E. coli STa enterotoxin is assumed to cause fluid secretion in the small intestine by stimulating chloride ion and fluid secretion into the lumen. Evidence comes mainly from in vitro studies in which STa causes a bumetanide-inhibitable increase in short-circuit current, the polarity of which is compatible with chloride secretion. For these reasons, the ability of bumetanide to restore fluid and ion transport to normal after STa exposure was investigated in vivo. Fluid absorption was measured in anaesthetised (70 mg kg^-1 I.P. Sagatal) Wistar rats using recirculated perfused loops by a volumetric method rather than measurement of indicator concentration. At the end of the experiments, the animals were humanely killed.

In jejunum perfused with 100 mM sodium bicarbonate, 50 mM sodium chloride buffer, fluid absorption was 107 ± 11 (6) µl cm^-1 h^-1 (mean ± S.E.M., number of animals in parentheses). With STa (80 ng ml^-1 perfusate), fluid absorption was significantly (P < 0.001) reduced to 13 ± 5 (8) µl cm^-1 h^-1. Significance was tested by Student’s t test, modified by Dunnett. Instead of restoration to control values, fluid absorption on perfusion of 0.1 mM bumetanide with STa was 1.0 ± 7 (6) µl cm^-1 h^-1, marginally less than with STa alone. STa reduced (P < 0.0005) net bicarbonate uptake from 20.3 ± 1.1 (6) to 9.8 ± 2.2 (7) µmol cm^-1 h^-1, net sodium ion uptake from 16.2 ± 1.54 (6) to 1.4 ± 1.5 (6) µmol cm^-1 h^-1 and net change in osmolarity from 25.6 ± 3.7 (6) µosmol cm^-1 h^-1 to nil. Net chloride entry into the lumen of normal jejunum was significantly different from zero at 2.5 ± 0.9 (6) µmol cm^-1 h^-1, unlike the value of -0.35 ± 1.0 (6) µmol cm^-1 h^-1 with STa. Bumetanide had no effect on any of these ion movements.

The action of STa on fluid absorption from isotonic and hypotonic saline was investigated to see whether the localised hypotonicity model (Lucas, 2001) was more likely to account for the action of STa than chloride secretion. In the presence of STa, fluid absorption from isotonic saline of 56.1 ± 10.0 (6) µl cm^-1 h^-1 was significantly (P < 0.05) reduced by STa to 28.6 ± 4.5 (7) µl cm^-1 h^-1. In hypotonic saline, fluid absorption of 77.4 ± 17.9 (7) was not significantly lower than the 90.9 ± 30.0 (5) µl cm^-1 h^-1 after STa. If STa causes secretion, then this should have offset the fluid absorption from the hypotonic solution. This inability is unlikely to result from denaturation since prior exposure of enterotoxin to hypotonic solution did not compromise its ability to reduce absorption in isotonic saline. In addition, STa in hypotonic saline still caused the characteristic alkalinisation of the lumen, despite a lack of effect of fluid absorption.

These results are incompatible with chloride-driven fluid secretion; final volumes at the end of the experiment never exceeded initial volumes, no increase in net luminal chloride concentration was detected after STa, bumetanide was completely without effect. These deficiencies and the inability of STa to reduce uptake from hypotonic solutions further support the mucosal hypotonicity model. In this model, fluid absorption is normally effected by the creation of mucosal hypotonicity, which STa prevents. When luminal hypotonicity is provided by an external gradient, STa is without effect.The authors gratefully acknowledge grant support from the Royal Society.

Lucas, M.L. (2001). J Appl. Microbiol. 90, 7-26.