Mucosa-associated but not luminal Escherichia coli is augmented in Crohn's disease and ulcerative colitis

Abstract

Background: Escherichia coli is believed to participate in the etiology of Crohn's disease (CD) and possibly of ulcerative colitis (UC), due at least in part to the observed rise in the number of these bacteria in the gut microbiota of CD and UC patients. Nevertheless, it is not fully understood whether this quantitative variation occurs equally throughout the mucosal and luminal spaces of the gut. To assess this question, stools and mucosa biopsies from distinct intestinal sites were cultured aiming at determining their E. coli concentration. The cultures were additionally screened for the presence of some virulence genes of pathogenic E. coli.Results: Analyses of clinical materials from 14 controls (38 biopsies and 14 stools samples), 11 CD (25 biopsies and 11 stools samples) and 7 UC patients (18 biopsies and 7 stools samples) indicated no significant variation in the number of E. coli present in stools, but a rise of at least one log(10) CFU/mg in biopsies from the ileum of CD patients and the sigmoid and rectum of CD and UC patients. The cultures were screened for the presence of E. coli attaching and effacing (eae), invasion plasmid antigen H (ipaH), aggregative adherence transcriptional activator (aggR), Shiga cytotoxins (stx), and heat labile enterotoxin (elt) and the following serine proteases autotransporters of Enterobacteriaceae (SPATE) genes: plasmid encoded toxin (pet), secreted autotransporter toxin (sat), Shigella extracellular protein (sepA), protein involved in intestinal colonization (pic) and Shigella IgA-like protease homolog (sigA). Six of the 10 genes screened were detected in the total of samples investigated: aggR, eae, pet, sat, sepA and sigA. No difference in the prevalence of any of these markers was observed in cultures from different clinical materials or groups of patients.Methods: Bacterial quantitation was carried out following cultures of diluted samples suspensions in MacConkey agar, Wilkins Chalgren agar for anaerobes, E. coli/coliform chromocult agar, and blood agar. Screening for E. coli virulence genes was performed by multiplex PCR of DNA purified from total MacConkey undiluted broth cultures.Conclusion: In CD and UC patients only the mucosa associated population of E. coli is augmented and the proliferation is prominent in the ileum of CD and rectum and sigmoid of both UC and CD patients which are sites where the lesions usually are observed. The augmented E. coli population in these sites presented a low number of the virulence markers, possibly meaning that they are not relevant for the disease process.