- Email: [email protected]
Isolation and Identification of Tetracycline Resistant Lactic Acid Bacteria from Pre-packed Sliced Meat Products
System. Appl. Microbial. 23, 279-284 (2000) © Urban & Fischer Verlag _htt-,-p_://w_w_w_.ur_ba_nf_is_ch_er_.de--'./jo_u_rn_als_/s_am_ _ _ _ _ _ _ _ _ _ _ _
SYSTEI'MTIC AND APPLIED MICROBIOLOGY
Isolation and Identification of Tetracycline Resistant Lactic Acid Bacteria from Pre-packed Sliced Meat Products DIRK GEVERS 1, GEERT Buys 1, FRANK DEVLIEGHERE3 , MIEKE UYTTENDAELE 3 , ] OHAN DEBEvERE3 , and] EAN SWINGS 1,2 1Laboratory of Microbiology, Department of Biochemistry, Physiology and Microbiology, University of Gent, Belgium 2BCCMlLMG Culture Collection, Department of Biochemistry, Physiology and Microbiology, University of Gent, Belgium 3Laboratory of Food Microbiology and Preservation, Department of Food Technology and Nutrition, University of Gent, Belgium Received April 26, 2000
Summary In recent years, the food chain has been recognised as one of the main routes for transmission of antibiotic resistant bacteria between the animal and human population. In this regard, the current study aimed to investigate if tetracycline resistant (tet R ) lactic acid bacteria (LAB) are present in ready-to-eat modified atmosphere packed (MAP) sliced meat products including fermented dry sausage, cooked chicken breast meat and cooked ham. From population graphs based on doubling tetracycline concentrations between 0 and 256 pg ml- 1, only fermented dry sausage was shown to contain a high-level tet R LAB population (5.10 1-2,23.10 4 CFU/g), and this in four out of ten examined sausages. From these four positive sausages, a total of 100 strains were isolated on de Man, Rogosa and Sharpe-sorbic acid (MRS-S) agar without tetracycline (n = 45) and on MRS-S agar supplemented with a tetracycline breakpoint concentration of 64 pg ml- 1 (n = 55). Using resistance histograms derived from the disc diffusion method, all these strains were grouped as sensitive to rifampicin, erythromycin and ampicillin. All strains from the tetracycline-containing MRS-S plates were resistant to tetracycline. Identification with whole-cell protein profiling revealed that the total strain set represented four different species: Pediococcus pentosaceus, Lactobacillus plantarum, Lactobacillus sakei subsp. carnosus and Lactobacillus curvatus. All species are commonly associated with fermented dry sausage, either as starter culture or as natural contaminants. The latter three species were found to comprise all tetracycline resistant strains. To our knowledge, this is the first report providing evidence for the presence of tet R LAB in final readyto-eat pre-packed fermented dry sausages. Key words: antibiotic resistance - lactic acid bacteria - fermented dry sausage - tetracycline - modified atmosphere packaging - SDS-PAGE - susceptibility testing
Introduction The occurrence and epidemiological spread of antimicrobial resistance genes throughout the human environment represents a major public health problem in developed and developing countries (LEVY, 1992). So far, research on the horizontal transfer of drug resistance determinants has mainly focussed on opportunistic and primary pathogenic bacteria (for review see COURVALIN, 1994). Less attention has been drawn to the possible role of human and animal commensal bacteria as reservoir organisms for drug resistance genes and their ability of transferring these genes to human pathogens (SALYERS, 1995). Such reservoir organisms could possibly be found in various foods and food products containing high densities of non-pathogenic bacteria as a result of their natu-
ral production process. The food chain can be considered as the main route of transmission of antibiotic resistant bacteria between the animal and human population (WITTE, 1997). More specifically, the indigenous microflora of dairy products and ready-to-eat meat products that are not heat-treated before consumption provide an increased risk that viable antibiotic resistant bacteria are ingested and end up in the human gastrointestinal tract.
Abbreviations: LAB, lactic acid bacteria; MAP, modified atmosphere packed; tetR, tetracycline resistant; tets, tetracycline sensitive 0723-2020/00/23/02-279 $ 12.00/0
280
D. GEVERS et al.
At present, modified atmosphere packing (MAP) of ready-to-eat meat products has become common practice as food manufacturers have attempted to meet consumer demands for fresh, refrigerated foods with extended shelf life (FARBER, 1991). In MAP meat products, the aerobic spoilage organisms are significantly suppressed by the presence of CO 2 which results in an autochthonous micro flora that is largely dominated by lactic acid bacteria (LAB) (BORCH et ai., 1996). Although most food-associated LAB have acquired the 'generally regarded as safe' (GRAS) status, the potential health risk due to the transfer of antibiotic resistance genes from LAB reservoir strains to bacteria in the resident microflora of the human gastrointestinal tract and hence to pathogenic bacteria has not been fully addressed. During the past decade, the incidence of antibiotic resistant LAB in food and food products has been reported. As such, PERRETEN and co-workers isolated a Lactococcus lactis subsp. lactis strain from a raw milk soft cheese that carried a plasmid conferring resistance to tetracycline, streptomycin, and chloramphenicol (PERRETEN et ai., 1997). Antibiotic-resistant LAB isolates have also been recovered from raw meat, including mainly enterococci (KNUDTSON and HARTMAN, 1993; WEGENER et ai., 1997; KLEIN et a!. 1998; QUEDNAU et a!., 1998) and lactobacilli (VIDAL and COLLINS-THOMPSON, 1987; TANNOCK et a!', 1994; LIN et a!', 1996). The horizontal transfer of antibiotic resistance genes located on conjugative transposons and plasmids by LAB has been reported in literature and is reviewed by TEUBER et a!. (1999). The incidence of resistance to the broad-spectrum antibiotic tetracycline is high in the above-mentioned literature. The emergence of tetracycline resistant strains has nowadays limited its wide use of the past decades. Because the molecular basis of tetracycline resistance is relatively well studied and documented, this agent was chosen as a model for the purpose of this study. The current study was undertaken to document the incidence of tetracycline resistant LAB in ready-to-eat MAP meat products including fermented dry sausage, cooked ham and cooked chicken breast meat.
Materials and Methods Processing of meat products Modified atmosphere packed sliced meat products including fermented dry sausage (7 types of 6 brands), cooked ham (3 types of 2 brands), and cooked chicken breast meat (2 types of 3 brands) were purchased from local supermarkets and stored at 4 °C until further research. In this study, a type was defined as a specific variety distributed by a specific commercial brand. At the end of its indicated shelf life, a 25 g sample was taken from each meat product, added to 225 ml sterile peptone physiological saline solution (PPS) (NaCl 8,5 gil, neutralised bacteriological peptone (Oxoid L34) 1 gil) and homogenised in a stomacher. Serial decimal dilutions (10-1-10-8 ) in PPS were prepared and 1 ml samples of appropriate dilutions were poured in triplicate on de Man, Rogosa and Sharpe-sorbic acid agar (MRS-S agar, Difco 0882-17-0) supplemented with or without a breakpoint concentration of tetracycline (as described below) (T-3383, Sigma, USA). Plates were incubated for five days at a temperature of 30°C under microaerophilic conditions.
Determination of tetracycline breakpoint concentration The tetracycline breakpoint concentration is the minimal concentration of tetracycline that has to be supplemented to MRS-S agar for the preparation of an isolation medium selective for high-level tetracycline resistant LAB. To determine the breakpoint concentration, appropriate dilutions of meat homogenates (obtained as described above) were poured on a series of MRS-S agar plates supplemented with tetracycline in doubling concentrations ranging between 0 and 256 pg ml-I. After incubation for five days at 30°C counts were performed manually. From these results, the breakpoint concentration was deduced, i.e. 64 pg ml-I. Selection and storage of strains Colonies were randomly selected from non-selective MRS-S plates (without antibiotics) and from selective MRS-S plates (supplemented with 64 pg ml- I tetracycline) and further purified under aerobic conditions on non-selective MRS-S plates. Isolates were stored in a bead storage system (Microbank system, Pro-Lab Diagnostics, Wirral, UK) at -80°C. Identification of LAB isolates Isolates were identified using sodiumdodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE) of whole-cell bacterial proteins. Preparations of protein extracts and polyacrylamide gel electrophoresis were done as described previously (POT et aI., 1994). Identification of the isolates was performed by comparison of their protein patterns with a laboratory database containing over 6000 reference strains encompassing all known LAB species. Pattern storage and database comparisons were performed using the software package GeiCompar (version 4.2; Applied Maths, Ghent, Belgium). Susceptibility testing A modified version of the Kirby-Bauer disc diffusion method (KIRBY et aI., 1966), in which Meuller-Hinton medium was replaced by MRS-S agar, was used for determination of antibiograms. Oxoid susceptibility test discs of ampicillin (25 pg), erythromycin (10 pg), rifampicin (30 pg) and tetracycline (30 pg) were applied on inoculated MRS-S plates using the Oxoid Disc Dispenser. Diameters of the respective inhibition zones were measured using a digital calliper (Mauser digital 2, Ludwigsburg, Germany) following a 16-18 h incubation of the antibiograms at 30°C and recorded. For each of the antibiotics tested, classification of the isolates into sensitive and resistant groups was based on resistance histograms (i.e. number of strains versus size of the inhibition zone). Cut-off values to differentiate among resistant and susceptible groups were defined on the basis of the bimodal distribution of the population in the resistance histograms.
Results and Discussion Three kinds of modified atmosphere packed (MAP) sliced meat products, i.e. fermented dry sausage, cooked chicken breast meat and cooked ham, were tested for the presence of high-level tetracycline resistant lactic acid bacteria (LAB) using breakpoint experiments. As shown in Fig. 1, fermented dry sausage clearly contains a highlevel tetracycline resistant (tetR ) LAB population (Fig. 1). Concentrations of tetracycline up to 32 ~g ml-1 have a moderate influence on the number of CFU that grow under standard conditions. A concentration of tetracycline of 64 ~g ml-1 diminishes the number of CFU with
Tetracycline Resistant Lactic Acid Bacteria in Meat Products
281
I,E+09
• cooked chicken breast meat I,E+08
o cooked ham
I E+07
o fennented dry sausage
I, +06
'iii u
E I,E+05 00
-.... ::> f.L, U
I,E+04
'II:
I,E+03 I,E+02 I,E+OI I,E+OO 0
2
8
4
16
32
64
128
256
Ilglml tetracycline in MR agar Fig. 1. Determination of the tetracycline breakpoint concentration for the lactic acid bacteria populations in cooked chicken breast meat (black), cooked ham (white) and fermented dry sausage (grey).
5 log units. For tetracycline concentrations as high as 2561lg ml- I (the upper limit of tested range), a significant number (2 to 3 log CFU/g) of LAB was observed after 5 days of incubation. These results indicate a tetracycline breakpoint concentration for the LAB population in fermented dry sausage between 32 and 64 Ilg ml-I. This finding was confirmed when screening other fermented dry sausages. Cooked ham and cooked chicken breast meat samples, on the other hand, did not contain a high-level tet R LAB population at the end of the shelf life (Fig. 1). In fact, a concentration of 8 and 16 Ilg ml- 1 tetracycline, respectively, was sufficient to inhibit growth of LAB on MRS-S agar. A possible explanation for the lack of highlevel resistant LAB in cooked ham and cooked chicken breast meat might be related to the heat treatment during the production process that eliminates most of the viable bacteria naturally present on the raw meat. In fact, the main cause of spoilage of these products lies with the environmental micro flora recontaminating the products after cooking, during slicing and packaging (BjoRKROTH and KORKEALA, 1997; SAMELIS et aI., 1998). In order to verify the breakpoint concentration as determined by the two-fold dilution technique, colonies were isolated from different MRS-S plates (poured with meat homogenates of sausage 1 and 2) supplemented with 0, 32 and 64 Ilg ml- 1 of tetracycline, and subjected to susceptibility testing using the disc diffusion method (Fig. 2). Strains isolated from non-selective MRS-S plates
(i.e. 0 Ilg ml- 1 tetracycline) were all tetracycline sensitive (tetS), whereas the isolates recovered from plates supplemented with 32 Ilg ml- 1 of tetracycline were divided in tet R and tetS strains. A concentration of 64 Ilg ml- I of tetracycline was chosen as the breakpoint concentration to prepare the selective isolation medium, because 100% of the strains isolated from plates with this concentration of tetracycline have an inhibition zone diameter smaller than 11mm indicating that they are resistant to tetracycline. Using a tetracycline breakpoint concentration set at 64 Ilg ml-I, a total of 10 types of fermented dry sausages were analysed for the presence of tet R LAB. The total number of LAB in the examined sausages typically ranged between 6 and 8 log CFU/g. Four out of these ten samples contained tet R LAB in different concentrations ranging between 1,70 and 4,35 log CFU/g (Table 1). Only the colonies from these four positive samples were selected for further research. A total of 100 colonies were randomly isolated from both non-selective (n = 45) and selective plates (n = 55) and stored in a bead storage system. Based on the distribution within the resistance histograms all strains were classified as sensitive to rifampicin, erythromycin and ampicillin, whereas 55 strains were found to be resistant to tetracycline. A comparison between MRS-S agar and Iso-Sensitest Agar (ISA; a welldefined media for susceptibility testing, like Meuller-Hinton medium) showed no influence of the medium on the size of the inhibition zone for the tested antibiotics (data
282
D. GEVERS et al.
60 ·r-----------------------------------------------------------------------~
o strains isolated from MR
- without antibiotics (n = 30)
strain isolated from MR -S supplemented with 32 ppm tet (n = 27)
50
• trains isolated from MR -
upplemented with 64 ppm tet (n = 42)
40
.~ '" 30
<...0
o
'$. 20
10
o 6
7
8
9
10
11
12
13
14
15
16
17
18
19
20 21
22
23
24
25
26
27
28
29
30
inhibitionzone in mm Fig. 2. Resistance histogram for tetracycline (tet). The histogram displays a bimodal distribution of the population enabling the differentiation among a resistant (R) and an intermediate (I) to susceptible (S) group.
not shown). Tested antibiotics were chosen on the basis of their spectrum of activity. Ampicillin, erythromycin and rifampicin are active against most Gram-positive bacteria and especially against enterococci and streptococci (YAO and MOELLERING, 1995). Moreover, ampicillin and erythromycin are broad-spectrum antibiotics and consequently widely used. Tetracycline susceptibility
was tested to verify the selectivity of the isolation medium and to determine the susceptibility of the strains isolated from the non-selective plates. The results clearly confirm the selectivity of the primary isolation medium supplemented with 64 }lg ml- 1 tetracycline for high-level tet R LAB. These high-level tetracycline resistances are most probably originating from acquired resistance genes
Table 1. Total number of LAB and of tet R LAB in four different fermented dry sausages after five days of incubation at 30 DC under microaerophilic conditions.
Table 2. Identification of isolates from four different sausages using sodiumdodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE) of whole-cell proteins.
Type
1 7
8
11
Total LAB (non-selective plates') (log cfu I g)
Total tet R LAB (selective plates b) (log cfu I g)
8,58 ± 7,93 8,31 ± 7,30 8,37 ± 7,02 8,51 ± 7,18
2,05 ± 1,67 1,70 ± 1,3 c 4,12 ±2,85 4,35 ± 2,83
• MRS-S plates without tetracycline. b MRS-S plates supplemented with 64 ].lg ml- 1 of tetracycline. c This number is below the statistical reliable limit for pour plates (i.e. 300 CFU/g of meat), but just above the lower limit of reliable counts for pour plates (i.e. 50 CFU/g of meat).
Type
1
8
7
11
No. of strains isolated and identified
20
24
20
36
Tetracycline resistantl sensitive Pediococcus pentosaceus Lactobacillus plantarum Lactobacillus sakei subsp. carnosus Lactobacillus curvatus
0/4 013 0117 010 1610 010 010 010 010 12/9 110 23/13 010 010 2/0 010
Division in tetracycline sensitive and resistant is based on susceptibility testing using the disc diffusion method.
Tetracycline Resistant Lactic Acid Bacteria in Meat Products r
,0
10
D
10
! 0/0
283
Sample nr.
..L. _ _--..I,~_T ~. ...........T~..........J,
DG 165 DG 173 DG 178 DG 181
II
DG 179 DG 180 DG 168 DG 143
7
Lb. sakeisubsp. carnOSU$
DG 038 DG 041 DG 035 DG 047 DG 046 DG 141 DG 142
Lb. curvalus
DG 015 DG 016 DG 017
Lb. planlarum
DG 023 DG 009
Fig. 3. Cluster analysis of the digitized protein profiles of a subset of the tetracycline resistant Lactobacillus (Lb.) strains isolated
from four different types of fermented dry sausage. Sample numbers correspond with type numbers throughout the text and Table 1.
because intrinsic high-level tetracycline resistance has not yet been reported in literature and non-lethal mutations to tetracycline resistance are unlikely to occur (SALYERS, 1995). So far, very few studies have attempted to demonstrate the presence of antibiotic resistance in LAB originating from fermented dry sausages. In TEUBER et a1.'s (1999) paper on food-associated enterococci, transferable tetracycline, erythromycine, lincomycin and penicillin resistance determinants were found in Enterococcus faecium and E. casseliflavus originating from fermented dry sausage. Moreover, the authors speculated that fermented foods made from raw milk and meat may contain antibiotic resistant LAB from the originating animals including enterococci, lactobacilli and lactococci. The 100 LAB isolates originating from the four samples were identified with SDS-PAGE of whole-cell bacterial proteins (Table 2; Fig. 3). All these strains could be unequivocally allocated to a specific species or subspecies by comparison of their unknown profiles with the SDSPAGE protein profiles of reference strains in the laboratory database. In total, four different LAB species were identified, i.e.: Pediococcus pentosaceus, Lactobacillus plantarum, Lactobacillus sakei subsp. carnosus and Lactobacillus curvatus. These four species are commonly used in starter culture preparations (JESSEN, 1995). Cluster analysis combined with visual inspection of digitised protein fingerprints clearly shows that isolates originating from the same sausage and belonging to the same species display highly similar if not identical patterns
(Fig. 3). In order to verify the clonality of these strains, the use of a genotypic fingerprinting technique such as PFGE (BJORKROTH et aI., 1996) or AFLP (GANCHEVA et aI., 1999) is certainly warranted. Strains of Lb. plantarum and Lb. curvatus were exclusively found in sausage 1 and 3, respectively. Interestingly, all isolates identified as Lb. plantarum were tetR whereas the number of Lb. sakei subsp. carnosus strains were well distributed over tet R and tet S groups. In this study, no tet R Pediococcus pentosaceus could be isolated from fermented dry sausages. More than three months after the first purchase, the same 10 types were bought again and re-examined for the presence of tetR LAB. Six out of ten samples contained high-level tet R LAB (data not shown). One type that originally exhibited high-level tetR LAB (i.e. sample 1 in Table 1) did not contain any tet R LAB following a second analysis. Three types previously lacking tet R LAB were now clearly positive. These data indicate that the presence of tet R LAB in a given fermented dry sausage is subject to variation. In order to explain this instability observed with the presence of tetR LAB, an epidemiological analysis encompassing the complete process line of a fermented dry sausage will be necessary. This approach may offer the possibility to determine the origin of tet R LAB in the fermented meat product, which could be the cause of the instability itself, and to study the ecology of the micro-organisms (i.e. natural or contaminating flora and the added starter culture) in relation to their envi-
284
D. GEVERS et a!.
ronment (i.e. from the raw meat until the fermented product). Such a survey requires differentiation of LAB isolates at the strain level. In conclusion, this study has provided evidence for the presence of tet R lactic acid bacteria in MAP ready-to-eat fermented dry sausage and the absence in cooked chicken breast meat and cooked ham sold in Belgian retail shops. All resistant strains were members of three Lactobacillus species. Within each of these species, protein profiling indicated a high clonality among the strains isolated from the same sausage, but this should be verified using genotypic fingerprinting technique. Likewise, the phenotypically determined resistance to tetracycline has to be confirmed by genetic analysis of tet R genes. In combination with gene transfer experiments, it will be possible to assess the potential impact of tet R LAB in meat products on the distribution of antibiotic resistance within the human compartment. Acknowledgements DIRK GEVERS was supported by a fellowship of the IWT (= Institute for the Promotion of Innovation by Science and Technology in Flanders). We thank URBAIN TORCK and GEERTRUI RASSCHAERT for their contributions to this work.
References B]6RKROTH, J., RID ELL, J., KORKEALA, H.: Characterization of Lactobacillus sake strains associated with production of ropy slime by randomly amplified polymorphic DNA (RAPD) and pulsed-field gel electrophoresis (PFGE) patterns. Int. J. Food Microbio!. 31, 59-68 (1996). Bj6RKROTH, J., KORKEALA, H.J: Use of rRNA gene restriction patterns to evaluate lactic acid bacterium contamination of vacuum-packaged sliced cooked whole-meat product in a meat processing plant. App!. Environ. Microbio!. 63, 448-453 (1997). BORcH, E., KANT-MuERMANS, M-L, BLIXT, Y.: Bacterial spoilage of meat and cured meat products. Int. J. Food Microbio!' 33, 103-120 (1996). COURVALIN, P.: Minireview: transfer of antibiotic resistance genes between Gram-positive and Gram-negative bacteria. Antimicrob. Agents Chemother. 38, 1447-1451 (1994). FARBER, J.M.: Microbiological aspects of modified-atmosphere packaging technology - a review. J. Food Protect. 54, 58-70 (1991). GANCHEVA, A:, POT, B., VANHONACKER, K., HOSTE, B. KERSTERS, K.: A polyphasic approach towards the identification of strains belonging to Lactobacillus acidophilus and related species. System. App!. Microbio!. 22, 573-585 (1999). JESSEN, B. : Starter cultures for meat fermentations, pp.130-154. In: Fermented meats (G. CAMPBELL-PLATT, P.E. COOK, eds.) Glasgow, Blackie academic & professional 1995. KIRBY, W.M.M., BAUER, A.W., SHERRIS, J.C., TURCK, M.: Antibiotic susceptibility testing by a standardised single disc method. Am. J. Clin. Patho!' 45,493-496 (1966). KLEIN, G., PACK, A., REUTER, G.: Antibiotic resistance patterns of enterococci and occurrence of vancomycin-resistant enterococci in raw minced beef and pork in Germany. Appl. Environ. Microbio!. 64,1825-1830 (1998).
KNUDTSON, L.M., HARTMAN, P.A.: Antibiotic resistance among enterococcal isolates from environmental and clinical sources. J. Food Protect. 56,489-492 (1993). LEVY, S.B.: The antibiotic paradox: how miracle drugs are destroying the miracle. New York, Plenum Press 1992. LIN, CH.E, FUNG, Z.E, Wu, CH.L., CHUNG, T.CH.: Molecular characterisation of a plasmid-borne (pTC82) chloramphenicol resistance determinant (cat-TC) from Lactobacillus reuteri G4. Plasmid 36, 116-124 (1996). PERRETEN, V., SCHWARZ, E, CRESTA, L., BOEGLIN, G.D., TEUBER, M.: Antibiotic resistance spread in food. Nature 389, 801-802 (1997). POT, B., VANDAMME, P., KERSTERS, K.: Analysis of electrophoretic whole-organisms protein fingerprints, pp. 493-521. In: Modern microbial methods; Chemical methods in prokaryotic systematics (GOODFELLOW, M., O'DONNEL, E.G., eds.) Chichester, Wiley 1994. QUEDNAU, M., AHRNE, S., PETERSSON, A.e., MOLIN, G.: Antibiotic-resistant strains of Enterococcus isolated from Swedish and Danish retailed chicken and pork. J. App!. Microbio!' 84,1163-1170 (1998). ROBERTS, M.e.: Genetic mobility and distribution of tetracycline resistance determinants, pp. 206-219. In: Antibiotic resistance: origins, evolution, selection and spread (D.J. CHADWICK, J. GOODE, eds.), Ciba foundation symposium 207, Chichester, Wiley 1997. SALYERS, A.A.: Antibiotic resistance transfer in the mammalian intestinal tract: implications for human health, food safety and biotechnology. Heidelberg, Springer-Verlag 1995. SAMELIS, J., KAKOURI, A., GEORGIADOU, K.G., METAXOPOULOS, J.: Evaluation of the extent and type of bacterial contamination at different stages of processing of cooked ham. J. App!. Microbiol. 84,649-660 (1998). TANNOCK, G.W., LUCHANSKY, J.B., MILLER, L., CONNELL, H., THODE-ANDERSEN, S., MERCER, A.A., KLAENHAMMER, T.R.: Molecular characterisation of a plasmid-borne (pGT633) erythromycin resistance determinant (ermGT) from Lactobacillus reuteri 100-163 . Plasmid 31, 60-71 (1994). TEUBER, M., MElLE, L., SCHWARZ, E: Acquired antibiotic resistance in lactic acid bacteria from food. Antonie van Leeuwenhoek 76,115-137 (1999). VIDAL, e.A., COLLINS-THOMPSON, D.L.: Resistance and sensitivity of meat lactic acid bacteria to antibiotics. J. Food Protect. 50,737-740 (1987). WEGENER, H .e., MADSEN, M., NIELSEN, N., AARESTRUP, EM.: Isolation of vancomycin resistant Enterococcus faecium from food. Int. J. Food Microbio!. 35, 57-66 (1997). WITTE, W.: Impact of antibiotic use in animal feeding on resistance of bacterial pathogens in humans, pp. 61-75. In: Antibiotic resistance: origins, evolution, selection and spread (D.J. CHADWICK, J. GOODE, eds.), Ciba foundation symposium 207, Chichester, Wiley 1997. YAO, J.D.e., MOELLERING, R.e. JR.: Antibacterial agents, pp. 1281-1307. In: Manual of Clinical Microbiology (MURRAY, P.R., BARON, E.J., PFALLER, M.A., eds.) Washington DC, ASM 1995.
Corresponding author: GEVERS DIRK, Laboratorium voor Microbiologie, Universiteit Gent, K.L. Ledeganckstraat 35, B-9000 Gent, Belgium Te!.: +32-9-264 52 49; Fax: +32-9-264 50 92; e-mail: [email protected]
SYSTEI'MTIC AND APPLIED MICROBIOLOGY
Isolation and Identification of Tetracycline Resistant Lactic Acid Bacteria from Pre-packed Sliced Meat Products DIRK GEVERS 1, GEERT Buys 1, FRANK DEVLIEGHERE3 , MIEKE UYTTENDAELE 3 , ] OHAN DEBEvERE3 , and] EAN SWINGS 1,2 1Laboratory of Microbiology, Department of Biochemistry, Physiology and Microbiology, University of Gent, Belgium 2BCCMlLMG Culture Collection, Department of Biochemistry, Physiology and Microbiology, University of Gent, Belgium 3Laboratory of Food Microbiology and Preservation, Department of Food Technology and Nutrition, University of Gent, Belgium Received April 26, 2000
Summary In recent years, the food chain has been recognised as one of the main routes for transmission of antibiotic resistant bacteria between the animal and human population. In this regard, the current study aimed to investigate if tetracycline resistant (tet R ) lactic acid bacteria (LAB) are present in ready-to-eat modified atmosphere packed (MAP) sliced meat products including fermented dry sausage, cooked chicken breast meat and cooked ham. From population graphs based on doubling tetracycline concentrations between 0 and 256 pg ml- 1, only fermented dry sausage was shown to contain a high-level tet R LAB population (5.10 1-2,23.10 4 CFU/g), and this in four out of ten examined sausages. From these four positive sausages, a total of 100 strains were isolated on de Man, Rogosa and Sharpe-sorbic acid (MRS-S) agar without tetracycline (n = 45) and on MRS-S agar supplemented with a tetracycline breakpoint concentration of 64 pg ml- 1 (n = 55). Using resistance histograms derived from the disc diffusion method, all these strains were grouped as sensitive to rifampicin, erythromycin and ampicillin. All strains from the tetracycline-containing MRS-S plates were resistant to tetracycline. Identification with whole-cell protein profiling revealed that the total strain set represented four different species: Pediococcus pentosaceus, Lactobacillus plantarum, Lactobacillus sakei subsp. carnosus and Lactobacillus curvatus. All species are commonly associated with fermented dry sausage, either as starter culture or as natural contaminants. The latter three species were found to comprise all tetracycline resistant strains. To our knowledge, this is the first report providing evidence for the presence of tet R LAB in final readyto-eat pre-packed fermented dry sausages. Key words: antibiotic resistance - lactic acid bacteria - fermented dry sausage - tetracycline - modified atmosphere packaging - SDS-PAGE - susceptibility testing
Introduction The occurrence and epidemiological spread of antimicrobial resistance genes throughout the human environment represents a major public health problem in developed and developing countries (LEVY, 1992). So far, research on the horizontal transfer of drug resistance determinants has mainly focussed on opportunistic and primary pathogenic bacteria (for review see COURVALIN, 1994). Less attention has been drawn to the possible role of human and animal commensal bacteria as reservoir organisms for drug resistance genes and their ability of transferring these genes to human pathogens (SALYERS, 1995). Such reservoir organisms could possibly be found in various foods and food products containing high densities of non-pathogenic bacteria as a result of their natu-
ral production process. The food chain can be considered as the main route of transmission of antibiotic resistant bacteria between the animal and human population (WITTE, 1997). More specifically, the indigenous microflora of dairy products and ready-to-eat meat products that are not heat-treated before consumption provide an increased risk that viable antibiotic resistant bacteria are ingested and end up in the human gastrointestinal tract.
Abbreviations: LAB, lactic acid bacteria; MAP, modified atmosphere packed; tetR, tetracycline resistant; tets, tetracycline sensitive 0723-2020/00/23/02-279 $ 12.00/0
280
D. GEVERS et al.
At present, modified atmosphere packing (MAP) of ready-to-eat meat products has become common practice as food manufacturers have attempted to meet consumer demands for fresh, refrigerated foods with extended shelf life (FARBER, 1991). In MAP meat products, the aerobic spoilage organisms are significantly suppressed by the presence of CO 2 which results in an autochthonous micro flora that is largely dominated by lactic acid bacteria (LAB) (BORCH et ai., 1996). Although most food-associated LAB have acquired the 'generally regarded as safe' (GRAS) status, the potential health risk due to the transfer of antibiotic resistance genes from LAB reservoir strains to bacteria in the resident microflora of the human gastrointestinal tract and hence to pathogenic bacteria has not been fully addressed. During the past decade, the incidence of antibiotic resistant LAB in food and food products has been reported. As such, PERRETEN and co-workers isolated a Lactococcus lactis subsp. lactis strain from a raw milk soft cheese that carried a plasmid conferring resistance to tetracycline, streptomycin, and chloramphenicol (PERRETEN et ai., 1997). Antibiotic-resistant LAB isolates have also been recovered from raw meat, including mainly enterococci (KNUDTSON and HARTMAN, 1993; WEGENER et ai., 1997; KLEIN et a!. 1998; QUEDNAU et a!., 1998) and lactobacilli (VIDAL and COLLINS-THOMPSON, 1987; TANNOCK et a!', 1994; LIN et a!', 1996). The horizontal transfer of antibiotic resistance genes located on conjugative transposons and plasmids by LAB has been reported in literature and is reviewed by TEUBER et a!. (1999). The incidence of resistance to the broad-spectrum antibiotic tetracycline is high in the above-mentioned literature. The emergence of tetracycline resistant strains has nowadays limited its wide use of the past decades. Because the molecular basis of tetracycline resistance is relatively well studied and documented, this agent was chosen as a model for the purpose of this study. The current study was undertaken to document the incidence of tetracycline resistant LAB in ready-to-eat MAP meat products including fermented dry sausage, cooked ham and cooked chicken breast meat.
Materials and Methods Processing of meat products Modified atmosphere packed sliced meat products including fermented dry sausage (7 types of 6 brands), cooked ham (3 types of 2 brands), and cooked chicken breast meat (2 types of 3 brands) were purchased from local supermarkets and stored at 4 °C until further research. In this study, a type was defined as a specific variety distributed by a specific commercial brand. At the end of its indicated shelf life, a 25 g sample was taken from each meat product, added to 225 ml sterile peptone physiological saline solution (PPS) (NaCl 8,5 gil, neutralised bacteriological peptone (Oxoid L34) 1 gil) and homogenised in a stomacher. Serial decimal dilutions (10-1-10-8 ) in PPS were prepared and 1 ml samples of appropriate dilutions were poured in triplicate on de Man, Rogosa and Sharpe-sorbic acid agar (MRS-S agar, Difco 0882-17-0) supplemented with or without a breakpoint concentration of tetracycline (as described below) (T-3383, Sigma, USA). Plates were incubated for five days at a temperature of 30°C under microaerophilic conditions.
Determination of tetracycline breakpoint concentration The tetracycline breakpoint concentration is the minimal concentration of tetracycline that has to be supplemented to MRS-S agar for the preparation of an isolation medium selective for high-level tetracycline resistant LAB. To determine the breakpoint concentration, appropriate dilutions of meat homogenates (obtained as described above) were poured on a series of MRS-S agar plates supplemented with tetracycline in doubling concentrations ranging between 0 and 256 pg ml-I. After incubation for five days at 30°C counts were performed manually. From these results, the breakpoint concentration was deduced, i.e. 64 pg ml-I. Selection and storage of strains Colonies were randomly selected from non-selective MRS-S plates (without antibiotics) and from selective MRS-S plates (supplemented with 64 pg ml- I tetracycline) and further purified under aerobic conditions on non-selective MRS-S plates. Isolates were stored in a bead storage system (Microbank system, Pro-Lab Diagnostics, Wirral, UK) at -80°C. Identification of LAB isolates Isolates were identified using sodiumdodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE) of whole-cell bacterial proteins. Preparations of protein extracts and polyacrylamide gel electrophoresis were done as described previously (POT et aI., 1994). Identification of the isolates was performed by comparison of their protein patterns with a laboratory database containing over 6000 reference strains encompassing all known LAB species. Pattern storage and database comparisons were performed using the software package GeiCompar (version 4.2; Applied Maths, Ghent, Belgium). Susceptibility testing A modified version of the Kirby-Bauer disc diffusion method (KIRBY et aI., 1966), in which Meuller-Hinton medium was replaced by MRS-S agar, was used for determination of antibiograms. Oxoid susceptibility test discs of ampicillin (25 pg), erythromycin (10 pg), rifampicin (30 pg) and tetracycline (30 pg) were applied on inoculated MRS-S plates using the Oxoid Disc Dispenser. Diameters of the respective inhibition zones were measured using a digital calliper (Mauser digital 2, Ludwigsburg, Germany) following a 16-18 h incubation of the antibiograms at 30°C and recorded. For each of the antibiotics tested, classification of the isolates into sensitive and resistant groups was based on resistance histograms (i.e. number of strains versus size of the inhibition zone). Cut-off values to differentiate among resistant and susceptible groups were defined on the basis of the bimodal distribution of the population in the resistance histograms.
Results and Discussion Three kinds of modified atmosphere packed (MAP) sliced meat products, i.e. fermented dry sausage, cooked chicken breast meat and cooked ham, were tested for the presence of high-level tetracycline resistant lactic acid bacteria (LAB) using breakpoint experiments. As shown in Fig. 1, fermented dry sausage clearly contains a highlevel tetracycline resistant (tetR ) LAB population (Fig. 1). Concentrations of tetracycline up to 32 ~g ml-1 have a moderate influence on the number of CFU that grow under standard conditions. A concentration of tetracycline of 64 ~g ml-1 diminishes the number of CFU with
Tetracycline Resistant Lactic Acid Bacteria in Meat Products
281
I,E+09
• cooked chicken breast meat I,E+08
o cooked ham
I E+07
o fennented dry sausage
I, +06
'iii u
E I,E+05 00
-.... ::> f.L, U
I,E+04
'II:
I,E+03 I,E+02 I,E+OI I,E+OO 0
2
8
4
16
32
64
128
256
Ilglml tetracycline in MR agar Fig. 1. Determination of the tetracycline breakpoint concentration for the lactic acid bacteria populations in cooked chicken breast meat (black), cooked ham (white) and fermented dry sausage (grey).
5 log units. For tetracycline concentrations as high as 2561lg ml- I (the upper limit of tested range), a significant number (2 to 3 log CFU/g) of LAB was observed after 5 days of incubation. These results indicate a tetracycline breakpoint concentration for the LAB population in fermented dry sausage between 32 and 64 Ilg ml-I. This finding was confirmed when screening other fermented dry sausages. Cooked ham and cooked chicken breast meat samples, on the other hand, did not contain a high-level tet R LAB population at the end of the shelf life (Fig. 1). In fact, a concentration of 8 and 16 Ilg ml- 1 tetracycline, respectively, was sufficient to inhibit growth of LAB on MRS-S agar. A possible explanation for the lack of highlevel resistant LAB in cooked ham and cooked chicken breast meat might be related to the heat treatment during the production process that eliminates most of the viable bacteria naturally present on the raw meat. In fact, the main cause of spoilage of these products lies with the environmental micro flora recontaminating the products after cooking, during slicing and packaging (BjoRKROTH and KORKEALA, 1997; SAMELIS et aI., 1998). In order to verify the breakpoint concentration as determined by the two-fold dilution technique, colonies were isolated from different MRS-S plates (poured with meat homogenates of sausage 1 and 2) supplemented with 0, 32 and 64 Ilg ml- 1 of tetracycline, and subjected to susceptibility testing using the disc diffusion method (Fig. 2). Strains isolated from non-selective MRS-S plates
(i.e. 0 Ilg ml- 1 tetracycline) were all tetracycline sensitive (tetS), whereas the isolates recovered from plates supplemented with 32 Ilg ml- 1 of tetracycline were divided in tet R and tetS strains. A concentration of 64 Ilg ml- I of tetracycline was chosen as the breakpoint concentration to prepare the selective isolation medium, because 100% of the strains isolated from plates with this concentration of tetracycline have an inhibition zone diameter smaller than 11mm indicating that they are resistant to tetracycline. Using a tetracycline breakpoint concentration set at 64 Ilg ml-I, a total of 10 types of fermented dry sausages were analysed for the presence of tet R LAB. The total number of LAB in the examined sausages typically ranged between 6 and 8 log CFU/g. Four out of these ten samples contained tet R LAB in different concentrations ranging between 1,70 and 4,35 log CFU/g (Table 1). Only the colonies from these four positive samples were selected for further research. A total of 100 colonies were randomly isolated from both non-selective (n = 45) and selective plates (n = 55) and stored in a bead storage system. Based on the distribution within the resistance histograms all strains were classified as sensitive to rifampicin, erythromycin and ampicillin, whereas 55 strains were found to be resistant to tetracycline. A comparison between MRS-S agar and Iso-Sensitest Agar (ISA; a welldefined media for susceptibility testing, like Meuller-Hinton medium) showed no influence of the medium on the size of the inhibition zone for the tested antibiotics (data
282
D. GEVERS et al.
60 ·r-----------------------------------------------------------------------~
o strains isolated from MR
- without antibiotics (n = 30)
strain isolated from MR -S supplemented with 32 ppm tet (n = 27)
50
• trains isolated from MR -
upplemented with 64 ppm tet (n = 42)
40
.~ '" 30
<...0
o
'$. 20
10
o 6
7
8
9
10
11
12
13
14
15
16
17
18
19
20 21
22
23
24
25
26
27
28
29
30
inhibitionzone in mm Fig. 2. Resistance histogram for tetracycline (tet). The histogram displays a bimodal distribution of the population enabling the differentiation among a resistant (R) and an intermediate (I) to susceptible (S) group.
not shown). Tested antibiotics were chosen on the basis of their spectrum of activity. Ampicillin, erythromycin and rifampicin are active against most Gram-positive bacteria and especially against enterococci and streptococci (YAO and MOELLERING, 1995). Moreover, ampicillin and erythromycin are broad-spectrum antibiotics and consequently widely used. Tetracycline susceptibility
was tested to verify the selectivity of the isolation medium and to determine the susceptibility of the strains isolated from the non-selective plates. The results clearly confirm the selectivity of the primary isolation medium supplemented with 64 }lg ml- 1 tetracycline for high-level tet R LAB. These high-level tetracycline resistances are most probably originating from acquired resistance genes
Table 1. Total number of LAB and of tet R LAB in four different fermented dry sausages after five days of incubation at 30 DC under microaerophilic conditions.
Table 2. Identification of isolates from four different sausages using sodiumdodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE) of whole-cell proteins.
Type
1 7
8
11
Total LAB (non-selective plates') (log cfu I g)
Total tet R LAB (selective plates b) (log cfu I g)
8,58 ± 7,93 8,31 ± 7,30 8,37 ± 7,02 8,51 ± 7,18
2,05 ± 1,67 1,70 ± 1,3 c 4,12 ±2,85 4,35 ± 2,83
• MRS-S plates without tetracycline. b MRS-S plates supplemented with 64 ].lg ml- 1 of tetracycline. c This number is below the statistical reliable limit for pour plates (i.e. 300 CFU/g of meat), but just above the lower limit of reliable counts for pour plates (i.e. 50 CFU/g of meat).
Type
1
8
7
11
No. of strains isolated and identified
20
24
20
36
Tetracycline resistantl sensitive Pediococcus pentosaceus Lactobacillus plantarum Lactobacillus sakei subsp. carnosus Lactobacillus curvatus
0/4 013 0117 010 1610 010 010 010 010 12/9 110 23/13 010 010 2/0 010
Division in tetracycline sensitive and resistant is based on susceptibility testing using the disc diffusion method.
Tetracycline Resistant Lactic Acid Bacteria in Meat Products r
,0
10
D
10
! 0/0
283
Sample nr.
..L. _ _--..I,~_T ~. ...........T~..........J,
DG 165 DG 173 DG 178 DG 181
II
DG 179 DG 180 DG 168 DG 143
7
Lb. sakeisubsp. carnOSU$
DG 038 DG 041 DG 035 DG 047 DG 046 DG 141 DG 142
Lb. curvalus
DG 015 DG 016 DG 017
Lb. planlarum
DG 023 DG 009
Fig. 3. Cluster analysis of the digitized protein profiles of a subset of the tetracycline resistant Lactobacillus (Lb.) strains isolated
from four different types of fermented dry sausage. Sample numbers correspond with type numbers throughout the text and Table 1.
because intrinsic high-level tetracycline resistance has not yet been reported in literature and non-lethal mutations to tetracycline resistance are unlikely to occur (SALYERS, 1995). So far, very few studies have attempted to demonstrate the presence of antibiotic resistance in LAB originating from fermented dry sausages. In TEUBER et a1.'s (1999) paper on food-associated enterococci, transferable tetracycline, erythromycine, lincomycin and penicillin resistance determinants were found in Enterococcus faecium and E. casseliflavus originating from fermented dry sausage. Moreover, the authors speculated that fermented foods made from raw milk and meat may contain antibiotic resistant LAB from the originating animals including enterococci, lactobacilli and lactococci. The 100 LAB isolates originating from the four samples were identified with SDS-PAGE of whole-cell bacterial proteins (Table 2; Fig. 3). All these strains could be unequivocally allocated to a specific species or subspecies by comparison of their unknown profiles with the SDSPAGE protein profiles of reference strains in the laboratory database. In total, four different LAB species were identified, i.e.: Pediococcus pentosaceus, Lactobacillus plantarum, Lactobacillus sakei subsp. carnosus and Lactobacillus curvatus. These four species are commonly used in starter culture preparations (JESSEN, 1995). Cluster analysis combined with visual inspection of digitised protein fingerprints clearly shows that isolates originating from the same sausage and belonging to the same species display highly similar if not identical patterns
(Fig. 3). In order to verify the clonality of these strains, the use of a genotypic fingerprinting technique such as PFGE (BJORKROTH et aI., 1996) or AFLP (GANCHEVA et aI., 1999) is certainly warranted. Strains of Lb. plantarum and Lb. curvatus were exclusively found in sausage 1 and 3, respectively. Interestingly, all isolates identified as Lb. plantarum were tetR whereas the number of Lb. sakei subsp. carnosus strains were well distributed over tet R and tet S groups. In this study, no tet R Pediococcus pentosaceus could be isolated from fermented dry sausages. More than three months after the first purchase, the same 10 types were bought again and re-examined for the presence of tetR LAB. Six out of ten samples contained high-level tet R LAB (data not shown). One type that originally exhibited high-level tetR LAB (i.e. sample 1 in Table 1) did not contain any tet R LAB following a second analysis. Three types previously lacking tet R LAB were now clearly positive. These data indicate that the presence of tet R LAB in a given fermented dry sausage is subject to variation. In order to explain this instability observed with the presence of tetR LAB, an epidemiological analysis encompassing the complete process line of a fermented dry sausage will be necessary. This approach may offer the possibility to determine the origin of tet R LAB in the fermented meat product, which could be the cause of the instability itself, and to study the ecology of the micro-organisms (i.e. natural or contaminating flora and the added starter culture) in relation to their envi-
284
D. GEVERS et a!.
ronment (i.e. from the raw meat until the fermented product). Such a survey requires differentiation of LAB isolates at the strain level. In conclusion, this study has provided evidence for the presence of tet R lactic acid bacteria in MAP ready-to-eat fermented dry sausage and the absence in cooked chicken breast meat and cooked ham sold in Belgian retail shops. All resistant strains were members of three Lactobacillus species. Within each of these species, protein profiling indicated a high clonality among the strains isolated from the same sausage, but this should be verified using genotypic fingerprinting technique. Likewise, the phenotypically determined resistance to tetracycline has to be confirmed by genetic analysis of tet R genes. In combination with gene transfer experiments, it will be possible to assess the potential impact of tet R LAB in meat products on the distribution of antibiotic resistance within the human compartment. Acknowledgements DIRK GEVERS was supported by a fellowship of the IWT (= Institute for the Promotion of Innovation by Science and Technology in Flanders). We thank URBAIN TORCK and GEERTRUI RASSCHAERT for their contributions to this work.
References B]6RKROTH, J., RID ELL, J., KORKEALA, H.: Characterization of Lactobacillus sake strains associated with production of ropy slime by randomly amplified polymorphic DNA (RAPD) and pulsed-field gel electrophoresis (PFGE) patterns. Int. J. Food Microbio!. 31, 59-68 (1996). Bj6RKROTH, J., KORKEALA, H.J: Use of rRNA gene restriction patterns to evaluate lactic acid bacterium contamination of vacuum-packaged sliced cooked whole-meat product in a meat processing plant. App!. Environ. Microbio!. 63, 448-453 (1997). BORcH, E., KANT-MuERMANS, M-L, BLIXT, Y.: Bacterial spoilage of meat and cured meat products. Int. J. Food Microbio!' 33, 103-120 (1996). COURVALIN, P.: Minireview: transfer of antibiotic resistance genes between Gram-positive and Gram-negative bacteria. Antimicrob. Agents Chemother. 38, 1447-1451 (1994). FARBER, J.M.: Microbiological aspects of modified-atmosphere packaging technology - a review. J. Food Protect. 54, 58-70 (1991). GANCHEVA, A:, POT, B., VANHONACKER, K., HOSTE, B. KERSTERS, K.: A polyphasic approach towards the identification of strains belonging to Lactobacillus acidophilus and related species. System. App!. Microbio!. 22, 573-585 (1999). JESSEN, B. : Starter cultures for meat fermentations, pp.130-154. In: Fermented meats (G. CAMPBELL-PLATT, P.E. COOK, eds.) Glasgow, Blackie academic & professional 1995. KIRBY, W.M.M., BAUER, A.W., SHERRIS, J.C., TURCK, M.: Antibiotic susceptibility testing by a standardised single disc method. Am. J. Clin. Patho!' 45,493-496 (1966). KLEIN, G., PACK, A., REUTER, G.: Antibiotic resistance patterns of enterococci and occurrence of vancomycin-resistant enterococci in raw minced beef and pork in Germany. Appl. Environ. Microbio!. 64,1825-1830 (1998).
KNUDTSON, L.M., HARTMAN, P.A.: Antibiotic resistance among enterococcal isolates from environmental and clinical sources. J. Food Protect. 56,489-492 (1993). LEVY, S.B.: The antibiotic paradox: how miracle drugs are destroying the miracle. New York, Plenum Press 1992. LIN, CH.E, FUNG, Z.E, Wu, CH.L., CHUNG, T.CH.: Molecular characterisation of a plasmid-borne (pTC82) chloramphenicol resistance determinant (cat-TC) from Lactobacillus reuteri G4. Plasmid 36, 116-124 (1996). PERRETEN, V., SCHWARZ, E, CRESTA, L., BOEGLIN, G.D., TEUBER, M.: Antibiotic resistance spread in food. Nature 389, 801-802 (1997). POT, B., VANDAMME, P., KERSTERS, K.: Analysis of electrophoretic whole-organisms protein fingerprints, pp. 493-521. In: Modern microbial methods; Chemical methods in prokaryotic systematics (GOODFELLOW, M., O'DONNEL, E.G., eds.) Chichester, Wiley 1994. QUEDNAU, M., AHRNE, S., PETERSSON, A.e., MOLIN, G.: Antibiotic-resistant strains of Enterococcus isolated from Swedish and Danish retailed chicken and pork. J. App!. Microbio!' 84,1163-1170 (1998). ROBERTS, M.e.: Genetic mobility and distribution of tetracycline resistance determinants, pp. 206-219. In: Antibiotic resistance: origins, evolution, selection and spread (D.J. CHADWICK, J. GOODE, eds.), Ciba foundation symposium 207, Chichester, Wiley 1997. SALYERS, A.A.: Antibiotic resistance transfer in the mammalian intestinal tract: implications for human health, food safety and biotechnology. Heidelberg, Springer-Verlag 1995. SAMELIS, J., KAKOURI, A., GEORGIADOU, K.G., METAXOPOULOS, J.: Evaluation of the extent and type of bacterial contamination at different stages of processing of cooked ham. J. App!. Microbiol. 84,649-660 (1998). TANNOCK, G.W., LUCHANSKY, J.B., MILLER, L., CONNELL, H., THODE-ANDERSEN, S., MERCER, A.A., KLAENHAMMER, T.R.: Molecular characterisation of a plasmid-borne (pGT633) erythromycin resistance determinant (ermGT) from Lactobacillus reuteri 100-163 . Plasmid 31, 60-71 (1994). TEUBER, M., MElLE, L., SCHWARZ, E: Acquired antibiotic resistance in lactic acid bacteria from food. Antonie van Leeuwenhoek 76,115-137 (1999). VIDAL, e.A., COLLINS-THOMPSON, D.L.: Resistance and sensitivity of meat lactic acid bacteria to antibiotics. J. Food Protect. 50,737-740 (1987). WEGENER, H .e., MADSEN, M., NIELSEN, N., AARESTRUP, EM.: Isolation of vancomycin resistant Enterococcus faecium from food. Int. J. Food Microbio!. 35, 57-66 (1997). WITTE, W.: Impact of antibiotic use in animal feeding on resistance of bacterial pathogens in humans, pp. 61-75. In: Antibiotic resistance: origins, evolution, selection and spread (D.J. CHADWICK, J. GOODE, eds.), Ciba foundation symposium 207, Chichester, Wiley 1997. YAO, J.D.e., MOELLERING, R.e. JR.: Antibacterial agents, pp. 1281-1307. In: Manual of Clinical Microbiology (MURRAY, P.R., BARON, E.J., PFALLER, M.A., eds.) Washington DC, ASM 1995.
Corresponding author: GEVERS DIRK, Laboratorium voor Microbiologie, Universiteit Gent, K.L. Ledeganckstraat 35, B-9000 Gent, Belgium Te!.: +32-9-264 52 49; Fax: +32-9-264 50 92; e-mail: [email protected]