pREP4) containing plasmid pQH1 (lane 2) or the control plasmid pQE32 (lane 1) were grown and induced as described in Methods. Total proteins (lanes 1 and 2) were separated by SDS-PAGE. The desired HbpS fusion protein was purified using affinity chromatography (lane 4). No protein could be eluted from the control strain (lane 3). Protein markers are in lane S.

 


View larger version (36K):
[in this window]
[in a new window]
 
Fig. 3. Detection of HbpS in Streptomyces strains. (a) S. lividans containing pWKS15 (lane 2) or the control plasmid pWKS17 (lane 1) was grown as described in Methods. The HbpS fusion protein was purified from the precipitated proteins using affinity chromatography (lane 4). No protein could be eluted from the control extracts (lane 3). Protein markers are in lane S. (b) Proteins were precipitated from the culture filtrate of S. reticuli wild-type (lane 1), S. reticuli hbpS disruption mutant (lane 2), S. lividans pWKS10 (lanes 3 and 6), S. lividans pWKS17 lacking hbpS (lane 4), S. lividans pWHM3 (lane 5) and S. lividans pWKS16 containing mutated hbpS (lane 7) or from a cytoplasmic extract of S. lividans pWKS10 (lane 8) and S. lividans pWKS16 containing mutated hbpS (lane 9). After their separation by SDS-PAGE proteins were analysed for the presence of HbpS (Western blot) using anti-HbpS antibodies. Monomeric (m), dimeric (d) and higher oligomeric (o) forms are indicated.

 
To determine the role of the twin arginine signature in HbpS, the codons for the twin arginines (R9 and R10) in the hbpS gene (located within the construct pWKS10) were replaced by two for lysine (see Methods). The resulting plasmid pWKS16 (Table 1) was transformed into S. lividans. The transformant S. lividans pWKS16 lacked any extracellular HbpS protein (Fig. 3b, lane 7), but it contained several immuno-reacting (using anti-HbpS antibodies) proteins intracellularly (Fig. 3b, lane 9), one dominant at 33 kDa and others at 41 kDa and 60 kDa. This result indicates that the secretion of HbpS depends on the presence of the RR-motif within the signal peptide.

To facilitate purification of larger amounts of HbpS for further analysis the hbpS gene was extended by six histidine codons (corresponding to the codon usage of Streptomyces). This was cloned in such a way that it replaced the original hbpS gene in pWKS10 and resulted in the plasmid pWKS15 (Table 1). The HbpS fusion protein with the His-tag predicted at the C-terminus could be concentrated by affinity chromatography from the culture filtrate of S. lividans pWKS15 (Fig. 3a, lane 4), but not from the control strain S. lividans pWKS17 (lacking the hbpS gene) (Fig. 3a, lane 3). The isolated protein has an apparent molecular mass of 17·2 kDa corresponding to the secreted form of the His-tag protein. Small quantities of its dimeric form (34 kDa) were also observed (Fig. 3a). Proteins corresponding to the non-mature form (20 kDa) and to the mature form (17·2 kDa) were purified from the cytoplasm of S. lividans pWKS15 (Fig. 4a, lane 3).



View larger version (26K):
[in this window]
[in a new window]
 
Fig. 4. Identification of HbpS as a haem-binding protein and influence of haemin on HbpS production. (a) The HbpS fusion protein purified from the extracellular (lane 1 and 2) or the cytoplasmic fraction (lane 3 and 4) of S. lividans pWKS15 was loaded (without reducing agents and without boiling) on to SDS-polyacrylamide gels and analysed. The lanes were stained with Coomassie brilliant blue (lane 1), silver (lane 3) or treated with tetramethylbenzidine/H2O2 (lane 2 and 4). The non-mature form of the cytoplasmic protein is indicated by I and the corresponding mature form by II. (b) The absorption spectrum of a solution containing purified HbpS fusion protein (100 µg in 10 mM HEPES buffer, pH 8·0). (c) Proteins from the culture filtrate from S. lividans pWKS10 containing hbpS (lane 2) and S. lividans pWKS17 lacking hbpS (lane 1) were analysed as described under (a) and treated with Coomassie brilliant blue (left), anti-HbpS antibodies (middle) or tetramethylbenzidine/H2O2 (right). (d) S. lividans pWKS10 was grown as described in Methods. Different concentrations of haemin (0, 10 or 50 µg ml–1) were added to the culture medium. After separation of proteins by SDS-PAGE, the proteins were transferred to a nylon membrane for Western analysis using anti-HbpS antibodies. The monomeric (m) and dimeric (d) forms of HbpS are indicated.

 
Identification of HbpS as a haem-binding protein
Domain analysis using the Pfam protein families database (Bateman et al., 2002) showed homologies with small stretches within the N-terminal regions from the NapC/NirT cytochrome c family, which bind four or five haem groups (Cartron et al., 2002). The haem-binding motif (CXXCH) of these regions was, however, not present in HbpS. Nevertheless, the ability of HbpS to bind haem was tested. The mature HbpS–His-tag protein which was purified from the culture filtrate of S. lividans pWKS15 (Fig. 3a) was shown to contain haem by using tetramethylbenzidine/H2O2 staining (see Methods) (Fig. 4a, lane 2). The mature HbpS–His-tag protein also had a characteristic ferric high-spin spectrum with a very strong Soret peak at 409 nm (Fig. 4b), which is typical for haem-containing proteins including the mycelium-associated S. reticuli CpeB (Zou & Schrempf, 2000), albumin and the mouse haem-binding protein p22 HBP (Taketani et al., 1998). Among the extracellular proteins of S. lividans pWKS10 carrying wild-type hbpS (without additional histidine codons) one protein was detected which immuno-reacted with anti-HbpS antibodies and was stained with tetramethylbenzidine/H2O2 (Fig. 4c, lane 2) indicating the presence of haem-containing HbpS. As expected, HbpS was absent in the culture filtrate of S. lividans pWKS17 (without hbpS) (Fig. 4c, lane 1).

To investigate haem-binding properties of the non-mature HbpS–His-tag protein, it was purified from the cytoplasmic fraction of S. lividans pWKS15 (Fig. 4a, lane 3). Tetramethylbenzidine/H2O2 staining showed that the non-mature (20 kDa) form of HbpS also contained haem (Fig. 4a, lane 4).

Analysis of HbpS production in S. reticuli wild-type and hbpS disruption mutant
Using antibodies, S. reticuli was found to secrete small amounts of HbpS (Fig. 3b, lane 1). As expected, the levels were considerably lower than those from S. lividans pWKS10 carrying the hbpS gene on the multicopy plasmid. As a basis for further investigations, the hbpS gene within the S. reticuli genome was disrupted according to the strategy outlined in Methods. Southern hybridizations using hbpS- and hyg-probes (data not shown) revealed that a double crossover between the genomic S. reticuli hbpS gene and the residual hbpS portions flanking the hyg in pUCH2 (Table 1) had occurred as desired (Fig. 1a, hbpS mutant). Immunological studies (Western blot analysis) revealed that the culture filtrate of the chromosomal S. reticuli hbpS disruption mutant lacked HbpS (Fig. 3b, lane 2).

Resistance to haemin correlates with HbpS production
It is well known that haemin, as a natural porphyrin, possesses significant antibacterial activity that is augmented by the presence of physiological concentrations of hydrogen peroxide or a reducing agent (Stojiljkovic et al., 2001). Increasing concentrations (100–400 µM) of haemin (the Fe3+ oxidation product of haem) in the culture medium led to a higher growth-inhibition of the S. reticuli hbpS mutant strain (Table 2) than of the S. reticuli wild-type strain. S. lividans pWKS10 also produced considerably greater amounts of HbpS in the presence of haemin (0–50 µg ml–1) than in the absence of haemin (Fig. 4d). The data imply that HbpS plays an important role in defence against the high toxicity of haemin, which catalyses free radical formation.


View this table:
[in this window]
[in a new window]
 
Table 2. Haemin sensitivity of S. reticuli strains

 
Effect of HbpS on cpeB expression
The hbpS gene is located downstream of the cpeB gene encoding the haem-containing catalase-peroxidase CpeB. Thus, it was speculated that HbpS might be required to form active, haem-containing CpeB. To test this assumption, S. reticuli wild-type and S. reticuli hbpS mutant were cultivated in minimal medium containing crystalline cellulose (1 % Avicel), in which high levels of the mycelium-associated enzyme are produced (Zou et al., 1999). Enzymic analysis (test for peroxidase activity; Fig. 5c) as well as immunological studies (using anti-CpeB antibodies; Fig. 5b) revealed that the amount of CpeB enzyme released from the mycelia (see Methods) of the S. reticuli wild-type was considerably greater than from the S. reticuli hbpS disruption mutant strain. Further hybridization studies (Northern blot) showed that among equivalent quantities of total RNA, the level of specific cpeB transcripts was considerably reduced in the S. reticuli hbpS disruption mutant (Fig. 5d). The wild-type strain also secreted a 45 kDa (Fig. 5a) protein, which was either absent or produced at undetectable levels in the hbpS mutant strain. This mutant strain, however, secreted a 28 kDa (Fig. 5a) protein which was not found in the wild-type strain. The nature of these two proteins still has to be characterized.



View larger version (50K):
[in this window]
[in a new window]
 
Fig. 5. Expression of cpeB in the S. reticuli hpbS mutant. S. reticuli wild-type (w) and S. reticuli hbpS mutant (m) were cultivated as described in Methods. Proteins were released from the mycelium, subjected to SDS-PAGE and either stained with Coomassie brilliant blue (a) or transferred to a nylon membrane for Western analysis using anti-CpeB antibodies (b). Protein-containing bands of 45 and 28 kDa are indicated in (a) by arrows. (c) Mycelium-associated proteins were loaded on to native polyacrylamide gel and tested for peroxidase activity as described in Methods. Protein markers are in lane S. (d) RNA was isolated from the same cultures and each aliquot (10 µg) was fractionated by gel electrophoresis, transferred to a nylon membrane and hybridized using a fragment of the cpeB gene as probe.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
We have described herein the molecular cloning and functional characterization of a novel S. reticuli extracellular protein, which was subsequently shown to bind haem and therefore was named HbpS (haem-binding protein from Streptomyces). HbpS contains an N-terminal signal peptide with a twin arginine signature within the sequence SRRTRV, which diverges from the consensus sequence of the Tat motif S/TRRXFL (Brink et al., 1998; Berks et al., 2000), with an additional arginine in position +3 and a valine in position +4. Results of previous experiments had revealed that not only the twin arginines (position 0 and +1) but also the hydrophobic residues at position +3 and +4 (Sargent et al., 1998) are important for translocation of substrates via the Tat pathway. The designed S. lividans transformant carrying an hbpS gene with substitutions of the two adjacent arginine codons no longer secreted HbpS but accumulated it in the cytoplasm, suggesting that HbpS may be secreted in a Tat-dependent manner. The most remarkable characteristic of the Tat pathway is that it apparently functions to transport folded proteins of variable dimensions across the cytoplasmic membrane. In most cases, the Tat substrates bind their corresponding cofactors in the cytoplasm and are thus already folded prior to export. Such proteins function predominantly in respiratory and photosynthetic electron transport chains and are vital for many types of bacterial energy metabolism (Berks et al., 2000). The Tat translocation pathway in S. lividans was found to be encoded by tatA, tatB and tatC genes (Schaerlaekens et al., 2001) encoding proteins which are close homologues to those characterized within E. coli (Bogsch et al., 1998). Until now, only two Streptomyces proteins have been experimentally proven to be secreted via the Tat pathway. The transactivator protein MelC1 of a tyrosinase (Streptomyces antibioticus) and xylanase C (S. lividans) were not secreted in the S. lividans tatC mutant carrying the corresponding genes on the cloned plasmids (Schaerlaekens et al., 2004). Further studies in a recently described S. lividans mutant lacking a functional Tat pathway (Schaerlaekens et al., 2001) could help to elucidate whether HbpS is indeed a Tat substrate. Computer analysis of the S. coelicolor A3(2) genome reveals the existence of 230 putative Tat substrates, which are members of a variety of protein classes including a high number functioning in degradation of macromolecules, in binding and transport, and in secondary metabolism (Schaerlaekens et al., 2004).

Whereas some haem-binding proteins (e.g. those of the NapC/NirT cytochrome c family) have the CXXCH motif, which is necessary for attaching of haem, others lack this motif. Albumin can bind several haem groups per molecule due to the presence of a hydrophobic region (Shin et al., 1994). Similarly, the mouse p22 haem-binding protein contains a hydrophobic region, which is speculated to bind haem (Taketani et al., 1998). The chaperone CcmE interacts with haem transiently in the periplasm of E. coli and delivers it to newly synthesized and exported c-type cytochromes. Alanine scanning mutagenesis of conserved amino acids revealed that only H130 is strictly required for haem-binding and delivery. Mutation of the hydrophobic amino acids (F37, F103, L127 and Y134) to alanine affected the interaction with haem of CcmE more than the mutation of other charged and polar amino acids. The data suggest that haem is bound to a hydrophobic platform at the surface of the protein and then attached to H130 by a covalent bond (Enggist et al., 2003). In this context, it is important to mention that HbpS is rich in hydrophobic residues and contains three histidine residues (H60, H83 and H187) surrounded by leucines and valines (Fig. 1b). The crystal structure of yeast cytochrome c peroxidase (CCP) shows that the histidine (H) residue at position 181 interacts with haem (Finzel et al., 1984). This residue is located within the motif LX2THLX10AA, which exhibits a similarity to the region LX3THLX10AA including the histidine residue at position 60 within HbpS (Fig. 1c).

The S. reticuli hbpS mutant strain was found to be more sensitive than the wild-type strain to higher concentrations of haemin (Fe3+-oxidized form of haem). It is known that haemin at higher concentrations is highly toxic because of its ability to catalyse free radical formation (Baker et al., 2003). The enhanced sensitivity to haemin in the mutant strain correlated with the lack of HbpS. This result suggested that within the wild-type strain haemin is titrated by HbpS, leading to a reduction of free haemin.

Haem has been found to be necessary for the accumulation and assembly of cytochrome c oxidase in Saccharomyces cerevisiae, which is able to synthesize haem (Woloszczuk et al., 1980). In addition, it was shown that haem is important not only for the formation of active catalases but also for the synthesis, or at least accumulation, of the apoproteins of catalases (A and T) in yeast. During these processes, haem could either act as a positive regulator of the synthesis of apocatalases or it could prevent rapid degradation of the enzyme precursors (Howe & Merchant, 1994).

Physiological studies showed that the amount of the catalase-peroxidase CpeB in the S. reticuli hbpS disruption mutant in comparison to S. reticuli wild-type was considerably lower. Within the mutant strain the amount of cpeB mRNA is reduced, suggesting that HbpS influences the expression of cpeB in a positive manner. These data lead to the following explanations: HbpS could act as chaperone that binds haem and delivers it to the mycelium-associated CpeB, or HbpS could be involved in a signal transduction cascade regulating the expression of cpeB. As HbpS is an extracellular protein, it could interact with extracellular or membrane-associated proteins involved in such signal transduction. Interestingly, downstream of hbpS two ORFs are located, which are divergently transcribed from hbpS and encode a predicted sensor kinase and a putative response regulator (data not shown). These proteins could be the components for the corresponding signal transduction system.


   ACKNOWLEDGEMENTS
 
We are very grateful to P. Zou and I. Borovok for analysing the DNA sequence and to A. Welzel and S. Gross-Hardt for supporting some cloning studies and biochemical analysis.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
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]

Baker, H. M., Anderson, B. F. & Baker, E. N. (2003). Dealing with iron: common structural principles in proteins that transport iron and heme. Proc Natl Acad Sci U S A 100, 3579–3583.[Abstract/Free Full Text]

Bateman, A., Birney, E., Cerruti, L. & 7 other authors (2002). The Pfam protein families database. Nucleic Acids Res 30, 276–280.[Abstract/Free Full Text]

Berks, B. C., Sargent, F. & Palmer, T. (2000). The Tat protein export pathway. Mol Microbiol 35, 260–274.[CrossRef][Medline]

Blake, M. S., Johnston, K. H., Russel-Jones, G. J. & Gotschlich, E. C. (1984). A rapid, sensitive method for detection of alkaline phosphatase-conjugated anti-antibody on Western blots. Anal Biochem 136, 175–179.[Medline]

Blondelet-Rouault, M. H., Weiser, J., Lebrihi, A., Branny, P. & Pernodet, J. L. (1997). Antibiotic resistance gene cassettes derived from the omega interposon for use in E. coli and Streptomyces. Gene 190, 315–317.[CrossRef][Medline]

Bogsch, E. G., Sargent, F., Stanley, N. R., Berks, B. C., Robinson, C. & Palmer, T. (1998). An essential component of a novel bacterial protein export system with homologues in plastids and mitochondria. J Biol Chem 273, 18003–18006.[Abstract/Free Full Text]

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

Braun, V. & Braun, M. (2002). Iron transport and signaling in Escherichia coli. FEBS Lett 529, 78–85.[CrossRef][Medline]

Brink, S., Bogsch, E. G., Edwards, W. R., Hynds, P. J. & Robinson, C. (1998). Targeting of thylakoid proteins by the {Delta}pH-driven twin-arginine translocation pathway requires a specific signal in the hydrophobic domain in conjunction with the twin-arginine motif. FEBS Lett 434, 425–430.[CrossRef][Medline]

Cartron, M. L., Roldan, M. D., Ferguson, S. J., Berks, B. C. & Richardson, D. J. (2002). Identification of two domains and distal histidine ligands to the four haems in the bacterial c-type cytochrome NapC; the prototype connector between quinol/quinone and periplasmic oxido-reductases. Biochem J 368, 425–432.[CrossRef][Medline]

Conyers, S. M. & Kidwell, D. A. (1991). Chromogenic substrates for horseradish peroxidase. Anal Biochem 192, 207–211.[Medline]

Dower, W. J., Miller, J. F. & Ragsdale, C. W. (1988). High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res 16, 6127–6145.[Abstract]

Enggist, E., Schneider, M. J., Schulz, H. & Thony-Meyer, L. (2003). Biochemical and mutational characterization of the heme chaperone CcmE reveals a heme binding site. J Bacteriol 185, 175–183.[Abstract/Free Full Text]

Finzel, B. C., Poulos, T. L. & Kraut, J. (1984). Crystal structure of yeast cytochrome c peroxidase refined at 1·7-Å resolution. J Biol Chem 259, 13027–13036.[Abstract/Free Full Text]

Fridovich, I. (1986). Biological effects of the superoxide radical. Arch Biochem Biophys 247, 1–11.[Medline]

Gold, L., Pribnow, D., Schneider, T., Shinedling, S., Singer, B. S. & Stormo, G. (1981). Translational initiation in prokaryotes. Annu Rev Microbiol 35, 365–403.[CrossRef][Medline]

Grimm, A. C. & Harwood, C. S. (1999). NahY, a catabolic plasmid-encoded receptor required for chemotaxis of Pseudomonas putida to the aromatic hydrocarbon naphthalene. J Bacteriol 181, 3310–3316.[Abstract/Free Full Text]

Higgins, D. G., Bleasby, A. J. & Fuchs, R. (1992). CLUSTAL V: improved software for multiple sequence alignment. Comput Appl Biosci 8, 189–191.[Abstract]

Hopwood, D. A., Bibb, M. J., Chater, K. F. & 7 other authors (1985). Genetic Manipulation of Streptomyces: a Laboratory Manual. Norwich: John Innes Foundation.

Howe, G. & Merchant, S. (1994). Role of heme in the biosynthesis of cytochrome c6. J Biol Chem 269, 5824–5832.[Abstract/Free Full Text]

Johansson, P. & Hederstedt, L. (1999). Organization of genes for tetrapyrrole biosynthesis in Gram-positive bacteria. Microbiology 145, 529–538.[Medline]

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

Moore, R. W., Welton, A. F. & Aust, S. D. (1978). Detection of hemoproteins in SDS-polyacrylamide gels. Methods Enzymol 52, 324–331.[Medline]

Nielsen, H., Engelbrecht, J., Brunak, S. & von Heijne, G. (1997). Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng 10, 1–6.[CrossRef]

Ortiz de Orué Lucana, D. & Schrempf, H. (2000). The DNA-binding characteristics of the Streptomyces reticuli regulator FurS depend on the redox state of its cysteine residues. Mol Gen Genet 264, 341–353.[CrossRef][Medline]

Ortiz de Orué Lucana, D., Troller, M. & Schrempf, H. (2003). Amino acid residues involved in reversible thiol formation and zinc ion binding in the Streptomyces reticuli redox regulator FurS. Mol Genet Genomics 268, 618–627.[Medline]

Panek, H. & O'Brian, M. R. (2002). A whole genome view of prokaryotic haem biosynthesis. Microbiology 148, 2273–2282.[Medline]

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

Sargent, F., Bogsch, E. G., Stanley, N. R., Wexler, M., Robinson, C., Berks, B. C. & Palmer, T. (1998). Overlapping functions of components of a bacterial Sec-independent protein export pathway. EMBO J 17, 3640–3650.[Abstract/Free Full Text]

Schaerlaekens, K., Schierova, M., Lammertyn, E., Geukens, N., Anné, J. & van Mellaert, L. (2001). Twin-arginine translocation pathway in Streptomyces lividans. J Bacteriol 183, 6727–6732.[Abstract/Free Full Text]

Schaerlaekens, K., van Mellaert, L., Lammertyn, E., Geukens, N. & Anné, J. (2004). The importance of the Tat-dependent protein secretion pathway in Streptomyces as revealed by phenotypic changes in tat deletion mutants and genome analysis. Microbiology 150, 21–31.[CrossRef][Medline]

Schlochtermeier, A., Niemeyer, F. & Schrempf, H. (1992a). Biochemical and electron microscopic studies of the Streptomyces reticuli cellulase (Avicelase) in its mycelium-associated and extracellular forms. Appl Environ Microbiol 58, 3240–3248.[Abstract]

Schlochtermeier, A., Walter, S., Schröder, J., Moorman, M. & Schrempf, H. (1992b). The gene encoding the cellulase (Avicelase) Cel1 from Streptomyces reticuli and analysis of protein domains. Mol Microbiol 6, 3611–3621.[Medline]

Shin, W. S., Yamashita, H. & Hirose, M. (1994). Multiple effects of haemin binding on protease susceptibility of bovine serum albumin and a novel isolation procedure for its large fragment. Biochem J 304, 81–86.[Medline]

Smalley, J. W., Charalabous, P., Birss, A. J. & Hart, C. A. (2001). Detection of heme-binding proteins in epidemic strains of Burkholderia cepacia. Clin Diagn Lab Immunol 8, 509–514.[Abstract/Free Full Text]

Stojiljkovic, I., Evavold, B. D. & Kumar, V. (2001). Antimicrobial properties of porphyrins. Expert Opin Investig Drugs 10, 309–320.[Medline]

Strohl, W. R. (1992). Compilation and analysis of DNA sequences associated with apparent streptomycete promoters. Nucleic Acids Res 20, 961–974.[Abstract]

Taketani, S., Adachi, Y., Kohno, H., Ikehara, S., Tokunaga, R. & Ishii, T. (1998). Molecular characterization of a newly identified heme-binding protein induced during differentiation of urine erythroleukemia cells. J Biol Chem 273, 31388–31394.[Abstract/Free Full Text]

Thony-Meyer, L. (1997). Biogenesis of respiratory cytochromes in bacteria. Microbiol Mol Biol Rev 61, 337–376.[Abstract]

Vara, J., Lewandowska-Skarbek, M., Wang, Y.-G., Donadio, S. & Hutchinson, C. R. (1989). Cloning of genes governing the deoxysugar portion of the erythromycin biosynthesis pathway in Saccharopolyspora erythraea (Streptomyces erythreus). J Bacteriol 171, 5872–5881.[Medline]

Villarejo, M. R., Zamenhof, P. J. & Zabin, I. (1972). {beta}-Galactosidase. In vivo {alpha}-complementation. J Biol Chem 247, 2212–2216.[Abstract/Free Full Text]

Woloszczuk, W., Sprinson, D. B. & Ruis, H. (1980). The relation of heme to catalase apoprotein synthesis in yeast. J Biol Chem 255, 2624–2627.[Abstract/Free Full Text]

Wood, D. W., Setubal, J. C., Kaul, R. & 48 other authors (2001). The genome of the natural genetic engineer Agrobacterium tumefaciens C58. Science 294, 2317–2323.[Abstract/Free Full Text]

Yrjala, K., Paulin, L. & Romantschuk, M. (1997). Novel organization of catechol meta-pathway genes in Sphingomonas sp. HV3 pSKY4 plasmid. FEMS Microbiol Lett 154, 403–408.[CrossRef][Medline]

Zou, P. & Schrempf, H. (2000). The heme-independent manganese-peroxidase activity depends on the presence of the C-terminal domain within the Streptomyces reticuli catalase-peroxidase CpeB. Eur J Biochem 267, 2840–2849.[Abstract/Free Full Text]

Zou, P., Borovok, I., Ortiz de Orué Lucana, D., Müller, D. & Schrempf, H. (1999). The mycelium-associated Streptomyces reticuli catalase-peroxidase, its gene and regulation by FurS. Microbiology 145, 549–559.[Medline]

Received 10 February 2004; revised 22 April 2004; accepted 20 May 2004.



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 Lucana, D. O. d. O.
Articles by Schrempf, H.
Articles citing this Article
PubMed
PubMed Citation
Articles by Lucana, D. O. d. O.
Articles by Schrempf, H.
Agricola
Articles by Lucana, D. O. d. O.
Articles by Schrempf, H.


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 © 2004 Society for General Microbiology.