We have previously shown that glutathione peroxidase (GPx) overexpressing mice (hGPx-tg) have reduced brain injury after neonatal hypoxia-ischemia (HI) as a consequence of reduced hydrogen peroxide accumulation. However, this protection is reversed with hypoxia preconditioning, raising the question of the roles of the genes regulated by hypoxia-inducible factor-1α (HIF-1α) and their transcription products, such as erythropoietin (EPO), in both the initial protection and subsequent reversal of protection. hGPx-tg and their wild-type (WT) littermates underwent the Vannucci procedure of HI brain injury at postnatal day 9 - left carotid artery ligation followed by exposure to 10% oxygen for 50 min. Brain cortices and hippocampi were subsequently collected 0.5, 4 and 24 h later for the determination of protein expression by Western blot for GPx, HIF-1α, HIF-2α, EPO, EPO receptor, ERK1/2, phospho-ERK1/2, spectrin 145/150 (as a marker of calpain-specific necrotic cell death), and spectrin 120 (as a marker of apoptotic cell death mediated via caspase-3). As expected, the GPx overexpressing mouse cortex had approximately 3 times the GPx expression as WT naïve. Also, GPx expression remained higher in the GPx overexpressing brain than WT at all time points after HI (0.5, 4, 24 h). HIF-1α was not significantly changed in hGPx-tg as a consequence of HI but decreased in the WT cortex 4 h after HI. HIF-2α decreased in the WT hippocampus after HI. EPO was higher in the GPx overexpressing cortex and hippocampus 30 min after HI compared to WT, but the EPO receptor was unchanged by HI. ERK1/2 phosphorylation increased in the hippocampus at 4 h after HI and in the cortex at 24 h after HI in both WT and hGPx-tg. Spectrin 145/150 was increased in the WT cortex at 4 and 24 h after HI, and spectrin 120 increased 24 h after HI, perhaps reflecting greater injury in the WT brain, especially at 24 h when brain injury is more evident. The effect of GPx overexpression does not appear to upregulate the HIF pathway, yet EPO was upregulated, perhaps via ERK. This might explain, in part, why cell death takes a necrotic or apoptotic path. This may also be an explanation for why the GPx overexpressing brain cannot be preconditioned. This information may prove valuable in the development of therapies for neonatal HI brain injury.