In the course of investigations on anorexia during infection, I found that B6-A(y) mice had significantly increased sensitivity to lipopolysaccharide (LPS)-induced lethality as compared with isogenic B6 mice. I also found that the sensitivity to the lethal effect of LPS dramatically increased in aged mice (age effect), both B6 and B6-A(y). However, the A(y) effect of enhancing sensitivity to LPS-induced lethality was still significant, suggesting that the A(y) effect is independent of age. In the absence of tumor necrosis factor alpha (TNFalpha), the A(y) effect was still significant, suggesting that the A(y) effect is independent of TNFalpha toxicity. A dose of LPS of 100 mug per mouse caused 15% lethality in B6, 65% in B6-A(y) (significantly higher than B6), and 100 % in leptin-deficient B6-ob/ob (significantly higher than B6 and B6-A(y)). The results support the hypothesis that endogenous leptin has a protective role against infection, and that a part of this leptin effect is mediated by alpha-melanocyte-stimulating hormone (alphaMSH). In contrast to the results of simple blockade at the melanocortin 4 receptor (MC4R), B6-A(y) suffered more severe LPS-induced anorexia than did B6; therefore, the pathway involving MC4R is not absolutely required for the LPS-induced anorexia, and the presence of pathways involving other melanocortin receptor types was suggested. Because alphaMSH is suggested to be an endogenous anti-inflammatory peptide, and because melanocortin 1 receptor (MC1R) is expressed in various cutaneous cell types, the A(y) effect might be caused via the pathway involving MC1R. Physiologic significance of alphaMSH-MC1R interaction in host defense against infection is discussed.