INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Lactobacillus helveticus is a homofermentative, thermophilic lactic acid bacterium used commercially for the manufacture of certain cheeses. It has GRAS (generally recognized as safe) status and could therefore be considered either as a suitable vehicle for genetic modification or as a cell factory for the synthesis of novel proteins. A prerequisite for any genetic manipulation is a method for introducing novel DNA sequences into the host bacteria. However, there is as yet no generalized method for transforming L. helveticus. Successful electroporation has been reported for certain strains, for example, CNRZ32 (2), isolated from whey from an artisan cheese starter (16); CNRZ1340, a derivative of NCDO766 which was isolated from a Finnish cheese starter culture (8); and S36.2, originally isolated from an Italian reggiano cheese starter culture (6). However, the transformation efficiency is low and the results are not always reproducible. In our laboratory, for example, transformants of strain CNRZ32 with plasmids pGK12 (10) and pSA3 (7) have been produced by using strain CNRZ32 obtained from Bhowmik and Steele's laboratory (2). By contrast, we were unable to effect electrocompetence in strain CNRZ32 obtained directly from the original culture collection. Barriers to successful electroporation may include restriction modification systems or physical features such as the thick gram-positive cell wall, which can be surrounded by a proteinaceous S-layer (13).

An alternative method for the introduction of novel DNA into L. helveticus exists. Conjugal transfer of certain broad-host-range (Tra⁺) plasmids can take place between species of mesophilic and thermophilic lactic acid bacteria (11, 17, 18) and can be used to mobilize a recombinant plasmid into strains of L. helveticus (20) by a delivery system whose utility had been demonstrated for lactococci (15). The success of this system for cloning depended on the generation of a stable cointegrate between the recombinant plasmid and the transmissible plasmid in order to overcome the low frequency of intergeneric transfer. In L. helveticus, a recombinant plasmid could replicate as a cointegrate and express a cloned bglA gene (from Bacillus amyloliquefaciens [4]). It was noted, however, that successful transfer of the cloned gene depended on its orientation when cloned into the EcoRI site of pSA3. In addition, the phospho--galactosidase gene (pbg) from Lactobacillus casei was also subcloned in both orientations into the PstI site of plasmid pSA3 (19); cotransfer of these recombinant plasmids between lactococci was detectable at a low frequency (up to ca. 10⁴) when mobilized by transmissible plasmid pVA797. Stable plasmid cointegrates were never recovered, however, and no transfer to L. helveticus could be detected (19, 20).

Recently, Mills et al. (14) noted the potential of oriT-containing vectors as delivery systems for the genetic modification of lactic acid bacteria, pointing out that plasmid size would not influence the outcome of a mating. The aim of the present study was to investigate the facility with which a pIP501-derived recombinant plasmid cointegrate could be transferred to a range of strains of L. helveticus and to assess the expression and stability of the cloned gene in different strains.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains, plasmids, and culture media. The bacteria and plasmids used during this investigation are listed in Tables 1 and 2, respectively. Strains of L. helveticus were grown in MRS medium (CM359 and CM361; Oxoid, Unipath Ltd., Basingstoke, United Kingdom) at 41°C under anaerobic conditions with a BBL GasPak anaerobic system (Becton Dickinson, Cockeysville, Md.). Strains of Lactococcus lactis subsp. lactis were grown in M17 medium (CM817; Oxoid) solidified with 1% (wt/vol) Oxoid Agar No. 1 when necessary and containing glucose (1% [wt/vol]; GM17) at 30°C in an atmosphere containing 5% (vol/vol) CO₂. GMYEA (glucose-milk-yeast extract agar) consisted of 1% (wt/vol) glucose, 5% (wt/vol) sterile skim milk powder (L31; Oxoid), and 1% (wt/vol) yeast extract (L21; Oxoid), solidified with 1% Oxoid Agar No. 1. Antibiotics were purchased from Sigma and used at the following concentrations, in micrograms per milliliter: L. helveticus, erythromycin at 2.5 or 5 and chloramphenicol at 5; for L. lactis subsp. lactis, erythromycin at 5, chloramphenicol at 5, streptomycin at 1,000, rifampin at 20, and spectinomycin at 200. Detection of -glucanase activity was performed by observing the hydrolysis of lichenin incorporated into MRS agar (0.2% [wt/vol] [5, 20]). The lichenin substrate and the Congo red that was used for detection of its hydrolysis (0.1% [wt/vol] aqueous solution) were purchased from Sigma.

Conjugation experiments. Intergeneric matings between lactococci and lactobacilli were performed as follows. An overnight culture of donor lactococci was diluted 1/50 into fresh GM17 with antibiotic selection and incubated at 30°C for 6 h. The recipient strains of L. helveticus were diluted 1/20 from an overnight culture and incubated for 6 h at 37°C. Bacteria from 5 ml of a donor culture were trapped on a membrane filter (type WCN; diameter, 47 mm, pore size, 0.45 µm; Whatman Ltd., Maidstone, United Kingdom) and washed by passing 10 ml of diluent (quarter-strength Ringer solution; BR52; Oxoid) through the filter to remove the selective antibiotic. Five milliliters of recipient bacteria was then trapped on the membrane, which was transferred aseptically to the GMYEA surface and maintained anaerobically at 37°C overnight. The cells were harvested by vortex mixing the filter (30 s with intermittent manual reciprocatory shaking) in 9 ml of diluent, and dilutions were spread plated onto appropriate media to determine the donor count (GM17 agar containing erythromycin and chloramphenicol at 30°C in 5% CO₂), recipient count (MRS agar maintained anaerobically at 41°C), transconjugant count (MRS agar containing chloramphenicol and maintained anaerobically at 41°C), and putative cointegrates (MRS agar containing erythromycin, with or without chloramphenicol and maintained anaerobically at 41°C). Colonies were counted after 3 days of incubation. The presence of plasmid cointegrates was confirmed by examining lysates by agarose gel electrophoresis. The plasmid harboring the -glucanase gene was identified by DNA hybridization with a probe for the gene obtained by PCR with primers based on the published gene sequence (GenBank accession no. M15674, coordinates 301 to 318 and the complement of 1310 to 1293). The PCR product was labelled by enhanced chemiluminescence (Amersham International plc, Little Chalfont, United Kingdom); Southern blotting and hybridization were performed as described by the manufacturer.

Plasmid stability. Transconjugant strains of L. helveticus were grown at 37°C in the presence of erythromycin as a selective antibiotic. A 1/1,000 dilution was then made into fresh, antibiotic-free MRS broth with the temperature adjusted to 37, 40, or 42°C in water baths. At 24-h intervals, samples were removed, dilutions were spread plated onto MRS agar with either erythromycin or MRS agar containing lichenin (for recombinant strains), and the plates were maintained anaerobically at 37°C for 48 h. For lichenin-containing agar, the colonies were counted, marked, and then transferred onto MRS agar with erythromycin to score for retention of the erythromycin resistance (Erm^r) marker before the plate was scored directly for colonies retaining -glucanase activity.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Transfer of a cointegrate plasmid to strains of L. helveticus. The range of strains of L. helveticus to which a plasmid cointegrate (pSA3b6::pVA797) could be exconjugated from L. lactis subsp. lactis LM2345 is shown in Table 3. A total of seven of nine strains were apparently able to act as recipients for the plasmid. (Note that in one instance where transfer was not detected, the viable counts for both the donor and the recipient bacteria were consistently low, which suggested that there was mutual antagonism between the donor and the recipient.) The putative transconjugants had acquired simultaneously both Erm^r and chloramphenicol resistance (Cmp^r). The presence of new plasmid bands in the Lactobacillus transconjugants was confirmed by agarose gel electrophoresis and DNA hybridization (Fig. 1). The mobility of the plasmid band giving the signal when the bglA gene PCR product was used as a probe was indistinguishable for the lactococcal donor strain and the L. helveticus recipient strains. (Faint bands also visible on the lanes may represent spontaneous deletions of the cointegrate.) Similar results were obtained for other transconjugants (data not shown). The ability of transconjugants to hydrolyze lichenin was confirmed (Fig. 2).

View larger version (34K): [in this window] [in a new window]   FIG. 1.   Confirmation of the presence of the plasmid pSA3b6::pVA797 cointegrate in L. helveticus. (A) The mobility of plasmid DNA from lysates of L. helveticus DSM20075 (lane 4) and an Ermr Cmpr BglA+ transconjugant (lane 5) and of L. helveticus NCDO1244 (lane 6) and a transconjugant (lane 7) was compared with that of pSA3b6 (lane 1), pVA797 (lane 2), and DNA from the donor strain, L. lactis subsp. lactis LM2345(pSA3b6::pVA797) (lane 3). (B) Plasmid DNA was blotted and probed with an enhanced chemiluminescence-labelled BglA gene determinant. Lanes 1 to 3 were exposed for 1 min, and lanes 4 to 7 were exposed for 60 min.

View larger version (127K): [in this window] [in a new window]   FIG. 2.   Detection of -glucanase activity in pSA3b6::pVA797 cointegrate transconjugants of strains of L. helveticus. Wild-type and transconjugant clones of the following six strains were spot inoculated onto MRS agar containing lichenin: CNRZ450 (a), CNRZ32 (Institut National de la Recherche Agronomique, Jouy-en-Josas, France) (b), CNRZ223 (c), DSM20075 (d), NCDO262 (e), and NCDO1244 (f). The positions of the inocula are arrowed. The left members of the pairs of inocula were the wild type; the right members were the transconjugants. Zones of clearing around the transconjugants due to the hydrolysis of lichenin could be seen following flooding of the plates with Congo red 48 h after incubation.

7

Stability of pSA3b6::pVA797 plasmid cointegrates in L. helveticus. Stability of the pSA3b6::pVA797 cointegrate containing the bglA gene in several L. helveticus strains at increasing temperatures is shown in Fig. 3. The cointegrate was stable at 37°C in two of the four strains tested after three serial transfers (up to 30 generations) without selective pressure. In all strains tested, the plasmid became unstable as the temperature of incubation was increased to 40°C and above. Growth of strain CNRZ450 with selective pressure (erythromycin or chloramphenicol) at the permissive (37°C) or restrictive (44°C) temperature for 100 generations did not result in the recovery of any Erm^r Cmp^s or Erm^s Cmp^r clones, suggesting that the cointegrate was structurally stable.

View larger version (28K): [in this window] [in a new window]   FIG. 3.   Stability of the plasmid pSA3b6::pVA797 cointegrate in four strains of L. helveticus at increasing incubation temperatures. Cultures of pSA3b6::pVA797-containing strains of L. helveticus were incubated at increasing temperatures in the absence of selection. The number of generations was estimated by comparing plate counts at 24-h intervals on MRS agar without selection following transfer of the culture onto fresh media (1/1,000 dilution). Data for the retention of the plasmid were obtained by comparing plate counts on agar with and without selection for the Ermr or BglA plasmid marker and were taken from up to three separate experiments. , 37°C; , 40°C; , 42°C.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Versatility of pSA3 in gene cloning for L. helveticus. Conjugal transfer appears to offer an approach that permits the introduction of new genetic material into a range of strains of L. helveticus, most of which have not been described as electrocompetent. The conditions selected for initial mating experiments were arbitrary. However, transfer frequencies were not enhanced by adjusting the donor/recipient ratios, growth phases, or contact times. This suggested that the crucial events for genetic transfer occurred during bacterial growth on the filter during mating. An identical range of strains could act as recipients when matings took place on GMYEA surfaces, although these experiments contrasted with filter matings in that few donor bacteria remained viable following incubation (data not shown). -Glucanase activity was detected for all transconjugants harboring the pSA3b6::pVA797 plasmid cointegrate, although the diameters of the zones of lichenin hydrolysis around individual colonies varied (data not shown). This could be due to differences in either the level of expression in different hosts or the ability of the bacteria to export the enzyme. It was notable that strain NCDO1244, which gave the smallest zones of lichenin hydrolysis around individual colonies, formed "sticky" colonies. A factor responsible for this phenotype (extracellular polysaccharide, for example) may have partially blocked the export of the -glucanase enzyme.