Association between intraepithelial Escherichia coli and colorectal cancer.

BACKGROUND & AIMS: Although multiple studies have focused on Helicobacter pylori, little is known about the mucosa-associated flora of the colon. The aim of this study was to detect bacteria directly in colonic mucosa from patients screened for colorectal cancer. METHODS: Bacteria were quantified with the polymerase chain reaction and identified by comparative sequence analysis in colonoscopic biopsy specimens from 31 asymptomatic and 34 symptomatic controls with normal colonoscopic findings, 29 patients with colonic adenoma, and 31 patients with colorectal carcinoma. In 41 patients, intra- and extracellular location of bacteria was confirmed with the gentamicin protection assay. RESULTS: No bacteria were detected in biopsy specimens from 97% of asymptomatic and 69% of symptomatic controls. In contrast, bacterial concentrations of 10(3)-10(5) colony-forming units per microliter were detected in biopsy specimens from both malignant and macroscopically normal tissue in 90% and 93% of patients with adenoma and carcinoma, respectively. E. coli and coli-like bacteria were shown to colonize the colonic mucosa in 82% of these patients. The gentamicin protection assay indicated that E. coli was partially intracellular in 87% of patients with adenoma and carcinoma and in none of the controls. CONCLUSIONS: The colonic mucosa of patients with colorectal carcinoma but not normal colonic mucosa is colonized by intracellular E. coli.

Time dependent changes in the concentration and type of bacterial sequences found in cholesterol gallstones.

The role of bacteria in gallstone formation could not be conclusively evaluated until bacterial presence or absence in a stone was consistently shown. Cultural bacteriologic investigations at the time of cholecystectomy, however, led to the assumption that cholesterol gallstones were free of bacteria. In this study, we used a culture independent, molecular genetic approach to detect, quantify, and identify bacteria in cholesterol gallstones from 100 patients at the time of cholecystectomy and 6 months following. Bacterial growth was recorded in the culture in 9 of 100 gallstones; bacterial DNA, however, was detected in 82 of 91 sterile gallstones. High concentrations corresponding to between 10(6) to 10(7) bacteria/g were detected in 11 stones and low concentrations of 10(5) bacteria/g were detected in 71 sterile stones. The infection in stones with a positive bacterial culture was characterized by the predominance of single bacterial sequence(s) of the bacteria cultured. A similar predominance, indicating a recent infection, was found in sterile gallstones with low DNA concentrations. A high diversity of non-repeating bacterial sequences, possibly arising from previous overlapping infections, was found in sterile gallstones with high concentrations of bacterial DNA. After 6 months concentrations of bacterial DNA fell significantly in all groups of gallstones. As bacterial DNA is quickly destroyed upon storage, but is nevertheless readily found in most gallstones at the time of cholecystectomy, there must be a mechanism by which it is replenished. One such mechanism is the frequently reoccurring, possibly self-terminating infection and another one is the permanent colonization of the gallstone with bacteria at low concentrations. Both can promote cholecystolithiasis.