Background: Intravenous administration of Concanavalin (Con) A leads to an immune-mediated hepatitis in mice. Con A hepatitis involves the activation of T cells and Kupffer cells, which produce the cytokines interferon (IFN)-γ and tumour necrosis factor (TNF)-α. The lipopolysaccharide (LPS)/Toll-like receptor (TLR) 4 signalling pathway is involved in several forms of liver injury. Furthermore, several forms of experimental liver injury involve increased gut permeability. Con A-induced hepatitis is attenuated in mice deficient in TLR4. In the current study, the effect of depleting gut bacteria on Con A-induced hepatitis was investigated. The effect of Con A on gut permeability was also examined. Methods: Groups of mice (C57BL/6) were treated for 5 days with Polymyxin B (150 mg/kg/day) and Neomycin (450/mg/kg/day) via drinking water to deplete Gram-negative bacteria in the gut. Control mice were given normal drinking water. Administration of Con A (20 mg/kg i.v) followed antibiotic treatment. Blood and tissue were harvested 8 or 24 hours later. Liver injury was assessed by measuring plasma ALT and histologically. In a separate group of mice, FITC-dextran was administered by oral gavage prior to Con A administration and its subsequent appearance in plasma was measured to assess gastrointestinal tract permeability. Permeability was further examined using an Ussing chamber to measure transepithelial resistance. Results: Control mice developed severe liver injury following Con A administration, indicated by elevated plasma ALT and confirmed by histology. However, mice treated with antibiotics prior to Con A administration showed no significant elevation in plasma ALT at either 8 or 24 hours and reduced injury was seen on histological inspection. Mice treated with Con A displayed increased gut permeability, indicated by an increase in plasma FITC-dextran levels compared control (1578 ±370 ng/mL in Con A treated vs. 160 ± 26 ng/mL in control mice). Furthermore, Con A treated mice displayed a 20% reduction in transepithelial resistance compared to control mice, which is consistent with increased permeability. Discussion: Our results show that Con A causes liver injury in mice with a normal bacterial population resident in the gastrointestinal tract. However, following the depletion of Gram-negative bacteria, Con A-induced injury is attenuated. Con A also increased gut permeability, indicating a disruption in the barrier function of the gastrointestinal tract. Together, these results suggest that the mechanism for Con A-induced liver injury involves increased translocation of LPS from the gastrointestinal tract by an as yet unknown mechanism and that the liver injury is dependent on activation of the LPS/TLR4 pathway.