Hyperoxic lung injury, believed to be mediated by reactive oxygen species, inflammatory cell activation, and release of cytotoxic cytokines, complicates the care of many critically ill patients. The cytokine tumor necrosis factor (TNF)-α is induced in lungs exposed to high concentrations of oxygen; however, its contribution to hyperoxia-induced lung injury remains unclear. Both TNF-α treatment and blockade with anti-TNF antibodies increased survival in mice exposed to hyperoxia. In the current study, to determine if pulmonary oxygen toxicity is dependent on either of the TNF receptors, type I (TNFR-I) or type II (TNFR-II), TNFR-I or TNFR-II gene- ablated [(-/-)] mice and wild-type control mice (WT; C57BL/6) were studied in >95% oxygen. There was no difference in average length of survival, although early survival was better for TNFR-I(-/-) mice than for either TNFR-II(-/-) or WT mice. At 48 h of hyperoxia, slightly more alveolar septal thickening and peribronchiolar and periarteriolar edema were detected in WT than in TNFR-I(-/-) lungs. By 84 h of oxygen exposure, TNFR-I(-/-) mice demonstrated greater alveolar debris, inflammation, and edema than WT mice. TNFR-I was necessary for induction of cytokine interleukin (IL)-1β, IL-1 receptor antagonist, chemokine macrophage inflammatory protein (MIP)-1β, MIP-2, interferon-χ-induced protein-10 (IP-10), and monocyte chemoattractant protein (MCP)-1 mRNA in response to intratracheal administration of recombinant murine TNF-α. However, IL-1β, IL-6, macrophage migration inhibitory factor, MIP-1α, MIP-2, and MCP-1 mRNAs were comparably induced by hyperoxia in TNFR-I(-/-) and WT lungs. In contrast, mRNA for manganese superoxide dismutase and intercellular adhesion molecule-1 were induced by hyperoxia only in WT mice. Differences in early survival and toxicity suggest that pulmonary oxygen toxicity is in part mediated by TNFR-I. However, induction of specific cytokine and chemokine mRNA and lethality in response to severe hyperoxia was independent of TNFR-I expression. The current study supports the prediction that therapeutic efforts to block TNF-α receptor function will not protect against pulmonary oxygen toxicity.