Journal of Postgraduate Medicine
 Open access journal indexed with Index Medicus & EMBASE  
     Home | Subscribe | Feedback  

Year : 1983  |  Volume : 29  |  Issue : 2  |  Page : 62-6  

Correlation between serotypes, enterotoxin production and presence of colonisation factor antigens (CFA) in E. coli isolates from acute diarrhoea in children.

LL Iyer, CK Deshpande, VV Kale, MK Jain 

Correspondence Address:
L L Iyer

How to cite this article:
Iyer L L, Deshpande C K, Kale V V, Jain M K. Correlation between serotypes, enterotoxin production and presence of colonisation factor antigens (CFA) in E. coli isolates from acute diarrhoea in children. J Postgrad Med 1983;29:62-6

How to cite this URL:
Iyer L L, Deshpande C K, Kale V V, Jain M K. Correlation between serotypes, enterotoxin production and presence of colonisation factor antigens (CFA) in E. coli isolates from acute diarrhoea in children. J Postgrad Med [serial online] 1983 [cited 2022 May 18 ];29:62-6
Available from:

Full Text


Esch. coli was established as a cause of childhood diarrhoeas since the 1940's when certain serotypes of this organism were isolated from serious epidemics.[1] The discovery of enterotoxigenic (ETEC) and enteroinvasive (EIEC) Esch. coli have thrown new light on diarrhoeal disease caused by Esch. coli. ETEC have emerged as the most important bacterial pathogens in diarrhoeas in children.[9], [13], [15], [21], [22] Enterotoxin produced by these organisms is plasmid mediated.[11] ETEC have also been found to possess two fimbrial antigens designated as colonisation factor antigen I (CFA I) and colonisation factor antigen II (CFA II). These factors help in the adherance of these organisms to the small bowel.[6], [10] A majority of the EPEC serotypes have not been found to be either toxigenic or invasive.[12], [14], [16], [18], [20], [23] Because the mechanism of pathogenicity of these strains is not clear there is a serious doubt regarding their aetiological role in sporadic diarrhoeas in children. The present study was undertaken to determine the serotype, enterotoxin production and CFA in the Esch. coli isolates from :cute diarrhoeas of children.


Faecal samples or rectal swabs were collected from children upto the age of 12 years suffering from acute diarrhoea. A total of 442 samples were obtained from out-patients and in-patients attending the Municipal General Hospitals, Bombay between November 1978 and April 1980. Faecal samples from 43 normal healthy children of the same age group who had not suffered from diarrhoea during the past one month were also studied. Only patients who had not taken antibiotics were studied. Each sample was collected in buffered glycerol saline and alkaline peptone water. The specimen was inoculated immediately on receipt at the laboratory, on MacConkey's agar, desoxycholate citrate agar and blood agar for the growth of Esch. coli as well as other pathogens.[2] Esch. coli isolates were identified by standard methods.[5] These were preserved in nutrient afar stabs sealed with paraffin wax. Serotyping of the Esch. coli isolates was done at the National Reference Centre at Kasauli.

Test for enterotoxin

The isolates were tested for heat labile (LT) enterotoxin production by the rabbit ileal loop method.3 The procedure was as follows:

Adult albino rabbits weighing 1.5-2.0 kg were fasted for 24-48 hours. The abdomen was opened under ether anaesthesia and ligatures were put in the proximal part of the ileum at distances of 8-10 cm. One ml of saline suspensions of different isolates of Esch. coli containing about 108 organisms were injected into separate loops. A negative saline control and a positive control (V. cholorae 569 B Inaba) was included with the test cultures. The abdomen was closed and the animal was sacrificed after 18 hours. Loops were examined for dilatation and fluid accumulation. Loops showing good fluid accumulation (at least 10 times the amount of original injection) were taken as positive for enterotoxin production.

Tests for colonisation factor antigens I and II were done by the haemagglutination methods of Evans et al[7] and Evans and Evans[6] respectively.


Esch. coli were isolated from 400 of 442 faecal samples of cases of acute diarrhoea. Of the 292 isolates serotyped, 46 (15.8%) were known enteropathogenic serotypes, the commonest being 0128, 020 and 018. 156 isolates (53.4%) belonged to other `O' serotypes, the most frequent being 017, 01, 060 and 011. Rough strains and untypable isolates comprised 30.8%. The details of the SPEC and Non-EPEC serotypes and the results of enterotoxin testing are shown in [Table I] and [Table II.] A total of 8 of 33 (24.24%) enteropathogenic serotypes tested were enterotoxigenic, and 43 of 125 (34.4%) other `O' types tested were enterotoxigenic.

E. coli were isolated from 32 of the 43 controls; 25 isolates were serotyped and 3 were EPEC (serotypes 044, 0124 and 0127). Four isolates were rough strains, and there were 3 each of serotypes 017, and 060, 2 each of 02 and 025 and 1 each of 05, 07, 040, 063 and 0115. Only 1 of the 21 isolates tested was positive for enterotoxin (Serotype 02).

CFAs were present on 21 of 47 (44.7%) ETEC, and also on 34 of 100 (34%) enterotoxin negative isolates. CFA I was found on 16, of 21 toxigenic isolates and 25 of 34 non-toxigenic isolates. The remaining isolates possessed CFA II. Twenty one E. Coli isolated from controls were tested and 2 non-ET'EC isolates were found to have CFA I.


In the present study 8 out of 33 (24.24 per cent) EPEC strains were found to release an enterotoxin. Earlier studies[3], [4], [13], [22] using the rabbit ileal loop model have described similar results. Recently, Tewari et al[25] have reported 82 per cent and Sarkar et al[24] have reported 66 percent of EPEC to be enterotoxigenic using the rabbit ileal loop technique and infant mouse technique. This is in contrast to the report of Panhotra et al[20] who, using the same techniques, found all the EPEC to be negative for enterotoxin. Several other reports[12], [14], [16], [18], [23] in the West have also shown all the EPEC serotypes tested to be consistently negative for enterotoxin using the Yl or Chinese hamster ovary cell assay for LT and the infant mouse test for ST. Unlike epidemic diarrhoeas, there is insufficient epidemiological evidence to prove the pathogenic role of Esch. coli in sporadic diarrhoea. Thus, until a mechanism of pathogenicity is established, EPEC can only be viewed as presumptive pathogens in sporadic diarrhoea. In such cases, a correlation with bacteriological findings of an almost pure growth of Esch. coli and clinical findings of characteristic EPEC disease may probably be useful.

In the present study results for enterotoxin production by the other 'O' serotypes showed absence of correlation with any particular serotype. Similar observations were made by Gorbach[18] and Sack.[22] Orskov et al[18] and other workers[21], [22] however reported ETEC to belong to only a few `O' serogroups. E. coli isolated from controls in the present study belonged to similar serotypes as those obtained from cases of diarrhoea. Only one was toxigenic. Therefore serotyping cannot be used as a test to diagnose ETEC infections, and a test for enterotoxin has to be done.

The presence of colonisation factor antigens did not correlate with enterotoxicity in the present study. CFAs were found on both ETEC and non-ETEC. Joseph[17] has described similar findings. Our results indicate that CFAs are probably not useful diagnostic markers for ETEC. However, Evans et al[8] have used CFA specific antisera and have found CFAs to be specifically associated with ETVC. Panthotra et al[9] have also found all ETEC strains to be positive for CFAs.


The authors would like to thank the Director, Central Research Institute, Kasauli, for serotyping of Esch. coli isolates.

This study was undertaken under the auspices of Research Society of K.E.M. Hospital, Bombay.


1Cooke, M. E.: "Escherichia coli and Man". Churchill-Livingstone, Edinburgh, London, 1974, pp 47-60.
2Cruickshank, R.: "Medical Microbiology" 12th Edn. Vol. II, Churchill Livingstone, Edinburgh, London, New York, 1975, pp 125-128.
3De, S. N., Bhattacharya and Sarkar, S. K.: A study of the pathogenicity of strains of Bacterium coli from acute and chronic enteritis. J. Path. & Bact., 71: 201-209, 1956.
4Drucker, M. M., Yeivin R. and Sack, T. G.: Pathogenesis of Esch. coli -enteritis in the ligated rabbit gut. Israel J. Med. Sci., 3: 445-452, 1967.
5Edwards, P. R. and Ewing, W. H.: "Identification of Enterobacteriaceae". 3rd Edition, Burgess Publishing Company, Minneapolis, 1972, pp 258.
6Evans, D. G. and Evans, D. J. Jr.: New surface associated, heat labile, colonisation factor antigen (CFA/II) produced by enterotoxigenic Esch. coli of serogroups 06 and 08. Infect. Immunol., 21: 638-647, 1978.
7Evans, D. G., Evans, D. J. Jr. and Tjoa, W.: Haemagglutination of human group A erythrocytes by enterotoxigenic Esch. coli isolated from adults with diarrhoea. Correlation with colonisation factor. Infect. Immunol., 18: 330-337, 1977.
8Evans, D. G.. Evans, D. J. Jr., Tjoa, V1 '. S. and DuPont, H. L.: Detection and characterisation of colonisation factor o enterotoxigenic Esch. coli isolated from adults with, diarrhoea. Infect. Immunol. 19: 727-736, 1978.
9Evans, D. G., Olarte, J., DuPont, H. L., Evans, D. J. Jr., Galindo, E., Portnoy, B. L. and Cantelin, R. H.: Enteropathogens associated ;with paediatric diarrhoea in Mexico City. J. Paediat., 91: 65-68, 1977.
10Evans, D. G., Silver, R. P., Evans, D. J. Jr., Chase, D. G., and Gorbach, S. L.: Plasmid controlled colonisation factor antigen associated with virulence in Esch. coli enterotoxigenic for humans. Infect. Immunol., 12: 656-667, 1975.
11Giles, C., Magdalene, S. and Falkow, S.: The enterotoxin plasmids of Esch. coli. J. Infect. Dis., 130: 40-49, 1974.
12Goldschmidt, M. C. and DuPont, H. L Enteropathogenic Esch. coli-lack of correlation of serotype with pathogenicity. J. Infect. Dis., 133, 153-156, 1976.
13Gorbach, S. L. and Khurana, L. M.: Toxigenic Esch. coli; A cause of diarrhoea in Chicago. New Engl. J. Med., 237: 791-795, 1972.
14Gross, R. J . , Scotland, S. M. and Rowe, B.: Enterotoxin testing of Esch. coli causing epidemic infantile enteritis in U.K.. Lancet, 1: 629-431, 1976.
15Guerrant, R. L., Dickens, M. D., Wenzel, R. P. and Kapikian, A. Z.: Toxigenic bacterial diarrhea-Nursery outbreak involving multiple bacterial strains. J. Paediatr., 89: 895-891, 1976.
16Gurwith, M. J., Hinde, D., Gross, R. and Rove-, B.: A prospective study of enteropathogenic Esch. coli in endemic diarrhoeal disease. J. Infect. Dis., 137, 292-297, 1978.
17Joseph, T.: Bacteriology of diarrhoea in infants and children. Ind. J. Microbiol., 20: 189-193, 1980.
18Orskov, F., Orskov, L, Evans, D. G. and Sack, R. B.: Special Esch. coli serotypes among enterotoxigenic strains from diarrhoea in adults and children, Med. Microbiol. & Immunol. 162: 73-80, 1976.
19Panhotra, B. R. and Agarwal, K. C.: Haemagglutination and buccal epithelial cell adherance of enterotoxigenic Esch. coli isolated from infants with acute diarrhoeal disease. Ind. J. Med. Res., 74: 187-191, 1981.
20Panhotra, B. R., Agarwal, K. C., Walla, B. N. S., Ayyagiri, A. and Joshi, P.: Enterotoxigenic Esch. coli from acute diarrhoea] disease in infants and children. Ind. J. Med. Res., 72: 347-351, 1980.
21Rowe, B., Scotland, S. M. and Gross. R. J.: Enterotoxigenic Esch. coli causing infantile enteritis in Brita n. Lancet, 1, 90-91, 1977 (Letter).
22Sack, R. B.: Human diarrhoeal disease caused by enterotoxigenic Esch. coli. Ann. Rev. Microbiol., 29: 333-353, 1975.
23Sack, R.. B.: (Letter) Serotyping of Esch. coli. Lancet, 1: 1132, 1976.
24Sarkar, R., Sehgal, H. and Roy Choudhary, A. N.: Clinical and laboratory assessment of the pathogenicity of serotyped enteropathogenic Esch. coli in children with diarrhoeal disease. Indian Paediatrics, 17: 29-35, 1980.
25Tewari, L., Agarwal, S. K., Khan, A. M., Kumar, A. and Mehrotra, R. M. L.: Enterotoxicity of Esch. coli isolated from infantile diarrhoea in rabbits and infant mice. Ind. J. Med. Res., 69: 231-239, 1979.

Wednesday, May 18, 2022
 Site Map | Home | Contact Us | Feedback | Copyright  and disclaimer