We examined the effect of infusion of lipopolysaccharide (LPS) on serum tumor necrosis factor alpha (TNF alpha) concentration and clinical attitude in 2- to 3-day-old colostrum-fed (CF) and colostrum-deprived (CD) foals. Eleven CF and 8 CD neonatal foals were given a bolus IV infusion of Escherichia coli 055:B5 lipopolysaccharide (0.5 microgram/kg of body weight) in sterile saline (0.9% NaCl) solution. Four CF and 2 CD foals were given saline solution alone. Serum IgG concentration and serum anti-LPS IGG(T) antibody titer were determined for each foal prior to infusion. A depression index was used to score clinical abnormalities. Serum TNF alpha concentration was estimated by use of an in vitro cytotoxicity bioassay that used WEHI 164 clone 13 cells as targets. The cytotoxic serum factor was identified as TNF alpha by immunoprecipitation with caprine antisera raised against the 15 NH2-terminal amino acids of human TNF alpha. Tumor necrosis factor alpha was not detected in any preinfusion serum samples nor in any samples from foals given saline solution alone. Serum TNF alpha concentration increased in all LPS-infused foals and peaked between 60 and 90 minutes after infusion. Serum TNF alpha concentrations, expressed as mean percentage of peak serum TNF alpha concentration, persisted longer in CD foals given LPS than in CF foals given LPS. All LPS-infused foals displayed clinical signs of endotoxemia, but mean depression index scores of the CF and CD foals given LPS were not significantly different at any time. Serum TNF alpha concentrations were correlated with depression index scores in both LPS-infused groups. Mean rectal temperature increased by 1 hour and remained high for 4 hours after infusion in CF foals given LPS. Mean rectal temperature in CD foals given LPS was significantly less than that for CF foals given LPS 1 and 2 hours after infusion and was higher than mean rectal temperature prior to infusion 3 and 4 hours after.