Invasiveness as a putative additional virulence mechanism of some atypical Enteropathogenic Escherichia coli strains with different uncommon intimin types

Abstract

Background: Enteropathogenic Escherichia coli (EPEC) produce attaching/effacing (A/E) lesions on eukaryotic cells mediated by the outer membrane adhesin intimin. EPEC are sub-grouped into typical (tEPEC) and atypical (aEPEC). We have recently demonstrated that aEPEC strain 1551-2 (serotype O non-typable, non-motile) invades HeLa cells by a process dependent on the expression of intimin sub-type omicron. In this study, we evaluated whether aEPEC strains expressing other intimin sub-types are also invasive using the quantitative gentamicin protection assay. We also evaluated whether aEPEC invade differentiated intestinal T84 cells.

Results: Five of six strains invaded HeLa and T84 cells in a range of 13.3%-20.9% and 5.8%-17.8%, respectively, of the total cell-associated bacteria. The strains studied were significantly more invasive than prototype tEPEC strain E2348/69 (1.4% and 0.5% in HeLa and T84 cells, respectively). Invasiveness was confirmed by transmission electron microscopy. We also showed that invasion of HeLa cells by aEPEC 1551-2 depended on actin filaments, but not on microtubules. In addition, disruption of tight junctions enhanced its invasion efficiency in T84 cells, suggesting preferential invasion via a non-differentiated surface.

Conclusion: Some aEPEC strains may invade intestinal cells in vitro with varying efficiencies and independently of the intimin sub-type.

Figure 1

Invasion of epithelial cells by aEPEC and tEPEC strains. A) Percent of invasion in HeLa cells. B) Percent of invasion in T84 cells. Monolayers were infected for 6 h (aEPEC) and 3 h (tEPEC). Results of percent invasion are expressed as the percentage of cell associated bacteria that resisted killing by gentamicin and are the means ± standard error from at least three independent experiments in duplicate wells. *significantly more invasive than prototype tEPEC E2348/69 (P < 0.05 by an unpaired, two-tailed t test).

Figure 2

Invasion of HeLa (epithelial) cells by aEPEC 1551-2 after treatment with cytoskeleton polymerization inhibitors. A) Cytochalasin D; B) Colchicine. Monolayers were infected for 6 h (aEPEC) and 3 h (tEPEC). S. enterica sv Typhimurium and S. flexneri were used as controls and monolayers were infected for 4 h and 6 h, respectively. Results as percent invasion are means ± standard error from at least three independent experiments performed in duplicate. * P < 0.05 by an unpaired, two-tailed t test.

Figure 3

Transmission electron microscopy of infected polarized and differentiated T84. A) aEPEC 1551-2, B) aEPEC 0621-6 and C) prototype tEPEC E2348/69. Monolayers were infected for 6 h (aEPEC) and 3 h (tEPEC). aEPEC 1551-2 and 0621-6 were selected because, according to the data in Fig. 1B, they presented an average invasion index as compared to the other strains studied. Arrows indicate bacterial-containing vacuoles.

Figure 4

Invasion of differentiated T84 cells by aEPEC 1551-2 after tight junction disruption by EGTA treatment. Monolayers were infected for 6 h (aEPEC) and 3 h (tEPEC). S. enterica sv Typhimurium and S. flexneri were used as controls and monolayers were infected for 4 h and 6 h, respectively. Results of percent invasion are the means ± standard error from at least three independent experiments performed in duplicate. * P < 0.05 by an unpaired, two-tailed t test.

Figure 5

Transmission electron microscopy of polarized and differentiated T84 cells infected via the basolateral side. A) aEPEC 1551-2. B) aEPEC 0621-6. C) prototype tEPEC E2348/69. Monolayers were infected for 6 h (aEPEC) and 3 h (tEPEC). Arrows indicate tight junction and (*) indicates a Transwell membrane pore.