C57BL/6J mice given low doses of lipopolysaccharide (LPS) (100 ng per mouse) plus D-galactosamine (8 mg per mouse) die within 24 h following LPS administration. We used this septic shock model to confirm the role of tumor necrosis factor in mortality using a monoclonal antibody to tumor necrosis factor to prevent lethality. Furthermore, we demonstrated that interleukin 6, rather than playing a lethal role, protected mice against death in this septic shock model. Antibody to interleukin 6 did not affect the fatal outcome in this low-LPS-dose model. However, pretreatment with antibody to tumor necrosis factor did protect the mice against death, in a dose-dependent manner. Moreover, mortality was enhanced by pretreatment with antibody to interleukin 6 when tumor necrosis factor was partly limited by anti-tumor necrosis factor treatment. Mortality was significantly reduced by pretreatment with both recombinant interleukin 6 and low doses of antibody to tumor necrosis factor; in the absence of the low dose of antibody to tumor necrosis factor, interleukin 6 alone did not confer any protection. These data demonstrate in vivo antagonistic activities of tumor necrosis factor and interleukin 6 and show that interleukin 6 can play a protective role against death from septic shock.