Abstract
Abbreviations: AFM, atomic force microscopy; MALDI-TOF MS, matrix-assisted laser desorption ionization time-of-flight mass spectrometry; RACE-PCR, rapid amplification of cDNA ends PCR
The GenBank/EMBL/DDBJ accession number for the sequence reported in this paper is AY842007.
It has been calculated that a rod-shaped bacterium of average size with a generation time of 20 min has to synthesize around 500 S-layer subunits per second to cover the cell surface completely (Sleytr & Messner, 1983). Accordingly, S-layer genes are expressed at extremely high levels, and S-layer proteins constitute 10 to 15 % of the total protein content of the cell (Messner & Sleytr, 1992). The high-level expression of S-layer genes is based not only on strong promoters and efficient transcription but also on highly stable mRNA molecules (Boot & Pouwels, 1996; Fisher et al., 1988; Kahala et al., 1997). However, S-layer proteins are seldom found in culture supernatants, suggesting that their synthesis is tightly regulated.
A limited number of studies have determined that the biosynthesis of S-layer proteins is controlled at the transcriptional and post-transcriptional level, involving DNA-binding transcription factors and feedback control mechanisms by the S-layer protein. The S-layer mRNAs of Bacillus anthracis, Lactobacillus brevis and Thermus thermophilus contain 5' untranslated regions of up to 300 bp in length that might be involved in stabilization of the transcript (Mignot et al., 2002; Vidgren et al., 1992). For instance, the C-terminal part of the S-layer protein SlpA of T. thermophilus binds to the 5' leader region of the slpA mRNA, providing a clue to the mechanism of post-transcriptional autoregulation during the expression of slpA (Fernandez-Herrero et al., 1997). The highly stable S-layer mRNAs of Caulobacter crescentus and L. brevis possess long half-lives of 1015 min and 32 min, respectively (Fisher et al., 1988; Kahala et al., 1997). On the other hand, S-layer proteins may act as transcription factors, regulating the expression of their own genes. For instance, the S-layer protein AbcA of Aeromonas salmonicida activates in trans the expression of the S-layer gene when analysed in the heterologous host Escherichia coli (Chu & Trust, 1993; Noonan & Trust, 1995). S-layer gene expression of B. anthracis is moreover controlled by alternative sigma factors and the transcriptional master regulator AtxA (Mignot et al., 2004).
The outermost surface of Corynebacterium glutamicum, a Gram-positive bacterium of great biotechnological importance (Hermann, 2003), also consists of a paracrystalline S-layer whose protomers are anchored in an outer-membrane-like structure (Chami et al., 1997). The S-layer of C. glutamicum possesses a hexagonal lattice symmetry (Chami et al., 1995), and has been classified into the M6C3-layer type by atomic force microscopy (Scheuring et al., 2002). The S-layer of C. glutamicum is formed by the so-called PS2 protein, which is encoded by the cspB gene (Peyret et al., 1993). Recently, a set of C. glutamicum isolates has been analysed systematically with respect to the presence of an S-layer that is found to be absent only in the completely sequenced type strain C. glutamicum ATCC 13032 (Hansmeier et al., 2004). Sequence-based analyses of 28 different PS2 proteins have revealed a direct coherence between the primary sequence of S-layer proteins and the corresponding morphology of S-layers (Hansmeier et al., 2004).
First hints regarding the regulation of S-layer gene expression in C. glutamicum were published by Chami et al. (1995), who showed that C. glutamicum cells grown on solid medium are completely covered with an S-layer, whereas cells cultivated in liquid medium are only partially covered by an ordered lattice. Subsequent studies have demonstrated that the amount of S-layer protein synthesized by C. glutamicum is dependent on the carbon source of the growth medium (Soual-Hoebeke et al., 1999). In particular, the addition of lactate to the growth medium has a stimulatory effect on PS2 production as well as on S-layer formation, whereas an inhibitory effect on PS2 synthesis is observed when glucose is used as carbon source. On the basis of these observations, a relationship between S-layer formation, carbohydrate metabolism and the physiological status of the cell in general has been suggested (Soual-Hoebeke et al., 1999).
In the present study, the S-layer gene region of C. glutamicum ATCC 14067 was identified, sequenced and subsequently compared with the genome sequence of C. glutamicum ATCC 13032 to get clues on why this strain is devoid of an ordered S-layer. Furthermore, we examined the transcriptional regulation of the C. glutamicum S-layer gene and identified a putative activator of the cspB gene.
Bacterial strains and growth conditions.C. glutamicum strains used in this study were obtained from the American Type Culture Collection. E. coli DH5αMCR (Grant et al., 1990) was used as host for recombinant plasmids. E. coli strains were routinely grown at 37 °C on solid Luria broth complex medium (Invitrogen). Selection for the presence of plasmids in E. coli was performed by addition of 50 µg kanamycin ml1 or 100 µg ampicillin ml1 to the growth medium. C. glutamicum strains were cultivated on solid Luria broth complex medium or in liquid minimal medium MM1 [MMYE without yeast extract (Katsumata et al., 1984)]. Antibiotics for the selection of plasmids in C. glutamicum were kanamycin (25 µg ml1) and chloramphenicol (5 µg ml1).
DNA preparation, manipulation and transformation.
Genomic DNA of C. glutamicum was isolated with the GenElute Bacterial Genomic DNA kit (Sigma-Aldrich) according to the protocol for Gram-positive bacteria. Plasmid and fosmid DNA was prepared from E. coli by the alkaline lysis technique using the QIAprep Spin Miniprep kit (Qiagen). Alkaline lysis of C. glutamicum cells was modified by using 20 mg lysozyme ml1 in resuspension buffer P1 at 37 °C for 2 h. All DNA manipulations were carried out as described by Sambrook et al. (1989). DNA transfer into competent C. glutamicum cells was performed by electroporation (Bonamy et al., 1990).
PCR techniques.
PCR experiments were performed with a PTC-100 thermocycler (MJ Research). Amplification of DNA was carried out with Pfx DNA polymerase according to the manufacturer's protocol (Invitrogen). Screening of the fosmid library by PCR was performed with the Qiagen Taq DNA polymerase. Oligonucleotides used in this study were purchased from Qiagen. PCR cycling times and temperatures were chosen according to the type of DNA polymerase, fragment length and oligonucleotide sequence. PCR products were purified by means of the QIAquick PCR Purification kit (Qiagen).
Construction of a cg2831 mutant of C. glutamicum.
To construct a cg2831 insertion mutant of C. glutamicum ATCC 13058, a 469 bp DNA fragment was amplified by PCR using the primer pair Int1 (AAGCCAGCACGTCCAACT) and Int2 (TCCGGTTGGTGTGAGTGA). The PCR product was cloned into the SmaI site of pK18mob (Schäfer et al., 1994), and the resulting plasmid pK18mob-Intcg2831 was transferred to C. glutamicum by electroporation. Gene disruption of cg2831 was confirmed by Southern hybridization (Sambrook et al., 1989).
Construction and screening of a C. glutamicum ATCC 14067 fosmid library.
A genomic library of C. glutamicum ATCC 14067 was prepared in the pCC1FOS fosmid vector by IIT Biotech (Bielefeld). Briefly, genomic DNA of C. glutamicum was randomly sheared to give 40 kb fragments that were ligated into pCC1FOS. The ligated DNA was packaged with MaxPlax Lambda Packaging Extracts and transduced into E. coli EPI300 cells that were spread on Luria broth agar plates containing 12·5 µg chloramphenicol ml1. The resulting fosmid library was screened by applying PCR experiments with the cspB-specific primers cspB_S1 (TGCTGGTCGAATCGCAAT) and cspB_S2 (AGAATGCTCGTCCGAACA). To facilitate DNA sequencing of the cspB gene region, fosmid clone pFOS-D1 was restricted either with EcoRI or with HindIII, and the resulting DNA fragments were subcloned in pUC19 (Yanisch-Perron et al., 1985). Subclones pUC19EcoRI-01 and pUC19HindIII-12 were selected for DNA sequencing by primer walking (IIT Biotech). The nucleotide sequence of the cspB gene region was assembled with the STADEN software package (Staden, 1996). Annotation of the sequence was performed with the help of BLAST algorithms and Conserved Domain Database searches (Altschul et al., 1997; Marchler-Bauer et al., 2003).
Separation and identification of C. glutamicum proteins by SDS-PAGE and matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS).
S-layer proteins were extracted from the cell surface of C. glutamicum as described previously (Hansmeier et al., 2004). The proteins were separated by one-dimensional 12·5 % SDS-PAGE using the technique of Laemmli (1970). Gels were stained using Coomassie brilliant blue R-250 and G-250 (Sambrook et al., 1989), and were briefly destained with 7 % acetic acid to visualize protein bands. Molecular masses were determined by using the Broad Range Precision Protein Standard (Bio-Rad Laboratories) or the Prestained Precision Protein Standard (MBI-Fermentas).
For the identification of proteins, bands were excised from Coomassie-stained SDS-PAGE gels. Then, tryptic digestions and MALDI-TOF MS analysis were applied to generate peptide mass fingerprints (Hansmeier et al., 2004). The Bruker Ultraflex MALDI-TOF mass spectrometer was used to generate mass spectra. Peptide mass fingerprints were compared with in silico-generated tryptic fingerprints by using the MASCOT software (Perkins et al., 1999).
DNA affinity purification assay.
To isolate proteins that bind to the putative promoter region of the C. glutamicum cspB gene, a DNA affinity purification method was used (Rey et al., 2003). DNA fragments pF1-2 and pF3-4 were generated by PCR using the biotin-labelled oligonucleotides pF1 (GTAGTCCGAGGTTAAGTG), pF2 (GCAGCCTGTCGTTGAGAABioTAG), pF3 (CAACGACAGGCTGCTAAG) and pF4 (CAGTGCGGATACGATTGTBioTAG). Unincorporated oligonucleotides and dNTPs were removed by means of the PCR Purification Spin kit (Qiagen). About 600 pmol of the biotin-labelled PCR products was immobilized to 200 µl streptavidin-coated magnetic particles (KMF Laborchemie). The magnetic particles were then equilibrated with protein binding buffer (20 mM Tris, 1 mM EDTA, 10 % (v/v) glycerol, 0·05 % (v/v) Triton X-100, 1 mM DTT, 100 mM NaCl, pH 8·0) and incubated with C. glutamicum crude cell extracts prepared from exponentially growing cultures. Unbound proteins were removed by using magnetic separation with a magnetic particle concentrator and by two washing steps with protein binding buffer. DNA-bound proteins were finally eluted with 20 µl protein binding buffer containing 1 M NaCl. Eluted protein fractions were collected and analysed by SDS-PAGE and MALDI-TOF MS.
Atomic force microscopy (AFM) of C. glutamicum cells.
C. glutamicum cells were cultivated on solid Luria broth medium for 2 days. The cells were washed twice in washing buffer (20 mM Tris/HCl, pH 7·5) and incubated on glass plates overnight. AFM images were recorded in tapping mode with silicon cantilevers (BS-Tap300Al, Budget Sensors) under ambient conditions by using a Nanoscope IIIa AFM system equipped with a Bioscope head (Veeco). AFM topographs and phase images were recorded simultaneously.
RNA isolation, rapid amplification of cDNA ends (RACE)-PCR and real-time RT-PCR.
To isolate total RNA from C. glutamicum cells, cultures were grown in Luria broth medium. Approximately 1x109 cells were harvested and total RNA was purified by using the RNeasy kit (Qiagen). To identify the transcription start site of the C. glutamicum cspB gene, 5' RACE-PCR experiments (Roche Diagnostics) were carried out with the specific oligonucleotides cspB_sp1a (CAACTGGCTGGATGGTGGAT), cspB_sp1b (GAAGCCGTTGGTGATGTTGA) and cspB_sp1c (GTTGGTCTCCTGAGCGAATG). Real-time RT-PCR was performed with the QuantiTect SYBR Green PCR kit (Qiagen) and the LightCycler system (Roche). Cycling times and temperatures were chosen according to the LightCycler manufacturer's protocols in dependence on the oligonucleotide sequences (cspBLC1, TGCGTAAGCAACGGACTC; cspBLC2, GGCTTCAACGATGCTGAT; cg2831LC1, AAGCCAGCACGTCCAACT; cg2831LC2, TCCGGTTGGTGTGAGTGA). Determination of crossing points was done by using the second derivative maximum data analysis method. The crossing point of the wild-type was used for determining relative gene expression (Pfaffl, 2001). All real-time RT-PCR experiments were carried out in duplicate with three biological replicates.
In order to identify components necessary for S-layer gene expression and formation, we cloned and sequenced the cspB gene region of the wild-type strain C. glutamicum ATCC 14067 (formerly named Brevibacterium flavum), which is capable of synthesizing the S-layer protomer PS2. A genomic library of randomly sheared chromosomal DNA fragments of C. glutamicum ATCC 14067 was constructed in fosmid vector pCC1FOS. About 1400 individual clones, representing an estimated 16-fold genome coverage based on the complete sequence of C. glutamicum ATCC 13032, were transferred into E. coli EPI300 and used for further experiments. The fosmid library was screened for the presence of the S-layer gene by PCR techniques, using the cspB-specific primer pair cspB_S1 and cspB_S2 and a defined combination of DNA pools of clones from microtitre plates (Capela et al., 1999). Despite the high genome coverage of the fosmid library, only four fosmids revealed a PCR fragment of the expected size of about 1 kb, representing an internal fragment of the cspB gene (data not shown). Fosmid pFOS-D1 was selected for further cloning, since it covered the entire cspB gene region. Fosmid DNA was digested with either EcoRI or HindIII, and restriction fragments were subcloned into pUC19. Two DNA fragments with sizes of approximately 5 and 6 kb overlapped the cspB gene region, and were finally chosen for nucleotide sequencing. A DNA region of 11 294 bp was determined on both strands by using a primer-walking strategy. The DNA sequence was analysed and annotated with the BLAST tools and by Conserved Domain Database searches. The resulting annotation and relevant characteristics of the predicted proteins are summarized in Table 1.
Table 1. Molecular features of the predicted proteins encoded by the cspB gene region of C. glutamicum ATCC 14067
A total of 13 coding sequences were predicted on the sequenced cspB gene region. Comparison with the known genome sequence of C. glutamicum ATCC 13032 (Kalinowski et al., 2003) showed that the deduced gene products of bf2708 to bf2713 are highly similar to proteins (Cg2708 to Cg2713) of the PS2 strain C. glutamicum ATCC 13032 (Table 1). Most interestingly, alignments of bf2714a with the genome sequence of C. glutamicum ATCC 13032 showed 99 % identity to the 5' region of cg2714 (nucleotides 1385), whereas part of bf2714f was nearly identical to the 3' region (nucleotides 378879) of cg2714. Consequently, the 5' and 3' ends of cg2714 are separated in C. glutamicum ATCC 14067 by an additional 5·97 kb DNA segment. This DNA segment is flanked by the 7 bp direct repeat CGCGATG, which represents the deduced border of the cspB gene region additionally present in the C. glutamicum ATCC 14067 chromosome. Accordingly, cg2714 is most likely a chimeric coding sequence, resulting from a defined event of gene loss in the chromosome of C. glutamicum ATCC 13032.
In addition to the cspB gene, six coding sequences, designated bf2714a to bf2714f, were identified on the sequenced DNA fragment of C. glutamicum ATCC 14067. All coding sequences were preceded by a ribosome-binding site in front of the predicted translational start codon. Palindromic stemloop structures resembling transcriptional terminators were predicted downstream of each coding sequence, with the exception of bf2714e. The deduced proteins for bf2714a to bf2714f revealed similarities to alcohol dehydrogenases, amidohydrolases, oxidoreductases and carboxylases of different organisms (Table 1). According to this bioinformatic prediction, no functional link of the six proteins to S-layer synthesis was apparent, suggesting that S-layer expression and formation might be possible in the PS2 strain C. glutamicum ATCC 13032 solely by transferring the cspB structural gene.
The cspB gene region is conserved in several C. glutamicum wild-type strains
Earlier examination of 28 different C. glutamicum isolates revealed a general occurrence of the S-layer gene cspB, with the exception of the sequenced type strain C. glutamicum ATCC 13032 (Hansmeier et al., 2004). To determine how conserved the genetic arrangement of the cspB gene region is in different C. glutamicum strains, a systematic PCR mapping approach was applied (Fig. 1a). According to a previously published classification scheme based on sequence and structural similarities of the corynebacterial S-layer, C. glutamicum wild-type strains are divided into five classes (Hansmeier et al., 2004). To gain a wide diversity of C. glutamicum isolates in the present study, one representative strain of each class was chosen for the PCR mapping approach, namely C. glutamicum ATCC 17965 (class 1), C. glutamicum 22243 (class 2), C. glutamicum ATCC 13058 (class 3), C. glutamicum ATCC 31808 (class 4) and C. glutamicum ATCC 14017 (class 5). The nucleotide sequence of the cspB gene region from C. glutamicum ATCC 14067 was used to design 13 primer pairs capable of amplifying a distinct set of DNA fragments that should completely cover structurally similar gene regions in other wild-type strains (Fig. 1a). Accordingly, chromosomal DNA of C. glutamicum ATCC 14067 (class 5) served as control in PCR experiments. PCR mapping of the hitherto unknown cspB gene region of C. glutamicum ATCC 13058, a member of S-layer class 3, is shown as an example in Fig. 1(b). The resulting PCR products were analysed by agarose gel electrophoresis, and all of the amplified DNA fragments revealed the expected sizes. Mapping analysis of representatives of other S-layer classes showed identical results (data not shown). Consequently, the PCR mapping approach demonstrated that the cspB gene region is conserved in different C. glutamicum wild-type strains capable of S-layer formation.
|
Transfer of the cspB gene to C. glutamicum ATCC 13032 is sufficient for S-layer formation
To restore the PS2 phenotype of the wild-type strain C. glutamicum ATCC 13032, plasmid pCGL824 was transferred by electroporation. This plasmid is a low-copy-number vector carrying the entire cspB gene of C. glutamicum ATCC 17965 under the control of its own promoter (Houssin et al., 2002; Peyret et al., 1993). The resulting strain C. glutamicum ATCC 13032 (pCGL824), as well as the control strains C. glutamicum ATCC 13032 and C. glutamicum ATCC 14067, were then cultivated on solid Luria broth complex medium. Cell surface proteins were extracted by using chaotropic detergents (Hansmeier et al., 2004), and were finally visualized by SDS-PAGE (Fig. 2). Comparison of the protein profile of C. glutamicum ATCC 13032 (pCGL824) with that of C. glutamicum ATCC 13032 showed an additional protein that was of a similar size to the most prominent protein of C. glutamicum ATCC 14067 (Fig. 2). Both proteins were subsequently identified as PS2 proteins by MALDI-TOF MS and peptide mass fingerprinting. These data demonstrated that the cspB gene is efficiently expressed in C. glutamicum ATCC 13032, resulting in a considerable amount of PS2 protein that can be extracted from the cell surface.
|
To assess whether, in addition, S-layer lattices were formed by C. glutamicum ATCC 13032 (pCGL824), images of the bacterial cell surface were taken by AFM. For this purpose, C. glutamicum ATCC 13032 (pCGL824) and the control strains C. glutamicum ATCC 13032 and C. glutamicum ATCC 14067 were cultivated on solid Luria broth medium at 30 °C for 48 h. The cells were then washed twice with Tris/HCl buffer before they were adsorbed on glass plates for AFM imaging. Fig. 3(a) shows AFM phase images of the corynebacterial cell surfaces that were recorded in tapping mode. The cell surface of C. glutamicum ATCC 13032 (Fig. 3a, 1) is devoid of any ordered structure that resembles hexagonal S-layers. On the other hand, C. glutamicum ATCC 13032 carrying pCGL824 with the cspB gene (Fig. 3a, 2) and C. glutamicum ATCC 14067 (Fig. 3a, 3) showed ordered cell surface lattices. This structure represents the typical hexagonal S-layer of C. glutamicum cells (Chami et al., 1997; Hansmeier et al., 2004; Scheuring et al., 2002). Morphological differences between the cell surfaces of C. glutamicum ATCC 13032 and the PS2+ strains C. glutamicum ATCC 13032 (pCGL824) and C. glutamicum ATCC 14067 were further emphasised by a more detailed view of selected areas of the images (Fig. 3b). Consequently, transfer of the cspB gene into C. glutamicum ATCC 13032 was sufficient for expression of the PS2 protein and formation of an ordered S-layer.
|
The promoter of the cspB gene was mapped by identifying the transcriptional start site
The DNA sequence determined on fosmid pFOS-D1 revealed that the intergenic region between cspB and bf2717b had a size of 637 nucleotides. To precisely map the promoter of the cspB gene within this DNA region, the 5' RACE-PCR method was applied. Total RNA was isolated from exponentially growing C. glutamicum cells and purified to amplify and sequence the 5' end of the cspB transcript. The resulting nucleotide sequence allowed us to determine the transcriptional start site of the cspB gene, which was located at an adenine residue 152 nucleotides upstream of the translational start codon of PS2 (Fig. 4a). Thus, the cspB mRNA of C. glutamicum contains a long 5' untranslated region, as reported previously for S-layer mRNAs of other bacteria (Castan et al., 2001; Mignot et al., 2002). The identified transcriptional start point was subsequently used to deduce the putative promoter sequence of the cspB gene according to the consensus sequence of C. glutamicum promoters defined by Patek et al. (1996). The deduced 35 (TGGCAA) and 10 (TACAGT) boxes of the cspB promoter are separated by a spacer of 17 nucleotides (Fig. 4a). The 10 box of the cspB promoter is almost identical with the consensus sequence of σA-dependent promoters of C. glutamicum [TA(T/C)AAT], whereas the 35 hexamer shares only four out of the six bases with the consensus sequence TTGCCA (Patek et al., 1996). These data suggested that the cspB gene of C. glutamicum is transcribed by the RNA polymerase-σA holoenzyme.
|
The LuxR-type transcriptional regulator Cg2831 was found to be involved in the activation of the cspB gene
The knowledge of the location of the cspB promoter enabled us to search for regulatory DNA-binding proteins interacting with this DNA region. For this purpose, DNA affinity purification assays with magnetic streptavidin beads were performed. The biotinylated DNA fragments pF1-2 and pF3-4, covering different portions of the cspB upstream region (Fig. 4a), were amplified and linked to streptavidin-coated magnetic beads. The immobilized DNA was then incubated with crude protein extracts from C. glutamicum cells grown in minimal medium MM1 with glucose as carbon source. Proteins were purified by binding to the DNA fragments, followed by elution with an appropriate buffer and separation by SDS-PAGE (Fig. 4b). The resulting protein gel showed a number of proteins binding to both the cspB promoter region present on pF1-2 and the pF3-4 region that served as control. Of special interest was a single protein band that was exclusively detected with the pF1-2 promoter fragment of the cspB gene (Fig. 4b). This protein band was excised from the gel, and the protein was digested with trypsin and analysed by peptide mass fingerprinting. This procedure resulted in the identification of protein Cg2831, which has been annotated in the course of the C. glutamicum genome project as a putative regulatory protein of the LuxR family (Kalinowski et al., 2003). An additional 14 protein bands were excised from each lane of the protein gel and analysed by MALDI-TOF MS and fingerprint analysis, but no further transcriptional regulator was detected. These other proteins were identified either as being subunits of RNA or DNA polymerase, or as having other DNA-binding activities, such as helicases, methyltransferases and single-strand binding protein (data not shown). In fact, the other proteins were purified by means of both the pF1-2 promoter region and the pF3-4 control fragment, apparently excluding a regulatory role for these proteins in cspB gene expression.
Inspection of the C. glutamicum genome sequence suggested that cg2831 represents a separate transcription unit that is located downstream of a gene (cg2830) encoding a putative adenosylcobalamin-dependent diol dehydratase and upstream of cg2833, which encodes a cysteine synthase (Rey et al., 2003). The cg2831 coding region has a length of 846 nucleotides and is preceded by a ribosome-binding site (AGGAGG) eleven nucleotides in front of the predicted translational initiation codon. Downstream of the coding region, a rho-independent transcriptional terminator (ΔG=12·8 kcal mol1, 53·6 kJ mol1) was predicted (Combet et al., 2000). The deduced protein of cg2831 consists of 281 amino acids with a molecular mass of 30·8 kDa that corresponds very well to the experimentally determined size of the protein when isolated by DNA affinity purification (Fig. 4b). A putative helixturnhelix motif of the C-terminaleffector-domain type was identified at the C-terminal end of Cg2831 (amino acids 235256), which might be indicative of a transcriptional activator function (Brune et al., 2005). InterProScan of the Cg2831 amino acid sequence revealed the presence at the N-terminus (amino acids 6109) of a putative GAF domain (encountered in cGMP-specific phosphodiesterases, adenylyl cyclases and formate hydrogen lyases), which is generally capable of binding second messenger molecules, such as cGMP, and might be required for triggering the regulatory activation of the protein (Ho et al., 2000). Furthermore, weak similarity (32 % identity within a stretch of 77 amino acids) was detected to the ATP-dependent transcriptional activator MalT of E. coli that binds to the asymmetrical operator sequence GGGGA(T/G)GAGG in front of several genes belonging to the maltose regulon (Vidal-Ingigliardi et al., 1991). Although the similarity of Cg2831 from C. glutamicum to MalT from E. coli is low, it is nevertheless worth mentioning in view of earlier observations that the expression of the S-layer is dependent on the carbon source of the growth medium (Soual-Hoebeke et al., 1999). Additionally, the sequence motif GGGGATGGGT, showing striking similarity to the MalT operator, was identified upstream of the cspB promoter region (Fig. 4a), which is consistent with the putative function of Cg2831 as transcriptional activator.
Inactivation of cg2831 in C. glutamicum ATCC 13058 abolished S-layer formation
To analyse the putative regulatory role of Cg2831 in S-layer formation, the cg2831 coding region was disrupted in C. glutamicum ATCC 13058. This strain was chosen because it produces an ordered S-layer (Hansmeier et al., 2004), and is easily accessible to molecular genetic engineering. An internal fragment of the cg2831 gene was amplified by PCR and cloned into the pK18mob vector, which allowed gene disruption to be performed in C. glutamicum (Schäfer et al., 1994). Disruption of the cg2831 gene in the chromosome was confirmed by Southern hybridization (data not shown). Subsequently, expression of the cspB gene was analysed on the mRNA and protein level in the resulting mutant, designated C. glutamicum Intcg2831, and the corresponding wild-type strain.
The effect of cg2831 disruption on cspB expression was first measured on the mRNA level. Both C. glutamicum ATCC 13058 and C. glutamicum Intcg2831 were therefore cultivated in complex Luria broth medium, and aliquots of the cells were harvested during exponential growth and stationary phase of the cultures. Subsequently, total RNA was isolated, purified, and applied in real-time RT-PCR experiments, using the cspB-specific primers cspBLC1 and cspBLC2. These assays were repeated three times with independently grown cultures. It was found that the relative expression of the cspB gene in C. glutamicum Intcg2831 was drastically decreased, by a factor of approximately 10 000-fold, compared with the expression of the wild-type strain. Moreover, a dependence on the growth phase of cspB gene expression was detected in the wild-type strain, since cspB expression was reduced ∼2 000-fold in the stationary phase compared with expression during exponential growth of the culture. A similar expression pattern was detected for cg2831, although transcription was reduced only fivefold when the culture entered stationary phase. On the other hand, expression of the cspB gene in C. glutamicum Intcg2831 remained unaffected by changes of the growth phase. These observations might indicate that the cspB gene is mainly expressed during exponential growth of the culture.
To determine the amount of PS2 protein synthesized by C. glutamicum Intcg2831 and the wild-type, cell surface proteins were extracted from both strains and analysed by SDS-PAGE. Equal amounts of bacterial cells were harvested, and cell surface proteins were separated according to the protocol of Hansmeier et al. (2004). The proteins were further characterized by SDS-PAGE combined with MALDI-TOF MS and peptide mass fingerprinting (Fig. 5a). The disruption of cg2831 resulted in an almost complete loss of the PS2 protein band. However, small amounts of PS2 were detected by MALDI-TOF MS in the mutant strain C. glutamicum Intcg2831 (Fig. 5a).
|
To assess whether S-layer lattices were formed on the cell surface of C. glutamicum Intcg2831, AFM images of the respective cells were recorded in tapping mode by phase imaging contrast, as described above. The cell surface of C. glutamicum Intcg2831 displayed only very small patches of the hexagonal S-layer that were primarily located at the bacterial cell poles (Fig. 5b). These patches covered only about 1 % of the cell surface of C. glutamicum Intcg2831. Consequently, the Cg2831 protein plays an important role as transcriptional activator of cspB gene expression, resulting in an enhanced transcription of the cspB gene and the formation of an ordered S-layer lattice on the surface of C. glutamicum cells. The PS2 phenotype of C. glutamicum ATCC 13032 is apparently caused by the loss of a 5·97 kb chromosomal DNA fragment
Recently, we investigated the occurrence of the S-layer gene cspB in 28 C. glutamicum wild-type strains (Hansmeier et al., 2004). Only the sequenced type strain C. glutamicum ATCC 13032 was devoid of the cspB gene along with a downstream ORF encoding a putative Zn-dependent alcohol dehydrogenase. It is a well-known phenomenon that bacteria can lose the ability for S-layer formation under specific culture conditions (Blaser et al., 1987; Fujimoto et al., 1989; Sleytr, 1997). In the present study, we demonstrated that C. glutamicum ATCC 13032 has apparently lost the ability for S-layer formation due to the chromosomal deletion of a 5·97 kb DNA region. The deleted DNA region is flanked by a 7 bp direct repeat, suggesting that illegitimate recombination was responsible for gene loss. Illegitimate recombination processes using very short stretches of homologous DNA have been observed before in C. glutamicum. Mateos et al. (1996) have shown that random integration of cloning vectors into the chromosome of C. glutamicum is dependent on the presence of homologous nucleotide sequences of only 812 bp in length. This type of illegitimate recombination is detectable with very low frequencies in C. glutamicum, and it is postulated that this process occurs in a RecA-independent way by specific enzymes that recognize very short regions of DNA homology. Similar mechanisms of illegitimate recombination have been described in several micro-organisms (Farabaugh et al., 1978; Kumagai & Ikeda, 1991; Lopez et al., 1984; Yi et al., 1988).
Loss of S-layer gene expression has also been analysed in more detail in Campylobacter fetus strains. In both Camp. fetus TK and Camp. fetus 23B, the promoter region of the S-layer gene sapA was found to be truncated (Fujita et al., 1997; Tummuru & Blaser, 1992). A Chi-like sequence located upstream of the S-layer gene, and most likely recognized by the RecBCD system of the cell, might be responsible for inactivation of the S-layer gene (Tummuru & Blaser, 1993). Loss of S-layer gene expression results in variation of antigenicity of Camp. fetus, and thus contributes to the virulence of this pathogenic bacterium (Garcia et al., 1995). The current lack of knowledge about the function of the C. glutamicum S-layer makes it difficult to relate the loss of the S-layer locus to the fitness of this bacterium. C. glutamicum is commonly found in soil, and the S-layer might be associated with adhesion of exoenzymes and substrates or with other surface recognition processes (Beveridge et al., 1997). However, the loss of the S-layer locus might simply be the result of prolonged cultivation of the type-strain on synthetic medium since its discovery in 1957 and of its extensive use in the fermentation industry (Hermann, 2003; Sleytr & Sara, 1997). Thus, loss of the cspB gene region under favourable culture conditions might reflect a mechanism of deactivation of a non-required surface structure that is, moreover, synthesized in considerable amounts (Sleytr, 1997). Accordingly, it would be interesting to analyse original C. glutamicum ATCC 13032 isolates from the late 1950s for the presence of the cspB gene region.
Cg2831, a member of the LuxR regulator family, is an activating element for the transcription of the S-layer gene cspB
To gain insight into transcriptional regulation of S-layer formation in C. glutamicum, we mapped the cspB promoter region and identified a DNA-binding transcriptional regulator. The cspB promoter shared nine out of 12 nucleotides with the promoters of the hom and gltA genes that are of medium strength in C. glutamicum and allow only weak gene expression in the heterologous host E. coli (Eikmanns et al., 1994; Mateos et al., 1994). Only a very low basal level of cspB expression was detected in C. glutamicum Intcg2831, indicating that the S-layer gene is expressed by a rather weak promoter. On the other hand, electron microscopic images of C. glutamicum show that the complete cell surface can be covered by an ordered hexagonal S-layer lattice (Peyret et al., 1993). To cope with the high amount of S-layer proteins necessary to cover the complete bacterial cell surface, high expression of the S-layer gene is mandatory. Generally, a high expression level of the S-layer gene is achieved by using very stable mRNAs and additional regulatory factors, such as DNA-binding transcriptional regulators (Mignot et al., 2002, 2004; Vidgren et al., 1992). In C. glutamicum, a high expression level of the cspB gene is apparently obtained by the LuxR-type transcriptional regulator Cg2831 that binds to the cspB promoter region. LuxR-type regulators act as transcriptional activators and control a wide variety of functions in various biological processes, for instance in biofilm and spore formation, cell division, plasmid transfer and bacterial virulence (Cui et al., 2005; Fuqua et al., 1994; Guvener & McCarter, 2003; Schweizer, 1991). Database searches revealed significant amino acid similarities of Cg2831 to putative LuxR-type regulatory proteins of other species, but none of the regulators has hitherto been functionally analysed. A weak similarity of Cg2831 was also observed to the LuxR-type activator MalT of in E. coli (Boos & Shuman, 1998). Expression of the malT gene is subject to catabolite repression, and therefore dependent on the intracellular cAMP level. Likewise, Cg2831 contains an N-terminal GAF domain that might be necessary for activation of the regulatory protein by second messenger molecules. Additionally, a DNA motif with similarity to the MalT operator consensus sequence was identified upstream of the cspB promoter. A binding site upstream of a promoter is consistent with an activating function of the respective regulatory protein (Madan Babu & Teichmann, 2003).
The role of Cg2831 as transcriptional activator of S-layer gene expression was deduced from the characterization of the mutant strain C. glutamicum Intcg2831 by real-time RT-PCR, SDS-PAGE and AFM: (i) relative expression of the cspB gene was drastically decreased in C. glutamicum Intcg2831 compared with expression in the wild-type strain; (ii) SDS-PAGE showed a nearly complete loss of the PS2 protomer in the mutant strain C. glutamicum Intcg2831 in comparison to the protein profile of the C. glutamicum wild-type; (iii) AFM images of C. glutamicum Intcg2831 showed that only the bacterial cell poles exhibited small S-layer patches, covering in total approximately 1 % of the cell surface. Furthermore, real-time RT-PCR revealed that cg2831 and the cspB gene were expressed at higher rates during exponential growth than during stationary phase. In earlier studies, Soual-Hoebeke et al. (1999) have demonstrated that the amount of S-layer protein is dependent on the growth phase of C. glutamicum and is related to the environmental stimulus of the carbon source available within the growth medium. For instance, the use of lactate as carbon source results in an enhanced formation of the S-layer, and, as a consequence thereof, C. glutamicum cells are densely covered by an ordered S-layer lattice. On the other hand, the use of glucose as carbon source shows the contrary effect on S-layer formation, and the cells are only partially covered by a hexagonal S-layer. Since LuxR regulators are known to act in dependence on distinct environmental stimuli, such as growth condition, cell density and stress (Boos & Shuman, 1998; Suzuki et al., 2002), it is likely that Cg2831 is the regulatory element responsible for the observed differences in S-layer gene expression and formation.
From these results, it is apparent that the Cg2831 protein plays an important role as transcriptional activator of cspB gene expression, resulting in an enhanced transcription of the cspB gene and formation of an ordered S-layer lattice on the surface of C. glutamicum cells. Since genes orthologous with cg2831 are present in other corynebacterial genome sequences that do not encode S-layer protomers, a more global role of Cg2831 in regulation of gene expression in corynebacteria is likely (Brune et al., 2005). The exact physiological function of the transcriptional regulator Cg2831 and the regulatory link with the carbohydrate metabolism of C. glutamicum remain to be elucidated.
While this manuscript was going to press, Cramer et al., (2006) provided further information on the LuxR-type regulator Cg2831 that activates transcription of the cspB gene. These authors named this regulator RamA (regulator of acetate metabolism A) and showed convincingly that it activates transcription of a genetic network involved in several aspects of acetate metabolism in C. glutamicum. In addition, they identified a consensus binding DNA motif that resembles very much the palindromic sequences double-underlined in Fig. 4. The authors thank Professor Dr Leblon and Dr Houssin (Université Paris-Sud) for providing plasmid pCGL824. The construction and sequencing of the C. glutamicum fosmid library by IIT Biotech (Bielefeld) and financial support from the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 613) is greatly acknowledged. In addition, we thank Professor Dr Bernhard J. Eikmanns, Ulm University, for sharing unpublished information.References
Beveridge, T. J., Pouwels, P. H., Sara, M. & 22 other authors (1997). Functions of S-layers. FEMS Microbiol Rev 20, 99149.[Medline]
Blaser, M. J., Smith, P. F., Hopkins, J. A., Heinzer, I., Bryner, J. H. & Wang, W. L. (1987). Pathogenesis of Campylobacter fetus infections: serum resistance with high-molecular-weight surface proteins. J Infect Dis 155, 696706.[Medline]
Bonamy, C., Guyonvarch, A., Reyes, O., David, F. & Leblon, G. (1990). Interspecies electro-transformation in Corynebacteria. FEMS Microbiol Lett 54, 263269.[Medline]
Boos, W. & Shuman, H. (1998). Maltose/maltodextrin system of Escherichia coli: transport, metabolism and regulation. Microbiol Mol Biol Rev 62, 204229.
Boot, H. J. & Pouwels, P. H. (1996). Expression, secretion and antigenic variation of bacterial S-layer proteins. Mol Microbiol 21, 11171123.[CrossRef][Medline]
Brune, I., Brinkrolf, K., Kalinowski, J., Pühler, A. & Tauch, A. (2005). The individual and common repertoire of DNA-binding transcriptional regulators of Corynebacterium glutamicum, Corynebacterium efficiens, Corynebacterium diphtheriae and Corynebacterium jeikeium deduced from the complete genome sequences. BMC Genomics 6, 86.[CrossRef][Medline]
Capela, D., Barloy-Hubler, F., Gatius, M.-T., Gouzy, J. & Galibert, F. (1999). A high-density physical map of Sinorhizobium meliloti 1021 chromosome derived from bacterial artificial chromosome library. Proc Natl Acad Sci U S A 96, 93579362.
Castan, P., de Pedro, M. A., Risco, C., Valles, C., Fernandez, L. A., Schwarz, H. & Berenguer, J. (2001). Multiple regulatory mechanisms act on the 5' untranslated region of the S-layer gene from Thermus thermophilus HB8. J Bacteriol 183, 14911494.
Chami, M., Bayan, N., Dedieu, J.-C., Leblon, G., Shechter, E. & Gulik-Krzywicki, T. (1995). Organization of the outer layers of the cell envelope of Corynebacterium glutamicum: a combined freeze-etch electron microscopy and biochemical study. J Biol Cell 83, 219229.
Chami, M., Bayan, N., Peyret, J. L., Gulik-Krzywicki, T., Leblon, G. & Shechter, E. (1997). The S-layer protein of Corynebacterium glutamicum is anchored to the cell wall by its C-terminal hydrophobic domain. Mol Microbiol 23, 483492.[CrossRef][Medline]
Chu, S. & Trust, T. J. (1993). An Aeromonas salmonicida gene which influences A-protein expression in Escherichia coli encodes a protein containing an ATP-binding cassette and maps beside the surface array protein gene. J Bacteriol 175, 31053114.
Combet, C., Blanchet, C., Geourjon, C. & Deleage, G. (2000). NPS@: network protein sequence analysis. Trends Biochem Sci 25, 147150.[CrossRef][Medline]
Cui, Y., Chatterjee, A., Hasegawa, H., Dixit, V., Leigh, N. & Chatterjee, A. K. (2005). ExpR, a LuxR homolog of Erwinia carotovora subsp. carotovora, activates transcription of rsmA, which specifies a global regulatory RNA-binding protein. J Bacteriol 187, 47924803.
Cramer, A., Gerstmeir, R., Schaffer, S., Bott, M. & Eikmanns, B. J. (2006). Identification of RamA, a novel LuxR-type transcriptional regulator of genes involved in acetate metabolism of Corynebacterium glutamicum. J Bacteriol (in press).
Eikmanns, B. J., Thum-Schmitz, N., Eggeling, L., Ludtke, K. U. & Sahm, H. (1994). Nucleotide sequence, expression and transcriptional analysis of the Corynebacterium glutamicum gltA gene encoding citrate synthase. Microbiology 140, 18171828.[Abstract]
Farabaugh, P. J., Schmeissner, U., Hofer, M. & Miller, J. H. (1978). Genetic studies of the Lac repressor. VII. On the molecular nature of spontaneous hotspots in the lacI gene of Escherichia coli. J Mol Biol 126, 847857.[CrossRef][Medline]
Fernandez-Herrero, L. A., Olabarria, G. & Berenguer, J. (1997). Surface proteins and a novel transcription factor regulate the expression of the S-layer gene in Thermus thermophilus HB8. Mol Microbiol 24, 6172.[CrossRef][Medline]
Fisher, J. A., Smit, J. & Agabian, N. (1988). Transcriptional analysis of the major surface array gene of Caulobacter crescentus. J Bacteriol 170, 47064713.
Fujimoto, S., Umeda, A., Takade, A., Murata, K. & Amako, K. (1989). Hexagonal surface layer of Campylobacter fetus isolated from humans. Infect Immun 57, 25632565.
Fujita, M., Moriya, T., Fujimoto, S., Hara, N. & Amako, K. (1997). A deletion in the sapA homologue cluster is responsible for the loss of the S-layer in Campylobacter fetus strain TK. Arch Microbiol 167, 196201.[CrossRef][Medline]
Fuqua, W. C., Winans, S. C. & Greenberg, E. P. (1994). Quorum sensing in bacteria: the LuxR-LuxI family of cell density-response transcriptional regulators. J Bacteriol 176, 269275.
Garcia, M. M., Lutz-Wallace, C. L., Denes, A. S., Eaglesome, M. D., Holst, E. & Blaser, M. J. (1995). Protein shift and antigenic variation in the S-layer of Campylobacter fetus subsp. venerealis during bovine infection accompanied by genomic rearrangement of sapA homologues. J Bacteriol 177, 19761980.
Grant, S. G., Jessee, J., Bloom, F. R. & Hanahan, D. (1990). Differential plasmid rescue from transgenic mouse DNAs into Escherichia coli methylation-restriction mutants. Proc Natl Acad Sci U S A 87, 46454649.
Guvener, Z. T. & McCarter, L. L. (2003). Multiple regulators control capsular polysaccharide production in Vibrio parahaemolyticus. J Bacteriol 185, 54315441.
Hansmeier, N., Bartels, F. W., Ros, R., Anselmetti, D., Tauch, A., Pühler, A. & Kalinowski, J. (2004). Classification of hyper-variable Corynebacterium glutamicum surface-layer proteins by sequence analyses and atomic force microscopy. J Biotechnol 112, 177193.[CrossRef][Medline]
Hermann, T. (2003). Industrial production of amino acids by coryneform bacteria. J Biotechnol 104, 155172.[CrossRef][Medline]
Ho, Y.-S. J., Burden, L. M. & Hurley, J. H. (2000). Structure of the GAF domain, a ubiquitous signaling motif and a new class of cyclic GMP receptor. EMBO J 19, 52885299.[CrossRef][Medline]
Houssin, C., Nguyen, D. T., Leblon, G. & Bayan, N. (2002). S-layer protein transport across the cell wall of Corynebacterium glutamicum: in vivo kinetics and energy requirements. FEMS Microbiol Lett 217, 7179.[CrossRef][Medline]
Kahala, M., Sovijoki, K. & Palva, A. (1997). In vivo expression of the Lactobacillus brevis S-layer gene. J Bacteriol 179, 284286.
Kalinowski, J., Bathe, B., Bartels, D. & 24 other authors (2003). The complete Corynebacterium glutamicum ATCC 13032 genome sequence and its impact on the production of L-aspartate-derived amino acids and vitamins. J Biotechnol 104, 525.[CrossRef][Medline]
Katsumata, R., Ozaki, A., Oka, T. & Furuya, A. (1984). Protoplast transformation of glutamate-producing bacteria with plasmid DNA. J Bacteriol 159, 306311.
Krogh, A., Larsson, B., von Heijne, G. & Sonhammer, E. L. (2001). Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J Mol Biol 305, 567580.[CrossRef][Medline]
Kumagai, M. & Ikeda, H. (1991). Molecular analysis of the recombination junctions of λbio transducing phages. Mol Gen Genet 230, 6064.[CrossRef][Medline]
Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680685.[CrossRef][Medline]
Lopez, P., Espinosa, M., Greenberg, B. & Lacks, S. A. (1984). Generation of deletions in pneumococcal mal genes cloned in Bacillus subtilis. Proc Natl Acad Sci U S A 81, 51895193.
Madan Babu, M. & Teichmann, S. A. (2003). Functional determinants of transcription factors in Escherichia coli: protein families and binding sites. Trends Genet 19, 7579.[CrossRef][Medline]
Marchler-Bauer, A., Anderson, J. B., DeWeese-Sott, C. & 24 other authors (2003). CDD: a curated Entrez database of conserved domain alignments. Nucleic Acids Res 31, 383387.
Mateos, L. M., Pisabarro, A., Patek, M., Malumbres, M., Guerrero, C., Eikmanns, B. J., Sahm, H. & Martin, J. F. (1994). Transcriptional analysis and regulatory signals of the hom-thrB cluster of Brevibacterium lactofermentum. J Bacteriol 176, 73627371.
Mateos, L. M., Schäfer, A., Kalinowski, J., Martin, J. F. & Pühler, A. (1996). Integration of narrow-host-range vectors from Escherichia coli into the genomes of amino acid-producing cornyebacteria after intergenic conjugation. J Bacteriol 178, 57685775.
Merino, S., Alberti, S. & Tomas, J. M. (1994). Aeromonas salmonicida resistance to complement-mediated killing. Infect Immun 62, 54835490.
Messner, P. & Sleytr, U. B. (1992). Crystalline bacterial cell-surface layers. Adv Microb Physiol 33, 213275.[Medline]
Mignot, T., Mesnage, S., Couture-Tosi, E., Mock, M. & Fouet, A. (2002). Developmental switch of S-layer protein synthesis in Bacillus anthracis. Mol Microbiol 43, 16151627.[CrossRef][Medline]
Mignot, T., Couture-Tosi, E., Mesnage, S., Mock, M. & Fouet, A. (2004). In vivo Bacillus anthracis gene expression requires PagR as an intermediate effector of the AtxA signalling cascade. Int J Med Microbiol 293, 619624.[CrossRef][Medline]
Noonan, B. & Trust, T. J. (1995). Molecular analysis of an A-protein secretion mutant of Aeromonas salmonicida reveals a surface layer-specific protein secretion pathway. J Mol Biol 248, 316327.[Medline]
Patek, M., Eikmanns, B. J., Patek, K. & Sahm, H. (1996). Promoters from Corynebacterium glutamicum: cloning, molecular analysis and search for a consensus motif. Micobiology 142, 12971309.[Abstract]
Perkins, D. N., Pappin, D. J., Creasy, D. M. & Cottrell, J. S. (1999). Probability-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 20, 35513567.[CrossRef][Medline]
Peyret, J. L., Bayan, N., Joliff, G., Gulik-Krzywicki, T., Mathieu, L., Shechter, E. & Leblon, G. (1993). Characterization of the cspB gene encoding PS2, an ordered surface-layer protein in Corynebacterium glutamicum. Mol Microbiol 9, 97109.[Medline]
Pfaffl, M. (2001). A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29, 45.
Rey, D. A., Pühler, A. & Kalinowski, J. (2003). The putative transcriptional repressor McbR, member of the TetR-family, is involved in the regulation of the metabolic network directing the synthesis of sulfur containing amino acids in Corynebacterium glutamicum. J Biotechnol 103, 5163.[Medline]
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Sara, M. & Sleytr, U. B. (2000). S-layer proteins. J Bacteriol 182, 859868.
Schäfer, A., Tauch, A., Jäger, W., Kalinowski, J., Thierbach, G. & Pühler, A. (1994). Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chromosome of Corynebacterium glutamicum. Gene 145, 6973.[CrossRef][Medline]
Scheuring, S., Stahlberg, H., Chami, M., Houssin, C., Rigaud, J.-L. & Engel, A. (2002). Charting and unzipping the surface layer of Corynebacterium glutamicum with the atomic force microscope. Mol Microbiol 44, 675684.[CrossRef][Medline]
Schneitz, C., Nuotio, L. & Lounatma, K. (1993). Adhesion of Lactobacillus acidophilus to avian intestinal epithelial cells mediated by the crystalline bacterial cell surface layer (S-layer). J Appl Bacteriol 74, 290294.[Medline]
Schweizer, H. P. (1991). The agmR gene, an environmentally responsive gene, complements defective glpR, which encodes the putative activator for glycerol metabolism in Pseudomonas aeruginosa. J Bacteriol 173, 67986806.
Sleytr, U. B. (1997). I. Basic and applied S-layer research: an overview. FEMS Microbiol Rev 20, 512.
Sleytr, U. B. & Messner, P. (1983). Crystalline surface layers on bacteria. Annu Rev Microbiol 37, 311339.[CrossRef][Medline]
Sleytr, U. B. & Sara, M. (1997). Bacterial and archeal S-layer proteins: structurefunction relationships and their biotechnological applications. Trends Biotechnol 15, 2026.[CrossRef][Medline]
Soual-Hoebeke, E., de Sousa-D'Auria, C., Chami, M., Baucher, M. F., Guyonvarch, A., Bayan, N., Salim, K. & Leblon, G. (1999). S-layer protein production by Corynebacterium strains is dependent on the carbon source. Microbiology 145, 33993408.
Staden, R. (1996). The Staden sequence analysis package. Mol Biotechnol 5, 233241.[Medline]
Suzuki, K., Wang, X., Weilbacher, T., Pernestig, A. K., Melefors, O., Georgellis, D., Babitzke, P. & Romeo, T. (2002). Regulatory circuitry of the CsrA/CrsB and BarA/UvrY systems in Escherichia coli. J Bacteriol 184, 51305140.
Tummuru, M. K. R. & Blaser, M. J. (1992). Characterization of the Campylobacter fetus sapA promoter: evidence that the sapA promoter is deleted in spontaneous mutant strains. J Bacteriol 174, 59165922.
Tummuru, M. K. R. & Blaser, M. J. (1993). Rearrangement of sapA homologues with conserved and variable regions in Campylobacter fetus. Proc Natl Acad Sci U S A 174, 59165922.
Vidal-Ingigliardi, D., Richet, E. & Raibaud, O. (1991). Two MalT binding sites in direct repeat: a structural motif involved in the activation of all the promoters of the maltose regulons in Escherichia coli and Klebsiella pneumoniae. J Mol Biol 218, 323334.[CrossRef][Medline]
Vidgren, G., Palva, I., Pakkanen, R., Lounatmaa, K. & Palva, A. (1992). S-layer gene of Lactobacillus brevis: cloning by polymerase chain reaction and determination of the nucleotide sequence. J Bacteriol 174, 74197427.
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, 103119.[CrossRef][Medline]
Yi, T. M., Stearns, D. & Demple, B. (1988). Illegitimate recombination in an Escherichia coli plasmid: modulation by DNA damage and a new bacterial gene. J Bacteriol 170, 28982903.
Received 8 November 2005; revised 6 December 2005; accepted 12 December 2005.