Heme oxygenase (HO), the rate limiting enzyme in the degradation of heme to bilirubin, has been speculated to serve in antioxidant defense. We have previously shown that a small decrease in HO-1 expression could decrease resistance to oxygen toxicity in previously oxygen resistant cells(1) and have more recently shown that HO-2 -/- mutant mice had increased susceptibility to oxygen toxicity compared with their wild type (WT) counterparts(2). We now wanted to determine whether partial or complete elimination of HO-1 in a mouse model would also be associated with decreased resistance to hyperoxia (>95% O2). We therefore exposed HO-1 -/- and HO-1 +/- mutant C57B1/6 mice to hyperoxia for 72 hours along with WT from the same strain. Lung HO mRNA, protein and activity in lung tissues were evaluated as well as lung protein oxidation and lipid peroxidation as markers of oxidative injury. To better understand the mechanism by which HO modulates antioxidant defenses, we also examined lung reactive iron, total iron, heme and ferritin content in all animals before and after hyperoxic exposure. Prior to hyperoxic exposure, all animals had similar lung HO activity since the HO-1 -/- and +/- had increased HO-2 mRNA and protein levels compared to WT. However, in hyperoxia, the HO-1 -/- and HO-1+/- had no increase in HO activity since HO-2 was now diminished and no HO-1 induction was noted, whereas the WT showed a 3 fold increase in HO activity due to HO-1 induction. As to markers of oxidative injury, surprisingly, the HO-1 mutants had significantly decreased lung protein oxidation and lipid peroxidation compared to the WT after hyperoxic exposure and showed no decreased survival in chronic hyperoxia. Although total lung iron was similar in all groups, reactive lung iron was significantly lower in the WT prior to hyperoxic exposure but increased 2 fold after hyperoxia whereas the HO-1 +/- had lowered reactive iron levels and the HO-1 -/- showed no differences in iron content after hyperoxia. These paradoxical observations suggest that increased HO-1 expression may not serve to protect against acute hyperoxic injury and that increased available reactive iron from the HO reaction may exacerbate hyperoxic lung injury. We speculate that constitutive HO expression is protective against oxygen toxicity but that hyperoxic induction of HO-1 may increase availability of redox-active iron and thereby worsen acute oxygen toxicity.