Veterinary Microbiology, 22 (1990) 277-290 Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands

277

D N A P r o b e s for the D e t e c t i o n of T o x i n Genes in E s c h e r i c h i a coli Isolated from D i a r r h o e a l D i s e a s e in Cattle and P i g s M.J. WOODWARD, R. KEARSLEY, C. WRAY and P.L. ROEDER

MAFF Central Veterinary Laboratory, New Haw, Weybridge, Surrey KT15 3NB (Great Britain) (Accepted 25 September 1989)

ABSTRACT Woodward, M.J., Kearsley, R., Wray, C. and Roeder, P.L., 1990. DNA probes for the detection of toxin genes in Escherichia coli isolated from diarrhoeal disease in cattle and pigs. Vet. Microbiol., 22: 277-290. DNA gene probes specific for genes coding for heat labile toxin (LT), heat stable toxins (STpa, STpb) and Vero-cell toxins (VT1, VT2 ) were used to examine 1031 diarrhoeal disease isolates of E. coli (345 from cattle and 686 from pigs). Of the bovine strains, 60 hybridized with the STpa probe and most possessed the K99 (F5) or F41 adhesin. Five bovine strains possessed STpb genes and five either VT1 or VT2 genes. Of the porcine strains, 245 hybridized with one or more gene probes. Of 160 K88 (F4) positive strains, 133 possessed both LT and STpb genes, whilst 17 possessed LT or STpb or STpa alone or in combination. Ten K88 strains did not possess toxin genes. Isolates bearing the K99 (F5) adhesin possessed either STpa, STpb and VT2 genes alone or in combination;in one isolate only the LT gene was detected. Isolates belongingto serogroup 0138:K81 were more heterogeneous as to their toxin genes; of the 60 strains, fourteen carried only VT2 genes, thirty-two carried VT2, STpa and STpb genes, one carried LT, VT2, STpa and STpb genes, two carried STpb gene, four carried STpa and STpb genes, one carried LT and VT2 genes, two carried LT and STpa genes, whilst four carried none. Twenty-four percent of all toxigenic strains apparently did not possess adhesins.

INTRODUCTION

Enterotoxigenic E. coli (ETEC) are an important cause of diarrhoeal disease in both man and animals (Sojka, 1971; Sack, 1975; Wray and Morris, 1984; Sussman, 1985; Robins-Browne, 1985). The heat-labile toxin (LT), antigenically similar to cholera toxin, comprises one 25.5 kD subunit and four or five 11.0 kD subunit (Dallas and Falkow, 1979) and stimulates protective immunity. The heat-stable toxins (ST), of which three have been identified from human (STh) and porcine (STpa and STpb ) sources, are poorly immunogenic 0378-1135/90/$03.50

© 1990 Elsevier Science Publishers B.V.

278

M.J. WOODWARD ET AL.

proteins of 5.0 kD. LT exerts its effect by stimulating production of adenylate cyclase in the upper small intestine (Evans et al., 1972 ) and ST activates guanylate cyclase in the intestinal mucosa (Alderete and Robertson, 1978; Hughes et al., 1978). Two further E. coli toxins, verocytotoxins VT1 and VT2, which like LT produce morphological changes in Veto cells, have also been identified in human enteropathogenic E. coli (Konowalchuk et al., 1977; Scotland et al., 1980). A verocytotoxin (SLT-IIV) produced by strains of E. coli responsible for edema disease of pigs is 94% homologous at the genetic level with VT2 (Weinstein et al., 1988). Conventional techniques for the detection of ST require bioassays (Dean et al., 1972; Smith and Halls, 1967), and several cell culture assays (Guerrant et al., 1974; Sack and Sack, 1975; Konowalchuk et al., 1977) have been used for LT and VT. These methods are time consuming and unsuitable for routine diagnosis. An important process in the pathogenesis of diarrhoea is the adhesion of ETEC to the intestinal mucosa by means of proteinaceous filaments (termed fimbriae or pili) which are structurally and antigenically distinct from each other (Mooi and De Graaf, 1985 ). Furthermore, adhesins tend to be associated with a host-species, e.g. K88(F4) (Orskov et al., 1964) and 987P(F6) (Nagy et al., 1976) on porcine ETEC, and CFA/I and CFA/II (Evans et al., 1975; Evans and Evans, 1978) on human ETEC, although K99 (F5) (Orskov et al., 1975) has been found on ETEC from calves, lambs and pigs. Thus, serological techniques for detecting adhesins on ETEC from animals have been used as an indicator of enterotoxigenicity (Merson et al., 1979). Genes encoding LT, ST and VT have been cloned (Dallas et al., 1979; So et al., 1979; Lathe et al., 1980; Willshaw et al., 1985 and 1987) and used as probes to detect enterotoxin genes in E. coli isolated from humans and animals with diarrhoea (Moseley et al., 1980; Patamoroj et al., 1983; Echeverria et al., 1984; Hill and Payne, 1984; Maas et al., 1985; Mainil et al., 1985; Willshaw et al., 1985; Hill et al., 1986). In a comparative study on methods for assaying LT, Vadivelu et al. (1987) demonstrated that DNA-DNA hybridization was a highly sensitive, specific and reproducible technique. On the basis of these findings and cogniscent of the need to replace animal tests, it was decided to develop DNA-DNA hybridization to a range of toxin genes for routine use in the diagnostic laboratory. The objectives were to subclone existing toxin gene probes so that probe DNA would be extracted simply and free from contaminating DNA sequences, to establish simple hybridization protocols and to apply probes to E. coli isolates from cases of animal enteritis. We report here our initial findings. MATERIALSAND METHODS Bacterial strains

One-thousand-and-thirty-one E. coli strains isolated from cattle and pigs suffering from intestinal infection during the period 1987-1988 were stored on

TOXIN GENES IN E. COLI FROM CATTLE AND PIGS

279

Dorset egg slopes (PM5, Oxoid) at room temperature. All cultures were subcultured every 6 months and before use in DNA-DNA hybridization experiments. All isolates were serotyped and tested for adhesins using standard methods based on those of Sojka (1965). A panel of reference strains which produced toxins as assessed by standard methods have been described (Sack and Sack, 1975; Nagy et al., 1976; Moseley et al., 1980; Willshaw et al., 1987) and were used as positive control strains in DNA-DNA hybridization experiments (Table 1 ). E. coli strain DH5alpha F - endA1 hsdR17 (r- m - ) supE44 thi-1 recA1 gyrA96 relAlO8OdlacZM15 (Gibco-BRL) was used as recipient in transformation experiments. Nucleic acid methods Plasmid DNA extraction, restriction endonuclease digestion, gel electrophoresis, electroelution of DNA fragments from agarose and polyacrylamide gels, Bal31 nuclease digestion, DNA polymerase infilling of recessed ends, ligation and transformation were performed using standard methods (Maniatis et al., 1982). Using pUC8 (Vieira and Messing, 1982) as the cloning vector, white transformants were selected on complete agar (Alaeddinoglu and Charles, 1979 ) supplemented with ampicillin (50ttg m1-1) and XGal (Miller, 1972) and screened for plasmids (Birnboim and Doly, 1979). To determine the size of insert, plasmid DNA was digested with the relevant restriction endonuclease before gel electrophoresis. D N A - D N A hybridization assay (a) Derivation and subcloning of toxin gene probes LTprobe. The LT probe was a 750bp HindIII fragment derived from pEWD299 (Dallas et al., 1979). STpa probe. The STpa probe was a 157bp HinfI fragment derived from pRIT10036, formerly pCLS2 (Lathe et al., 1980; Moseley et al., 1980) purification of which was hindered by a 154bp HinfI fragment derived from pBR322 into which the STpa region had been originally cloned (Lathe et al., 1980). Therefore, to simplify probe preparation, the pSTpa region was recloned. Plasmid pRIT10036 DNA was digested to completion with HaeIII and the largest fragment of approximately lkb in which the STpa region was centrally located (Lathe et al., 1980 ) was purified. Prior to recloning, this fragment was digested with Bal31 nuclease removing about 200bp from each end. This removed extraneous DNA sequences but left the STpa probe excision sites (So and McCarthy, 1980) intact. The reduced fragment was infilled, ligated with SmaI cut pUC8 and the DNA was used to transform E. coli K12 strain DH5alpha.

280

abc

M.J. WOODWARDET AL.

d

bp probe Op

Fig. 1. Restriction endonuclease digests of pRIT10036 and pVWIST run through 5% polyacrylamide gels and visualized by UV transilluminationafter ethidium bromide staining, pBR322 digested with HaeIII as molecular weight markers (lanes a and d); pRIT10036 digested with Hinfl (lane b ); pVWIST digested with HinfI (lane c).

Transformants were screened for plasmid content and restriction digests were run on 5% polyacrylamide gels to verify presentation of the 157bp HinfI STpa probe DNA fragment free from adjacent bands; one transformant harbouring a recombinant plasmid, designated pVWIST, had an easily recoverable STpa probe (Fig. 1). STpb probe. The STpb probe was a 460bp Hinfl fragment derived from the smaller of two HindIII fragments of pRAS, formerly pCHL6 (Lee et al., 1983 ). To simplify probe preparation, the STpb probe was recloned. The STpb HinfI probe DNA fragment was infilled, ligated with Sinai cut pUC8 and the DNA used to transform E. coli 12 strain DH5alpha. The SmaI site was located immediately between EcoRI and BamHI restriction endonuclease sites in the multiple cloning site of the pUC8 vector (Vieira and Messing, 1982). Thus, one transformant harboured a plasmid, designated pVWIIST, of approximately 3.1kb which when cut with EcoRI and BamI together gave fragments of 2.7kb (linear vector) and 460bp (STpb probe ).

TOXIN GENES IN E. COLI FROM CATTLE AND PIGS

281

VT1 probe. The VT1 probe was a 530bp HindIII/AccI fragment derived from pNTP705 (Willshaw et al., 1985). VT2 probe. The VT2 probe was a 660bp EcoRV/HincII fragment derived from pNTP707 (Willshaw et al., 1987). (b) Probe preparation Probe fragments were purified by electroelution of DNA-containing bands cut from agarose (if more than 500bp) or polyacrylamide (if less than 500bp ) gels of restriction enzyme digests. After electroelutions, probe DNA fragments were washed successively in phenol, phenol: chloroform (1 : 1) chloroform (Maniatis et al., 1982) and concentrated by precipitation under 2.5 volumes of 100% ethanol at - 7 0 ° C after addition of sodium acetate to 150mM. Probe DNA was radiolabelled to about 1X109 cpm /lg -1 specific activity with alpha32PdCTP by random hexanucleotide primer DNA synthesis (Multiprime, Amersham) as described by Feinberg and Volgestein (1983).

(c) Colony lysis Cultures for testing were grown overnight in brain-heart infusion broth containing 15% (w/v) glycerol at 37°C with agitation. From each culture 200~tl were transferred to separate wells of a sterile microtitre plate (Falcon). A multipoint inoculator (Don Whitley Scientific ) was used to transfer cultures from microtitre plates to the surface of a nylon membrane (Hybond-N, Amersham) overlaid on a dried MacConkey agar plate. Inoculated microtitre plates were stored at - 70 ° C and thawed for reinoculation of nylon membranes when required. Inoculated filters were incubated at 37 ° for 4h. Colonies were lysed by laying filters colony side up sequentially on blotting paper saturated with 5% SDS (5 min), 0.5 M NaOH, 1.5 M NaC1 (15 min) and 1.5 M NaC1, 1.0 M TrisHC1 pH7 (2 X 7 min) based on the technique of Mainil et al. (1985). Air dried filters were wrapped in Saran wrap and exposed to ultra-violet irradiation (302nm) by inversion over a transilluminator (ultra violet product inc. ) for 4 min to fix the DNA.

(d) Hybridization When probing for LT, STpa or STpb genes, filters were prehybridized for at least l h at 65°C in 6 x S S C ( 1 x S S C is 0.15M NaC1, 0.015M sodium citrate, pH7), 5 × Denhardt's solution ( 1 × Denhardt's is 0.02% (w/v) bovine serum albumin (Pentax fraction V), 0.02% (w/v) Ficoll Mr 400 000, 0.02% polyvinylpyrolidone Mr 360 000) and 0.5% (w/v) SDS containing 20/1g m1-1 sonicated heat denatured salmon sperm DNA (Sigma). When probing for VT1 or VT2 gene filters were prewashed at 42 °C for 2h in 50mM Tris/HC1 (pH8.0), 1M NaC1, l m M EDTA and 0.1% SDS followed by prehybridization for 4h at 42 ° C in 50% v/v formamide, 5 X SSPE (1 X SSPE

282

M.J. W O O D W A R D E T AL.

is 0.18M NaC1, 10mM NaH2PO4, l m M EDTA), 5 ×Denhardt's solution and 0.1% (w/v) SDS containing 100 ttg m1-1 sonicated heat denatured salmon sperm DNA. Radiolabelled probes were denatured at 100 °C for 2 min then immediately plunged in ice and added to the filters (10ng probe DNA per 10cm × 10cm filter) in fresh hybridization solution. The probes were hybridized for 16h; those against LT, STpa and STpb being performed at 65 oC and those against VT1 and VT2 at 42°C. Post-hybridization stringency washes for all probes were in 2 × SSC, 0.1% SDS (4× 10 min) at room temperature, I × S S C , 0.1% SDS (2h) at 68°C and 0.2 × SSC, 0.1% SDS ( l h ) at 68°C. After air drying, filters were exposed to Xray film (Fuji RX) with intensifying screens (Cronex) at - 7 0 ° C for periods ranging from 30 min to 4h.

(e) Storage and re-use o/filters Filters were stored at 4 oC wrapped in Saran wrap within a light-proof container. Probe was removed from filters by washing in 0.4M NaOH (30 mins) at 65°C followed by 0.1xSSC, 0.1% SDS, 0.2M Tris-HC1 (pH7.5) (lh) at 65 oC. Filters were partially dried and stored damp as described above. Hybridization procedures reusing filters were as described except that prehybridization was reduced to 30 min. RESULTS

Validation of gene probes The gene probes were used in DNA-DNA hybridization experiments against a panel of reference strains maintained at the Central Veterinary Laboratory (Table 1 ). Each reference strain had been tested for the production of each of five toxins by animal tests, by tissue culture tests and by immunological tests (see Methods). There was an exact correlation between positive probe response and production of toxin (Table 1 ).

0 serogroups of E. coli strains from diarrhoeal disease A total of 1031 strains were probed, 345 from cattle and 686 from pigs. Of the bovine strains 113 were untypable, 56 were 0101 serogroup, 44 were 08 serogroup, 23 were 017 serogroup whilst the remainder belonged to 48 different serogroups. Of the porcine strains 158 were untypable, 72 were 0149 serogroup, 61 were 0157 serogroup, 60 were 0138 serogroup, 42 were 08 serogroup, 27 were 0101 serogroup, 26 were 0147 serogroup whilst the remainder belonged to 69 serogroups.

283

TOXIN GENES IN E. COLI FROM CATTLE AND PIGS TABLE 1 Results of hybridization of toxin gene probes against reference strains Source

Strain

Toxins produced

Serotype

Probe response LT

STpa

STpb

VT1

VT2

Calf

B44 B42 B41 Blll B79

0 9 : K 3 0 : K 9 9 F41 0 9 : K 3 5 : K 9 9 F41 0 1 0 1 : K - : K 9 9 F41 0 1 0 1 : K 3 0 : K 9 9 F41 0 1 0 1 : K 3 2 : K 9 9 F41

STpa STpa STpa STpa STpa

-

+ + + + +

-

-

-

Pig

G7 G205 V50 V79 1751 Vl13 E57 E4 E681 Gl108 Abbotstown

08:K87:K88ab 08:K87:K88ac 010:K103 035:K'V79' 0101:F41 0119:K'V113' 0138:K81 0139:K82 0141:K85ab 0141:K85ab:K88ab 0149:K91:K88ac

LT STpb LT STpb none STpb STpa none STpa VT2 none LT STpb VT2 LT STpb VT2 LT STpa STpb

+ +

-

-

+

-

+

+ + +

+

+ + + + + +

-

+ + + -

H19 E32511

026:H19 0157:H-

VT1 VT2

-

+ -

+

Human

TABLE 2 S e r o t y p e a n d t o x i n g e n e p a t t e r n a m o n g s t 72 b o v i n e s t r a i n s i d e n t i f i e d as t o x i g e n i c b y g e n e p r o b e s Serotype of toxigenic strain

N u m b e r of isolates b e a r i n g toxin genes STpa

015 K ' R V C 4 7 6 0 ' 071 ( g r o u p ) 093 ( g r o u p ) 0113 ( g r o u p ) 08:K25 K99 08:K84ab K99 0 9 : K 3 2 K 9 9 F41 0 9 : K 3 5 K 9 9 F41 020:K17 K99 0101:K- K99 0101: K - K 9 9 F41 0101:K28 K99 0101:K30 K99 0 1 0 1 : K 3 0 K 9 9 F41 0101:K32 K99 0 1 0 1 : K 3 2 K 9 9 F41 0 1 0 1 : K 2 7 F41 0 1 0 1 : K 3 0 F41 u n t y p a b l e F41

STpa/STpb

STpb

Other

0 0 0 0

0 0 0 0

1 0 1 0

0 1,VT1 0 1,VT2

1 11 1 2 1 4 1 31 1 1 0 1

0 0 0 0 0 0 0 1 1 0 0 0

0 0 1 0 0 0 0 0 0 0 1 0

0 0 0 0 0 0 0 0 0 0 0 0

2 2 1

0 0 0

0 0 1

0 0 3,VT2

60

2

5

5

284

M.J. WOODWARDET AL.

TABLE 3 Serotype a n d toxin gene p a t t e r n amongst 245 porcine strains identified as toxigenic by gene probes Serotype of toxigenic strain

N u m b e r s of isolates bearing toxin genes LT

LT/STpb

STpb

STpa

Other 1,STpb/STpa 0 0 9,VT2 1,STpa/STpb 2,VT2 0 0 6,VT2 29,VT2/STpa/STpb 1,VT2 2,VT2 ] ,STpa/STpb 1,VT2/STpa/STpb 1,LT/STpa

08 (group) 09 (group) 020:K101 026 (group) 053:KV142 065:K071 (group) 0115:KV165 0138:K81

0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 1 0 0

0 0 0 0 1 0 0 2 0

0 1 1 0 1 0 0 0 0

rough untypable

0 1

0 2

0 3

0 5

08:K27 K88ab 045:KE565 K88ab 0141:K85ab K88ac 0141:K91 K88ac 0147:K89 K88ac 0149:K- K88ac 0149:K89 K88ac 0149:K92 K88ac 0157:KV17 K88ac

1 0 1 1 1 0 0 1 3

4 6 1 0 19 1 1 68 33

0 0 0 0 1 0 0 0 3

0 0 0 0 0 0 0 0 1

0 1,STpa/STpb 2,STpa/STpb 0 1,LT/STpa/STpb 0 0 0 0

0 2 : K - K99 987P 08: K - K99 021 (group) K99 987P F41 064:K- K99 987P F41 0101:K30 K99 0120:K30 K99 0120:K- K99 0138:K81 K99

0 0

0 0

0 0

0 1

I,VT2 0

1 0 0 0 0

0 0 0 0 0

0 0 0 0 0

0 1 3 1 2

0 0 0 0

0138:K81 K99 F41

0

1

0

0

0138:K81 F41

0

1

0

0

10

138

10

17

1 ,VT 2,VT2/STpa/STpb 2,STpa/STpb 1,VT2/STpa/STpb 1,LT/VT2/STpa/STpb 1,VT2 1,LT/VT2 2,STpa/STpb

70

TOXIN GENES IN E. COLI FROM CATTLE AND PIGS

285

Gene probing of bovine E. coli isolates Seventy-two bovine E. coli strains hybridized with one or more toxin gene probes {Table 2). Sixty of the positive strains harboured the STpa gene, all of which possessed either K99 or F41 adhesins. Serogroup 0101 accounted for 46 of the toxin positive strains but two 0101 strains, although K99 positive, did not hybridize with any probe. Five strains were STpb positive of which two apparently lacked adhesins and a further two were STpa and STpb positive. Verocytotoxin gene, VT2, as detected in four strains, three of which were untypable but possessed F41 adhesin and one belonged to serogroup 0113 but lacked detectable adhesin. Only one strain, belonging to serogroup 071 hybridized VT1 and this strain lacked detectable adhesin. Two strains belonging to serogroup 0101 and possessing K99 adhesins failed to hybridize any probe. Gene probing of porcine E. coli isolates Two-hundred-and-forty-five of 692 porcine E. coli isolates hybridized one or more toxin gene probes (Table 3 ). Of strains possessing K88 adhesin, 133 were LT and STpb positive, 8 were LT positive, 4 were STpb positive, 1 was STpa positive, 3 were STpa and STpb positive and 1 was LT, STpa and STpb positive. Ten K88 bearing strains did not hybridize any probe. Of fourteen strains possessing K99 adhesin but not belonging to serogroup 0138, 8 were STpa positive, 1 was LT positive, 1 was VT2 positive and 4 did not hybridize any probe. Verocytotoxin gene, VT2, as detected in 58 strains belonging to four 0 serogroups of which 0138 accounted for 42 strains and 026 accounted for 9 strains. Serogroup -0138 was heterogenous with respect to adhesin and toxin gene possession with, for example, one strain possessing K99 and F41 adhesins and LT, STpa, STpb and VT2 toxin genes. Two 0138 strains failed to hybridize any gene probe. A total of 72 toxin probe positive strains did not express any detectable adhesin; within this group were 50 strains bearing the VT2 toxin gene of which 35 belonged to 0138 serogroup and 9 belonged to 026 serogroup. DISCUSSION The use of gene probes for the detection of LT and ST in human and animal E. coli isolates has become an established laboratory technique (Moseley et al., 1980; Patamoroj et al., 1983; Mainil et al., 1985; Olsvik et al., 1985; Monckton and Hasse, 1988). In the present study, the range ofE. coli toxins was extended by the use of VT probes and simple reproducible protocols suitable for the routine laboratory were established. Whether toxins were produced by all the strains found to be positive by gene probing was not tested; in view of the correlation between probing and toxin production demonstrated by Vadivelu et al. (1987) it seems likely that the majority of hybridization positive strains were producing toxin. Genes coding for STpa, STpb, VT1 and VT2 toxins were detected in the

286

M.J. WOODWARD ET AL.

bovine E. coli isolates. With a few exceptions, the presence of toxin genes correlated with the possession of K99 or F41 adhesins. No LT genes were detected. Harnett and Gyles (1985) found that ST and K99 often transferred together and it is possible that these virulence determinants may be on the same plasmid. Both ST and K99 were found to be associated within a limited range of E. coli 0 groups, 08 and 0101 predominating. Mainil and Kaeckenbeeck (1986) suggested that plasmid instability may occur in other E. coli serotypes which may result in them not elaborating toxins and adhesins. Most of the bovine ETEC encoded STpa, although STpb alone was detected in five isolates, of which two did not possess either the K99 or F41 adhesins. Mainil et al. (1985) also detected STpb in bovine isolates and suggested that these strains may be accidental pig strains but in our experience 093 and 015 serogroups are uncommon porcine isolates and strains belonging to the 015 serogroup are recognized calf pathogens ( Sojka, 1965). A wider range of toxin genes were detected in the porcine E. coli isolates. Although most of the K88 positive strains produced both LT and STpb, this was not invariably the case and some cultures produced only one or other of the two toxins. The presence of the STpb gene alone was, however, uncommon which agrees with the findings of Harnett and Gyles (1985). The greater diversity of genotypes may be explained by the findings of Franklin et al. (1981) who showed that in some strains ST and K88 genes were on the same plasmid and in others on different plasmids. Porcine K99 positive E. coli were similar to those isolated from cattle in that they usually encoded STpa. E. coli harbouring verocytotoxin genes were uncommon with the exception of isolates of the 0138:K81 serotype many of which also encoded STpa and STpb and, in three cases, LT. Some of these strains have been shown to be enteroinvasive and associated with post-weaning diarrhoea (Linggood and Thompson, 1987 ). Although the VT2 probe was derived from a verocytotoxin gene of human origin, it is assumed that it hybridized SLT-11v, sharing 94% homology, expression of which caused edema disease in pigs (Marques et al., 1987 ). A significant number of strains carrying toxin genes apparently lacked K99, K88, 987P and F41 adhesins; poor expression in vitro or loss of these genes may explain this result. To overcome this problem, others (Lanser and Anargyros, 1985; Mainil et al., 1987; Monckton and Haase, 1988) have used gene probes specific for adhesin genes in ETEC surveys. Alternatively, it is possible that these strains elaborate newly described or previously undescribed adhesins; for example F17 has been reported (Lintermans et al., 1988) and may be common in ETEC although electron microscopy did not reveal fimbriae in any of the 61 isolates examined in this work (unpublished observations). It is possible that E. coli may acquire toxin genes alone by genetic exchange in vivo and thus act as a reservoir for dissemination of these genes despite not being actively pathogenic. That toxin genes were present in the less frequently observed serogroups of E. coli supports the findings of Fairbrother et al. (1988)

TOXIN GENES IN E. COLI FROM CATTLE AND PIGS

287

w h o also d e m o n s t r a t e d t h e o c c u r r e n c e o f m a n y a d h e s i n p o s i t i v e t o x i n n e g a t i v e strains. A t o t a l of t e n a d h e s i n p o s i t i v e t o x i n n e g a t i v e s t r a i n s were identified in t h i s survey; loss of p l a s m i d s e n c o d i n g t o x i n g e n e s u p o n p r o l o n g e d s t o r a g e m a y a c c o u n t for these. N e w t o x i n s s u c h as L T I I ( S e r i w a t a n a et al., 1988) h a v e b e e n d e s c r i b e d a n d , if it is a s s u m e d t h a t a c q u i s i t i o n o f a d h e s i n a n d t o x i n genes is a d y n a m i c p r o c e s s , t h e n t h e c o n t i n u e d use of E. coli v a c c i n e s b a s e d o n a l i m i t e d r a n g e of a d h e s i n s will select for E T E C b e l o n g i n g to t h e less f r e q u e n t l y o b s e r v e d s e r o g r o u p s a n d e n h a n c e t h e i r p r e v a l e n c e in t h e p o p u l a t i o n . ACKNOWLEDGEMENTS T h e a u t h o r s are g r a t e f u l to Dr. S. M o s e l e y for t h e gifts of p R I T 1 0 0 3 6 , p R A S , p E W D 2 9 9 a n d to Dr. H. S m i t h for t h e gifts of p N T P 7 0 5 a n d p N T 7 0 7 . T h e a u t h o r s also a c k n o w l e d g e t h e helpful d i s c u s s i o n w i t h Drs. G.A. W i l l s h a w a n d H. S m i t h c o n c e r n i n g V T p r o b e s a n d Mr. G. Sullivan for his t e c h n i c a l assistance.

REFERENCES Alaeddinoglu, N.G. and Charles, H.P., 1979. Transfer of a gene for sucrose utilization in Escherichia coli K12 and consequent failure of expression of genes for D-serine utilization. J. Gen. Microbiol., 110: 47-59. Alderete, J.F. and Robertson, D.C., 1978. Purification and chemical characterization of the heatstable enterotoxin produced by porcine stains of enterotoxigenic Escherichia coli. Infect. Immun., 19: 1021-1030. Birnboim, H.C. and Doly, J., 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acid Res., 1: 1515-1523. Dallas, W.S. and Falkow, S., 1979. The molecular nature of heat-labile enterotoxin (LT) of Escherichia coll. Nature, 277: 406-407. Dallas, W.S., Gill, D.M. and Falkow, S., 1979. Cistrons encoding Escherichia coli heat-labile toxin. J. Bacteriol., 139: 850-858. Dean, A.G., Ching, Y.-C., Williams, R.G. and Harden, L.B., 1972. Test for Escherichia coli enterotoxin using infant mice. Application in a study of diarrhoea in children in Honolulu. J. Infect. Dis., 15: 407-411. Echeverria, P., Seriwatana, J., Leksomboon, U., Tirapat, C., Chaicumpa, W. and Rowe, B., 1984. Identification by DNA hybridization of enterotoxigenic Escherichia coli in homes in children with diarrhoea. Lancet, 1: 63-66. Evans, D.G. and Evans, D.J., 1978. New surface-associated heat-labile colonisation factor antigen (CFA/II) produced by enterotoxigenic Escherichia coli of serogroups 06 and 08. Infect. Immun., 21: 638-647. Evans, D.J., Chen, L.C., Curlin, G.T. and Evans, D.G., 1972. Stimulation of adenyl cyclase by Escherichia coli enterotoxin. Nature, 236: 137-138. Evans, D.G., Silver, R.P., Evans, D.J., Chase, D.G. and Gorbach, S.L., 1975. Plasmid-controlled colonisation factor associated with virulence in Escherichia coli enterotoxigenic for humans. Infect. Immun., 12: 656-666. Fairbrother, J.M., Lariviere, S. and Johnson, W.M., 1988. Prevalence of fimbrial antigens and enterotoxins in nonclassical serogroups of Escherichia coli isolated from newborn pigs with diarrhoea. Am. J. Vet. Res., 49: 1325-1328.

288

M.J. WOODWARDET AL.

Feinberg, A.P. and Vogelstein, B., 1983. A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem., 132: 6-13. Franklin, A., Soderlind, 0. and Molby, R., 1981. Plasmids coding for enterotoxins, K88 antigens and colicins in porcine Escherichia coil strains of 0 group 149. J. Med. Microbiol. Immunol., 170: 63-72. Guerrant, R.L., Brunton, L.L., Schnaitman, T.C., Rebhun, I. and Gilman, A.G., 1974. Cyclic adenosine monophosphate and alteration of Chinese hamster ovary cell morphology: a rapid, sensitive in vivo assay for the enterotoxins of Virbio cholerae and Escherichia coli. Infect. Immun., 10: 320-327. Harnett, N.M. and Gyles, C.L., 1985. Enterotoxin plasmids in bovine and porcine enterotoxigenic Escherichia coli of 0 groups 9, 20, 64 and 101. Can. J. Comp. Med., 49: 79-87. Hill, W.E. and Payne, W.L., 1984. Genetic methods for the detection of microbial pathogens. Identification of enterotoxigenic Escherichia coli by DNA colony hybridization. Collaborative study. J. Assoc. Off. Anal. CHem., 67: 801-807. Hill, E.W., Wentz, B.A., Payne, W.L., Jagow, J.A. and Zon, G., 1986. DNA colony hybridization method using synthetic oligonucleotides to detect enterotoxigenic Escherichia coll. J. Assoc. Off. Anal. Chem., 69: 531-536. Hughes, J.N., Murad, F., Chang, B. and Guerrant, R.L., 1978. Role of cyclic GMP in the action of heat stable enterotoxin of Escherichia coll. Nature, 271: 755-756. Konowalchuk, J., Spiers, J.I. and Stavric, S., 1977. VERO cell response to a cytotoxin of Esche richia coll. Infect. Immun., 18: 755-779. Lanser, J.A. and Anargyros, P.A., 1985. Detection of Escherichia coli adhesins with DNA probes. J. Clin. Microbiol., 22: 425-427. Lathe, R., Hirth, P., DeWilde, M., Harford, N. and Lecocq, J.-P., 1980. Cell-free synthesis of enterotoxin ofE. coli from a cloned gene, Nature, 284: 483-474. Lee, C.H., Moseley, S.L. and Moon, H.W., 1983. Characterization of the gene encoding heat-stable toxin II and preliminary molecular epidemiological studies of enterotoxigenic Escherichia coli heat-stable toxin II producers. Infect. Immun., 42: 264-268. Linggood, M.A. and Thompson, J.M., 1987. Verotoxin production among porcine strains of Esch erichia coli and its association with oedema disease. J. Med. Microbiol., 25: 359-362. Lintermans, P., Pohl, P., Deboeck, F., Bertels, A., Schlicker, C., Vandekerckhove, J. Van Damme. J. Van Montage, M. and De Greve, H., 1988. Isolation and nucleotide sequence of the F17-A gene encoding the structural protein of F17 fimbriae in bovine enterotoxigenic Escherichia coll. Infect. Immun., 56: 1475-1484. Maas, R., Silva, R.M., Gomes, T.A.T., Trabulsi, L.R. and Maas, W.K., 1985. Detection of genes for heat-stable enterotoxin I in Escherichia coli strains isolated in Brazil. Infect. Immun., 49: 46-51. Mainil, J. and Kaekenbeeck, A., 1986. Les souches d'Echerichia coli enterotoxigenes du veau produisent un seul type de toxine active chez le souriceau: STaP. Ann. Med. Vet., 130: 531534. Mainil, J., Bex, F., Coutorier, M. and Kaekenbeeck, A., 1985. Hybridization of bovine enterotoxigenic Escherichia coli with the heat-stable enterotoxin gene probes. Am. J. Vet. Res., 46: 2582--2584. Mainil, J., Sadowski, P.L., Tarsio, M. and Moon, H.W., 1987. In vivo emergence of enterotoxigenic Escherichia coli variants lacking genes for K99 fimbriae and heat-stable enterotoxin. Infect. Immun., 55: 3111-3116. Maniatis, T., Fritsch, C.F. and Sambrook, J., 1982. Molecular cloning: A laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring harbor, New York, NY. Marques, L.R.M., Pairis, J.S.M., Cryz, S.J. and O'Brien, A.D., 1987. Escherichia coli strains isolated from pigs with edema disease produce a variant of Shiga-like toxin II. FEMS Letters, 44: 33-38.

TOXIN GENES IN E. COLI FROM CATTLE AND PIGS

289

Merson, M.H., Orskov, F., Orskov, I., Sack, R.B., Huq, I. and Koster, F.T., 1979. relationship between enterotoxin production and serotype in enterotoxigenic Escherichia coll. Infect. Immun., 23: 325-329. Miller, J.H., 1972. Experiments in Molecular Genetics, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, pp. 73-79. Monckton, R.P. and Haase, D., 1988. Detection of enterotoxigenic Escherichia coli piggeries in Victoria by DNA hybridization using K88, K99, LT, ST1 and ST2 probes. Vet. Microbiol., 16: 273-281. Mooi, F.R. and De Graaf, F.K., 1985. Molecular biology in fimbriae of enterotoxigenic Escherichia coli. Curr. Top. Microbiol. Immunol., 118: 199-138. Moseley, S.L., Huq, I., Alim, A.R.M.A., So, M., Samaadpour-Motalebi, M. and Falkow, S., 1980. Detection of enterotoxigenic Escherichia coli by DNA colony hybridization. J. Infect. Dis., 142: 892-898. Nagy, B., Moon, H.W. and Issacson, R.E., 1976. Colonisation of porcine small intestine by Escherichia coli: ileal colonisation and adhesin by pig enteropathogens that lack K88 antigen and by some acapsular mutants. Infect. Immun., 13: 124-1220. Olsvik, 0., Solberg, R. and Bergan, T., 1985. Characterisation of enterotoxigenic Escherichia coll. Acta Pathol. Microbiol. Immunol. Scand., 93: 255-262. Orskov, I., Orskov, F., Sojka, W.J. and Wittig, W., 1964. K antigens K88ab (L) and K88ac (L) in E. coli. A new 0 antigen: 0147 and a new K antigen: K89(B). Acta Pathol. Microbiol. Scand., 62: 439-477. Orskov, I., Orskov, F., SMith, H.W. and Sojka, W.J., 1975. The establishment of K99, a thermolabile, transmissible Escherichia coli K antigen previously called "KCO", possessed by calf and lamb enteropathogenic strains. Acta Pathol. Microbiol. Scand., 83:31-36. Patamoroj, U., Seriwatana, J. and Echeverria, P., 1983. Identification of enterotoxigenic Escherichia coli isolated from swine with diarhoea in Thailand by colony hybridization, using three enterotoxin gene probes. J. Clin. Microbiol., 18: 1429-1431. Robins-Browne, R.M., 1985. Pathogenesis of bacterial diarrhoea - an overview. In: S. Tzipoli (Editor), Infectious diarrhoea in young. Elsevier, Amsterdam, pp. 7-23. Sack, R.B., 1975. Human diarrheal disease caused by enterotoxigenic Escherichia coll. Ann. Rev. Microbiol., 29: 333-353. Sack, D.A. and Sack, R.B., 1975. Test for enterotoxigenic Escherichia coli using Y1 adrenal cells in miniculture. Infect. Immun., 11: 334-336. Scotland, S.M., Day, N.P. and Rowe, B., 1980. Production of a cytotoxin affecting vero cells by strains of Escherichia coli belonging to traditional enteropathogenic serogroups. FEMS Microbiol. Letters, 7: 15-17. Seriwatana, J., Echeverria, P., taylor, D.N., Rasrinaul, L., Brown, L.E., Peiris, J.S.M. and Clayton, C.L., 1988. Type II heat-labile enterotoxin-producing Escherichia coli isolated from animals and humans. Infect. Immun., 56: 1158-1161. Smith, H.W. and Halls, S., 1967. Observations by the ligated intestinal segment and oral inoculation methods in Escherichia coli infections in pigs, calves, lambs and rabbits. J. Pathol. Bacteriol., 93: 531-545. So, M. and McCarthy, B.J., 1980. Nucleotide sequence of bacterial transposon Tn1681 encoding a heat stabile toxin (ST) and its identification in enterotoxigenic E. coli strains. Proc. Natl. Acad. Sci. U.S.A., 77: 4011-4015. So, M., Heffron, F. and McCarthy, B.J., 1979. The Escherichia coli gene encoding heat stable toxin is a bacterial transposon flanked by inverted repeats of ISI. Nature, 277: 453-456. Sojka, W.J., 1965. Escherichia coli in domestic animals and poultry. Commonwealth Agricultural Bureaux. Farnham Royal. Sojka, W.J., 1971. Enteric diseases in newborn piglets, calves and lambs due to Escherichia coli infection. Vet. Bull., 41: 509-522.

290

M.J. WOODWARDET AL.

Sussman, M., 1985. The virulence of Escherichia coli. Reviews and Methods. Society of General Microbiology Special publication, Academic Press, pp. 7-45. Vadivelu, J., Dunn, D.T., Feachem, R.G., Drasar, B.S., Cox, N.P., Harrison, T.J. and Lloyd, B.J., 1987. Comparison of five assays for the heat-labile enterotoxin of Escherichia coll. J. Med. Microbiol., 23: 221-226. Vieira, J. and Messing, J., 1982. The pUC plasmids, and M13mp-7 derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene, 19:259 268. Weinstein, D.L., Jackson, M.P., Samuel, J.E., Holmes, R.K. and O'Brien, A.D., 1988. Cloning and sequencing of a Shiga-like toxin type II variant from an Escherichia coli strain responsible for edema disease of swine. J. Bacteriol., 170: 4223-4230. Willshaw, G.A., Smith, H.R., Scotland, S.M. and Rowe, B., 1985. Cloning genes determining the production of Vero Cytotoxin by Escherichia coll. J. Gen. Microbiol., 131: 3047-3053. Willshaw, G.A., Smith, H.R., Scotland, S.M., Field, A.M. and Rowe, B., 1987. Heterogeneity of Escherichia coli phages encoding Vero cytotoxin: Comparison of clones sequences determining VT1 and VT2 and development of specific gene probes. J. Gen. Microbiol., 133: 1309-1317. Wray, C. and Morris, J.A., 1984. Enteric diseases in newborn piglets, calves and lambs caused by enterotoxigenic Escherichia coli. J. Hyg., 95: 577-585.