Adherence of Escherichia coli and Proteus mirabilis to human transitional cells.

This study utilizes a light microscopy assay for bacterial adherence to human male transitional cells. Prior light microscopy studies have used voided squamous cells, periurethral cells or scraped vaginal cells, which are less representative of the cells lining the majority of the urinary tract. Using a modification of previous bacterial adherence assays, the mean adherence for 28 strains of E. coli in 92 bacteria-cell incubations was 10.2 +/- 11.5 (standard deviation) bacteria per cell. The mean adherence for 20 strains of P. mirabilis in 60 bacteria cell incubations was 8.1 +/- 11.4. No statistically significant difference in adherence between E. coli and P. mirabilis was found (p greater than 0.05). Studies comparing the adherence of E. coli isolated from the urine of patients with pyelonephritis (eight strains), cystitis (10 strains) and anal swabs of females without urinary tract infections (10 strains), showed no statistically significant differences in mean adherence (p greater than 0.05). However, there was a trend toward higher adherence in the more virulent groups. Experiments comparing the adherence of P. mirabilis isolated from infected renal stones to P. mirabilis isolated from anal swabs of female patients without history of P. mirabilis UTI revealed no statistically significant differences in mean adherence between the two groups (p greater than 0.05). These data do not support previous contentions that P. mirabilis adhere poorly to human transitional cells. The absence of a significant difference in adherence among strains of E. coli and P. mirabilis that differ in clinical pathogenicity suggests that factors other than adherence contribute to their virulence.