Physical and genetic map of the Lactobacillus sakei 23K chromosome

Anne-Marie Dudez1, Stéphane Chaillou1, Lionel Hissler1, Régis Stentza,1, Marie-Christine Champomier-Vergès1, Carl-Alfred Alpertb,1 and Monique Zagorec1

Unité Flore Lactique et Environnement Carné, INRA, Domaine de Vilvert, F-78350 Jouy-en-Josas, France1

Author for correspondence: Monique Zagorec. Tel: +33 1 34 65 22 89. Fax: +33 1 34 65 21 05. e-mail: zagorec{at}diamant.jouy.inra.fr


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The Lactobacillus sakei 23K chromosome was analysed by pulsed-field gel electrophoresis after digestion with the restriction enzymes AscI, NotI and SfiI. The chromosome size was estimated to be 1845±80 kb. The use of I-CeuI, specific for rrn genes encoding 23S rRNAs, showed that seven rrn loci were present, on 40% of the chromosome. The seven rrn clusters were mapped and their orientation was determined, allowing the position of the replication origin to be estimated. Partial I-CeuI digestions were used to construct a backbone and the different restriction fragments obtained with AscI, NotI and SfiI were assembled to a physical map by Southern hybridization. Eleven L. sakei gene clusters previously identified were mapped, as well as 25 new loci located randomly on the chromosome and 11 regions flanking the rrn gene clusters. A total of 47 clusters were thus mapped on L. sakei chromosome. The new loci were sequenced, allowing the identification of 73 complete or incomplete coding sequences. Among these 73 new genes of L. sakei, the function of 36 could be deduced from their similarity to known genes described in databases. However, 10 genes had no homologues, 10 encoded proteins similar to proteins of unknown function and 17 were similar to hypothetical proteins.

Keywords: lactic acid bacteria, meat, PFGE

Abbreviations: LM-PCR, ligation-mediated polymerase chain reaction

The GenBank accession numbers for the sequences reported in this paper are given in Table 2 and the legend to Fig. 3.

a Present address: Institute of Food Research, Norwich Research Park, Norwich NR4 7UA, UK.

b Present address: Unité Ecologie et Physiologie du Système Digestif, INRA, Domaine de Vilvert, F78350 Jouy-en-Josas, France.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Lactobacillus sakei is a non-pathogenic bacterium that occurs naturally on meat and meat products. This species, like the closely related Lactobacillus curvatus, is used for the fermentation of meat and belongs to the group of facultative heterofermentative lactic acid bacteria. Both were classified in the phylogenic group of Lactobacillus casei, a species which is mostly used in the dairy industry (Sneath et al., 1986 ). During the last decade, several chromosomal genes encoding housekeeping, bacteriocin-production or stress-resistance functions of L. sakei have been cloned, sequenced and well characterized (for a recent review see Champomier-Vergès et al., 2001 ). However, no information was available on the size or structure of the L. sakei genome. The chromosome size of other lactic acid bacteria such as Lactobacillus plantarum, Oenococcus oeni and Lactococcus lactis, and of other bacteria important for the meat industry such as Staphylococcus carnosus, has been determined by pulsed-field gel electrophoresis (PFGE) using various restriction enzymes. It ranged from 1857–1932 kb for O. oeni (Zé-Zé et al., 1998 , 2000 ) to 3300–3400 kb for L. plantarum (Chevallier et al., 1994 ). The estimated chromosome size of S. carnosus was 2590 kb (Wagner et al., 1998 ). The size of the L. lactis IL1403 chromosome, estimated to be 2420 kb by PFGE, was precisely demonstrated to be 2365589 bp when the complete sequence was determined (Le Bourgeois et al., 1992 ; Bolotin et al., 2001 ).

The present study reports the determination of the chromosome size of a plasmid-cured L. sakei strain (Berthier et al., 1996 ). Several loci previously cloned from various L. sakei strains were used to construct a genetic map but additional probes were necessary to complete the map. For this purpose a collection of plasmids containing DNA fragments randomly cloned from the L. sakei chromosome (Leloup et al., 1997 ) was used, as well as additional L. sakei probes obtained by various strategies.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, media and growth conditions.
The plasmid-cured strain L. sakei 23K (Berthier et al., 1996 ), from our laboratory stock was used for chromosomal DNA preparation. The probes used for mapping experiments were either cloned from the L. sakei 23K chromosome or amplified by PCR from the chromosome of the same strain. Escherichia coli strains TG1 or DH5{alpha} (Sambrook et al., 1989 ) were used for plasmid propagation. L. sakei was grown in MRS medium (De Man et al., 1960 ) at 30 °C. E. coli was grown in LB medium (Sambrook et al., 1989 ) at 37 °C.

PFGE analysis.
Bacteria were harvested from MRS cultures when the OD600 reached 0·5. Cells were collected from 1 ml culture samples by centrifugation at room temperature, suspended in 0·5 ml TEE buffer (Tris/HCl 10 mM, pH 9·0, EDTA 100 mM, EGTA 10 mM), incubated for 10 min at 37 °C and embedded in an equal volume of 2 % low-melting-point agarose. Bacteria were lysed by incubating plugs for 2 h at 37 °C in TEE buffer containing lysozyme (5 mg ml-1) and N-lauroylsarcosine (0·05%) and then overnight at 55 °C in TEE buffer containing proteinase K (1 mg ml-1) and SDS (1%). Plugs were dialysed for 1 h at room temperature in TE buffer (Tris/HCl 10 mM, pH 8·0, EDTA 10 mM) containing PMSF (0·1 mM), then twice in TE buffer, and kept at at 4 °C in EDTA (0·5 M) until use. For restriction enzyme digestion the plugs were first dialysed against sterile water, then equilibrated for 3 h in 200 µl of specific buffer and at the temperature recommended by the supplier for each enzyme. Restriction enzymes (20 U, or 5 U for I-CeuI) were added and the plugs were incubated at 4 °C for 4 h (overnight for I-CeuI) and then overnight (2 h for I-CeuI) at the temperature recommended for each enzyme. The restriction enzymes AscI and I-CeuI, NotI and SfiI were from New England Biolabs, MBI Fermentas and Boehringer, respectively. After digestion, the plugs were rinsed for 1–3 h in Tris/HCl 10 mM, pH 8·0, EDTA 100 mM and then submitted to PFGE on a CHEF-DR III apparatus (Bio-Rad). PFGE was performed in 1% GTG SeaKem agarose in 0·5x TBE buffer (Sambrook et al., 1989 ). To separate the large DNA fragments, migration was carried out at 14 °C for 22 h at 6 V cm-1 and an angle of 120°. The switch time was 50 s to 90 s. Then, a second migration was performed for 13 h, with the same parameters except a switch time of 1 s to 12 s, allowing the separation of small fragments. The molecular mass markers used were yeast chromosomal DNA (2·03–194 kb) and {lambda} concatemers (48·5–727 kb). After migration, gels were stained in ethidium bromide and photographed under UV light, or analysed on a FluorImager (Molecular Dynamics), with a filter at 488 nm, after staining with Vistra-Green (Amersham) according to the recommendation of the manufacturer.

DNA manipulation and hybridization.
Plasmid DNA was prepared from E. coli by classical methods (Sambrook et al., 1989 ). Chromosomal DNA used for PCR amplification was prepared from L. sakei 23K by the method of Anderson & McKay (1983) . PCR amplification of DNA probes was performed on 1 µg chromosomal DNA, as previously described (Stentz et al., 1997 ). Southern hybridization experiments were performed on chromosomal DNA transferred to Nytran-N nylon membranes (Schleicher & Schuell). DNA probes were labelled by the use of the ECL kit from Amersham or with [{alpha}-32P]dCTP by random priming (Sambrook et al., 1989 ). When probes were PFGE restriction fragments, bands were excised from PFGE gels, extracted from agarose and ethanol-purified before labelling. For large fragments, DNA in agarose bands cut from gels was digested with EcoRI before extraction.

Ligation-mediated PCR of the rrn flanking regions.
The DNA regions flanking rrn loci were amplified by PCR as described by Prod’hom et al. (1998) . The 16S-specific primer OLS1 (5'-CCGCCACTCACTCAAATGTTTATCAATGAG-3') was derived from the 16S rRNA gene of L. sakei available in the GenBank database (accession number M58829) and the 5S-specific primer OLS2 (5'-GAAGGATACACCTGTTCCCATGCCGAACAC-3') was derived from a conserved region in the consensus sequence of an alignment of known 5S rRNA genes from lactobacilli (Lactobacillus brevis, X02026; L. casei partial 5S rRNA gene sequence, AF098108; L. plantarum, X12886; Lactobacillus delbrueckii, X15246; and Lactobacillus viridescens, X01950). The EcoRI linker was composed of the two primers OLS36 (5'-TAGCTTATTCCTCAAGGCACGAGC-3') and OLS37 (5'-AATTGCTCGTGC-3'). Chromosomal DNA (100 µg) was digested with EcoRI, purified, and subsequently ligated to the EcoRI linker (generated by OLS36+OLS37) by incubating for 1 h at 16 °C. PCR amplifications were then conducted with OLS36/OLS1 or OLS36/OLS2 pairs of primers and OLS36 or OLS1/2 alone as controls. The Expand Long Template PCR system (Roche Diagnostics) was used. PCR products were purified by Geneclean kit (Bio 101) and cloned in pGEMt-easy vector (Promega).

Cloning or PCR amplification of the ackB, gal, argF, pur, glpK, pdc and gpm clusters.
Cloning of the ackB partial sequence and upstream region was previously described by Stentz (1998) . Briefly, a 100 bp PCR product obtained with degenerate primers based on an alignment of several bacterial ack genes was used as probe to clone a 950 bp NsiI fragment in pBluescriptSK+II (pBSKII+, Stratagene). Further inverse PCR experiments, performed with primers deduced from the sequence of the NsiI fragment, allowed the amplification of two overlapping fragments corresponding to the upstream (2·3 kb) and downstream (4·8 kb) regions. The two strategies yielded together a 6890 bp region of the ackB cluster. The galETM cluster was cloned in pBSKII+ as a 2·4 kb EcoRI–HindIII DNA fragment which hybridized with a probe comprising the galT gene from L. casei (Bettenbrock & Alpert, 1998 ), yielding plasmid pRV522. The argF gene cluster was cloned in pBSKII+ as a 4·3 kb EcoRV fragment (plasmid pRV411) by hybridization to a 100 bp PCR probe obtained with degenerate primers 5'-TTC(T)ATGCAC(T)TGC(T)TTA(G)CC-3' and 5'-(A,T,C,G)CGG(A)TTT(C)TC(A,T,C,G)GCT(C)TC-3'. These two primers were based on conserved motifs of ornithine transcarbamylases. The purB, purFMN and glpK clusters were cloned in pBSKII+ as a 1·55 kb HindIII fragment (plasmid pRV30), a 2·3 kb ClaI fragment (pRV29) and a 0·4 kb HindIII fragment (pRV31), respectively. These plasmids were isolated as false-positive clones hybridizing to rbsK and rbsR probes (Stentz & Zagorec, 1999 ). The pdc cluster was isolated as a 940 bp PCR fragment. The gpm cluster was cloned by hybridization in pBSKII+, as a 3·8 kb EcoRV fragment, adjacent to a peptidase gene (plasmid pRV416, M.-C. Champomier-Vergès, unpublished results).

Construction of a mini-shotgun library.
Genomic DNA from cells harvested at the end of the exponential growth phase was diluted to a concentration of 200 ng µl-1 in ice-cold TE buffer and sonicated briefly. Several time-pulses were tested in order to produce several ranges of DNA fragment sizes (from 0·1–2 kb to 1–10 kb). DNA from these experiments was pooled, then treated with RNase A for 20 min at 37 °C, purified by phenol/chloroform/isoamyl alcohol extraction and precipitated with ethanol. End-repairing of sheared DNA was performed in several batches of 40 µl in T4 DNA polymerase buffer (Biolabs) containing 4 µg DNA. The end-repairing reaction was performed with dNTPs at final concentration of 0·075 mM and a mixture of T4 DNA polymerase (6 U)/Klenow enzyme (10 U) for 30 min at room temperature. The polymerization reaction was stopped by heating the samples at 75 °C for 15 min followed by chilling on ice for 15 min. T4 polynucleotide kinase (10 U) and ATP (final concentration 0·04 mM) were then added and the samples were incubated for 30 min at 37 °C. DNA was further purified by phenol/chloroform/isoamyl alcohol extraction and precipitated with ethanol before loading on agarose gel for size selection. The range from 1 to 2 kb was excised from the gel and purified with Geneclean purification kit. The blunt-ended DNA was then ligated to a 10-fold excess of BstXI adaptators (InVitrogen). The ligation mixture was loaded on to agarose gel to purify the 1–2 kb fraction from unligated adaptators or adaptor doublet. The 1–2 kb fragments were finally ligated to BstXI-digested pcDNA2.1 vector (InVitrogen) and transformed into E. coli DH5-{alpha}. Ten white clones were chosen randomly and named pRV322, pRV323, pRV326, pRV327, pRV328, pRV329, pRV330, pRV331, pRV370 and pRV371.

DNA sequencing.
All plasmids and PCR products were sequenced with BigDye terminators according to the manufacturer (Perkin-Elmer). For chromosomal DNA sequencing, genomic DNA was first denatured for 5 min at 100 °C and chilled on ice. The DNA was then treated with RNase A for 20 min at 37 °C prior to purification with phenol/chloroforme/isoamyl alcohol extraction and precipitation with ethanol. The DNA was then washed with ethanol (70%), dried at room temperature and solubilized in sterile water. Sequencing reaction mixtures contained 6 µg high-quality treated DNA, 20 pmol 25-mer primers and 16 µl BigDye terminator mix in a total volume of 40 µl. The cycle conditions were an initial denaturation at 95 °C for 5 min followed by 99 cycles (95 °C for 30 s, 55 °C for 20 s and 60 °C for 4 min). Excess dye terminators were removed by passage through a Sephadex G50 column and reaction mixtures were dried in a SpeedVac system.

Bioinformatics.
Assembling projects of all clusters were carried out with the phred/phrap/consed software package (Gordon et al., 1998 ; http://bozeman.mbt.washington.edu). Annotation of the various clusters was done with the Artemis annotation tool from the Sanger Centre (Rutherford et al., 2000 ; http://www.sanger.ac.uk/Software/artemis). These softwares were run on a LINUX workstation. DNA and protein properties were analysed with the STADEN-2000 software package and/or GCG programs via the genome server from the INRA. tRNA genes were detected with the program tRNAscan-SE (Lowe & Eddy, 1997 ; http://genome.wustl.edu/eddy/tRNAscan-SE).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Genome size estimation and determination of rrn loci copy number
The size of the genome was evaluated by PFGE of L. sakei 23K chromosomal DNA digested with the restriction enzymes AscI, NotI and SfiI, for which recognition sites are rare in the L. sakei genome, which has a G+C content of 42–44 mol% (Hammes & Vogel, 1995 ). The intron-encoded endonuclease I-CeuI, whose recognition site is located in the 23S rRNA gene in bacteria, was also used to determine the number of rrn operons. The results are presented in Fig. 1. Seven bands were observed after I-CeuI digestion, indicating that seven rrn operons are present on the L. sakei chromosome, two of which are presumably contiguous since the fragment C7 (5 kb) corresponds to the size of a classical bacterial rrn operon. Table 1 summarizes the size of the 12 AscI (A1 to A12), 11 NotI (N1 to N11), 11 SfiI (S1 to S11) and 7 I-CeuI (C1 to C7) restriction fragments obtained. Fragments N8 and S4 appeared broader and had a higher staining intensity relative to other bands. A computer-assisted analysis of peak area quantification indicated that these two bands were both doublets (data not shown); this was confirmed by hybridization experiments (see below). The doublets were named N8/N8bis and S4/S4bis, respectively. Thus, the genome size of strain 23K deduced from these values is 1845±80 kb.



View larger version (47K):
[in this window]
[in a new window]
 
Fig. 1. PFGE of AscI-, NotI-, SfiI- and I-CeuI-digested L. sakei 23K DNA. {lambda} concatemers used as size markers are shown on both sides of the gel and their size is indicated in kb. Running conditions in two steps and gel staining with Vistra green were as described in Methods.

 

View this table:
[in this window]
[in a new window]
 
Table 1. Fragment sizes of restriction enzyme digests with AscI, NotI, SfiI and I-CeuI of the L. sakei 23K chromosome

 
Construction of an I-CeuI backbone
The digestion of the L. sakei 23K chromosomal DNA with I-CeuI resulted in fewer fragments compared to the three other enzymes. Therefore, prior to complete mapping experiments, partial I-CeuI digestions were performed in order to construct a backbone of the chromosome and to organize the C1 to C7 fragments. After partial digestion a total of eight additional fragments (Ca to Ch) were observed (Fig. 2). From the sizes of C1 to C7 reported in Table 1, we could deduce that fragment Ca (1265 kb) could only be obtained with the combination of the doublet C1+C3 (1104 kb+170 kb) and/or the triplet C1+C4+C6 (1104 kb+128 kb+36 kb). According to this, the other fragments can correspond only to the combinations of C2+C3+C5 (331 kb+170 kb+98 kb) for Cb (590 kb); C2+C4+C5 (331 kb+128 kb+98 kb) for Cc (550 kb); C2+C4+C6 (331 kb+128 kb+36 kb) for Cd (485 kb); C2+C4 (331+128) for Ce (455 kb); C2+C5 (331 kb+98 kb) for Cf (420 kb); C3+C5 (170 kb+98 kb) for Cg (265 kb); and C4+C6 (128 kb+36 kb) for Ch (164 kb). Therefore, the results unequivocally link the fragments in the order -C1-C3-C5-C2-C4-C6-C1-.



View larger version (26K):
[in this window]
[in a new window]
 
Fig. 2. PFGE of I-CeuI-digested (1) or I-CeuI-partially digested (2) L. sakei 23K DNA. {lambda} concatemers are shown in lane 3. I-CeuI fragments C1 to C7 are indicated on the left; partial I-CeuI fragments are indicated on the right. Running conditions in two steps and gel staining with Vistra green were as described in Methods.

 
Because of the small size of C7 (5 kb) no doublet or triplet including C7 could be detected. However, the doublet C6+C7 would have been distinguished from C6 itself in our electrophoresis conditions; this indicates that C7 and C6 are not linked. Similarly, the doublet C1+C6 could not be detected but was probably not separated from C1 itself. C7 was subsequently positioned by hybridization after cloning of the rrn flanking regions (see below).

Identification of the rrn flanking regions
Eleven fragments corresponding to the flanking regions of the rrn operons were isolated in order to orientate these clusters with respect to the origin of replication (Cole & Saint-Girons, 1994 ). As described above, the backbone of the L. sakei 23K chromosome was first defined by the position of the rrn loci given by the I-CeuI cutting sites. To isolate the seven rrn flanking regions we used the technique of ligation-mediated PCR (LM-PCR) based on the use of an EcoRI-digested DNA ligated to an EcoRI adapter (see Methods). This adapter-ligated DNA was then used in long-range PCR experiments with either primer OLS1, pointing out of the 16S rRNA gene and thus designed for upstream regions, or primer OLS2, pointing out of the 5S rRNA gene, for downstream regions. Results of the LM-PCR and cloning experiments are shown in Fig. 3.



View larger version (28K):
[in this window]
[in a new window]
 
Fig. 3. Schematic representation of the LM-PCR amplification conducted to identify the rrn operon flanking regions. Lane 1, OLS36/OLS1 amplification products (upstream from the 16S rRNA gene); lane 2, {lambda}BstEII size marker; lane 3, OLS36/OLS2 amplification products (downstream from the 5S rRNA gene). Bands marked with an asterisk (*) indicate an atypical amplification product (OLS36/OLS36). Products which were successfully cloned are underlined with an arrow labelled with their respective plasmid name. Gene description is summarized in Table 2. The tRNA genes are shown with the one-letter amino acid code. The accession number of the insert of pRV340 is AF401673.

 
Only five of the seven PCR products obtained with OLS2 could be cloned. Unsuccessful cloning presumably resulted from a high instability of the PCR products in the host E. coli since several strategies (including various strains and low-copy-number cloning vectors) were used. The two other products were partially sequenced by direct sequencing of the PCR product and/or direct sequencing with chromosomal DNA as template. The seven OLS2-derived products were characterized as the rrn downstream regions including the C7 inter-operon region (pRV354, Fig. 3). Four of the L. sakei rrn downstream regions were revealed to contain tRNA clusters. A substantial level of duplication and rearrangement was observed between these four tRNA clusters, and 70% of the genetic code was represented in these tRNA genes (see Fig. 3). Therefore, it appeared necessary to further sequence one of the rrn downstream regions (pRV352) by using chromosomal DNA as template in order to identify a unique region for probing. The largest (7 kb) PCR product, which could not be cloned, was only partially sequenced since the 2 kb downstream from the 5S rRNA gene of this region offered enough unique non-repetitive sequence for probing. In order to locate the C7 band, the rrn inter-operon region of pRV354 was subcloned in pRV302, which was subsequently used as a probe.

LM-PCR with OLS1 also resulted in the amplification of seven products. However, only two products containing the rrn upstream regions could be cloned. The other products were then directly sequenced but only two of them were identified as an rrn upstream region, whereas the others were atypical amplification products (OLS36/OLS36). More LM-PCR experiments were conducted with OLS1 and DNA digested with other restriction enzymes but they did not result in the characterization of additional 16S rRNA gene upstream regions. Sequencing of the 2·5 kb OLS1-derived PCR product or chromosomal DNA sequencing of this region was tedious and only 500 bp could be identified upstream of the 16S rRNA gene. Thus, only five of the rrn upstream regions could be identified out of seven, including the small C7 region in between two close rrn loci. These regions were subsequently included in the hybridization experiments to confirm the orientation of the rrn operons determined with the use of the seven rrn downstream regions. Furthermore, the flanking regions of the rrn genes were sequenced, allowing the determination of new genetic clusters (see below).

Construction of the gene map
To assemble the AscI-, NotI- and SfiI-digested fragments on the I-CeuI backbone, we hybridized the digested fragments with unique gene probes. This strategy helped to localize genetic markers in addition to the construction of the whole genome map. For this study we used 11 gene clusters which had been previously cloned and characterized from L. sakei 23K or other strains. The probes specific for these loci were either plasmids, when the DNA originated from L. sakei 23K, or PCR fragments amplified from the L. sakei 23K chromosome with primers designed from the published sequences when these were derived from other strains. The lacLM locus (Obst et al., 1995 ) could not be mapped as several bands hybridizing to this probe were detected. In addition, we used 10 randomly cloned fragments from a mini-shotgun library of L. sakei 23K, 15 loci obtained from our laboratory plasmid collection or by collection of PCR products of L. sakei 23K, and the 11 probes corresponding to rrn flanking regions. Details of the gene probes used for this purpose, with data summarizing sequencing results of new loci characterized in this study, are listed in Table 2.


View this table:
[in this window]
[in a new window]
 
Table 2. Description of the probes used for the construction of the genetic map of the L. sakei 23K chromosome

 
Results of hybridization with genetic markers
All restriction fragments gave hybridization signals with at least one of the 47 gene probes, except fragments S11, S7, A12 and N11. These small fragments were then excised and purified from the PFGE gel and used directly as probes. In addition fragments N3 and S4 were treated similarly to confirm that band S4 was a doublet. Analysis of these hybridization assays led us to link unambiguously most of the various AscI, NotI, SfiI and I-CeuI fragments. These results also demonstrated that the L. sakei 23K chromosome is circular, as commonly observed in most other prokaryotic organisms. However, the true order of the bands A11/A12, A5/A8 and S9/S10 could not be determined since they could exchange their position with each other. Hybridization with I-CeuI fragments was performed with only some of the gene probes. Nevertheless the results confirmed the order -C1-C3-C5-C2-C4-C6-C1- already determined with the I-CeuI partial digestion. Hybridization with pRV302, which includes the rrn inter-operon region of the C7 fragment, unequivocally located the rrn doublet between C5 and C2. It should be noted that hybridization with pRV302 gave some substantial cross-hybridization signals with other bands such as C2. This phenomenon can be explained by the relative similarity of the rrn inter-operon region with some of the 16S rRNA gene upstream regions characterized in this study. Indeed, within the 500 bp of the rrn inter-operon region, about 170 bp showed 50–70% identity to the four 16S rRNA gene upstream regions identified by LM-PCR. None of the six probes specific for the rrn downstream regions hybridized with C2 whereas two of them (labM and labO) hybridized with C1. These data together with the hybridization results obtained for the other rrn-specific probes revealed that the rrn operons are directed outward in both directions of the C2 fragments, and then proceed in the same direction on each side to the C1 fragment. It can thus be concluded from this observation that the replication origin of the L. sakei 23K chromosome is located within the region covered by the C2 fragment. Additional observations based on the hybridization pattern of our gene collection support this view. Indeed, about 50% of the genetic markers chosen for this study hybridized within the C5-C7-C2-C4 region, which covers only 30% of the genome. Most of these probes consisted of housekeeping genes or loci previously characterized, or clones from our laboratory collection which were studied for their assignment to physiological functions in L. sakei. In contrast, clones representing randomly picked-up clusters (clones from the mini-shotgun library, or from a previous random plasmid collection; Leloup et al., 1997 ), were found scattered all over the rest of the genome map. Only two gene probes out of thirty-six hybridized with N3, suggesting a region of lower gene-density compared to the C2 region located opposite N3.

As described in Table 2, we could identify duplicated acetate kinase genes. The two paralogues, ackA and ackB, which display only 65% identity at the DNA level, did not show cross-hybridization under stringent hybridization conditions and were therefore localized at two distinct positions on the genome map.

Furthermore, the annotation of plasmid pRV3002 indicated the presence of an IS3-like insertion element. An internal fragment of the IS was amplified by PCR and used subsequently as probe in order to estimate the copy number of recurrent elements. The IS3-like element hybridized with three AscI and three I-CeuI fragments (A1, A2 and A4; C1, C2 and C3), four NotI fragments (N2, N4, N7 and N11) and five SfiI fragments (S1, S2, S4, S5 and S6). According to the hybridization pattern of this experiment, the IS3-like element is located in at least five different positions on the genome.

Positioning of AscI, NotI, SfiI and I-CeuI skeletons
Some of the AscI, NotI and SfiI restriction sites were placed on the I-CeuI backbone by using double digestion of C2, C3 or C5 fragments. The results are shown in Table 3. Some of the double-restricted fragments could be combined unambiguously to produce AscI, NotI or SfiI fragments overlapping I-CeuI bands. Bands corresponding to N7, N8 and N11 could be identified amongst the fragments produced by NotI digestion of C2. Moreover, the 150 kb fragment could be combined with the 20 kb NotI digestion of C5 to give a fragment of 170 kb which could be N5. Indeed, the fragment order established by gene probe hybridization indicated that C2 and C5 are contiguous and that N5 might overlap the two I-CeuI bands. Similarly, the 85 kb fragment obtained with NotI digestion of C5 could be combined with the 50 kb fragment of NotI-cut C3 band to produce a fragment of 135 kb which could be N6. Band N8bis could be identified in the NotI digestion products of C4. Another 5 kb product could be combined with the 20 kb fragment obtained with the NotI digestion of C2 to give a 25 kb fragment which could be N11 (located within the C2–C4 region according to hybridization results). The remaining 70 kb fragment resulting from NotI digestion of C4 is presumably a fraction of the broad N1 fragment. Also, the remaining 130 kb fragment of NotI digestion products of C3 is presumably a fraction of the broad N2 fragment. Bands A10, A11 and A12 could be identified in the AscI digest of C5, together with a 10 kb fragment which is a part of either A7 or A1 (see Table 1). Finally, bands S5 and S11 could be identified in the SfiI digest of C2. In the same digest, a 75 kb fragment appeared broader and with a higher staining intensity than the 130 kb S5 band, indicating that it might be a doublet or a triplet. This 75 kb band could correspond to S6 and to other partial bands of higher size flanking the C2 region (such as S4 and S2; see Table 1). From these data, the position of AscI, NotI, SfiI and I-CeuI skeletons could be precisely determined. Consequently, the physical and genetic map of L. sakei 23K strain was constructed (Fig. 4). The origin of replication was arbitrarily placed at the beginning of band S6, which is located in middle of band C2.


View this table:
[in this window]
[in a new window]
 
Table 3. Size (kb) and analysis of fragments generated by digestion of the I-CeuI fragments with NotI, AscI or SfiI

 


View larger version (50K):
[in this window]
[in a new window]
 
Fig. 4. Physical and genetic map of the L. sakei 23K chromosome. The restriction fragments are designated as indicated in the text and Table 1. The gene clusters are described in Table 2. The relative order of A11/A12, S9/S10 and A5/A8 is not known. The chromosome map was oriented based on the orientation of rrn operons. The scale is in kb.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The results reported here allowed the construction of a physical and genetic map of the L. sakei chromosome. The size of the chromosome was estimated to be 1845±80 kb. Several gene probes were necessary to assemble the various restriction fragments. Among these gene probes, 11 were derived from previously characterized DNA regions of L. sakei, and the others were new DNA fragments originating from L. sakei. These 36 new loci, corresponding to approximately 60 kb, were sequenced; coding sequences were searched and analysed, generating a total of 73 partial or complete new genes of L. sakei. For half of them (36) a putative function was assigned, based on their similarity to known genes. As mentioned above, we identified a second acetate kinase gene, 65% identical to the ackA gene previously described (Stentz & Zagorec, 1999 ). Other genes involved in carbon metabolism were identified, such as scrA encoding EIIscr of the phosphotransferase system and responsible for sucrose transport, a gluconate permease gene (gntP) and the galETM operon involved in galactose catabolism. The presence of these genes confirmed previous physiological features of L. sakei sugar utilization (Lauret et al., 1996 ). However, a partial putative glpK gene encoding glycerol kinase was found, although L. sakei does not grow easily on this carbon source (Stentz et al., 2001 ). Also, the presence of a pyrAB gene encoding one of the two subunits of carbamoyl-phosphate synthase was in accordance with recent data obtained on uracil prototrophy of L. sakei, showing that an active pyrimidine biosynthetic pathway exists in this species (Marceau et al., 2001 ). Among the newly identified gene probes used in this study, a mutated gene cluster putatively encoding enzymes of arginine biosynthesis was also found. As L. sakei is auxotrophic for arginine, it is likely that this operon is not functional. Indeed, the putative xasA gene located in the operon has a frameshift mutation, which was confirmed by the sequencing of several independent PCR fragments. As only the 3'-end of argF was identified in this study, we cannot exclude that additional mutations in the promoter region or the upstream part of argF are also present.

Ten coding sequences had no homologies with any proteins from databases, 10 were similar to proteins of unknown function, and 17 had homologies with hypothetical proteins such as ABC transporters, hypothetical membrane proteins or transcriptional regulators.

One IS3-like element was observed in our plasmid library. It is present in at least five copies in the chromosome. Some other IS3-like transposons have already been described in L. sakei as mobile elements responsible for mutations in bacteriocin-production operons (Skaugen & Nes, 1994 , 2000 ). Further studies would be necessary to establish the exact copy number of these elements in the L. sakei chromosome. From all the DNA sequences used in this work, a mean G+C content of the chromosome of 42 mol% was found, with a minimal value of 35·4 mol% and a maximal value of 50·1 mol%, which is in accordance with the G+C content previously estimated (Hammes & Vogel, 1995 ).

Finally, the observation of seven copies of rrn clusters in a 1845 kb genome might indicate that L. sakei could be designated as a fast-growing organism according to Krawiec & Riley (1990) . This high copy number of rrn clusters might be linked to the bacterium’s ability to colonize the meat environment, hence reflecting an ecological and physiological role for rrn copies as was demonstrated for E. coli (Condon et al., 1995 ) and soil bacteria (Klappenbach et al., 2000 ). The upstream and downstream regions flanking the rrn gene clusters were sequenced. Four of the seven regions located downstream from the 5S rRNA genes were shown to encode tRNAs. In total 45 tRNAs were identified, corresponding to 19 of the 20 amino acids.

The genetic map, as well as the sequence of the 47 loci used in this study, will be of great help for the final assembly of the sequences for the genome sequencing project that has been undertaken in our laboratory.


   ACKNOWLEDGEMENTS
 
We would like to thank Véronique Martin and Christophe Caron for help, suggestions and setup of the LINUX workstation, Christian Hertel for the gift of dnaK probe and Marie-Claude Muller for her advice with PFGE.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Anderson, D. G. & McKay, L. L. (1983). Simple and rapid method for isolating large plasmid DNA from lactic streptococci. Appl Environ Microbiol 46, 549-552.[Medline]

Berthier, F., Zagorec, M., Champomier-Vergès, M. C., Ehrlich, S. D. & Morel-Deville, F. (1996). Efficient transformation of Lactobacillus sake by electroporation. Microbiology 142, 1273-1279.

Bettenbrock, K. & Alpert, C. A. (1998). The gal genes for the Leloir pathway of Lactobacillus casei 64H. Appl Environ Microbiol 64, 2013-2019.[Abstract/Free Full Text]

Bolotin, A., Wincker, P., Mauger, S., Jaillon, O., Malarme, K., Weissenbach, J., Ehrlich, S. D. & Sorokin, A. (2001). The complete genome sequence of the lactic acid bacterium Lactococcus lactis ssp. lactis IL1403. Genome Res 11, 731-753.[Abstract/Free Full Text]

Champomier-Vergès, M. C., Chaillou, S., Cornet, M. & Zagorec, M. (2001). Lactobacillus sakei: recent developments and future prospects. Res Microbiol 152, 839-848.[Medline]

Chevallier, B., Hubert, J.-C. & Kammerer, B. (1994). Determination of chromosome size and number of rrn loci in Lactobacillus plantarum by pulse-field gel electrophoresis. FEMS Microbiol Lett 120, 51-56.[Medline]

Cole, S. T. & Saint-Girons, I. (1994). Bacterial genomics. FEMS Microbiol Rev 14, 139-160.[Medline]

Condon, C., Liveris, D., Schwartz, L. & Squires, C. L. (1995). rRNA operon multiplicity in Escherichia coli and the physiological implications of rrn inactivation. J Bacteriol 177, 4152-4156.[Abstract]

De Man, J. C., Rogosa, M. & Sharpe, M. E. (1960). A medium for the cultivation of lactobacilli. J Appl Bacteriol 23, 130-135.

Gordon, D., Abajian, C. & Green, P. (1998). Consed: a graphical tool for sequence finishing. Genome Res 8, 198-202.

Hammes, W. P. & Vogel, R. F. (1995). The genus Lactobacillus. In The Genera of Lactic Acid Bacteria , pp. 18-54. Edited by B. J. B. Wood & W. H. Holzapfel. Glasgow, UK:Blackie Academic & Professional.

Klappenbach, J. A., Dunbar, J. & Schmidt, T. M. (2000). rRNA operon copy number reflects ecological strategies of bacteria. J Bacteriol 66, 1328-1333.

Knauf, H. J., Vogel, R. F. & Hammes, W. P. (1992). Cloning, sequence, and phenotypic expression of katA, which encodes the catalase of Lactobacillus sake LTH677. Appl Environ Microbiol 58, 832-839.[Abstract]

Krawiec, S. & Riley, M. (1990). Organization of the bacterial chromosome. FEMS Microbiol Rev 54, 502-539.

Lauret, R., Morel-Deville, F., Berthier, F., Champomier-Vergès, M. C., Postma, P., Ehrlich, S. D. & Zagorec, M. (1996). Carbohydrate utilization in Lactobacillus sake. Appl Environ Microbiol 62, 1922-1927.[Abstract]

Le Bourgeois, P., Lautier, M., Mata, M. & Ritzenthaler, P. (1992). Physical and genetic map of the chromosome of Lactococcus lactis subsp. lactis IL1403. J Bacteriol 174, 6752-6762.[Abstract]

Leloup, L., Ehrlich, S. D., Zagorec, M. & Morel-Deville, F. (1997). Single-crossover integration in the Lactobacillus sake chromosome and insertional inactivation of the ptsI and lacL genes. Appl Environ Microbiol 63, 2117-2123.[Abstract]

Lowe, T. M. & Eddy, S. R. (1997). tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res 25, 955-964.[Abstract/Free Full Text]

Malleret, C., Lauret, R., Ehrlich, S. D., Morel-Deville, F. & Zagorec, M. (1998). Disruption of the sole ldhL gene in Lactobacillus sakei prevents the production of both L- and D-lactate. Microbiology 144, 3327-3333.[Abstract]

Marceau, A., Méra, T., Zagorec, M. & Champomier-Vergès, M. C. (2001). Protein expression under uracil privation. FEMS Microbiol Lett 200, 49-52.[Medline]

Morel-Deville, F., Fauvel, F. & Morel, P. (1998). Two-component signal-transducing systems involved in stress responses and vancomycin susceptibility in Lactobacillus sakei. Microbiology 144, 2873-2883.[Abstract]

Obst, M., Meding, E. R., Vogel, R. F. & Hammes, W. P. (1995). Two genes encoding the ß-galactosidase of Lactobacillus sake. Microbiology 141, 3059-3066.[Abstract]

Prod’hom, G., Lagier, B., Pelicic, V., Hance, A. J., Gicquel, B. & Guilhot, C. (1998). A reliable amplification technique for the characterization of genomic DNA sequences flanking insertion sequences. FEMS Microbiol Lett 158, 75-81.[Medline]

Rutherford, K., Parkhill, J., Crook, J., Horsnell, T., Rice, P., Rajandream, M.-A. & Barrell, B. (2000). Artemis: sequence visualization and annotation. Bioinformatics 16, 944-945.[Abstract]

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

Schmidt, G., Hertel, C. & Hammes, W. P. (1999). Molecular characterization of the dnaK operon of Lactobacillus sakei LTH681. Syst Appl Microbiol 22, 321-328.[Medline]

Skaugen, M. & Nes, I. F. (1994). Transposition in Lactobacillus sake and its abolition of lactocin S production by insertion of IS1163, a new member of the IS3 family. Appl Environ Microbiol 60, 2818-2825.[Abstract]

Skaugen, M. & Nes, I. F. (2000). Transposition in Lactobacillus sakei: inactivation of a second lactocin S operon by the insertion of IS1520, a new member of the IS3 family of insertion sequences. Microbiology 146, 1163-1169.[Abstract/Free Full Text]

Sneath, P. H., Mair, N. S., Sharpe, M. E. & Holt J. G. (editors) (1986). Bergey’s Manual of Systematic Bacteriology, vol 2. Baltimore: Williams & Wilkins.

Stentz, R. (1998). Etude de la régulation globale du métabolisme des sucres chez Lactobacillus sakei. Thèse de doctorat de l’Université Paris VII.

Stentz, R. & Zagorec, M. (1999). Ribose utilization in Lactobacillus sakei: analysis of the regulation of the rbs operon and putative involvement of a new transporter. J Mol Microbiol Biotechnol 1, 165-173.[Medline]

Stentz, R, Lauret, R., Ehrlich, S. D., Morel-Deville, F. & Zagorec, M. (1997). Molecular cloning and analysis of the ptsHI operon in Lactobacillus sake. Appl Environ Microbiol 63, 2111-2116.[Abstract]

Stentz, R., Cornet, M., Chaillou, S. & Zagorec, M. (2001). Adaptation of Lactobacillus sakei to meat: a new regulatory mechanism of ribose utilization? Lait 81, 131-138.

Wagner, E., Doskar, J. & Götz, F. (1998). Physical and genetic map of the genome of Staphylococcus carnosus TM300. Microbiology 144, 509-517.[Abstract]

Zé-Zé, L., Tenreiro, R., Brito, L., Santos, M. A. & Paveia, H. (1998). Physical map of the genome of Oenococcus oeni PSU-1 and localization of genetic markers. Microbiology 144, 1145-1156.[Abstract]

Zé-Zé, L., Tenreiro, R. & Paveia, H. (2000). The Oenococcus oeni genome: physical and genetic mapping of strain GM and comparison with the genome of a ‘divergent’ strain, PSU-1. Microbiology 146, 3195-3204.[Abstract/Free Full Text]

Zuñiga, M., Champomier-Vergès, M. C., Zagorec, M. & Pérez-Martínez, G. (1998). Structural and functional analysis of the gene cluster encoding the enzymes of the arginine deiminase pathway of Lactobacillus sake. J Bacteriol 180, 4154-4159.[Abstract/Free Full Text]

Received 30 July 2001; revised 8 October 2001; accepted 12 October 2001.