We tested the hypothesis that laminarin (LAM), a beta (1-3) polysaccharide extracted from brown algae, can modulate the response to a systemic inflammation. Male Wistar rats (n=7 per group) were fed a standard diet (control) or a diet supplemented with LAM for 25 days (5% during 4 days followed by 10% during 21 days). Thereafter, Escherichia coli lipopolysaccharides (LPS; 10 mg/kg i.p.) were injected and the animals were sacrificed 24 h after LPS challenge. The hypothermia, hyperglycemia and hypertriglyceridemia occurring early after LPS administration were less pronounced in LAM-treated rats than in controls. The increase in serum alanine aminotransferase (ALT), aspartate aminotransferase (AST) and lactate dehydrogenase (LDH) activities - reflecting hepatic alterations - was lessened after LPS injection in LAM-treated rats compared to control rats. LAM treatment decreased serum monocytes number, nitrite (NO2) and tumor necrosis factor-alpha (TNF-alpha). LAM also modulated intra-hepatic immune cells: it lowered the occurrence of peroxidase-positive cells (corresponding to monocytes/neutrophils) and, in contrast, it increased the number of ED2-positive cells, corresponding to resident hepatic macrophages, i.e. Kupffer cells. In conclusion, the hepatoprotective effect of marine beta (1-3) glucan during endotoxic shock may be linked to its immunomodulatory properties. We propose that both lower recruitment of inflammatory cells inside the liver tissue and lower secretion of inflammatory mediators play a role in the tissue protective effect of LAM. These effects could be due to a direct effect of beta-glucan on immune cells, or to an indirect effect through their dietary fibre properties (fermentation in the gut).