VETERINARSKI ARHIV 68 (6), 191-198, 1998

ISSN 0372-5480
Printed in Croatia

Serotyping of enterotoxigenic isolates of Escherichia coli from suckling and weaned piglets with colibacillosis

Andrea Humski¹*, Vitomir Bilic¹, Danko Hajsig², and Boris Habrun¹

¹Department of Bacteriology, Croatian Veterinary Institute, Zagreb, Croatia

²Pliva d.d., Zagreb, Croatia

* Contact address:
Mr. sc. Andrea Humski,
Department of Bacteriology, Croatian Veterinary Institute, 10000 Zagreb, Savska cesta 143, P.O. Box 883, Croatia,
Phone: 385 1 61 90 841; Fax: 385 1 61 90 838

HUMSKI, A., V. BILIC, D. HAJSIG, B. HABRUN: Serotyping of enterotoxigenic isolates of Escherichia coli from suckling and weaned piglets with colibacillosis. Vet. arhiv 68, 191-198, 1998.

ABSTRACT

The aim of the study was to determine the presence, serogroup and haemolytic activity of fimbrial antigens in 929 E. coli isolates from suckling (n=338) and weaned (n=591) piglets with clinical manifestation of colibacillosis. Haemolytic activity was verified in 823 (89%) isolates, with a two-fold rate in the isolates from weaned piglets (69%) as compared to those from suckling piglets (32%). Serogroups were identified in 18% of the isolates from suckling piglets. Serovars O149:K91 and O8:K87 were found to predominate (48% and 30% of the isolates, respectively). In the weaned piglets the prevailing serovars were O8:K87, O149:K91 and O157:K"V17" (27%, 26% and 19% of the isolates, respectively). Fimbrial adhesin K88ac was found in 61% of the serotyped isolates, most frequently with serovars O149:K91 (29%) and O8:K87 (23%). Their prevalence in the serotyped isolates from suckling and weaned piglets was 23% and 42%, respectively. Concurrence of K88ac fimbrial adhesin and haemolytic activity was recorded in 65% of the serotyped E. coli isolates. 987P fimbrial adhesin was detected in 27% of E. coli haemolytic isolates with an identical prevalence (50%) in the suckling and weaned piglets. In the weaned animals, K99 fimbrial adhesin was found in only 4 (1%) serotyped isolates, which were also capable of producing haemolysin.

Key words: colibacillosis, piglets, haemolytic activity, fimbrial adhesins, serogroups

Introduction

Gastrointestinal diseases appear to be the most common cause of morbidity and mortality in pig agglomerations in Croatia (ANON., 1994, 1995). Infections with enterotoxigenic strains of Escherichia coli (ETEC), along with occasional outbreaks of enzootic viral enteritis, are one of the prevailing etiologic factors.

Escherichia coli (E. coli) strains colonising the intestine and producing enterotoxins that stimulate the secretion of electrolytes and water in the intestinal mucosa, play a key role in the pathogenesis of colibacillosis in pigs and piglets (WHIPP et al., 1981; ROSE et al., 1987).

Two groups of laboratory methods have been adopted world-wide for laboratory demonstration of isolated E. coli strain pathogenicity. One group is aimed at demonstrating the E. coli strain's capacity for producing heat-labile and heat-stable enterotoxins. Some immunologic methods are widely employed in the identification of these enterotoxins, e.g., ELISA (YOLKEN et al., 1977; SVENNERHOLM and HOLMGREN, 1978), coagglutination test (RONNBERG and WADSTROM, 1983; RUDENSKY et al., 1988) and latex agglutination test (FINKELSTEIN et al., 1983; ITO et al., 1983). In spite of their exactness, these methods of enterotoxin identification have proved inappropriate for use in routine diagnosis, as they are costly, time-consuming and sensitive.

The other group of methods are used to determine the antigenic structure of isolated strain, with special reference to fimbrial adhesins, i.e. K88 (F4), K99 (F5), 987P (F6), F41, F42 and F165 (SOJKA, 1965; SOJKA, 1973; ØRSKOV and ØRSKOV, 1983; ØRSKOV and ØRSKOV, 1984). Their presence in isolated strains is directly associated with the ability of porcine enterotoxigenic strains of E. coli to adhere to intestinal mucosa, and with intra-intestinal bacterial proliferation in piglets (MOON et al., 1979; GAASTRA and DE GRAAF, 1982; ØRSKOV and ØRSKOV, 1983). Thus, these adhesins have been designated colonisation factor antigens (CFA) (JONES and RUTTER, 1972; MOON et al., 1977; NAGY et al., 1977; MORRIS et al., 1983; YANO et al., 1986; FAIRBROTHER et al., 1986).

Due to a high correlation of the presence of fimbrial adhesins and production of haemolysin with the ability of isolated E. coli strain to produce enterotoxins, a majority of the pathogenic strains can be diagnosed by serotyping. The aim of our study was to assess the prevalence of ETEC strains in suckling and weaned piglets with colibacillosis by determination of their serotype, presence of fimbrial adhesins and haemolysin production.

Materials and methods

During 1995 the presence of ETEC strains was determined in 929 piglets with clinical and pathomorphologic manifestations indicative of colibacillosis. Samples were obtained of two age groups, i.e., 338 suckling pigs aged <1 month and 591 weaned piglets aged >1 month. Portions of the intestine (jejunum and ileum) were taken from dead animals, and rectal swabs from diseased animals. Most animals originated from the nine largest pig-breeding farms in Croatia, and only some from small breeders.

The following two solid nutrient media were used for E. coli isolation and identification: neutral agar (CM3, Oxoid, Basingstoke, Hampshire, England); agar with the addition of 5% (w/w) defibrinated sheep blood and 1% esculin; XLD (1000I6DEBF, Becton Dickinson, Cockeysville, USA) as a selective medium, and TSI agar (CM277, Oxoid, Basingstoke, Hampshire, England) for fast determination of biochemical properties.

Agglutination antisera were prepared from standard E. coli strains (Statens Seruminstitut, Copenhagen, Denmark) by rabbit immunisation. Serum was prepared and absorbed as described elsewhere (SOJKA, 1965; SOJKA, 1973).

Serotyping of isolated E. coli strains was performed by rapid agglutination on a micro-slide. As some antigen O serogroups may occur in combination with certain K antigens, serotyping of pathogenic strains in pigs can be performed with a smaller number of antisera (SOJKA, 1965; SOJKA, 1973).

The following polyvalent antisera were used in the study: O8:K87, O9:K30; O45:K"E65"; O101:K99:F41; O108:K"V189"; O138:K81; O141:K85ab; O141:K85ac; O147:K89; O149:K91 and O157:K"V17". Absorbed sera were used for identification of the K88, K99, F41 and 987P adhesive antigens.

Results and discussion

Of 929 E. coli isolates from the suckling and weaned piglets, the serotyping procedure failed in 536 (58%) isolates, 188 (56%) of them from the suckling pigs and 348 (59%) from the weaned piglets. Serotyping failure may be due to a number of factors (e.g., treatment with antimicrobials of the whole population or, exceptionally, of some individual animals, use of unsafe animal feed, presence of mixed bacterial and viral infections), which could account for the high percentage of untyped isolates.

Haemolytic virulence was detected in 823 (89%) of all E. coli isolates, i.e., 259 (77%) from the suckling pigs and 564 (95%) from the weaned piglets. In the group of serologically untyped isolates, 45% were found to be capable of producing bacterial haemolysin. The most common haemolytic isolates with antigenic formulas O8:K87, O147:K89 and O149:K91 were present in all groups of animals at a slightly varying rate. Haemolytic isolates of the O45:K"E65", O108:K"V189", O138:K81 and O157:K"V17" serotypes were more common in the weaned piglets. It should be emphasised that in the total number of serotyped isolates from suckling animals (n=150), less than a half (71 isolates) were haemolytic, which should by no means be neglected on isolation and identification of the potentially pathogenic E. coli. The number of haemolytic and serotyped isolates in the weaned animals was almost equal (216 H+ vs. 243 ST+).

Our study demonstrated the presence of K88ac fimbrial adhesin in 234 (60%) serotyped isolates, 154 (66%) of them haemolytic. Of these, 53 (23%) isolates were found in suckling pigs and 101 (43%) in weaned piglets. K99 antigen was found in only 3 (0.8%) haemolytic isolates, 2 of them also having F41 fimbrial adhesin.

The presence of 987P antigen was determined in 59 subsequently isolated haemolytic strains. This fimbrial adhesin was detected in 16 (27%) isolates, 8 (50%) of them isolated from suckling pigs with both O149:K91 and K88ac antigens. The isolates from weaned piglets had the adhesive 987P antigen bound to O8:K87, O149:K91 and O157:K"V17" antigens. Among them, 75% of the isolates also had the K88ac fimbrial antigen. In the remaining 43 isolates from this group, the presence of 987P fimbrial adhesin could not be demonstrated. A phasic variation of this antigen in some of them, as reported by MOON et al. (1980), could only be speculated.

These results are partially inconsistent with those reported by other Croatian authors (BILIC and ZUTIC, 1984; VRBANAC et al., 1991) on the prevalence of K88, K99 and 987P fimbrial adhesins in piglets with colibacillosis.

The results described by BILIC and ZUTIC (1984) indicate a more pronounced presence of K88 fimbrial adhesin in the isolates from suckling pigs (57%) than in those from weaned piglets (42%). K99 fimbrial adhesin was not found in their study. VRBANAC et al. (1991) report on an identical prevalence of K88 fimbrial adhesin in the isolates from suckling pigs (46%) and weaned (46%) piglets, whereas the prevalence of K99 fimbrial adhesin in suckling pigs and weaned piglets was 36% and 15%, respectively. 987P fimbrial adhesin was not present in E. coli isolates from weaned piglets, while 9% of the isolates from suckling pigs had both K88 and 987P fimbrial adhesins.

Variable data have also been reported in international literature. EVANS et al. (1986) report on a lower than usual prevalence of K88 antigen (27%) and a relatively high prevalence of K99 adhesin (44%) in the isolates from suckling pigs. MOON et al. (1980) describe a high prevalence of K99 adhesin in E. coli isolates from suckling pigs (36%) and a considerably lower prevalence (15%) in those from weaned piglets. The prevalence of 987P adhesin in suckling pigs was 14%. The study of NAKAZAWA et al. (1987) showed an equal prevalence (45%) of K88 adhesin in the isolates from the two groups of piglets. Only few isolates had K99 or 987P fimbrial adhesin (7% and 1%, respectively).

In the present study, 393 (42%) of 929 E. coli isolates were successfully serotyped, i.e., 150 (38%) isolates from the suckling pigs and 243 (62%) isolates from the weaned piglets. Overall serotyping results showed a variety of identified serogroups, both within and between the study groups of piglets. Most of the serotyped isolates of E. coli from the suckling pigs with colibacillosis belonged to the group with somatic antigens O149 (48%) and O8 (30%). In the group of weaned piglets, the prevalence of isolates with O8 and O149 somatic antigens was nearly identical (27% and 26%, respectively). Table 1 shows the prevalence of each serogroup in the two groups of piglets.

Table 1. Serogroups of E. coli isolates from the suckling pigs and
weaned piglets with colibacillosis

The results of our, and other, related studies published by BILIC and ZUTIC (1985) are significantly different. The prevalence of E. coli isolates from suckling pigs differed in the groups with O8, O141 and O157 somatic antigens, and in those isolated from weaned piglets in the groups with O8, O138, O149 and O157 somatic antigens. The prevalence of each serogroup in the years 1985 (BILIC and ZUTIC, 1985) and 1995 (the present study) is shown in Table 2.

Table 2. Prevalence of particular serogroups in serotyped E. coli isolates
from suckling pigs and weaned piglets in 1985 and 1995 (%)

* from BILI] and @UTI] (1985)

International reports point to a specific geographical distribution and prevalence of particular E. coli groups, and to the considerable variations in their incidence over a certain period of time (DAM and KNOX, 1974). Additional studies on the issue in Croatia might contribute to these observations.

Results of the studies conducted so far appear to suggest that the O149:K91 and O8:K87 serogroups of the E. coli isolates play a major role in the etiology of piglet colibacillosis on Croatian pig-breeding farms. The data obtained indicate a change in the prevalence of particular serogroups in the serotyped E. coli isolates, as well as in the presence of K88 adhesive antigen in the isolates from various piglet age groups, which is directly related to systematic immunoprophylactic measures taken in Croatian pig-breeding farms.

References

ANONYMOUS (1994): Annual report of the Croatian Veterinary Institute for the year 1993. Croatian Veterinary Institute. Zagreb.

ANONYMOUS (1995): Annual report of the Croatian Veterinary Institute for the year 1994. Croatian Veterinary Institute. Zagreb.

BILIC, V., M. ZUTIC (1984): Incidence of K88 and K99 antigens of Escherichia coli strains isolated from wilde and domestic animals. Praxis vet. 32, 193-196.

BILIC, V., M. ZUTIC (1985): Incidence of serologic OK groups of Escherichia coli in pigs. Praxis vet. 33, 199-207.

DAM, A., B. KNOX (1974): Hemolytic Escherichia coli associated with enteritis and enterotoxemia in pigs in Denmark with particular reference to the spread of serogroup O149:K91. Nord Vet. Med. 26, 219-225.

EVANS, M. G., G. L. WAWLER, P. J. NEWMAN (1986): Prevalence of K88, K99 and 987P pili of Escherichia coli in neonatal pigs with enteric colibacillosis. Am. J. Vet. Res. 47, 2431-2434.

FAIRBROTHER, J. M., S. LARIVIERE, R. LALLIER (1986): New fimbrial antigen F165 from Escherichia coli serogroup O115 strains isolated from piglets with diarrhea. Infect. Immunol. 51, 10-15.

FINKELSTEIN, R. A., Z. YANG, S. L. MOSELEY, H. W. MOON (1983): Rapid latex particle agglutination test for Escherichia coli strains of porcine origin producing heat-labile enterotoxin. J. Clin. Microbiol. 18, 1417-1418.

GAASTRA, W., F. K. DE GRAAF (1982): Host-specific fimbrial adhesins on noninvasive enterotoxigenic Escherichia coli strains. Microbiol. Rev. 46, 129-161.

ITO, T., S. KUWAHARA, T. YOKOTA (1983). Automatic and manual latex agglutination tests for measurment of cholera toxin and heat-labile enterotoxin of Escherichia coli. J. clin. Microbiol. 17, 7-12.

JONES, G. W., J. M. RUTTER (1972): Role of the K88 antigen in the pathogenesis of neonatal diarrhea caused by Escherichia coli in piglets. Infect. Immun. 6, 918-927.

MOON, W. H., R. E. ISAACSON, J. POHLENZ (1979): Mechanism of association of enteropathogenic Escherichia coli with intestinal epithelium. Am. J. Clin. Nutr. 32, 119-127.

MOON, W. H., E. M. KOHLER, R. A. SCHNAIDER (1980): Prevalence of pilus antigens, enterotoxins types and enteropathogenicity among K88 negative enterotoxigenic Escherichia coli from neonatal pigs. Infect. Immunol. 27, 222-230.

MOON, W. H., B. NAGY, R. E. ISAACSON, I. ØRSKOV (1977): Occurrence of K99 antigen of Escherichia coli isolated from pigs and colonization of pig ileum by K99+ enterotoxigenic E. coli from calves and pigs. Infect. Immunol. 15, 614-620.

MORRIS, J. A., C. J. THORNS, G. A. H. WELLS, A. C. SCOTT, W. J. SOJKA (1983): The production of F41 fimbriae by piglet strains of enterotoxigenic Escherichia coli that lack K88, K99 and 987P fimbriae. J. Gen. Microbiol. 129, 2753-2759.

NAGY, B., W. H. MOON, R. E. ISAACSON (1977): Colonization of porcine intestine by enterotoxigenic Escherichia coli: selection of piliated forms in vivo, adhesion of piliated forms to epithelial cells in vitro, and incidence of a pilus antigen among porcine enteropathogenic E. coli. Infect. Immunol. 16, 344-352.

NAKAZAWA, M., C. SUGIMOTO, Y. ISAYAMA, M. KASHIWAZAKI (1987): Virulence factor in Escherichia coli isolated from piglets with neonatal and post-weaning diarrhea in Japan. Vet. Microbiol. 13, 291-300.

ØRSKOV, I., F. ØRSKOV (1983): Serology of Escherichia coli fimbriae. Prog. Allergy 33, 80-105.

ØRSKOV, I., F. ØRSKOV (1984): Serotyping of E. coli. In: Methods in Microbiology (Bergan, T., J. R. Norris, Eds.). Acad. Press, N. Y. pp. 43-112.

RONNBERG, B., T. WADSTROM (1983): Rapid detection by a coagglutination test for heat-labile enterotoxin in cell lysates from blood agar grown Escherichia coli. J. Clin. Microbiol. 17, 1021-1025.

ROSE, R., S. C. WHIPP, H. W. MOON (1987): Effects of Escherichia coli heat-stable enterotoxin b on small intestinal villi in pigs, rabbits and lambs. Vet. Pathol. 24, 71-79.

RUDENSKY, B., M. ISACSOHN, A. ZILBERBERG (1988): Improved detection of heat-labile enterotoxin of enterotoxigenic Escherichia coli by using a commercial coagglutination test. J. Clin. Microbiol. 26, 2231-2232.

SOJKA, W. J. (1965): Escherichia coli in domestic animals and poultry. Commonwealth Agricultural Bureaux. Farnham Royal, UK.

SOJKA, W. J. (1973): Enteric colibacillosis in pigs. Lectures in Mexico City. Central Veterinary Laboratory. Weybridge, Surrey, England.

SVENNERHOLM, A. M., J. HOLMGREN (1978): Identification of Escherichia coli heat-labile enterotoxin by means of a ganglioside immunosorbent assay. (GMI-ELISA) procedure. Curr. Microbiol. 1, 19-23.

VRBANAC, I., I. UDOVICIC, I. VALPOTIC (1991): Prevalence of K88 (F4), K99 (F5) and 987P (F6) fimbrial antigens of Escherichia coli isolated from pigs with colibacillosis. Vet. arhiv 61, 339-354.

WHIPP, S. C., H. W. MOON, R. A. ARGENZIO (1981): Comparison of enterotoxic activities of heat-stable enterotoxins from class 1 and class 2 Escherichia coli of swine origin. Infect. Immunol. 31, 245-251.

YANO, T., D. DA S. LEITE, I. J. B. DE CAMARGO, A. F. DE CASTRO (1986): A probable new adhesive factor (F42) produced by enterotoxigenic Escherichia coli isolated from pigs. Microbiol. Immunol. 30, 495-508.

YOLKEN, R. H., H. B. GREENBERG, M. H. MERSON, R. B. SACK, A. Z. KAPIKIAN (1977): Enzyme-linked immunosorbent assay for detection of Escherichia coli heat-labile enterotoxin. J. Clin. Microbiol. 6, 439-444.

Received: 1 October 1997
Accepted: 26 October 1998

HUMSKI, A., V. BILIC, D. HAJSIG, B. HABRUN: Serotipizacija enterotoksigenih izolata bakterije Escherichia coli iz sisajuce i odbite prasadi s kolibacilozom. Vet. arhiv 68, 191-198, 1998.

SAZETAK

Istrazena je hemoliticka aktivnost, pripadnost seroloskoj skupini i prisutnost adhezivnih antigena u 929 izolata E. coli izdvojenih iz sisajuce (338 izolata) i odbite prasadi (591 izolat) s klinickim manifestacijama kolibaciloze. Hemoliticka aktivnost ustanovljena je u 823 (89%) izolata uz dvostruko vecu ucestalost (69%) medu izolatima izdvojenim iz odbite nego sisajuce prasadi (32%). Pripadnost seroloskim skupinama ustanovljena je u 18% izolata izdvojenih iz sisajuce prasadi uz dominantnu prisutnost serovarova O149:K91 i O8:K87 (48% i 30%). U odbite prasadi dominantno su bili prisutni serovarovi O8:K87, O149:K91 i O157:K"V17" (27%, 26% i 19% sojeva). Adhezivni antigen K88ac dokazan je u 61% seroloski tipiziranih izolata, a najcesce uz serovarove O149:K91 (29%) i O8:K87 (23%). U sisajuce prasadi dokazan je u 23%, a u odbijene prasadi u 42% tipiziranih izolata. Istodobna prisutnost adhezivnog antigena K88ac i hemoliticke aktivnosti ustanovljena je u 65% seroloski tipiziranih izolata E. coli. Prisutnost adhezivnog antigena 987P ustanovljena je u 27% hemolitickih izolata vrste E. coli, uz jednaku ucestalost (50%) u sisajuce i odbite prasadi. Adhezivni antigen K99 dokazan je kod odbite prasadi u samo 4 seroloski tipizirana izolata (1%), koji su istodobno imali sposobnost tvorbe hemolizina.

Kljucne rijeci: kolibaciloza, prasad, hemoliticka aktivnost, fimbrijski adhezini, seroloska skupina