Haemolytic uraemic syndrome caused by Shiga toxin-producing E. coli (STEC) is dependent on release of Shiga toxins (Stxs) during intestinal infection and subsequent absorption into the bloodstream. An understanding of Stx-related events in the human gut is limited due to lack of suitable experimental models. In this study, we have used a vertical diffusion chamber system with polarized human colon carcinoma cells to simulate the microaerobic (MA) environment in the human intestine and investigate its influence on Stx release and translocation during STEC O157:H7 and O104:H4 infection. Stx2 was the major toxin type released during infection. Whereas microaerobiosis significantly reduced bacterial growth as well as Stx production and release into the medium, Stx translocation across the epithelial monolayer was enhanced under MA versus aerobic conditions. Increased Stx transport was dependent on STEC infection and occurred via a transcellular pathway other than macropinocytosis. While MA conditions had a similar general effect on Stx release and absorption during infection with STEC O157:H7 and O104:H4, both serotypes showed considerable differences in colonization, Stx production, and Stx translocation which suggest alternative virulence strategies. Taken together, our study suggests that the MA environment in the human colon may modulate Stx-related events and enhance Stx absorption during STEC infection.

Hemolytic uremic syndrome (eHUS) is a severe complication of human infections with Shiga toxins (Stxs)-producing Escherichia coli. A key step in the pathogenesis of eHUS is the interaction of Stxs with blood components before the targeting of renal endothelial cells. Here, we show that a single proteolytic cleavage in the Stx2a A-subunit, resulting into two fragments (A1 and A2) linked by a disulfide bridge (cleaved Stx2a), dictates different binding abilities. Uncleaved Stx2a was confirmed to bind to human neutrophils and to trigger leukocyte/platelet aggregate formation, whereas cleaved Stx2a was ineffective. Conversely, binding of complement factor H was confirmed for cleaved Stx2a and not for uncleaved Stx2a. It is worth noting that uncleaved and cleaved Stx2a showed no differences in cytotoxicity for Vero cells or Raji cells, structural conformation, and contaminating endotoxin. These results have been obtained by comparing two Stx2a batches, purified in different laboratories by using different protocols, termed Stx2a(cl; cleaved toxin, Innsbruck) and Stx2a(uncl; uncleaved toxin, Bologna). Stx2a(uncl) behaved as Stx2a(cl) after mild trypsin treatment. In this light, previous controversial results obtained with purified Stx2a has to be critically re-evaluated; furthermore, characterisation of the structure of circulating Stx2a is mandatory to understand eHUS-pathogenesis and to develop therapeutic approaches.