Pharmacological effects of dietary amino acids (AA) and peptides must be compared to an isonitrogenous control that is as inert as possible. To establish a rationale for the choice of such a control, potential metabolic and nutritional effects of three currently used nitrogenous controls (glycine, alanine, and casein) were evaluated in an endotoxemic rat model that has well-defined alterations in AA and protein metabolism. Five-week-old male Sprague-Dawley rats (113 +/- 1 g) were randomly assigned to four groups and received at d 0 an intraperitoneal injection of endotoxin (3 mg/kg). After withdrawal of food for 24 h, the rats were enterally refed for 48 h with a liquid diet (Osmolite((R))) supplemented with 0.19 g N. kg(-1). d(-1) in the form of glycine [lipopolysaccharide (LPS)-GLY group], alanine (LPS-ALA group) or casein (LPS-CAS group). One group (LPS group) received only Osmolite((R)). Plasma, two skeletal muscles, the liver and the intestine were then removed. Body and tissue weights and tissue protein contents did not differ among the four groups. Intestine histomorphometry showed no significant difference among groups. Jejunal hydrolase activities were significantly affected by the nitrogenous supplementations, but no effect was observed in the ileum. Only limited significant effects were observed on plasma and tissue-free AA concentrations, except for an accumulation of glycine in the plasma and tissues from the LPS-GLY group, compared to other groups. Overall, whereas glycine as a nitrogenous control should be used with care, either alanine or casein may be used as the "placebo," with the choice depending on the study to be performed.