International Journal of Food Microbiology 39 (1998) 231236

Short communication

Characterization of plasmids from Listeria monocytogenes and Listeria innocua strains isolated from short-ripened cheeses
* Abelardo Margolles, Clara G. de los Reyes-Gavilan
Instituto de Productos Lacteos de Asturias ( CSIC). Crta. de Inesto s / n, 33300 Villaviciosa, Asturias, Spain Received 30 June 1997; received in revised form 2 October 1997; accepted 10 November 1997

Abstract The plasmid content of 30 isolates of Listeria monocytogenes and 18 isolates of Listeria innocua obtained from short-ripened cheeses was analysed. The isolates of L. monocytogenes serogroup 1 harboured a single plasmid, pLM33 (33.2 kbp), whereas the serogroup 4 isolates did not contain plasmids. One group of L. innocua strains harboured the plasmid pLI71 (71 kbp) and another one contained two plasmids: pLI59 (59.5 kbp) and pLI56 (56.5 kbp). These plasmid groups were in accordance with clusters previously dened by pulsed-eld gel electrophoresis analysis of the chromosomal DNA of Listeria isolates. Plasmids pLM33, pLI71 and pLI59 shared homology regions of at least 20 kbp. Plasmid pLI56 did not encode genes for any known character (such as carbohydrate fermentation, resistance to antibiotics, heavy metals or disinfectants, growth at low pH, NaCl tolerance or thermal inactivation by pasteurisation) and displayed different characteristics to the other three plasmids. It was also the only one cured from the parent strain and the sole plasmid not digested by the restriction enzyme Pst I. In addition, its lack of homology with pLM33, pLI71 and pLI59 enhanced the possibility of a different origin for plasmid pLI56. 1998 Elsevier Science B.V. Keywords: Listeria monocytogenes ; Listeria innocua ; Plasmid; Cheeses

1. Introduction Listeria monocytogenes is an opportunistic pathogen of humans and animals which in the last decade has been implicated in several outbreaks and sporadic cases of listeriosis traced to contaminated food (Farber and Peterkin, 1991). Listeria innocua, a species closely related to L. monocytogenes, is nonhemolytic and nonpathogenic. Both of them can
*Corresponding author. Tel.: 1 34 85892131; fax: 1 34 85892233; e-mail: iplacsic@airastur.es

be frequently isolated from cheeses, dairy products and other foods. The presence of plasmids in Listeria et al. (1982) and was rst reported by Perez-Daz since this time, research has focused mainly on clinical isolates of L. monocytogenes. Most plasmids in Listeria are cryptic but several authors have recently shown the involvement of plasmids and transposons in cadmium and antibiotic resistance in Listeria (Poyart-Salmeron et al., 1990; Lebrun et al., 1992; Poyart-Salmeron et al., 1992; Facinelli et al., 1993; Hadorn et al., 1993; Lebrun et al., 1994a,b). However, despite several studies on food and en-

0168-1605 / 98 / $19.00 1998 Elsevier Science B.V. All rights reserved. PII S0168-1605( 97 )00132-3

232

/ International Journal of Food Microbiology 39 (1998) 231 236 A. Margolles, C.G. de los Reyes-Gavilan

vironmental strains (Fistrovici and Collins-Thompson, 1990; Kolstad et al., 1990; Peterkin et al., 1992; Facinelli et al., 1993), there is still very little data about plasmids in L. monocytogenes from nonclinical sources such as foods, as well as in other Listeria species. Recently, we reported the isolation of several L. monocytogenes and L. innocua strains from some regional (Asturias, northern Spain) short-ripened cheeses (Margolles et al., 1996). Analysis by pulsedeld gel electrophoresis (PFGE) of Apa I and Sma Idigested chromosomal DNA dened ve clusters in L. monocytogenes (m1 to m5 ) and two main clusters in L. innocua (i1 and i2 ). Clusters m1, m2 and m3 of L. monocytogenes harboured strains of serogroup 1 whereas clusters m4 and m5 contained strains of serogroup 4 (Margolles et al., 1997). In the present work we report on the preliminary characterization of the plasmid content of these Listeria isolates.

2. Material and methods

2.1. Plasmid proles

Listeria strains were cultured on tryptonesoya broth (TSB) (Adsa-Micro, Barcelona, Spain) at 308C. Small-scale plasmid DNA preparations of Listeria were made as described by Simon et al. (1985) with several modications. Cells were harvested by centrifugation from 10 ml of exponential-growth-phase cultures and protoplasts were prepared with lysozyme (nal concentration: 10 mg / ml) and incubated for 1 h at 378C. In addition, after NaCl precipitation, several phenol and chloroform extractions were made. DNA was examined on 0.6% horizontal agarose gels (Seakem, FMC Bioproducts, Rockland, Maine) run at 3.5 V/ cm during 5 h in TAE buffer (40 mM Trisacetate, pH 8.0, and 2 mM EDTA). Plasmids from Enterococcus faecalis BM 4100WT (Courvalin et al., 1980) were used as molecular size standards (70.1, 67.3, 49 and 2 kbp, respectively).

containing novobiocin at a subinhibitory concentration (0.2 mg / ml). The culture was incubated for 24 h at high temperature (408C), followed by 9 subcultures (1% inoculum) in the same medium containing novobiocin. After growth under these conditions, the cultures were plated on tryptone soya agar (TSA) and individual colonies were subsequently picked, puried and submitted to a plasmid extraction and endonuclease restriction analysis. Parent and cured strains were phenotypically characterised. In brief, lactose and melezitose fermentation ability were tested in an appropriate liquid medium (peptone of meat 1%, sodium chloride 0.5%, bromocresol purple 0.02g / l, pH 6.8). Minimal inhibitory concentrations against several antibiotics (penicillin G, ampicillin, cephalotin, streptomycin, gentamicin, kanamycin, neomycin, chloramphenicol, tetracycline, erythromycin, rifampicin and phosphomycin: 0.031 to 512 mg / ml) and heavy metals salts (C 6 H 5 FeO 7 ? H 2 O, CuSO 4 , HgCl 2 , ZnSO 4 ? 7H 2 O, Pb(NO 3 ) 2 , and CdSO 4 ? 8H 2 O: 2 to 2040 mg / ml) were determined on MuellerHinton agar and on TSA, respectively. Susceptibility to several disinfectants (active agents: chlorine, peroxide, acid or base) was evaluated by the Association of Ofcial Analytical Chemists (1984) method. Thermal inactivation by HTST pasteurisation was tested on overnight cultures washed and suspended in skim milk. Heat treatment of milk was carried out in a Linus Dualcycler (Linus, Madrid, Spain) and survivors were subsequently determined on TSA. The ability to grow at low pH (3.5 to 5.5) produced by inorganic (HCl) or different organic acids (acetic, citric and lactic acid: 50 and 150 mM) as well as NaCl tolerance (5 to 15%) were also evaluated using microtiter plates with different concentrations of acids or NaCl in TSB.

2.3. Restriction endonuclease analysis and DNA DNA hybridisation

Restriction endonuclease digestions were performed as recommended by the supplier (Boehringer Mannheim, Germany). DNA was electrophoresed on TAE buffer in 1% agarose gels at 1.2 V/ cm during 16 h. DNA fragments from agarose gels were vacuum blotted to Hybond-N nylon membranes (Amersham, Buckinghamshire, UK) in a Bio-Rad

2.2. Plasmid curing experiments and phenotypic characterization of parent and cured derivative strains
An overnight culture of several Listeria strains in TSB was inoculated (0.1 ml in 10 ml) into TSB-

 / International Journal of Food Microbiology 39 (1998) 231 236 A. Margolles, C.G. de los Reyes-Gavilan

233

785 vacuum blotter (Bio-Rad, Richmond, CA) according to the manufacturer instructions. Southern hybridisations were done using as probes individual plasmids isolated from Listeria strains. Ten mg of Eco RI digests of the DNA probe were labelled with digoxigenin-dUTP (DIG DNA Labelling and Detection Kit, Boehringer). Hybridisation and washing steps were performed at 658C. Subsequent colorimetric detection with NBT (nitroblue tetrazolium salt) and X-Phosphate was performed as recommended by the supplier (Boehringer).

3. Results and discussion Plasmid proles of 30 isolates of L. monocytogenes and 18 of L. innocua were analysed. All L. monocytogenes isolates serogroup 1 harboured a single plasmid (pLM33) whereas serogroup 4 isolates did not contain plasmids (Fig. 1). However, plasmids were present in all L. innocua strains: isolates of one group (PFGE cluster i1 ) harboured one plasmid (pLI71) and the isolates of the other group (PFGE cluster i2 ) contained two plasmids (pLI59 and pLI56). As previously reported by other et al., 1982; Flamm et al., 1984; authors (Perez-Daz Fistrovici and Collins-Thompson, 1990; Peterkin et

Fig. 1. Plasmids of Listeria strains. Lanes: 1, pLM33 (L. monocytogenes Lm1 strain); 2, pLI71 (L. innocua Li17 strain); 3, pLI59 and pLI56 (L. innocua Li16 strain); 4, pLI59 (L. innocua Li16c strain).

al., 1992), our Listeria spp. isolates contained no more than two different plasmids per cell. In addition, no multiplasmic strains were found. On the other hand, 75.7% of our strains isolated from shortripened cheeses contained plasmids. This was a similar level to that reported by Kolstad et al. (1990) in Listeria spp. isolates from different origins, although the rather lower incidence of plasmid positive strains (between 14 and 30%) has been et al., 1982; reported by other workers (Perez-Daz Flamm et al., 1984; Fistrovici and Collins-Thompson, 1990; Peterkin et al., 1992). These differences can be explained if we take into account the different origins of the isolates. Several studies indicate that the percentage of plasmid positive strains in Listeria spp. is higher in strains of food and environmental et al., origin than in clinical isolates (Perez-Daz 1982; Nocera et al., 1990). Several authors have found higher levels of strains harbouring plasmids in L. innocua (93%) than in L. monocytogenes (20%) (Peterkin et al., 1992), as was also the case with our Listeria isolates. Curing experiments were carried out with novobiocin and at a high temperature as curing agents on isolates representative of all plasmid groups. After several attempts, only pLI56 was cured from strain Li16, giving rise to the cured derivative Li16c which only harboured the plasmid pLI59 (Fig. 1). No differences were found between Li16 and Li16c based on the phenotypic characters tested in this work (lactose and melezitose fermentation, susceptibility to antibiotics, heavy metals and disinfectants, growth at low pH and with different organic acids, tolerance to NaCl and thermal inactivation by pasteurisation). This indicates that pLI56 did not encode genes for the above mentioned properties. On the other hand, as it was reported above, all our serogroup 1 L. monocytogenes isolates contained a unique plasmid (pLM33), whereas serogroup 4 isolates did not contain plasmids. This suggests a possible correlation between the serogroup and the presence of plasmid in this species. Since no cured variants lacking plasmid were obtained from serogroup 1 strains, this hypothesis could not be proved. To this respect, Lebrun et al. (1992) indicated that the conjugative introduction of plasmids in L. monocytogenes strains did not cause a change of its serogroup and Kolstad et al. (1990) did not nd a relation between serotype and plasmid content in their work.

234

/ International Journal of Food Microbiology 39 (1998) 231 236 A. Margolles, C.G. de los Reyes-Gavilan

It is also interesting to note that Nocera et al. (1990) reported an absence of extrachromosomal DNA in L. monocytogenes serotype 4b strains related to the Swiss epidemic of listeriosis (19831987) and et al. (1982) did not nd plasmids in Perez-Daz clinical serogroup 4 strains. It is also necessary to take into account that the plasmid content in Listeria strains isolated from foods could be underestimated because isolates are usually obtained by enrichment in selective media containing acriavin, a potential curing agent (Slade et al., 1988). Total plasmid content of Lm1 (pLM33), Li17 (pLI71) Li16 (pLI59 and pLI56) and Li16c (pLI59) strains was analysed by endonuclease restriction. For plasmids pLM33, pLI71 and pLI59, the molecular sizes calculated from Eco RI digestions were 33.2, 71 and 59.5 kbp, respectively (Fig. 2) which were similar to those of other Listeria strains (Kolstad et al., 1990). However, the molecular size of pLI56 estimated by comparing Eco RI digestion proles of

Li16 strain and its derivative Li16c was quite lower than that expected from the electrophoretic migration of the undigested plasmid. Because of that, the enzyme Hin dIII was used for this purpose giving a size of 56.5 kbp for pLI56 (data not shown). On the other hand, restriction analysis with Eco RI and Pst I proved that the plasmid content of all isolates shared numerous bands with identical molecular size, suggesting that homology regions could exist between plasmids. Furthermore, comparison of Li16 and Li16c Pst I restriction proles indicated that this enzyme did not cut the DNA of pLI 56 probably by absence of recognising sequences for Pst I. The Eco RI digests of the plasmid content of strains from each plasmid group were hybridised to digoxigenin-labelled probes prepared from plasmids pLM33, pLI59, and pLI71. Using the Eco RI digest of pLI71 as a probe, homology was found with all fragments of pLM33 and pLI59 [Fig. 2B(a)]. When the probe was prepared from pLI59, all fragments of

Fig. 2. (A) Eco RI and Pst I restriction proles of the different plasmid groups. Lanes 1 and 6, l DNA digested with Pst I (1) and Hin dIII (6) (molecular sizes in kbp). Plasmids pLM33, pLI71, pLI59 1 pLI56, and pLI59 digested with Eco RI (lanes 2 to 5) and with Pst I (lanes 7 to 10). (B) Respective hybridisations of lines 1 to 6 obtained using the Eco RI digests of pLI71 (a), pLI59 (b), and pLM33 (c) as a probe.

 / International Journal of Food Microbiology 39 (1998) 231 236 A. Margolles, C.G. de los Reyes-Gavilan

235

pLM33 and most of pLI71 fragments showed hybridisation [Fig. 2B(b)]. Also, the probe from pLM33 displayed homology with most but not all fragments of pLI59 and pLI71 [Fig. 2B(c)]. Using each one of these three plasmids as a probe, ve common hybridisation bands to pLM33, pLI59 and pLI71 were always obtained whose molecular sizes were 9.0, 8.3, 1.5, 0.8 and 0.5 kbp. This data corroborated the genetic proximity between plasmids pLM33, pLI59 and pLI71 which have homology regions of at least 20 kbp. The presence of homology zones between native plasmids in Listeria has already been shown by other authors (Kolstad et al., 1990; Flamm et al., 1984). On the other hand, using each one of the above three plasmids as a probe, hybridisation proles of the strain Li16 and its curing derivative Li16c were identical, indicating the lack of homology between pLI56 and the other three plasmids of L. monocytogenes and L. innocua. Different characteristics of pLI56 to the other plasmids (pLM33, pLI71 and pLI59) strengthened the hypothesis of a different origin for this plasmid. It has been suggested that Listeria spp. can acquire plasmids in certain environments such as the fecal tract (Fistrovici and Collins-Thompson, 1990). In addition, several authors have effectively demonstrated that mobile genetic elements such as plasmids and transposons are responsible for the acquisition of antibiotic resistance in L. monocytogenes and L. innocua (Flamm et al., 1984; Vicente et al., 1988; PoyartSalmeron et al., 1990, 1992; Facinelli et al., 1993; Charpentier et al., 1994). Due to its ubiquitous nature, Listeria as well as the genetically related enterococci and Streptococcus spp. can be present in many different environments. Thus, a genetic transfer from other microorganisms could explain the presence of pLI56 in our L. innocua isolates. Finally, hybridisation of the different probes with the total DNA of the strains studied did not show additional hybridisation bands different from those of the owner plasmid indicating that no homology exists between plasmids and chromosome (data not shown).

Spain (Grant ALI93-0114). A. Margolles was the recipient of a predoctoral fellowship from the Fun para la Investigacion Cientca dacion y Tecnica (Asturias, Spain). We thank Ian Bytheway for his helpful comments about the English usage on this manuscript.

References
Association of Ofcial Analytical Chemists, 1984. Disinfectants. In: Williams, S. (Ed.), Ofcial Methods of Analysis of the Association of Ofcial Analytical Chemists. Arlington, VA, p. 70. Charpentier, E., Gerbaud, G., Courvalin, P., 1994. Presence of the Listeria tetracycline resistance gene tet( S) in Enterococcus faecalis. Antimicrob. Agents Chemother. 38, 23302335. Courvalin, P., Carlier, C., Collatz, E., 1980. Plasmid-mediated resistance to aminocyclitol antibiotics in group D streptococci. J. Bacteriol. 143, 541551. Facinelli, B., Roberts, M.C., Giovanetti, E., Casolari, C., Fabio, U., Varaldo, P.E., 1993. Genetic basis of tetracycline resistance in food-borne isolates of Listeria innocua. Appl. Environ. Microbiol. 59, 614616. Farber, J.M., Peterkin, P.I., 1991. Listeria monocytogenes, a foodborne pathogen. Microbiol. Rev. 55, 476511. Fistrovici, E., Collins-Thompson, D.L., 1990. Use of plasmid proles and restriction endonuclease digest in environmental studies of Listeria spp. from raw milk. Int. J. Food Microbiol. 10, 4350. Flamm, R.K., Hinrichs, D.J., Thomashow, M.F., 1984. Introduction of pAMb1 into Listeria monocytogenes by conjugation and homology between native Listeria monocytogenes plasmids. Infect. Immun. 44, 157161. Hadorn, K., Hachler, H., Schaffner, A., Kayser, F.H., 1993. Genetic characterization of plasmid-encoded multiple antibiotic resistance in a strain of Listeria monocytogenes causing endocarditis. Eur. J. Clin. Microbiol. Infect. Dis. 12, 928937. Kolstad, J., Rrvik, L.M., Granum, P.E., 1990. Characterization of plasmics from Listeria sp. Int. J. Food Microbiol. 12, 123 132. Lebrun, M., Audurier, A., Cossart, P., 1994. Plasmid-borne cadmium resistance genes in Listeria monocytogenes are similar to cadA and cad C of Staphylococcus aureus and are induced by cadmium. J. Bacteriol. 176, 30403048. Lebrun, M., Audurier, A., Cossart, P., 1994. Plasmid-borne cadmium resistance genes in Listeria monocytogenes are present on Tn5422, a novel transposon closely related to Tn917. J. Bacteriol. 176, 30493061. Lebrun, M., Loulergue, J., Chaslus-Dancla, E., Audurier, A., 1992. Plasmids in Listeria monocytogenes in relation to cadmium resistance. Appl. Environ. Microbiol. 58, 3183 3186. C., 1996. Margolles, A., Rodrguez, A., de los Reyes-Gavilan, Some chemical and bacteriological characteristics of regional cheeses from Asturias. Spain. J. Food Prot. 59, 509515.

Acknowledgements This work was nancially supported by the Comi of sion Interministerial de Ciencia y Tecnologa

236

/ International Journal of Food Microbiology 39 (1998) 231 236 A. Margolles, C.G. de los Reyes-Gavilan Poyart-Salmeron, C., Trieu-Cuot, P., Carlier, C., MacGowan, A., McLauchlin, J., Courvalin, P., 1992. Genetic basis of tetracycline resistance in clinical isolates of Listeria monocytogenes. Antimicrob. Agents Ch. 36, 463466. Simon, D., Rouault, A., Chopin, M.C., 1985. Protoplast transformation of group N streptococci with cryptic plasmids. FEMS Microbiol. Lett. 26, 239241. Slade, P.J., Collins-Thompson, D.L., Fletcher, F., 1988. Incidence of Listeria species in Ontario raw milk. Can. Inst. Food Sci. Technol. J. 21, 425429. Vicente, M.F., Baquero, F., Perez-Daz, J.C., 1988. Conjugative acquisition and expression of antibiotic resistance determinants in Listeria spp. J. Antimicrob. Chemother. 21, 309318.

C., 1997. Margolles, A., Mayo, B., G. de los Reyes-Gavilan, Polymorphism of Listeria monocytogenes and Listeria innocua strains isolated from short-ripened cheeses. J. Appl. Microbiol. (in press). Nocera, D., Bannerman, E., Rocourt, J., Jaton-Ogay, K., Bille, J., 1990. Characterization by DNA restriction endonuclease analysis of Listeria monocytogenes strains related to Swiss epidemic of listeriosis. J. Clin. Microbiol. 28, 22592263. Perez-Daz, J.C., Vicente, M.F., Baquero, F., 1982. Plasmids in Listeria. Plasmid 8, 112118. Peterkin, P.I., Gardiner, M-A, Malik, N., 1992. Plasmids in Listeria monocytogenes and other Listeria species. Can. J. Microbiol. 38, 161164. Poyart-Salmeron, C., Carlier, C., Trieu-Cuot, P., Courtieu, A-L, Courvalin, P., 1990. Transferable plasmid-mediated antibiotic resistance in Listeria monocytogenes. Lancet 335, 14221426.