Survey in Vanuatu on enterotoxigenic Escherichia coli in children and infants with and without acute diarrhea.

We have studied the incidence of enterotoxigenic Escherichia coli (ETEC) strains isolated from infants with and without diarrheal diseases in Vanuatu, South Pacific. Over a period of 5 months we have isolated enterotoxigenic E. coli strains from 29 (26.6%) of 109 children with acute diarrhea and from 13 (21.6%) of 60 children of the control group. In the group with diarrhea, 7 (6.4%) strains released heat-labile toxin, 7 (6.4%) released heat-stable toxin, and 15 (13.7%) produced both heat-labile and heat-stable toxin. In the control group, only one strain (1.6%) produced heat-stable toxin, 12 (20%) produced heat-labile toxin, and none produced both. Association of strains releasing heat-stable toxin or both heat-labile and heat-stable toxin with diarrhea was highly significant as shown by statistical analysis. The O serogroups and colonization factors CFA/I and CFA/II are presented.     

Relationship between enterotoxin production and serotype in enterotoxigenic Escherichia coli.

We examined the relationship between serotype and enterotoxin production in 109 enterotoxigenic Escherichia coli strains isolated from 109 patients with severe cholera-like diarrhea in Dacca, Bangladesh. Of 69 strains producing both heat-labile and heat-stable toxins, 59 (86%) belonged to the one of four O serogroups, and 56 (81%) of these strains belonged to one of six O:K:H serotypes. In contrast, 34 strains producing only heat-stable toxin were distributed among 15 O serogroups, and six strains producing only heat-labile toxin were distributed among six O serogroups. Twelve strains producing heat-labile and heat-stable toxins and five strains producing heat-stable toxin were found which had the same serotype (O78:K-:H12) and biotype. It appears that at least in one geographic setting E. coli strains producing both heat-labile and heat-stable toxins are more restricted in their O groups and O:K:H serotypes than E. coli that produce only heat-stable toxin and that certain serobiotypes may commonly include strains which produce both toxin types.     

O serogroups, biotypes, and eae genes in Escherichia coli strains isolated from diarrheic and healthy rabbits.

A total of 305 Escherichia coli strains isolated from diarrheic and healthy rabbits in 10 industrial fattening farms from different areas of Spain were serotyped, biotyped, and tested for the presence of the eae gene and toxin production. The characteristics found in strains isolated from healthy rabbits were generally different from those observed in E. coli strains associated with disease. Thus, strains with the eae gene (74% versus 22%); strains belonging to serogroups O26, O49, O92, O103, and O128 (64% versus 12%); rhamnose-negative strains (51% versus 5%); and rhamnose-negative O103 strains with eae genes present (41% versus 1%) were significantly (P < 0.001 in all cases) more frequently detected in isolates from diarrheic animals than in those from healthy rabbits. Whereas a total of 35 serogroups and 17 biotypes were distinguished, the majority of the strains obtained from diarrheic rabbits belonged to only four serobiotypes, which in order of frequency were O103:B14 (72 strains), O103:B6 (16 strains), O26:B13 (12 strains), and O128:B30 (12 strains). These four serobiotypes accounted for 48% (112 of 231) and 5% (4 of 74) of the E. coli strains isolated from diarrheic and healthy rabbits, respectively. Only six strains were toxigenic (three CNF1+, two CNF2+, and one VT1+). We conclude that enteropathogenic E. coli strains that possess the eae gene are a common cause of diarrhea in Spanish rabbit farms and that the rhamnose-negative highly pathogenic strains of serotype O103:K-:H2 and biotype B14 are especially predominant. Detection of the eae gene is a useful method for the identification of enteropathogenic E. coli strains from rabbits. However, a combination of serogrouping and biotyping may be sufficient to accurately identify the highly pathogenic strains for rabbits.     

Does enteropathogenic Escherichia coli produce heat-labile enterotoxin, heat-stable enterotoxins a or b, or cholera toxin A subunits?

Although most enteropathogenic Escherichia coli strains do not produce recognized enterotoxins, we wished to examine whether they produce any factors like heat-stable enterotoxin b or cholera toxin active subunits that might be missed by conventional assay methods. E. coli strains E851 (O142) and E2348 (O127) that had caused diarrhea in volunteers were negative for heat-labile enterotoxin and heat-stable enterotoxin a in Chinese hamster ovary cell and suckling mouse assays, failed to cause secretion in ligated small bowel loops from 6- to 8-week-old pigs after 4 to 5 h (used to show heat-stable enterotoxin b), and did not activate adenylate cyclase in pigeon erythrocyte lysates (used to demonstrate cholera toxin A subunit). We conclude that crude, unconcentrated culture filtrates and sonicates do not mimic heat-labile or heat-stable enterotoxins or cholera toxin or its A subunit and that enteropathogenic strains of E. coli probably have yet another mechanism or group of mechanisms by which they cause diarrhea.     

Etiology of childhood diarrhea in Korea.

To assess the role of recently recognized enteropathogens in childhood diarrhea in Korea, 231 children with diarrhea admitted to and 104 children without diarrhea seen at the well-baby clinic or the outpatient department of Hanyang University Hospital in Seoul, Korea, were evaluated during a 14-month period. Stools were cultured for bacterial pathogens, including enterotoxigenic (heat-labile and heat-stable enterotoxin-producing) and enteroadherent organisms. Only those stools obtained from patients with diarrhea were examined for rotavirus. All Escherichia coli isolates were screened for Shiga-like toxin (SLT) I, SLT-II, enterohemorrhagic E. coli fimbriae, and enteroinvasiveness by colony hybridization. One or more pathogens were identified in 75.8% of the children with diarrhea. Rotavirus was the most frequently identified pathogen, accounting for 47% of the cases. Other major enteropathogens were enterotoxigenic E. coli (22%), Clostridium difficile (16%), enteroadherent E. coli (15%), and enteropathogenic E. coli (6%). Shigella spp., Campylobacter jejuni, Salmonella spp., SLT-I-and enterohemorrhagic-E. coli-fimbria-probe-positive E. coli serotype O26:H11 and enteroinvasive E. coli were isolated from only a few patients. Aeromonas hydrophila and E. coli O157 were not isolated. Compared with those of the controls, the isolation rates of heat-stable-enterotoxin-producing E. coli (P less than 0.05), C. difficile (P less than 0.025), and enteroadherent E. coli (P less than 0.05) were significantly higher in the patients with diarrhea. The greatest number of rotavirus, enterotoxigenic E. coli, and C. difficile cases were identified during the cool, dry months of October and November.     

Outbreak of infantile enteritis caused by enterotoxigenic Escherichia coli O6.H16.

Ten out of 18 babies at risk developed enteritis in an outbreak in a special-care baby unit. Salmonella, Shigella, and Escherichia coli belonging to the traditional infantile enteropathogenic serogroups were not found in faeces from the babies and the staff, and no virus particles were found by electron microscopy. Detailed serotyping of E. coli showed that five of the 10 babies with diarrhoea and one of the eight without diarrhoea were excreting E. coli O6.H16. All six isolates of this serotype produced both heat-labile and heat-stable enterotoxin. Enterotoxigenic strains of E. coli O6.H16 have caused outbreaks of enteritis in adults in the USA and Japan.     

Frequency of Escherichia coli strains producing heat-labile toxin or heat-stable toxin or both in children with and without diarrhea in São Paulo.

Enterotoxigenic Escherichia coli strains were isolated from 32 (13.4%) of 245 children with diarrhea and from 11 (11.4%) of 96 children of the control group. Strains producing heat-labile toxin were found more frequently in normal children than in children with diarrhea. Strains producing heat-stable toxin and both heat-labile and heat-stable toxins were isolated only from children with diarrhea. Association of these strains with diarrhea was highly significant as shown by statistical analysis. The O:H types and the colonization factors of strains producing heat-stable toxin and both heat-labile and heat-stable toxins are presented.     

Natural and experimental infection with an attaching and effacing strain of Escherichia coli in calves.

Gnotobiotic calves were inoculated with an O5:K4:H-, urease-positive strain of Escherichia coli isolated from a 2-day-old calf with diarrhea. The calves developed elevated temperatures and passed loose mucoid feces, with or without blood. The E. coli strain was negative for heat-stable and heat-labile enterotoxins but produced high levels of Shiga-like toxin. Bacteria attached diffusely to the epithelium of the large intestine and multifocally to the epithelium of the ileum. The duodenum and jejunum were not affected. At the sites of bacterial attachment, microvilli were effaced, enterocytes were degenerate, and necrosis and exfoliation had occurred. These results confirm a previous report from England that calves may naturally contract infections similar to those caused by enteropathogenic E. coli strains pathogenic to humans or rabbits. This suggests that the calf bacterial strains, like some enteropathogenic E. coli strains, produce high levels of Shiga-like toxin and cause attachment and effacement lesions in the colonic epithelium of the infected host.     

Colonization factors of enterotoxigenic Escherichia coli isolated from children with diarrhea in Argentina.

A prospective study was performed to evaluate the presence of colonization factor antigens (CFAs) in enterotoxigenic Escherichia coli (ETEC) strains isolated from 1,211 children with diarrhea in Argentina. One hundred nine ETEC strains that were isolated from seven different laboratories in various regions of the country were tested for CFAs by using monoclonal antibodies against CFA/I and E. coli surface antigens CS1, CS2, and CS3 of CFA/II and CS4 and CS5 of CFA/IV; a polyclonal antiserum against CS6 was used. The CFAs searched for were found in 52% of the ETEC strains: 23% of the strains carried CFA/I, 17% carried CFA/IV, and 12% carried CFA/II. All of the CFA/I strains produced heat-stable enterotoxin, and several of them were of the prevalent serotypes O153:H45 and O78:H12. Among the 19 strains expressing CFA/IV, 16 expressed CS5 and CS6 and produced the heat-stable enterotoxin and most were of serotype O128:H21; the remaining 3 strains produced CS6 only. No ETEC strains expressing CS4 were found. Most (11 of 13) of the CFA/II-carrying ETEC strains expressed CS1 and CS3, and 10 of them were of the O6:K15:H16 serotype and produced both heat-labile and heat-stable toxins. As many as 24 of the 109 CFA-negative ETEC strains gave mannose-resistant hemagglutination with erythrocytes from different species; 4 strains had high surface hydrophobicity, suggesting the presence of additional, as yet undefined, colonization factors in up to 25% of the ETEC isolates.     

Laboratory investigation of enteroaggregative Escherichia coli O untypeable:H10 associated with a massive outbreak of gastrointestinal illness.

A massive outbreak of gastrointestinal illness occurred in Tajimi city, Gifu prefecture, in June of 1993 in which 2,697 children in elementary and junior high schools developed severe diarrhea. Stool specimens from 30 children with severe protracted diarrhea were studied. Twenty-seven strains of enteroaggregative Escherichia coli (EAggEC) isolated from 12 of 30 patients all belonged to the same serotype, O untypeable (OUT):H10, and showed the same biochemical characteristics and antibiotic susceptibility pattern. These strains were negative for the virulence factors of the four standard categories of diarrheagenic E. coli (enterotoxigenic, enteropathogenic, enteroinvasive, and enterohemorrhagic). However, the isolates showed an aggregative pattern of adherence to HEp-2 cells and had a 60-MDa plasmid and an astA gene, which encodes heat-stable enterotoxin-1 production. These data suggested that the EAggEC serotype OUT:H10 was associated with this massive outbreak of gastrointestinal illness.     

Nationwide surveillance program to identify diarrhea-causing Escherichia coli in children in Thailand.

Escherichia coli strains isolated from children with diarrhea were collected from 16 hospitals in different districts in Thailand during 1985 and 1986 and submitted to the National Reference Laboratory. Isolates were identified by serogrouping or as enterotoxigenic E. coli (ETEC), enteroinvasive E. coli (EIEC), enteropathogenic E. coli (EPEC) adhesin factor (EAF) E. coli, or Shiga-like-toxin (SLT)-producing E. coli by DNA hybridization. EPEC strains of known serogroups were isolated from 10%, ETEC strains were isolated from 6%, EAF E. coli strains were isolated from 4%, EIEC strains were isolated from less than 1%, and SLT-producing E. coli strains were isolated from none of 393 children with diarrhea. Among 278 children whose ages were recorded, the highest rate of isolation of EAF E. coli was 11% (9 of 85) from children less than 6 months old. ETEC was isolated from 5% (4 of 85) of children less than 6 months old, from 10% (12 of 118) of children 6 to 23 months old, and from 1% (1 of 75) of children greater than 23 months old. EPEC strains of known serogroups were isolated from 18% (15 of 85) of children less than 6 months old, from 11% (13 of 118) of children 6 to 23 months old, and from 9% (7 of 75) of children greater than 23 months old. E. coli strains that hybridized with the EIEC probe were isolated from three children who were 20, 36, and 48 months old. Examining E. coli for hybridization with DNA probes for virulence determinants is a practical way of conducting nationwide surveillance of diarrhea-causing E. coli. Since only 33% (13 of 39) of EPEC serogroups hybridized with the EAF probe and none hybridized with the SLT probes, identification of EPEC by serogroups analysis, followed by serotyping, should continue to be used in the identification of EPEC.     

Multiplicity of serogroups and adhesins in enteropathogenic and enterotoxigenic Escherichia coli isolated from acute diarrhea in Senegal.

Escherichia coli strains were isolated from 228 children with diarrhea in Senegal from 1982 to 1984. Among these E. coli involved in cases of diarrhea, we found that 20.3% were enteropathogenic E. coli. Only 3.9% of the strains adhered to the brush borders of human intestinal enterocytes, and they belonged to different serotypes. All these adhesion-positive strains possessed genes encoding for the heat-stable enterotoxin, but their adhesive factors were different regarding serology with anti-colonization factor sera, hemagglutination patterns, electron microscopy structures, or major surface protein subunits.     

Controlled study of cytolethal distending toxin-producing Escherichia coli infections in Bangladeshi children.

The role of cytolethal distending toxin (CDT)-producing Escherichia coli, a newly described category of E. coli, in the causation of diarrhea was studied by screening E. coli isolates from 546 children < 5 years of age with diarrhea and 215 matched controls without diarrhea by using a specific DNA probe. Although CDT-positive E. coli strains were isolated from more children with diarrhea than from healthy controls (3.1 versus 0.93%), this difference did not reach statistical significance (P = 0.082). All CDT-positive strains also possessed the virulence factors of enteropathogenic E. coli or enteroaggregative E. coli isolates.     

Controlled study of Escherichia coli diarrheal infections in Bangladeshi children.

Diarrheal diseases are highly prevalent in Bangladesh. However, the relative contribution of diarrheagenic Escherichia coli organisms--those that are enterotoxigenic (ETEC), enteropathogenic (EPEC), enteroinvasive, enterohemorrhagic, enteroaggregative, and diffuse adherent--to diarrhea in Bangladeshi populations is not known. With DNA probes specific for these diarrheagenic E. coli strains, we analyzed fecal E. coli from 451 children up to 5 years of age with acute diarrhea seeking treatment at a Dhaka hospital and from 602 matched control children without diarrhea from July 1991 to May 1992. Enteroinvasive E. coli was not isolated from any children; enterohemorrhagic E. coli was not isolated from any diarrheal children but was isolated from five control children; enteroaggregative and diffuse adherent E. coli strains were isolated with similar frequencies from children with and without diarrhea, thereby showing no association with diarrhea; ETEC was significantly associated with diarrhea in the diarrheal children as a whole and especially in the age groups of 0 to 24 months and 37 to 48 months (further analysis suggests an association with diarrhea for the heat-stable toxin only and for both heat-labile- and heat-stable-toxin-producing ETEC only); and EPEC was significantly associated with diarrhea in the diarrhea group as a whole and particularly in infants up to 1 year of age. Further analysis suggested that EPEC strains of only the traditional serogroups were significantly associated with diarrhea. ETEC and EPEC infections peaked during warm months. Our data thus suggest that EPEC and ETEC are important causes of acute diarrhea in children in this setting.     

Virulence factors of lactose-negative Escherichia coli strains isolated from children with diarrhea in Somalia

Lactose-negative Escherichia coli strains were isolated at high frequency from children with diarrhea in Somalia during a 2-year study on diarrheal diseases. Sixty-four of these strains, considered to be a representative sample, were characterized for virulence factors, plasmid profiles, and antibiotic resistance. Of these strains, 5 were recognized as enteroinvasive E. coli (they were serotyped as O135:K-:H-), 6 belonged to classical enteropathogenic E. coli serotypes, 9 were able to adhere to tissue culture cells (of these, 4 showed a pattern of localized adherence and 1 was an enteropathogenic strain), 18 were both adherent and hemolytic, and 8 were simply hemolytic. None hybridized with 32P-labeled heat-labile or heat-stable (a and b) enterotoxin gene probes or produced moderate or high-level cytotoxic effects on HeLa cells. Of the 64 strains examined, 24 produced mannose-resistant hemagglutination with human, chicken, and monkey erythrocytes. One of these was serotyped as O4:K-:H8, and a rabbit O antiserum raised against this strain allowed us to establish that 23 strains had the same O antigen. The 23 O4 strains were hemolytic and were not enterotoxic for rabbit ileal loops, and intact bacteria were able to destroy tissue culture cell monolayers very rapidly. The uniformity of the antibiotic resistance pattern and of the plasmid DNA content, together with the fact that they were isolated in different years and in different children, suggests that the O4 strains must be epidemiologically relevant in Somalia. A possible diarrheagenic role for the adherent-hemolytic E. coli strains is also discussed.     

Characterization of Escherichia coli O157:H7 and other Shiga toxin-producing E. coli serotypes isolated from sheep.

The isolation and characterization of Escherichia coli O157:H7 and non-O157 Shiga toxin-producing E. coli (STEC) strains from sheep are described. One flock was investigated for E. coli O157:H7 over a 16-month period that spanned two summer and two autumn seasons. Variation in the occurrence of E. coli O157:H7-positive sheep was observed, with animals being culture positive only in the summer months but not in the spring, autumn, or winter. E. coli O157:H7 isolates were distinguished by pulsed-field gel electrophoresis (PFGE) of chromosomal DNA and toxin gene restriction fragment length polymorphism (RFLP) analysis. Ten PFGE patterns and five RFLP patterns, identified among the isolates, showed that multiple E. coli O157:H7 strains were isolated from one flock, that a single animal simultaneously shed multiple E. coli O157:H7 strains, and that the strains shed by individuals changed over time. E. coli O157:H7 was isolated only by selective enrichment culture off 10 g of ovine feces. In contrast, strains of eight STEC serotypes other than O157:H7 were cultured from feces of sheep from a separate flock without enrichment. The predominant non-O157 STEC serotype found was O91:NM (NM indicates nonmotile), and others included O128:NM, O88:NM, O6:H49, and O5:NM. Irrespective of serotype, 98% of the ovine STEC isolates possessed various combinations of the virulence-associated genes for Shiga toxin(s) and the attaching-and-effacing lesion (stx1, stx2, and eae), suggesting their potential for human pathogenicity. The most common toxin-eae genotype was positive for stx1, stx2, and eae. A Vero cell cytotoxicity assay demonstrated that 90% of the representative STEC isolates tested expressed the toxin gene. The report demonstrates that sheep transiently shed a variety of STEC strains, including E. coli O157:H7, that have potential as human pathogens.     

Toxin detection after storage or cultivation of enterotoxigenic with colicinogenic Escherichia coli: a possible mechanism for toxin-negative pools.

Of 100 non-enterotoxigenic Escherichia coli isolated from children with diarrhea in Bangkok, Thailand, 24 were found to produce colicin(s). Of these, 87% were active against one or more enterotoxigenic E. coli isolated from the same population. Storage of nine enterotoxigenic E. coli with known inhibitory colicin-producing E. coli in different proportions caused 51 of 96 pools to become negative in the suckling mouse assay (heat-stable toxin) and 17 of 52 to become negative in the Y-1 adrenal cell assay (heat-labile toxin). Cocultivation of the same strains, without prior storage, caused 12 of 96 pools to become negative for heat-stable toxin and 1 of 52 pools to become negative for heat-labile toxin. Storage or cultivation of E. coli in pools may cause negative results in the suckling mouse and Y-1 adrenal cell assays if any of the isolates in the pool produces colicin(s).     

Correlation between adherence to HeLa cells and serogroups, serotypes, and bioserotypes of Escherichia coli.

Four hundred fifty Escherichia coli strains of 45 O serogroups and subgroups and 112 serotypes were studied to determine their patterns of adherence to HeLa cells. Adherence was exhibited by strains of 17 O serogroups and subgroups, but within these groups more than one adherence pattern was frequently observed. However, within each serotype, the adherence pattern was highly consistent. Localized adherence (LA) was observed much more frequently in serotypes that we considered to be enteropathogenic E. coli serotypes (93%) than in other serotypes (14%), whereas diffuse adherence (DA) occurred predominantly among nonenteropathogenic E. coli strains. Determination of biochemical characteristics showed that within O serogroups, nonmotile strains tended to have the same behavior as motile strains with the LA adherence pattern, suggesting that they were derived from these motile strains. LA and non-LA strains of the same serotype differed biochemically. LA appears to be a property of most E. coli commonly considered to be enteropathogenic and should assist attempts to determine which E. coli are enteropathogenic and to elucidate their pathogenic mechanisms.     

Enzyme-linked immunosorbent assay for products of the 60-megadalton plasmid of Escherichia coli serotype O157:H7.

Eighty strains of pathogenic Escherichia coli, representing each of the major diarrheal disease-causing groups, were examined by direct enzyme-linked immunosorbent assay (ELISA) for the presence of proteins associated with a 60-MDa plasmid from E. coli serotype O157:H7. Antiserum specific for plasmid-encoded proteins was prepared by immunizing a rabbit with a wild-type E. coli O157:H7 strain (strain 7785) and absorbing the serum with a plasmid-cured derivative (strain 2-45). Use of this antiserum in Western immunoblot analysis detected two proteins of 82 and 92 kDa in strain 7785 but not in strain 2-45. All 16 wild-type E. coli O157:H7 strains and all 10 Shiga-like toxin (SLT)-producing E. coli strains of serotypes other than O157 were ELISA positive. Thirteen of 14 enterotoxigenic and all of 24 enteroinvasive E. coli strains were ELISA negative, as were all of 16 E. coli strains isolated from healthy persons. Of 16 traditional enteropathogenic E. coli (EPEC) serotypes, 10 were ELISA positive, including 10 of 12 strains carrying the EPEC adherence factor gene. Absorption of the serum with an EPEC adherence factor-positive EPEC eliminated EPEC reactivity. This study demonstrates that two plasmid-mediated proteins are common to E. coli O157:H7 strains and to SLT-producing strains of other serotypes. Detection of these proteins by ELISA provides a sensitive and specific screening test for identifying SLT-producing E. coli of both O157 and non-O157 serotypes. Identification of the cross-reactive proteins found in EPEC could provide the basis for a single assay to detect both EPEC and SLT-producing E. coli.     

Serotype, antimicrobial resistance, and adherence properties of Escherichia coli strains associated with outbreaks of diarrheal illness in children in the United States.

Since most recorded outbreaks of diarrhea in U.S. infants attributed to Escherichia coli occurred before currently available pathogenicity assays existed, we examined the characteristics of nonenterotoxigenic E. coli strains isolated from 50 outbreaks of diarrheal disease in U.S. infants between 1934 and 1987. We assayed the strains for enteropathogenic E. coli (EPEC) serotype, localized adherence (LA) and diffuse adherence to tissue cultures, the presence of EPEC adherence factor genes, Shiga-like (Vero) toxin production, and antimicrobial resistance. EPEC serotypes were identified in 28 outbreaks (56%). LA to HeLa cells was found in 23 outbreak strains and correlated 100% with the EPEC adherence factor probe. LA was observed in 21 of 28 EPEC and 2 of 22 non-EPEC strains; however, 5 of 23 strains that were LA positive for HeLa cells did not adhere to HEp-2 or HL cells. One strain was diffuse adherence positive, and none was Shiga-like toxin positive. Multiple resistance was common in EPEC (64%), LA-positive (74%), and LA-positive EPEC (76%) strains but not in others (10%). EPEC serotypes or LA was found in 60% (n = 30) of the outbreak strains. The remaining E. coli strains may represent nonpathogenic normal flora, as-yet-undefined pathogens, or pathogens that have lost virulence-associated traits during storage or subculturing.