Vol. 29, No. 10
JOURNAL OF CLINICAL MICROBIOLOGY, OCt. 1991, p. 2250-2252 0095-1137/91/102250-03$02.00/0 Copyright © 1991, American Society for Microbiology
Immunodot Detection of Escherichia coli Heat-Stable Enterotoxin b by Using Enhanced Chemiluminescence Reaction LOUIS-ANDRE LORTIE, JOSEE HAREL, JOHN M. FAIRBROTHER, AND J. DANIEL DUBREUIL* Department of Pathology and Microbiology, Faculty of Veterinary Medicine, University of Montreal,
3200 Sicotte, P.O. Box 5000, Saint-Hyacinthe, Quebec, Canada J2S 7C6 Received 29 April 1991/Accepted 15 July 1991
An indirect immunodot assay with rabbit antibodies raised against purified heat-stable enterotoxin type b (STb) and with a Western blotting (immunoblotting) detection system (ECL; Amersham International plc, Amersham, United Kingdom) was developed for the detection of STb toxin. Culture supernatants of 62 Escherichia coli isolates from pigs with diarrhea were blotted onto nitrocellulose and incubated with anti-STb serum. The chemiluminescence produced by the action of horseradish peroxidase with luminol and H202 was recorded by exposure of X-ray film. Over 90% correlation was observed between the rat or pig intestinal ligated loop assay and a radioactive DNA probe and the ECL immunodot assay for the detection of STb. In addition, using this new and sensitive technique, we could detect STb in the feces of a newborn pig inoculated with an STb-producing E. coli strain. Detection of STb-producing E. coli in pigs with diarrhea will be greatly facilitated by the use of this convenient and rapid diagnostic assay.
Enterotoxigenic Escherichia coli causes diarrhea in newborn animals, travelers, and children in underdeveloped countries by its capacity to produce enterotoxins which have been classified on the basis of their heat stability (3, 14). Heat-stable enterotoxin type b (STb, or ST-II) is methanol insoluble and induces intestinal secretion in neonatal pigs but does not affect infant mice (15). The STb gene (estB) codes for a 71-amino-acid peptide, including a signal sequence, and is totally different from that coding for STa (9, 13). The biochemistry and mode of action of STb are poorly understood because of the lack of a convenient toxin assay and the very low amount of toxin produced by wild-type strains. Until recently, detection of STb has mostly been accomplished by the laborious, expensive, and poorly reproducible pig intestinal loop assay. More recently, a rat intestinal loop assay in the presence of trypsin inhibitor was reported (4, 16). In vitro assays for the detection of estB (10, 11), such as colony hybridization with nonisotopic probes and an STbspecific DNA amplification assay by use of the polymerase chain reaction, are useful for the diagnosis of STb-positive enterotoxigenic E. coli but do not ascertain the production of the STb toxin (10, 11). The enhanced chemiluminescence Western blotting (immunoblotting) detection system (ECL; Amersham International plc, Amersham, United Kingdom), a recently described rapid and highly sensitive nonisotopic method for the detection of proteins, appears to be more than 10 times as sensitive as colorimetric methods, thus allowing the detection of very low abundance proteins (1). Use of STb-specific antibody in this system could improve and facilitate the quantitative detection of STb. In this study we evaluated the ECL technique described here by using antibodies raised against purified STb to detect the production of STb toxin directly in culture supernatants, and we compared it with the standard bioassays and the colony hybridization assay by using a radioactive DNA probe. *
MATERIALS AND METHODS Bacterial strains. Wild-type E. coli strains were isolated from pigs with diarrhea (Faculte de Medecine Veterinaire, St-Hyacinthe, Qudbec, Canada) (6). The E. coli K-12 (pRASI) strain contained a multicopy recombinant plasmid carrying an STb toxin-encoding gene (9, 12) and was used as a positive control. Biological assay. Detection of STb toxin was done by using the rat intestinal ligated loop assay as described previously (4, 6). Some isolates were tested by the pig intestinal ligated loop assay as described previously (6, 7). Colony hybridization assay. The probe used for the detection of the STb gene was derived from the recombinant plasmid pRASI (2, 12) and was used for the colony hybridization assay as described previously (2, 6, 12). Preparation of samples. E. coli isolates were inoculated into 5 ml of Trypticase soy broth (Difco Laboratories, Detroit, Mich.), along with 50 p.g of ampicillin per ml for strain K-12(pRASI), and were incubated under agitation for 16 h at 37°C to an optical density at 540 nm nearly equal to 1.2. A volume of 1.5 ml of each culture was centrifuged in a microcentrifuge tube for 5 min to separate bacterial cells. Supernatants were kept at -70°C until analysis. ECL immunodot assay. All incubations were performed at room temperature with slow agitation (35 rpm). A volume of 200 ,ul of each supernatant was dot inoculated onto a nitrocellulose membrane (Trans-Blot transfer medium; BioRad Laboratories, Richmond, Calif.) by using a Bio-Dot apparatus (Bio-Rad Laboratories) following the instructions of the manufacturer. The membrane was then incubated in 1 ml of blocking solution per cm2 for at least 1 h. The blocking solution was TBS-T-4% gelatin (TBS-T is 20 mM Tris hydrochloride [pH 7.6], 137 mM NaCl, and 0.2% [vol/vol] Tween 20). The membrane was washed once for 15 min and then twice for 5 min in TBS-T and was subsequently incubated for 1 h with the appropriate dilution of rabbit antibodies raised against purified STb (4). The washing steps were repeated, and the membrane was then incubated for 1 h with goat anti-rabbit immunoglobulin G (Daymar Laboratories Inc., Toronto, Ontario, Canada) coupled with horseradish perox-
Corresponding author. 2250
IMMUNOBLOT DETECTION OF E. COLI STh
VOL. 29, 1991 TABLE 1. Detection of STh-positive E. coli isolates by the enhanced chemiluminescence reaction
1
2
a
2251
*
No. of isolates
Serogroup
Toxin production'
STb positive by:
Total (n = 62)
Probe and
ECLc
LTp
2 4
2 4
2 3 (1)
STap
1
1
1
1
0
0
1 2
0 2
0 2
064
2
0
2
0115
17
12
0119
2
0
0
STap, VT VT
1 1 1
1 1 0
1 1 0
VT
3
1
2e
0141
2
0
0
0147
STap LTp
4 1 2
4 1 2
4 1 2
LTp VT
6 3 1
4 2 1
2f (1) 2 1
LTp
2 3
1 3
1 1(2)
08
09 010
020 STap
0138
0139
0149
0157
bioassayb
11d (0)
a Toxin production was determined as described previously (6). LTp, porcine heat-labile toxin; STap, porcine heat-stable toxin type a; VT, verotoxin. b STb colony hybridization and production of STb as observed in pig or rat ligated gut loops. c Enhanced chemiluminescence reaction by using culture supernatant. Values in parentheses are the number of probe and bioassay STb-positive isolates that were ECL negative for STh in culture supematants but that were ECL positive after bacterial sonication. d These were the same isolates as those that were positive by probe and bioassay. I These isolates were positive by probe and bioassay. f One probe and bioassay STb-negative isolate was positive when it was tested by ECL.
idase appropriately diluted in TBS-T (1 mi/cm2 of membrane). The membrane was finally washed for 15 min and then four times for 5 min each time with TBS-T before performing the enhanced chemiluminescence reaction. All subsequent procedures were performed in a dark room. The membrane was incubated with revealing solution (containing equal volumes of detection reagents 1 and 2 [ECL Western blotting detection system; Amersham]) for 1 min. To detect chemiluminescence, the membrane was then wrapped with a plastic sheet and exposed to X-ray film (Hyperfilm ECL; Amersham) at room temperature for 15, 30, 60, and 120 s. The film was then developed according to the instructions of the manufacturer.
FIG. 1. Immunodot assay by using the enhanced chemiluminescence reaction. 1, STh-negative control strain; 2, supernatant of bacterial culture from the feces of a pig infected with STb-negative strain 87-4725; 3, STb-positive control strain; 4, supernatant of bacterial culture from the feces of a pig infected with STb-positive strain 80-3531.
Detection of STb in stools. Two 3-day-old, colostrumdeprived, hysterotomy-derived pigs were experimentally inoculated with E. coli by the oral route. One pig received 2 x 10O° bacteria of strain 80-3531 (0149:K91:K88:STb+: STa-:LT-). The other pig received a similar number of bacteria of E. coli 87-4725 (045:K"E65":K88- :STb-:STa-: LT-). Sixteen hours later, the rectal contents of each pig were removed and kept at -70°C until analysis. For both pigs, 500 ,ul of rectal contents was mixed with 200 RI1 of phosphate-buffered saline (pH 7.4). After centrifugation at 15,000 x g for 5 min, the supernatant was dot blotted as described above for the other samples. In addition, for both pigs 5 p.l of rectal contents was inoculated into 5 ml of Trypticase soy broth and incubated as described above. Culture supernatants were then dot blotted. RESULTS A total of 62 E. coli isolates were tested for the production of STb toxin by the ECL immunodot assay (Table 1). The bioassay and the colony hybridization assay identified 42 isolates as STb positive. The ECL immunodot assay detected the presence of STb in culture supernatants of 40 of these 42 STb-positive isolates and did not detect STb in culture supernatants of 16 of the 20 STb-negative isolates. Thus, the correlation between the ECL immunodot assay and standard assays was over 90% (P < 0.0001, chi-square test). Initially, STb production was not detected by the ECL immunodot assay in the culture supernatants of six STb probe- and bioassay-positive isolates of serogroups 08, 0115, 0149, and 0157. It has been shown that STb is sometimes not secreted but remains inside the bacterial cell (8). Thus, these six isolates were lysed by sonication and retested by the ECL immunodot assay. Only one 0115 isolate and one 0149 isolate remained negative after lysis by sonication. Both isolates were negative for the production of other toxins. The four probe- and bioassay-negative, ECLpositive isolates belonged to serogroups 064, 0139, and 0149. The pig inoculated with the STb-positive isolate had severe diarrhea 16 h after infection, while the other pig remained apparently healthy. The ECL immunodot assay detected STb toxin directly in the fecal material of the diarrheic pig but not in that of the control pig. The same result was obtained when supernatants of bacterial cultures from the feces of the pigs were used (Fig. 1). DISCUSSION STb detection in culture supernatants of different E. coli isolates by the enhanced chemiluminescence reaction correlated well (P < 0.0001, chi-square test) with the results obtained by the standard assays. We found that most of the
2252
J. CLIN. MICROBIOL.
LORTIE ET AL.
STb-positive isolates could be detected by examination of culture supernatants by the ECL immunodot assay. However, culture supernatants of six isolates identified as STb positive by using the radioactive DNA probe and the bioassay were negative by the ECL immunodot assay. It has been demonstrated that secretion of STb by enterotoxigenic E. coli does not always occur (8). After lysis of cells by sonication, four of the six ECL immunodot assay-negative isolates were positive for STb by the immunodot method, thus indicating that STb was present in these isolates and that the toxin was probably not secreted efficiently. Thus, we suggest that isolates that are STb negative by ECL immunodot assay examination of culture supernatants should always be retested after cell lysis by sonication. The finding that the culture supernatants of these same four isolates were active in the bioassay could be explained by the greater amount of culture supernatant used in the bioassay and/or the greater sensitivity of the bioassay. The fact that two STb-positive isolates remained negative by the ECL immunodot assay could indicate that they produced less STb than the four other sonicated isolates did. Culture supernatants of four STb-negative isolates reacted with our STbspecific antibodies raised against purified toxin by the ECL immunodot technique. This nonspecific reaction could be due to the lack of specificity of the antiserum raised against the purified toxin. However, when these four isolates were tested by the ECL immunodot assay by using STb-specific antibodies raised against the P-galactosidase-STb fusion protein (adsorbed against P-galactosidase) (5), the same results were obtained. The nonspecific reaction did not appear to be due to a cross-reaction with heat-labile enterotoxin, STa, or verotoxin. It could be inherent to the ECL immunodot technique or it could be due to the detection of another secreted toxin or cell product that is antigenically similar to STb. In an experimental infection study, we detected STb in the fecal material from pigs inoculated with an STb-producing isolate. This observation indicates that the ECL immunodot technique could be used as an alternative method for the diagnosis of STb-positive enterotoxigenic E. coli in animal feces. This technique is inexpensive, extremely rapid, and simple to perform. In addition, it would not be necessary to isolate enterotoxigenic E. coli from feces for the identification of STb producers. The detection of STb-positive enterotoxigenic E. coli isolates will be greatly facilitated by the use of this convenient and rapid diagnostic assay. ACKNOWLEDGMENTS We thank C. Handl and J.-I. Flock (Karolinska Institute, Stockholm, Sweden) for providing p-galactosidase-STb antibodies. We thank Stephane Bolduc for technical assistance. This study was supported in part by grants from the Natural Sciences and Engineering Research Council of Canada (OGP0046569) and Fonds de la Recherche en Santd du Quebec (Etablissement de Jeune Chercheur-901527) (to J.D.D.), from the Conseil des Recherches en Peche et Agroalimentaire du Quebec (2243) and Fonds de la Recherche en Sante du Qudbec (Etablissement de Jeune Chercheur-870046) (to J.H.), and from the Ministere de l'Enseignement Supdrieur et de la Science (to L.A.L.).
1.
2.
3.
4.
5.
6.
7.
8. 9.
10.
11.
12.
13. 14.
15.
16.
REFERENCES Amersham International plc. 1990. Amersham bulletin 1990. ECL Western blotting detection system. Amersham International plc, Amersham, United Kingdom. Broes, A., J. M. Fairbrother, J. Mainil, J. Harel, and S. Larivi&re. 1988. Phenotypic and genotypic characterization of enterotoxigenic Escherichia coli serotype 08: K"2829" strains isolated from piglets with diarrhea. J. Clin. Microbiol. 26:24022409. Burgess, M. N., R. J. Bywater, C. M. Cowley, N. A. Mullan, and P. M. Newsome. 1978. Biological evaluation of a methanolsoluble, heat-stable Escherichia coli enterotoxin in infant mice, pigs, rabbits, and calves. Infect. Immun. 21:526-531. Dubreuil, J. D., J. M. Fairbrother, R. Lallier, and S. Lariviere. 1991. Production and purification of heat-stable enterotoxin b from a porcine Escherichia coli strain. Infect. Immun. 59:198203. Handl, C., B. Ronnberg, B. Nilsson, E. Olsson, H. Jonsson, and J.-I. Flock. 1988. Enzyme-linked immunosorbent assay for Escherichia coli heat-stable enterotoxin type II. J. Clin. Microbiol. 26:1555-1560. Harel, J., H. Lapointe, A. Fallara, L. A. Lortie, M. BigrasPoulin, S. Lariviere, and J. M. Fairbrother. 1991. Detection of genes for fimbrial antigens and enterotoxins associated with Escherichia coli serogroups isolated from pigs with diarrhea. J. Clin. Microbiol. 29:745-752. Lariviere, S., C. L. Gyles, and D. A. Barnum. 1972. A comparative study of the rabbit and pig gut loop systems for the assay of Escherichia coli enterotoxin. Can. J. Comp. Med. 36:319328. Lawrence, R. M., P.-T. Huang, J. Glick, J. D. Oppenheim, and W. K. Maas. 1990. Expression of the cloned gene for enterotoxin STb of Escherichia coli. Infect. Immun. 58:970-977. Lee, C. H., S. L. Moseley, H. W. Moon, S. C. Whipp, C. L. Gyles, and M. So. 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. Lortie, L. A., J. D. Dubreuil, and J. Harel. 1991. Characterization of Escherichia coli strains producing heat-stable enterotoxin b (STb) isolated from humans with diarrhea. J. Clin. Microbiol. 29:656-659. Lortie, L. A., J. Harel, J. M. Fairbrother, and J. D. Dubreuil. 1991. Evaluation of three new techniques for the detection of STb-positive Escherichia coli strains. Mol. Cell. Probes 5:271275. Mainil, J. G., S. L. Moseley, R. A. Schneider, K. Sutch, T. A. Casey, and H. W. Moon. 1986. Hybridization of bovine Escherichia coli isolates with gene probes for four enterotoxins (STaP, STaH, STb, LT) and one adhesion factor (K99). Am. J. Vet. Res. 47:1145-1148. Picken, R. N., A. J. Mazaitis, W. K. Maas, M. Rey, and H. Heyneker. 1983. Nucleotide sequence of the gene for heat-stable enterotoxin II of Escherichia coli. Infect. Immun. 42:269-275. Robichaud, N., R. Lallier, and S. Lariviere. 1978. Comparison of the various test systems for the detection of Escherichia coli enterotoxin. In S. D. Acres (ed.), Proceeding of the 2nd International Symposium on Neonatal Diarrhea. Saskatoon, Saskatchewan, Canada. Weikel, C. S., H. N. Nellans, and R. L. Guerrant. 1986. In vivo and in vitro effects of a novel enterotoxin, STh, produced by Escherichia coli. J. Infect. Dis. 153:893-901. Whipp, S. C. 1990. Assay for enterotoxigenic Escherichia coli heat-stable toxin b in rats and mice. Infect. Immun. 58:930-934.
JOURNAL OF CLINICAL MICROBIOLOGY, OCt. 1991, p. 2250-2252 0095-1137/91/102250-03$02.00/0 Copyright © 1991, American Society for Microbiology
Immunodot Detection of Escherichia coli Heat-Stable Enterotoxin b by Using Enhanced Chemiluminescence Reaction LOUIS-ANDRE LORTIE, JOSEE HAREL, JOHN M. FAIRBROTHER, AND J. DANIEL DUBREUIL* Department of Pathology and Microbiology, Faculty of Veterinary Medicine, University of Montreal,
3200 Sicotte, P.O. Box 5000, Saint-Hyacinthe, Quebec, Canada J2S 7C6 Received 29 April 1991/Accepted 15 July 1991
An indirect immunodot assay with rabbit antibodies raised against purified heat-stable enterotoxin type b (STb) and with a Western blotting (immunoblotting) detection system (ECL; Amersham International plc, Amersham, United Kingdom) was developed for the detection of STb toxin. Culture supernatants of 62 Escherichia coli isolates from pigs with diarrhea were blotted onto nitrocellulose and incubated with anti-STb serum. The chemiluminescence produced by the action of horseradish peroxidase with luminol and H202 was recorded by exposure of X-ray film. Over 90% correlation was observed between the rat or pig intestinal ligated loop assay and a radioactive DNA probe and the ECL immunodot assay for the detection of STb. In addition, using this new and sensitive technique, we could detect STb in the feces of a newborn pig inoculated with an STb-producing E. coli strain. Detection of STb-producing E. coli in pigs with diarrhea will be greatly facilitated by the use of this convenient and rapid diagnostic assay.
Enterotoxigenic Escherichia coli causes diarrhea in newborn animals, travelers, and children in underdeveloped countries by its capacity to produce enterotoxins which have been classified on the basis of their heat stability (3, 14). Heat-stable enterotoxin type b (STb, or ST-II) is methanol insoluble and induces intestinal secretion in neonatal pigs but does not affect infant mice (15). The STb gene (estB) codes for a 71-amino-acid peptide, including a signal sequence, and is totally different from that coding for STa (9, 13). The biochemistry and mode of action of STb are poorly understood because of the lack of a convenient toxin assay and the very low amount of toxin produced by wild-type strains. Until recently, detection of STb has mostly been accomplished by the laborious, expensive, and poorly reproducible pig intestinal loop assay. More recently, a rat intestinal loop assay in the presence of trypsin inhibitor was reported (4, 16). In vitro assays for the detection of estB (10, 11), such as colony hybridization with nonisotopic probes and an STbspecific DNA amplification assay by use of the polymerase chain reaction, are useful for the diagnosis of STb-positive enterotoxigenic E. coli but do not ascertain the production of the STb toxin (10, 11). The enhanced chemiluminescence Western blotting (immunoblotting) detection system (ECL; Amersham International plc, Amersham, United Kingdom), a recently described rapid and highly sensitive nonisotopic method for the detection of proteins, appears to be more than 10 times as sensitive as colorimetric methods, thus allowing the detection of very low abundance proteins (1). Use of STb-specific antibody in this system could improve and facilitate the quantitative detection of STb. In this study we evaluated the ECL technique described here by using antibodies raised against purified STb to detect the production of STb toxin directly in culture supernatants, and we compared it with the standard bioassays and the colony hybridization assay by using a radioactive DNA probe. *
MATERIALS AND METHODS Bacterial strains. Wild-type E. coli strains were isolated from pigs with diarrhea (Faculte de Medecine Veterinaire, St-Hyacinthe, Qudbec, Canada) (6). The E. coli K-12 (pRASI) strain contained a multicopy recombinant plasmid carrying an STb toxin-encoding gene (9, 12) and was used as a positive control. Biological assay. Detection of STb toxin was done by using the rat intestinal ligated loop assay as described previously (4, 6). Some isolates were tested by the pig intestinal ligated loop assay as described previously (6, 7). Colony hybridization assay. The probe used for the detection of the STb gene was derived from the recombinant plasmid pRASI (2, 12) and was used for the colony hybridization assay as described previously (2, 6, 12). Preparation of samples. E. coli isolates were inoculated into 5 ml of Trypticase soy broth (Difco Laboratories, Detroit, Mich.), along with 50 p.g of ampicillin per ml for strain K-12(pRASI), and were incubated under agitation for 16 h at 37°C to an optical density at 540 nm nearly equal to 1.2. A volume of 1.5 ml of each culture was centrifuged in a microcentrifuge tube for 5 min to separate bacterial cells. Supernatants were kept at -70°C until analysis. ECL immunodot assay. All incubations were performed at room temperature with slow agitation (35 rpm). A volume of 200 ,ul of each supernatant was dot inoculated onto a nitrocellulose membrane (Trans-Blot transfer medium; BioRad Laboratories, Richmond, Calif.) by using a Bio-Dot apparatus (Bio-Rad Laboratories) following the instructions of the manufacturer. The membrane was then incubated in 1 ml of blocking solution per cm2 for at least 1 h. The blocking solution was TBS-T-4% gelatin (TBS-T is 20 mM Tris hydrochloride [pH 7.6], 137 mM NaCl, and 0.2% [vol/vol] Tween 20). The membrane was washed once for 15 min and then twice for 5 min in TBS-T and was subsequently incubated for 1 h with the appropriate dilution of rabbit antibodies raised against purified STb (4). The washing steps were repeated, and the membrane was then incubated for 1 h with goat anti-rabbit immunoglobulin G (Daymar Laboratories Inc., Toronto, Ontario, Canada) coupled with horseradish perox-
Corresponding author. 2250
IMMUNOBLOT DETECTION OF E. COLI STh
VOL. 29, 1991 TABLE 1. Detection of STh-positive E. coli isolates by the enhanced chemiluminescence reaction
1
2
a
2251
*
No. of isolates
Serogroup
Toxin production'
STb positive by:
Total (n = 62)
Probe and
ECLc
LTp
2 4
2 4
2 3 (1)
STap
1
1
1
1
0
0
1 2
0 2
0 2
064
2
0
2
0115
17
12
0119
2
0
0
STap, VT VT
1 1 1
1 1 0
1 1 0
VT
3
1
2e
0141
2
0
0
0147
STap LTp
4 1 2
4 1 2
4 1 2
LTp VT
6 3 1
4 2 1
2f (1) 2 1
LTp
2 3
1 3
1 1(2)
08
09 010
020 STap
0138
0139
0149
0157
bioassayb
11d (0)
a Toxin production was determined as described previously (6). LTp, porcine heat-labile toxin; STap, porcine heat-stable toxin type a; VT, verotoxin. b STb colony hybridization and production of STb as observed in pig or rat ligated gut loops. c Enhanced chemiluminescence reaction by using culture supernatant. Values in parentheses are the number of probe and bioassay STb-positive isolates that were ECL negative for STh in culture supematants but that were ECL positive after bacterial sonication. d These were the same isolates as those that were positive by probe and bioassay. I These isolates were positive by probe and bioassay. f One probe and bioassay STb-negative isolate was positive when it was tested by ECL.
idase appropriately diluted in TBS-T (1 mi/cm2 of membrane). The membrane was finally washed for 15 min and then four times for 5 min each time with TBS-T before performing the enhanced chemiluminescence reaction. All subsequent procedures were performed in a dark room. The membrane was incubated with revealing solution (containing equal volumes of detection reagents 1 and 2 [ECL Western blotting detection system; Amersham]) for 1 min. To detect chemiluminescence, the membrane was then wrapped with a plastic sheet and exposed to X-ray film (Hyperfilm ECL; Amersham) at room temperature for 15, 30, 60, and 120 s. The film was then developed according to the instructions of the manufacturer.
FIG. 1. Immunodot assay by using the enhanced chemiluminescence reaction. 1, STh-negative control strain; 2, supernatant of bacterial culture from the feces of a pig infected with STb-negative strain 87-4725; 3, STb-positive control strain; 4, supernatant of bacterial culture from the feces of a pig infected with STb-positive strain 80-3531.
Detection of STb in stools. Two 3-day-old, colostrumdeprived, hysterotomy-derived pigs were experimentally inoculated with E. coli by the oral route. One pig received 2 x 10O° bacteria of strain 80-3531 (0149:K91:K88:STb+: STa-:LT-). The other pig received a similar number of bacteria of E. coli 87-4725 (045:K"E65":K88- :STb-:STa-: LT-). Sixteen hours later, the rectal contents of each pig were removed and kept at -70°C until analysis. For both pigs, 500 ,ul of rectal contents was mixed with 200 RI1 of phosphate-buffered saline (pH 7.4). After centrifugation at 15,000 x g for 5 min, the supernatant was dot blotted as described above for the other samples. In addition, for both pigs 5 p.l of rectal contents was inoculated into 5 ml of Trypticase soy broth and incubated as described above. Culture supernatants were then dot blotted. RESULTS A total of 62 E. coli isolates were tested for the production of STb toxin by the ECL immunodot assay (Table 1). The bioassay and the colony hybridization assay identified 42 isolates as STb positive. The ECL immunodot assay detected the presence of STb in culture supernatants of 40 of these 42 STb-positive isolates and did not detect STb in culture supernatants of 16 of the 20 STb-negative isolates. Thus, the correlation between the ECL immunodot assay and standard assays was over 90% (P < 0.0001, chi-square test). Initially, STb production was not detected by the ECL immunodot assay in the culture supernatants of six STb probe- and bioassay-positive isolates of serogroups 08, 0115, 0149, and 0157. It has been shown that STb is sometimes not secreted but remains inside the bacterial cell (8). Thus, these six isolates were lysed by sonication and retested by the ECL immunodot assay. Only one 0115 isolate and one 0149 isolate remained negative after lysis by sonication. Both isolates were negative for the production of other toxins. The four probe- and bioassay-negative, ECLpositive isolates belonged to serogroups 064, 0139, and 0149. The pig inoculated with the STb-positive isolate had severe diarrhea 16 h after infection, while the other pig remained apparently healthy. The ECL immunodot assay detected STb toxin directly in the fecal material of the diarrheic pig but not in that of the control pig. The same result was obtained when supernatants of bacterial cultures from the feces of the pigs were used (Fig. 1). DISCUSSION STb detection in culture supernatants of different E. coli isolates by the enhanced chemiluminescence reaction correlated well (P < 0.0001, chi-square test) with the results obtained by the standard assays. We found that most of the
2252
J. CLIN. MICROBIOL.
LORTIE ET AL.
STb-positive isolates could be detected by examination of culture supernatants by the ECL immunodot assay. However, culture supernatants of six isolates identified as STb positive by using the radioactive DNA probe and the bioassay were negative by the ECL immunodot assay. It has been demonstrated that secretion of STb by enterotoxigenic E. coli does not always occur (8). After lysis of cells by sonication, four of the six ECL immunodot assay-negative isolates were positive for STb by the immunodot method, thus indicating that STb was present in these isolates and that the toxin was probably not secreted efficiently. Thus, we suggest that isolates that are STb negative by ECL immunodot assay examination of culture supernatants should always be retested after cell lysis by sonication. The finding that the culture supernatants of these same four isolates were active in the bioassay could be explained by the greater amount of culture supernatant used in the bioassay and/or the greater sensitivity of the bioassay. The fact that two STb-positive isolates remained negative by the ECL immunodot assay could indicate that they produced less STb than the four other sonicated isolates did. Culture supernatants of four STb-negative isolates reacted with our STbspecific antibodies raised against purified toxin by the ECL immunodot technique. This nonspecific reaction could be due to the lack of specificity of the antiserum raised against the purified toxin. However, when these four isolates were tested by the ECL immunodot assay by using STb-specific antibodies raised against the P-galactosidase-STb fusion protein (adsorbed against P-galactosidase) (5), the same results were obtained. The nonspecific reaction did not appear to be due to a cross-reaction with heat-labile enterotoxin, STa, or verotoxin. It could be inherent to the ECL immunodot technique or it could be due to the detection of another secreted toxin or cell product that is antigenically similar to STb. In an experimental infection study, we detected STb in the fecal material from pigs inoculated with an STb-producing isolate. This observation indicates that the ECL immunodot technique could be used as an alternative method for the diagnosis of STb-positive enterotoxigenic E. coli in animal feces. This technique is inexpensive, extremely rapid, and simple to perform. In addition, it would not be necessary to isolate enterotoxigenic E. coli from feces for the identification of STb producers. The detection of STb-positive enterotoxigenic E. coli isolates will be greatly facilitated by the use of this convenient and rapid diagnostic assay. ACKNOWLEDGMENTS We thank C. Handl and J.-I. Flock (Karolinska Institute, Stockholm, Sweden) for providing p-galactosidase-STb antibodies. We thank Stephane Bolduc for technical assistance. This study was supported in part by grants from the Natural Sciences and Engineering Research Council of Canada (OGP0046569) and Fonds de la Recherche en Santd du Quebec (Etablissement de Jeune Chercheur-901527) (to J.D.D.), from the Conseil des Recherches en Peche et Agroalimentaire du Quebec (2243) and Fonds de la Recherche en Sante du Qudbec (Etablissement de Jeune Chercheur-870046) (to J.H.), and from the Ministere de l'Enseignement Supdrieur et de la Science (to L.A.L.).
1.
2.
3.
4.
5.
6.
7.
8. 9.
10.
11.
12.
13. 14.
15.
16.
REFERENCES Amersham International plc. 1990. Amersham bulletin 1990. ECL Western blotting detection system. Amersham International plc, Amersham, United Kingdom. Broes, A., J. M. Fairbrother, J. Mainil, J. Harel, and S. Larivi&re. 1988. Phenotypic and genotypic characterization of enterotoxigenic Escherichia coli serotype 08: K"2829" strains isolated from piglets with diarrhea. J. Clin. Microbiol. 26:24022409. Burgess, M. N., R. J. Bywater, C. M. Cowley, N. A. Mullan, and P. M. Newsome. 1978. Biological evaluation of a methanolsoluble, heat-stable Escherichia coli enterotoxin in infant mice, pigs, rabbits, and calves. Infect. Immun. 21:526-531. Dubreuil, J. D., J. M. Fairbrother, R. Lallier, and S. Lariviere. 1991. Production and purification of heat-stable enterotoxin b from a porcine Escherichia coli strain. Infect. Immun. 59:198203. Handl, C., B. Ronnberg, B. Nilsson, E. Olsson, H. Jonsson, and J.-I. Flock. 1988. Enzyme-linked immunosorbent assay for Escherichia coli heat-stable enterotoxin type II. J. Clin. Microbiol. 26:1555-1560. Harel, J., H. Lapointe, A. Fallara, L. A. Lortie, M. BigrasPoulin, S. Lariviere, and J. M. Fairbrother. 1991. Detection of genes for fimbrial antigens and enterotoxins associated with Escherichia coli serogroups isolated from pigs with diarrhea. J. Clin. Microbiol. 29:745-752. Lariviere, S., C. L. Gyles, and D. A. Barnum. 1972. A comparative study of the rabbit and pig gut loop systems for the assay of Escherichia coli enterotoxin. Can. J. Comp. Med. 36:319328. Lawrence, R. M., P.-T. Huang, J. Glick, J. D. Oppenheim, and W. K. Maas. 1990. Expression of the cloned gene for enterotoxin STb of Escherichia coli. Infect. Immun. 58:970-977. Lee, C. H., S. L. Moseley, H. W. Moon, S. C. Whipp, C. L. Gyles, and M. So. 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. Lortie, L. A., J. D. Dubreuil, and J. Harel. 1991. Characterization of Escherichia coli strains producing heat-stable enterotoxin b (STb) isolated from humans with diarrhea. J. Clin. Microbiol. 29:656-659. Lortie, L. A., J. Harel, J. M. Fairbrother, and J. D. Dubreuil. 1991. Evaluation of three new techniques for the detection of STb-positive Escherichia coli strains. Mol. Cell. Probes 5:271275. Mainil, J. G., S. L. Moseley, R. A. Schneider, K. Sutch, T. A. Casey, and H. W. Moon. 1986. Hybridization of bovine Escherichia coli isolates with gene probes for four enterotoxins (STaP, STaH, STb, LT) and one adhesion factor (K99). Am. J. Vet. Res. 47:1145-1148. Picken, R. N., A. J. Mazaitis, W. K. Maas, M. Rey, and H. Heyneker. 1983. Nucleotide sequence of the gene for heat-stable enterotoxin II of Escherichia coli. Infect. Immun. 42:269-275. Robichaud, N., R. Lallier, and S. Lariviere. 1978. Comparison of the various test systems for the detection of Escherichia coli enterotoxin. In S. D. Acres (ed.), Proceeding of the 2nd International Symposium on Neonatal Diarrhea. Saskatoon, Saskatchewan, Canada. Weikel, C. S., H. N. Nellans, and R. L. Guerrant. 1986. In vivo and in vitro effects of a novel enterotoxin, STh, produced by Escherichia coli. J. Infect. Dis. 153:893-901. Whipp, S. C. 1990. Assay for enterotoxigenic Escherichia coli heat-stable toxin b in rats and mice. Infect. Immun. 58:930-934.
