Appl Microbiol Biotechnol (1990) 33:547-552

Applied Microbiology Biotechnology © Springer-Verlag1990

Occurrence of aminoglycoside phosphotransferase subclass I and II structural genes among Enterobacteriaceae spp. isolated from meat samples A. H. G. P. Jayaratne, D. L. Collins-Thompson, and J. T. Trevors Department of Environmental Biology,Universityof Guelph, Guelph, Ontario, Canada, N1G 2W1 Received 8 December 1989/Accepted 6 March 1990

Summary. 3'-Aminoglycoside phosphotransferase [APH(3')] enzymes are a group responsible for resistance to the antibiotics kanamycin (Km) and neomycin (Nm) in bacteria. Escherichia coli ECT24, originally isolated from a meat sample, harboured an 83-kb conjugative R-plasmid (pRPJ24) that carries transferable resistance to Km and Nm. Plasmid pRPJ24 was transferred by conjugation to Enterobacter cloacae 94R, which was used as the source of plasmid DNA in development of a probe for the Krn-resistance determinant. Random cloning of B a m H I and HindIII double-digest restriction fragments of pRPJ24 in the pUC18 vector plasmid produced clones resistant to both Nm and Km carrying a 1.9-kb DNA insert. Southern hybridization of pRPJ24 cloned chimeric plasmid DNA (pKPJ94) showed homology with the APH(Y)II gene from transposon TnS. A PstI digest of pKPJ94 produced a 920-bp fragment which hybridized with the APH(3')II structural gene, and was used as a DNA probe for the APH(3')II subclass gene. A 980-bp B a m H I fragment from plasmid pGH54 carrying the APH(3')I gene from transposon Tn903 was used as a subclass I probe. Total DNA from 206 randomly screened Kin-resistant Enterobacteriaceae isolates from raw ground beef and chicken meat samples were examined for the occurrence of APH(3') subclass I and II using non-radioactively-labelled DNA probes. Thirty-six percent and 60% of the isolates examined carried subclass I and II resistances, respectively, in the isolates from chicken meat samples. The corresponding values for bacterial strains from raw ground beef samples were 51% and 72%, respectively; Four percent of the resistant bacterial isolates from chicken samples did not display homology to either probe. This value was 28% for bacterial isolates from ground beef. Three percent of bacterial isolates from chicken samples and 44% from ground beef samples displayed homology to both APH(3') I and II DNA probes.

Offprint requests to: A. H. G. P. Jayaratne

Introduction A number of studies have dealt with acquisition of drug-resistant bacteria by human beings through handling, preparing, and eating of meat and poultry products (Feinman 1984). Such organisms, particularly Salmonella spp., acquired from meat products have been implicated in a number of outbreaks (Holmberg et al. 1984; Spika et al. 1987). Relatively little is known about the occurrence and behaviour of drug resistance determinants in natural bacterial populations. However, much can be deduced by studying the molecular and genetic properties of such resistance determinats. Studies of antibiotic resistance genes using nucleic acid probes can determine the distribution and movement of specific resistance determinants not only in community-acquired infections, but also in environments such as food, which may be the source of acquisition of those infectious antibiotic-resistant organisms. Resistance to aminoglycoside antibiotics in bacteria is usually due to enzymes within bacterial cells that chemically modify the drugs (Bryan 1984; Brzezinska and Davies 1973; Davies and Smith 1978). For example, aminoglycoside phosphotransferase (APH[3']) mediates resistance to neomycin (Nm) and kanamycin (Km) by transfer of a terminal phosphate group from ATP to the 3'-hydroxyl group of the aminoglycoside antibiotic (Goldman and Northrop 1976; Matsuhashi et al. 1975). The APH (3') enzymes are the most widely distributed aminoglycoside-modifying enzymes in bacteria (Miller et al. 1980; Young et al. 1985). Many of the genes for these enzymes are located on plasmids (Courvalin et al. 1978; Davies and Smith 1978; Rubens et al. 1979) and transposable elements (Berg et al. 1978; Labigne-Roussel et al. 1983), which has led to their widespread dissemination. Neomycin in addition to its use in human therapy, is routinely incorporated into animal feeds, for therapeutic and prophylactic purposes (Siegel et al. 1974). It is known to cause selection for multiple antibiotic-resistant R-plasmids (Pohl 1977). Direct selection pressure of this antibiotic has caused an increased occur-

548 r e n c e o f N m r e s i s t a n c e in p a t h o g e n i c a n d n o n - p a t h o g e n i c E n t e r o b a c t e r i a c e a e o f a n i m a l o r i g i n (Siegel et al. 1974). The resistance determinant of APH(3') subclass I o c c u r s m o r e f r e q u e n t l y t h a n s u b c l a s s I I ( C o u r v a l i n et al. 1978; Y o u n g et al. 1985). It is also c a r r i e d o n t r a n s p o s o n s , T n 9 0 3 (Berg et al. 1975; D a v i e s et al. 1977; O k a et al. 1981), T n 6 (Berg et al. 1975; D a v i e s et al. 1977), a n d T n 1 5 2 5 ( L a b i g n e - R o u s s e l et al. 1983); w h e r e a s t h e A P H ( 3 ' ) I I s t r u c t u r a l g e n e is c a r r i e d o n l y o n T n 5 ( B e c k et al. 1982; Berg et al. 1978; C o u r v a l i n et al. 1978). T h e r e h a v e b e e n n o a t t e m p t s to d e t e r m i n e t h e d i s t r i b u tion of APH(3') resistant determinants among Enterob a c t e r i a c e a e e n v i r o n m e n t a l isolates. E n z y m e a s s a y s inv o l v i n g s u b s t r a t e u t i l i z a t i o n are c o m m o n l y u s e d to i d e n t i f y p h o s p h o t r a n s f e r a s e activity ( B r e z i n s k a a n d D a v i e s 1973; D a v i e s a n d S m i t h 1978)). This p r o c e d u r e is u n s u i t a b l e f o r t e s t i n g a l a r g e n u m b e r o f b a c t e r i a l isolates. T h e p r e s e n c e o f s e v e r a l A P H i s o z y m e s in b a c t e rial i s o l a t e s ( D a v i e s a n d S m i t h 1978; Y o u n g et al. 1985) also m a k e s t h e t a s k m o r e difficult. Since D N A p r o b e s h a v e p r o v e d to b e h i g h l y s p e c i f i c a n d u s e f u l in t h e d e t e c t i o n o f s p e c i f i c D N A s e q u e n c e s in b a c t e r i a ( H o u v i e n e n et al. 1988; N g et al. 1987; T e n o v e r et al. 1984), the development of DNA probes for APH(3') structural genes w o u l d b e useful. In the present study, the construction of a non-radioactive DNA probe for the APH(3')II determinant f r o m a n R - p l a s m i d i s o l a t e d f r o m Escherichia coli, a n d the occurrence of the APH(3')I and II stuctural gene subclasses among Enterobacteriaceae isolated from g r o u n d b e e f a n d c h i c k e n m e a t s a m p l e s are d e s c r i b e d . Furthermore, such information on the occurrence and d i s t r i b u t i o n o f r e s i s t a n c e d e t e r m i n a n t s is u s e f u l in tracing t h e m o v e m e n t o f a n t i b i o t i c r e s i s t a n c e g e n e s in t h e environment.

Isolation of Km-resistant isolates. A total of 50 subsamples of each type of raw ground beef or chicken were taken from ten samples purchased from five different retail outlets in Guelph, Ontario, Canada. Kanamycin-resistant isolates were isolated by surface plating 100 ixl aliquots of serial dilutions of blended meat samples on VRBA supplemented with 1% (w/v) glucose and 25 ~tg/ml kanamycin sulphate as described previously (Jayaratne et aL 1989). One hundred randomly selected colonies were obtained from each sample source and maintained in LB broth containing 15% (v/v) sterile glycerol at - 2 0 ° C.

Materials and methods Bacterial strains. The E. coli ECT24 harbouring pRPJ24 (Km R, Nm R, Tc R) strain was isolated from a turkey meat sample (Jayarathe et al. 1989). The Enterobacter cloacae 94R recipient was isolated from a ground beef sample (Jayaratne et al. 1989). Reference bacterial strains, E. coli PS1323 and E. coli PRC321 harbouring plasmids pGH54 and pSAY16, respectively, were provided by Dr. F. C. Tenover, Veterans Administration, Medical Center, Seattle, Wash, USA. All other Km- and Nm-resistant Enterobacteriaceae strains were isolated from samples of raw ground beef and chicken meat.

Preparation of APH(3') subclass I and H DNA probes. Approximately 10 Ixg purified plasmid DNA of pGH54 (APH[3']I) and pKPJ94 (APH[3']II) were digested with BamHI and PstI, respectively. The restriction enzyme digests were subjected to electrophoresis in 0.8% (w/v) agarose gels; the furthest bands (highest mobility) from the wells were excised with a sterile scalpel and the DNA fragments isolated and purified using Geneclean (Bio 101, La Jolla, Calif, USA) as recommended by the manufacturer. Each DNA fragment was non-radioactively labelled with biotin-ll-d uridine triphosphate (UTP) (Bethesda Research Laboratories, Gaithersbury, Md, USA) or digoxigenin-dUTP (Boehringer Mannheim) according to the manufacturer's procedure.

Media and chemicals. Violet red bile agar (VRBA) (Difco Laboratories, Detroit, Mich, USA) was used in the initial isolation of Nm- and Km-resistant enterobacteria from meat. Sterile 0.1% (w/ v) peptone water (Oxoid L 37, Oxoid, Basingstoke, Hants, UK) was used as the diluent. Luria-Bertani (LB) broth (Gibco Laboratories, Madison, Wisc, USA) was used as the growth medium for all bacterial strains at 37°C with shaking at 100 rpm. All antibiotics were purchased from Sigma Chemical Company, St. Louis, Mo, USA. All other chemicals used were of molecular biology grade, and purchased from Sigma, Fisher Scientific Co., Fair Lawn, NJ, USA, and International Biotechnologies, New Haven, Corm, USA. All restriction endonucleases were obtained from Boehringer Mannheim (Dorval, PQ, Canada).

DNA blotting and hybridization. After electrophoresis of plasmid DNA samples and their enzyme digests, Southern transfer of DNA onto nitrocellulose BA85 (Scheicher and Schuell, Keene, NH, USA) was carried out using a Transvac vaccuum blotter (Hoefer Scientific Instruments, San Francisco, Calif, USA) and 10X SSC transfer buffer (1.5 M NaC1, 150 mM sodium citrate, pH 7.0), according to the manufacturer's instructions. The transfer time to nitrocellulose was 30 min. Prior to transfer, DNA on agarose gels was depurinated, denatured, and neutralized by the method of Maniatis et al. (1982). Total genomic DNA samples were diluted to 200 lxl with 6X SSC (pH 7.0); and 100 p~l samples per slot were blotted using a Hybri-Dot blotting apparatus (Bethesda Research Laboratories).

DNA isolation and agarose gel electrophoresis. Purified plasmid DNA used for restriction enzyme digests, cloning, and probe construction was prepared using the alkaline sodium dodecyl sulphate lysis method followed by ultracentrifugation in a CsC1 gradient (Maniatis et al. 1982). The total genomic DNA used for slotblot hybridization was isolated as described by Lewington et al. (1986), with modifications. After NaC1 precipitation of proteins and cell debris, 150 p~l supernatant was treated with 30 lxg/ml RNAse A for 15 min at 37 ° C, and then with 50 Ixg/ml proteinase K for 30 min at 37 ° C. The DNA was precipitated with sodium acetate and 95% ethanol (Lewington et al. 1986), and redissolved in 25 Ix1 TE buffer (10 mM TRIS-HC1, 1 mM disodium ethylenediaminetetraacetate [NazEDTA], pH 8.0). Total DNA samples (5 ~tl) were electrophoresed in a 0.8% (w/ v) agarose gel at 60 V in a horizontal Mini-Sub cell apparatus (Bio-Rad Laboratories, Richmond, Calif, USA) using TAE buffer (40 mM TRIS-acetate, 1 mM NazEDTA, pH 8.2) as the gel and reservoir buffer. The DNA was stained in an aqueous solution of ethidium bromide (0.5 l~g/ml) for 30 min, followed by destaining for 30 min in deionized water before visualization under a 302 nm UV transiUuminator. Cloning of APH(3')H structural gene. Enterobacter cloacae 94R harbouring plasmid pRPJ24 which contained the APH(3')II structural gene was used as the source organism. Plasmid DNA was double digested with HindIII and BamHI restriction enzymes. DNA fragments obtained by enzymatic digestion were cloned into the pUC 18 vector using the pUC 18 cloning kit (Boehringer Mannhelm). Cloned plasmid DNA was transformed into competent Escheriehia coli JM83 cells using the procedure recommended by the manufacturer. Clones carrying Nm and Km resistances were obtained by plating on selective LB agar supplemented with these antibiotics at 25 ~tg/ml concentrations.

549 The nitrocellulose filters were air dried and DNA denatured and neutralized as described by Young et al. (1985). The blots were prepared in duplicate. AH nitrocellulose filters were first air dried and then baked in a vaccuum oven at 80° C for 2 h. Prehybridization and hybridization of all DNA blots with biotin-11-dUTP and digoxigenin-dUTP-labelled probe DNA were carried out according to instructions provided with the BluGene (Bethesda Research Laboratories) and Boehringer Mannheim non-radioactive nucleic acid labelling and detection systems, respectively. Probe DNA was nick-translated using biotin-11-dUTP and a nick-translation kit (Bethesda Research Laboratories); labelled DNA was purified by Sephadex G-50 spun-column chromatography (Maniatis et al. 1982). Digoxigenin-dUTP was incorporated into probe DNA by the random prime labelling method using the Klenow enzyme (large fragment of DNA polymerase I). The labelled DNA was purified by adding 0.1 vol LiC1 (4 mol/1) followed by ethanol precipitation at - 20° C for 2 h.

Results and discussion The restriction endonuclease digests of pRPJ24 with BamHI and H i n d l I I produced 13 D N A fragments ranging in sizes from about 20 kb to 1 kb. A fragment of 1.9 kb was inserted into the corresponding restriction sites of the pUC18 cloning vector (Fig. 1). This fragment was used to prepare a D N A p r o b e as it contained a resistance determinant for kanamycin. The resulting chimeric plasmid (pKPJ94) when transformed into E. coli JM83, resulted in clones resistant to K m and N m at concentrations o f 25 Ixg/ml. A probe developed by labelling 1.9-kb cloned D N A fragment carrying K m and N m resistance determinants displayed h o m o l o g y only

Fig. 1. Agarose gel electrophoresis of cloned, vector plasmid DNA and respective restriction endonuclease BamHI and HindlII digests: lane 1, pRPJ24 DNA; lane 2, BamHI and HindlII digest of pRPJ24; lane 3, pKPJ94 DNA; lane 4, BamHI and Hindill digest of pKPJ94; lane 5, pUC18 DNA; lane 6, BamHI and HindlII digest of pUC18. Open circular and multimeric forms of plasmid DNA are present in lanes 3 and 5. The • indicates the linearized vector DNA and ~ shows the position of the 1.9-kb insert DNA

Fig. 2. Agarose gel electrophoresis (A), and Southern hybridization filter (B), containing plasmid DNA and their respective restriction endonuclease digests. Lanes in A and B have corresponding numbers. A cloned 1.9-kb DNA fragment carrying kanamycin (Km) resistance, was used as the probe for B. Lane 1, BamHI and HindlII digest of pRPJ24; lane 2, pKPJ94 DNA; lane 3, BamHI and HindlII digest of pKPJ94; lane 4, PstI digest of pSAY16; lane 5, BamHI and HindlII digest of pGH54; lane 6, DNA molecular weight marker (B011 digest of pBR328+Hinfl digest of pBR328). The +- denotes the position of the 920-bp 3'aminoglycoside phosphotransferase subclass II [APH(3')II] probe DNA fragment; • shows the position of the 980-bp APH(3')I probe DNA fragment; > indicates the position of the 1.9-kb cloned DNA fragment. Open circular and dimeric plasmid DNA are present in lane 2

with A P H ( 3 ' ) I I and not with the APH(3')I structural gene (Fig. 2). Digestion of pKPJ94 with P s t I resulted in a 920-bp fragment that hybridized with the A P H ( 3 ' ) I I structural gene (results not shown). This 920-bp D N A fragment was later used as the probe for the A P H ( Y ) I I determinant. Screening of 206 randomly selected Kin-resistant bacterial isolates from a m o n g 425 Km-resistant Enterobacteriaceae isolates obtained from chicken meat and ground b e e f samples, showed marked differences in the presence of plasmid DNA. Eighty-eight percent of the isolates obtained from chicken samples, and 30% o f the bacterial isolates from ground beef samples carried plasmid D N A , as determined by agarose gel electrophoresis (Table 1). Only the Km-resistant bacterial isolates that hybridized with the APH(3') gene probes showed resistance to 25 ~tg/ml Nm. Slot-blot hybridization of total D N A obtained from these isolates with the APH(3')I and A P H ( 3 ' ) I I gene probes, demonstrated a variability between sources. In general, the occurrence of the structural gene coding for APH(Y) subclass I I was greater than that of APH(3') subclass I (Fig. 3). Positive (subclass I) and negative (subclass II) signals for the digoxigenin-dUTP labelled probe are displayed in Fig. 3A; positive (subclass II) and negative (subclass I) signals for the labelled p r o b e are also shown in Fig. 3B for comparison. However, in bacterial isolates f r o m ground b e e f sampies, the probe D N A hybridization indicated a higher percentage (44%) of cross-hybridizations of total D N A samples with both probes, than, with samples obtained from chicken meat (3%) (Table 1). Moreover, there were

550 Table 1. Occurrence of 3'-aminoglycosidephosphotransferase (APH[3']) subclasses I and II among Enterobacteriaceae spp. isolated from meat~ DNA probe homology (%)b Meat source

Isolates carrying plasmid DNA (%)

A P H ( 3 ' ) I APH(3')II APH(3')I APH(3')II

Neither probe

Chicken Ground beef

88 30

36 51

4 28

60 72

3 44

One hundred kanamycin-resistant isolates from chicken meat samples and 106 isolates from ground beef samples were examined for plasmid DNA and probe homology. Not all isolates examined for DNA probe homology contained plasmid(s) b Percentages do not total 100%because some isolates hybridized to both probes

Fig. 3. Slot-blot hybridization duplicate filters of total genomic

DNA of Km-resistant isolates with APH(3')I (A), and APH(3')II (B), digoxigenin-labelled DNA probes. The negative controls were // (genomic DNA of isolates carrying APH[3']I determinants), and I (genomic DNA of isolates carrying APH[3']II determinants) in blots A and B, respectively.The ~ shows the position of genomic DNA carryingthe APH(3')I probe; 41 denotes the position of genomic DNA carrying the APH(3')II probe. Samples showing equal or less intensity than negative controls were considered negative a considerable number of isolates (28%) from ground beef that did not exhibit a hybridization signal with either probe (Table 1). In the present study, an 83-kb plasmid originally isolated from E. coli ECT24 and later transferred by conjugation to an Enterobacter cloacae 94R strain, carried the structural gene coding for APH(3')II enzyme activity. Subsequently, the resistance determinant was cloned and the resultant chimeric plasmid was used to

develop a non-radioactive DNA probe. Non-radioactive biotin-labelled (Carter et al. 1989; Wetherall et al. 1988; Zeph and Stotzky 1989) and acetylaminoflurenelabelled (Chevrier et al. 1989) DNA probes have been used to detect specific sequences in bacterial and viral DNA. The sensitivity, safety, long storage life, reusability, and low background hybridization signals have made these probes a convenient alternative to radiolabelled DNA probes. Non-specific binding of the streptavidin-alkaline phosphate conjugates to cellular components other than labelled DNA can be a problem when using biotinylated DNA probes with target DNA obtained from crude bacterial preparations (Zwadyk et al. 1986). This was evident in a preliminary study where genomic DNA samples, showed non-specific hybridization with the biotinylated probe. Use of digoxigeninlabelled DNA probes do not pose such a problem, are more convenient to use, exhibit a lower background signal and equal sensitivity of detection. This should make these probes extremely valuable in diagnostic microbiology procedures. The genetic determinants for APH(3')II activity is encoded on transposon Tn5, which has been sequenced (Beck et al. 1982). Comparison of the physical map of Tn5 (Jorgensen et al. 1979) with the cloned 1.9-kb B a m H I and HindlII fragment of pKPJ94, indicated that the APH(3')II gene of pRPJ24 is of Tn5 type. This would permit an accelerated dissemination of the resistance genes among native bacterial populations in the environment. The probe developed for APH(3')II (920 bp, PstI fragment of pKPJ94) contains about 90% of the APH(3')II structural gene, as well as approximately 300 bp of DNA sequences external to the coding region (Beck et al. 1982). This region may have contributed to the low level of non-specific hybridization between the probe DNA and extraneous DNA sequences (probably from the bacterial chromosome) observed in hybridization experiments. However, the APH(3')I probe (980 bp, B a m H I fragment of pGH54) is known to contain only sequences within the open reading frame of the APH(3')I structural gene (Young et al. 1985). Furtherm o r e , it also behaved in a similar manner to the APH(3')II probe with a low, non-specific hybridization signal. This eliminated the possibility of non-specific hybridization due to a probe sequence external to the

551 gene. When purified plasmid preparations of pKPJ94 carrying APH(3')II and pGH54 carrying APH(3')I were probed instead of total DNA samples, the APH(3') subclass I and II probes did not display any cross-hybridization (results not shown). This suggested that such signals detected in hybridizations of total DNA were probably homologous sequences from chromosomal DNA that hybridized with the probe DNA. Previous studies have revealed that the APH(3')I gene of Tn903 does not cross-hybridize with the APH(3')II gene of Tn5 (Courvalin et al. 1978; Young et al. 1985). The APH(3') enzymes are also known to be immunologically different (Matsuhashi et al. 1976). However, in a study by Thompson and Gray (1983) the pairwise comparison of the Streptomyces fradiae APH(3') gene with the APH(3') genes encoded by Tn5 and Tn903 were shown to have significant nucleofide and amino acid homologies, which indicated a common evolutionary relationship of these antibiotic-resistant genes. This finding is also supported by the research of Beck et al. (1982). These researchers observed a high degree of protein homology in sequences of the corresponding regions of Tn5 and Tn903. They also detected significant homology at the nucleotide level in the carboxy-terminal parts of the two genes. This evidence suggested a common origin of apparently unrelated APH(3') subclass I and II enzymes. Similar APH(3') DNA probes have been previously used to characterize APH(3')I and APH(3')II enzyme activities in the Enterobacteriaceae and Pseudomonas spp. (Young et al. 1985). However, the present study examined uncharacterized environmental isolates. Young et al. (1985), reported that approximately 92% of the organisms examined showed probe homology with APH(3')I, and only 8% with the APH(3')II probe. In the present investigation a higher percentage of homology was detected with the APH(3')II than with the APH(3')I subclass. This high occurrence of the APH(3')II determinant may be the result of the initial bacterial isolation process, which may have repeatedly isolated the predominant resistant strains, while not detecting bacterial isolates present in very low numbers. However, the present investigation provides information on type I and II APH(3') resistant determinants and their distribution among Enterobacteriaceae populations in a food product such as meat. Furthermore, a considerable number of bacterial isolates obtained from ground beef samples exhibited probe homologies with both subclasses (APH[3']I and APH[3']II) or with neither of them. Although Young et al. (1985) did not report on the presence of DNA homology to both subclasses of APH(3') genes, many clinical bacterial isolates are known to contain several aminoglycoside-modifying enzymes (Bryan 1984). This would explain the probe homology of some DNA isolates which hybridized with both probes. In addition to aminoglycoside phosphotransferases, other modifying enzymes, particularly aminoglycoside acetyltransferases (AAC), are ~responsible for antibiotic resistance in many Enterobacteriaceae species such as Escherichia coli and Klebsiella spp.

(Bryan 1984). Aminoglycoside nucleotidetransferases (AAD) are another group of enzymes known to modify substrates such as Km but not Nm (Bryan 1984). The presence of such an AAD enzyme system in the isolates that did not display probe homology with APH(3') gene probes is supported by the fact that they failed to grow on neomycin even at the low concentration of 25 Ixg/ ml. All the other isolates which showed probe homology with APH(3') genes were able to grow in the presence of neomycin even at the high concentration of 200 p~g/ml. In summary, the digoxigenin-labelled probe DNA prepared from APH(3') structural gene subclasses I and II were successfully used to determine the occurrence of those subclasses of aminoglycoside-modifying enzymes in Enterobacteriaceae isolates obtained from meat samples. Types I and II of the APH(3') enzyme are the most frequently occurring modifying enzymes responsible for Km and Nm antibiotic resistance among such bacterial isolates. Another group of modifying enzymes, presumably of aminoglycoside adenyltransferase, which may encode for Km resistance also seems to be present among many of the ground beef isolates. Further investigations are being conducted to determine this potentially unidentified group of enzymes.

Acknowledgements. This research was supported by Natural Sciences and Engineering Research Council (Canada) grants to D. L. Collins-Thompsonand J. T. Trevors. A. H. G. P. Jayaratne was supported by a Commonwealth Graduate Scholarship. Appreciation is expressed to Dr. F. C. Tenover for providing reference bacterial strains and plasmids.

References Beck E, Ludwig G, Auerswald EA, Reiss B, Schaller H (1982) Nucleotide sequence and exact localization of the neomycin phosphotransferase gene from transposon Tn5. Gene 19:327336 Berg DE, Davies J, Allet B, Rochaix JD (1975) Transposition of R-factor genes to bacteriophages. Proc Natl Acad Sci USA 72:3628-3632 Berg DE, Jorgensen R, Davies J (1978) Transposable kanamycinneomycin resistance determinants. In: SchlessingerD (ed) Microbiology-1978. American Society for Microbiology, Washington, DC, pp 13-15 Bryan LE (1984) Aminoglycosideresistance. In: Bryan LE (ed) Antimicrobial drug resistance. Academic Press, Orlando, F1, pp 241-277 Brzezinska M, Davies J (1973) Two enzymeswhich phosphorylate neomycin and kanamycin in Escherichia coli strains carrying R factors. Anfimicrob Agents Chemother 3:266-269 Carter GI, Towner KJ, Pearson NJ, Slack RCB (1989) Use of nonradioactive hybridization assay by direct detection of Gramnegative bacteria carrying TEM fl-lactamase genes in infected urine. J Med Microb~Lol28:113-117 Chevrier D, Larzul D, Megraud F, Guesdon J (1989) Identification and classification of Campylobacter strains by using nonradioactive DNA probes. J Clin Microbiol 27:321-326 Courvalin P, Fiandt M, Davies J (1978) DNA relationship between genes coding for aminoglycoside modifying enzyme from antibiotic-producingbacteria and R plasmids. In: Schlessinger D (ed) Microbiology-1978.American Societyfor Microbiology, Washington, DC, pp 262-266

552 Davies J, Smith DI (1978) Plasmid determined resistance to antimicrobial agents. Ann Rev Microbiol 32:469-518 Davies J, Berg D, Jorgenson R, Fiandt M, Huang TSR, Courvalin P, Schloff J (1977) Transposable neomycin phosphotransferases. In: Drews J, Hogenauer G (eds) R-Factors: their properties and possible control. Springer, Berlin, Heidelberg, New York, pp 101-114 Feinman SE (1984) The transmission of antibiotic-resistant bacteria to people and animals. In: Thomas HJ, Dupont HL, Crawford LM (eds) Handbook series in zoonoses, vol I, section D. Antibiotics and public health. CRC Press, Boca Raton, Fla, pp 151-171 Goldman RP, Northrop DB (!976) Purification and spectrophotometric assay of neomycin phosphotransferase II. Biochem Biophys Res Comm 69:230-236 Holmberg SD, Oesterholm MT, Senger KA, Cohen ML (1984) Drug-resistant Salmonella from animals fed antimicrobials. N Engl J Med 311:617-622 Houvienen S, Houvienen P, Jacoby GA (1988) Detection of plasmid-mediated fl-lactamases with DNA probes. Antimicrob Agents Chemother 32:175-179 Jayaratne AHGP, Collins-Thompson DL, Trevors JT (1989) In vitro and in situ conjugal transfer of an R-plasmid among Enterobacteriaceae species isolated from meat samples. FEMS Microbiol Ecol 62:79-86 Jorgensen RA, Rothstein SJ, Reznikoff WS (1979) Restriction enzyme cleavage map of Tn5 and location of a region encoding neomycin resistance. Mol Gen Genet 177:65-72 Labigne-Roussel A, Briaux-Gerbaud S, Courvalin P (1983) Tn1525, a kanamycin R determinant flanked by two direct copies of IS 15. Mol Gen Genet 189:90-101 Lewington J, Greenaway SD, Spillane BJ (1986) Rapid small scale preparation of bacterial genomic DNA, suitable for cloning and hybridization analysis. Lett Appl Microbiol 5:51-53 Maniatis T, Fritsch EF, Sambrook J (1982) Molecular cloning: a laboratory manual, 1st edn. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY Matsuhashi H, Yagisawa M, Kondo S, Takeuchi T, Umezawa H (1975) Aminoglycoside 3'-phosphotransferase I and II in Pseudomonas aeruginosa. J Antibiot 28:442-447 Matsuhashi Y, Sawa T, Takeuchi T, Umezawa H (1987) Immunological studies of aminoglycoside 3'-phosphotransferase. J Antibiot 29:1127-1128 Miller GH, Sabatelli FJ, Hare RS, Waitz JA (1980) Survey of aminoglycoside resistance patterns. Dev Ind Microbiol 21:91104

Ng L, Stiles ME, Taylor DE (1987) DNA probe for identification of tetracycline resistance genes in Campylobacter species isolated from swine and cattle. Antimicrob Agents Chemother 31 : 1669-1674 Oka A, Sugisaki H, Takanami M (1981) Nucleotide sequence of the kanamycin resistance transposon Tn903. J Mol Biol 147:217-226 Pohl P (1977) Relationship between antibiotic feeding in animals and emergence of bacterial resistance in man. J Antimicrob Chemother 3 (Suppl C):67-70 Rubens CE, McNeill WF, Farrar WE Jr (1979) Evolution of multiple antibiotic resistance plasmids mediated by transposable plasmid deoxyribonucleic acid sequences. J Bacteriol 140:713-719 Siegel D, Huber WG, Enloe F (1974) Continuous non-therapeutic use of antimicrobial drugs in feed and drug resistance of the Gram-negative enteric florae of food-producing animals. Antimicrob Agents Chemother 6:697-701 Spika JS, Waterman SH, SooHoo GW, St Louis ME, Pacer RE, James SM, Bissett ML, Mayer LW, Chiu JY, Hall B, Greene K, Potter ME, Cohen ML, Blake PA (1987) Chloramphenicolresistant Salmonella newport traced through hamburger to dairy farms. N Engl J Med 316:565-570 Tenover FC, Gootz TD, Gordon KP, Tompkins LS, Young SA, Plorde JJ (1984) Development of a DNA probe for the structural gene of the 2"-0-adenyltransferase aminoglycoside-modifying enzyme. J Infect Dis 150:678-687 Thompson CJ, Gray GS (1983) Nucleotide sequence of a streptomycete aminoglycoside phosphotransferase gene and its relationship to phosphotransferase encoded by resistance plasraids. Proc Nat1 Acad Sci USA 80:5190-5194 Wetherall BL, McDonald PJ, Johnson AM (1988) Detection of Campylobacterpylori DNA by hybridization with non-radioactive probes in comparison with a 32p-labelled probe. J Med Microbiol 26:257-263 Young SA, Tenover FC, Gootz TD, Gordon KP, Plorde JJ (1985) Development of two DNA probes for differentiating the structural genes of subclasses I and II of the aminoglycoside-modifying enzyme 3'-aminoglycoside phosphotransferase. Antimicrob Agents Chemother 27:739-744 Zeph LR, Stotzky G (1989) Use of biotinylated DNA probe to detect bacteria transduced by bacteriophage P1 in soil. Appl Environ Microbiol 55:661-665 Zwadyk Jr P, Cooksey RC, Thornsberry C (1986) Commerical detection methods for biotinylated gene probes: comparison with 32p-labelled DNA probes. Curr Microbiol 14:95-100