Epidemiologic studies of Escherichia coli diarrheal infections in a low socioeconomic level peri-urban community in Santiago, Chile.

The incidence of diarrhea due to six categories of diarrheogenic Escherichia coli was determined in two pediatric cohorts in a low socioeconomic level community in Santiago, Chile, with access to chlorinated water. An age cross-sectional cohort of 340 children aged birth to 47 months was assembled. A newborn cohort was assembled by enrolling 10-12 newborns monthly for 12 months. Episodes of diarrhea were detected by twice weekly household visits. E. coli from stool cultures of cases and matched controls were hybridized with DNA probes specific for enterotoxigenic, enteroinvasive, enteropathogenic, enterohemorrhagic, enteroaggregative, and diffuse adherence E. coli. Overall, the incidence of diarrhea was low (2.1 episodes/infant/year). Nevertheless, a putative E. coli enteropathogen was found in a large proportion of diarrheal episodes, particularly during the summer. In both cohorts, enterotoxigenic E. coli were important pathogens. Enteropathogenic E. coli were incriminated during the first year of life in the newborn cohort, where they were found significantly more often in cases (p = 0.021) than in controls; beyond this age, isolation rates were similar. In contrast, the relative risk of isolation of diffuse adherence E. coli increased with age in the age cross-sectional cohort, where, overall, the difference in rate of isolation between cases and controls was significant (p = 0.0024). Enteroinvasive and enterohemorrhagic E. coli were isolated infrequently. Enteroaggregative E. coli were encountered equally in cases and controls. Facile transmission of E. coli enteropathogens is occurring in this community despite the availability of potable water.

PIP: Researchers conducted an age cross sectional cohort analysis of 340 0-47 month old children and newborn cohort analysis of 144 newborns to determine the diarrheogenic Escherichia coli incidence in Santa Julia, a low socioeconomic community in Santiago, Chile. Children in the age cross sectional cohort had age, sex, and sector matched controls. The newborns had sex matched controls. A public health nurse or nurse auxiliary visited the household of each subject 2 times a week to detect diarrhea episodes. Between December 1986 and February 1990, the age cross sectional cohort had 1178 episodes of diarrhea and the newborn cohort had 674 episodes. The overall diarrhea incidence was only 2.1 episodes/child/year. An E. coli enteropathogen was isolated in many of these episodes, especially during the summer (e.g. enterotoxigenic E. coli [ETEC], 2.2 cases/month in summer vs. 0.4 cases/month in winter; p = .00001). Diffuse adherence E. coli (DAEC) and enteropathogenic E. coli (EPEC) infections also peaked in the summer. ETEC contributed greatly to diarrheal episodes in both cohorts. Among newborns, EPEC was isolated significantly more often in cases than controls during the 1st 12 months of life (6.7% vs. 2.5%; p = .021). After 1 year, however, E. coli isolation rates were essentially the same. On the other hand, in the age cross sectional cohort, the relative risk of isolation of DAEC rose with age (e.g., 1.1 for 0.11 months, 1.4 for 36-47 months, and 2.1 for = or 48 months). In the same cohort, DAEC infections were much more common in cases than controls (16.6% vs. 11.9%; p = .0024). Enteroinvasive and enterohemorrhagic E. coli were the most rarely isolated E. coli types. No difference in the isolation rate of enteroaggregative E. coli existed between cases and controls. Since most households in Santa Julia have access to potable water (68%) and an indoor toilet (64%), food contamination were likely the vehicles of E. coli transmission because more than 50% of households do not have a refrigerator.

Practical and economical method for using biotinylated DNA probes with bacterial colony blots to identify diarrhea-causing Escherichia coli.

A simple and economical method was developed for using biotinylated DNA probes to hybridize with bacterial colonies belonging to the various categories of diarrhea-causing Escherichia coli. Simplification and cost containment were achieved by using Whatman no. 541 filter papers instead of nitrocellulose, by minimizing the concentration of proteinase K (an expensive but necessary reagent used to pretreat the colony blots prior to hybridization with biotin-labeled DNA probes) and by reusing hybridization solution containing labeled probe DNA. After exposing the colony blots to lysing solution and steam, followed by lysozyme (1.5 mg/ml), sucrose (25%), and proteinase K (10 micrograms/ml) treatments, biotinylated probes were used to detect enterotoxigenic, enteropathogenic, enterohemorrhagic, diffuse adherence, and enteroinvasive categories of diarrhea-causing E. coli with a high level of sensitivity and specificity. Three independent observers who were experienced in reading DNA blots recorded remarkably similar results, while less satisfactory results were obtained when the blots were read by an inexperienced observer. This technique will be useful in laboratories in which radioactive isotopes are unavailable or impractical and in which budgets are restricted.

Role of plasmid-encoded adherence factors in adhesion of enteropathogenic Escherichia coli to HEp-2 cells.

Plasmid-encoded adherence factors have been shown to be important for the full expression of enteropathogenic Escherichia coli (EPEC) pathogenicity and for EPEC adhesion to cultured HEp-2 cells. EPEC strain E2348 (O127) shows localized HEp-2 cell adhesion and possesses a 60-megadalton plasmid, pMAR2. When E2348 is cured of pMAR2 it loses the ability to adhere to HEp-2 cells, while nonadherent E. coli K-12 strains P678-54 and HB101 acquire HEp-2 adhesiveness after they gain the plasmid. By electron microscopy, E2348 was seen to adhere to HEp-2 cells in a manner that closely resembled EPEC adhesion to intestinal mucosa; bacteria were intimately attached to projections of the apical HEp-2 cell membrane and caused localized destruction of microvilli. The plasmid-containing K-12 strains, on the other hand, did not show intimate attachment and there was no modification of cell surface architecture. It is concluded that plasmid pMAR2 codes for an adhesin, possibly fimbrial in nature, that promotes HEp-2 adhesion but that other chromosomally encoded factors are required for EPEC to achieve the characteristic mode of intimate cell attachment.

Localization of a determinant for HEp-2 adherence by enteropathogenic Escherichia coli.

pMAR2, a 60-megadalton plasmid encoding localized HEp-2 adherence in enteropathogenic Escherichia coli, was mapped with BamHI, HindIII, and SalI. Deletion and insertion mutants were constructed and used to define a potential DNA probe. Preliminary results indicate that this probe is sensitive and specific for the genes encoding the enteropathogenic E. coli adherence factor.

The diarrheal response of humans to some classic serotypes of enteropathogenic Escherichia coli is dependent on a plasmid encoding an enteroadhesiveness factor.

Isolates of the most common O serogroups of enteropathogenic Escherichia coli (EPEC) associated with infant diarrhea (designated class I) adhere to Hep-2 cells; the genes for this adhesin, termed EPEC adherence factor (EAF), are located on plasmids 50-70 MDa in size. Volunteers ingested 10(10) organisms of an O127:H6 Hep-2-adhesive class I strain (E2348/69) or its plasmid-minus, nonadhesive derivative. Diarrhea occurred in nine of 10 volunteers who ingested the parent strain (mean, 1,178 ml) but in only two of nine who took the plasmid-minus variant (mean, 433 ml; P less than .006). All volunteers ill from strain E2348/69 mounted serum IgA and IgG responses to a 94-kDa plasmid-associated outer membrane protein of E2348/69; this protein was found in other class I EPEC but not in enterotoxigenic or meningitic strains. The 50-70-MDa EAF plasmid seems necessary for full expression of pathogenicity in EPEC that exhibit Hep-2 adhesiveness. EPEC isolates of certain other, less common, O serogroups (O44, O86, and O114) are rarely Hep-2 adhesive. These EPEC, designated class II, possess distinct 50-70 MDa plasmids lacking EAF genes. Diarrhea was caused by 10(8) or 10(10) organisms of an O114:H2 class II EPEC strain (mean, 1,156 ml) in six of 11 volunteers. This result confirmed that class II EPEC are pathogenic by a mechanism not involving Hep-2 adhesiveness.

Detection of an adherence factor of enteropathogenic Escherichia coli with a DNA probe.

A DNA probe to detect genes conferring localized adherence of enteropathogenic Escherichia coli (EPEC) to Hep-2 cells was evaluated by using E. coli isolates from the stools of Peruvian infants with and without diarrhea. The probe was both sensitive and specific and revealed that Hep-2 adherence (because of the EPEC adherence factor [EAF] was more frequent in some O serogroups of EPEC than in others. Those serogroups in which EAF is almost always found have been designated class I EPEC; serogroups in which EAF is rarely found have been designated class II. Both class I (EAF-positive) and class II EPEC are associated with diarrheal disease.

Cloning of the thermostable direct or Kanagawa phenomenon-associated hemolysin of Vibrio parahaemolyticus.

Genes encoding the thermostable direct or Kanagawa phenomenon hemolysin of Vibrio parahaemolyticus were cloned in Escherichia coli. DNA hybridization experiments with the cloned genes showed that none of the five Kanagawa phenomenon-negative environmental isolates tested possessed DNA sequences homologous to the probe.

Recombinant nontoxinogenic Vibrio cholerae strains as attenuated cholera vaccine candidates.

An ideal vaccine does not yet exist to prevent cholera, a significant health problem in many less developed countries. Vibrio cholerae, the agent of epidemic and endemic cholera, colonizes the small bowel and secretes a potent enterotoxin that consists of a single A subunit, which stimulates adenylate cyclase activity, and five identical B subunits which bind to the ganglioside GM1 receptor of intestinal mucosal cells. Previous studies in man indicate that toxoid-derived antitoxic immunity by itself is insufficient to provide effective, long-lasting protection against cholera. Using recombinant DNA techniques we have now constructed a live, attenuated V. cholerae strain by deleting genes encoding the enterotoxin. Restriction enzyme fragments encoding cholera toxin were deleted in vitro from cloned vibrio chromosomal DNA and the resulting mutations introduced into the chromosome of a vibrio strain of proven immunogenicity. Recently, Mekalanos and coworkers have reported attenuated V. cholerae strains constructed by similar methods. It appears that recombinant DNA techniques offer a promising approach to the development of effective cholera vaccines.

Plasmid-mediated adhesion in enteropathogenic Escherichia coli.

A survey of classical serotype enteropathogenic Escherichia coli has been made with respect to their plasmid profile and ability to adhere to HEp-2 cells. Thirty-one of the 32 strains examined contained a 50-70 Md plasmid, and many exhibited HEp-2 adherence. Strain E2348 (0127:H6), which causes diarrhea in volunteers and is HEp-2-adhesive, was chosen for further study. The large 55 Md plasmid in E2348, pMAR2, has been marked with a transposon coding for ampicillin resistance. E2348 that has been cured of pMAR2 loses the ability to adhere to HEp-2 cells, while HB101, a nonadherent E. coli K12, acquires HEp-2 adhesiveness after gaining this plasmid. Plasmid presence was also shown to correlate with in vivo adhesion to intestine, using the colostrum-deprived piglet model.

Ambient-temperature primary nonselective enrichment for isolation of Salmonella spp. from an estuarine environment.

A primary, nonselective, ambient-temperature enrichment procedure for isolation of Salmonella spp. is described. The procedure was superior to elevated-temperature selective enrichment for Salmonella when estuarine water samples were examined. Five Chesapeake Bay stations were monitored, over an 8-month period, for the presence of salmonellae. Of 72 water and sediment samples collected, 17 (23.6%) yielded Salmonella spp. Seven serotypes were identified among the isolates. A seasonal pattern was noted for the incidence of the salmonellae. A most probable number procedure, performed by membrane filtration and nonselective enrichment, yielded Salmonella most probable number indices as high as 110 per 100 g of sediment. The results suggest that new methods, such as the one described in this report, are required for isolation of human intestinal pathogens from estuaries and coastal waters.