Heme oxygenase-1-mediated neuroprotection in subarachnoid hemorrhage via intracerebroventricular deferoxamine

Abstract

Background: Subarachnoid hemorrhage (SAH) is a devastating disease that affects over 30,000 Americans per year. Previous animal studies have explored the therapeutic effects of deferoxamine (DFX) via its iron-chelating properties after SAH, but none have assessed the necessity of microglial/macrophage heme oxygenase-1 (HO-1 or Hmox1) in DFX neuroprotection, nor has the efficacy of an intracerebroventricular (ICV) administration route been fully examined. We explored the therapeutic efficacy of systemic and ICV DFX in a SAH mouse model and its effect on microglial/macrophage HO-1. Methods: Wild-type (WT) mice were split into the following treatment groups: SAH sham+vehicle, SAH+vehicle, SAH+intraperitoneal (IP) DFX, and SAH+ICV DFX. For each experimental group, neuronal damage, cognitive outcome, vasospasm, cerebral and hematogenous myeloid cell populations, cerebral IL-6 concentration, and mitochondrial superoxide anion production were measured. HO-1 co-localization to microglia was measured using confocal images. Trans-wells with WT or HO-1-/- microglia and hippocampal neurons were treated with vehicle, red blood cells (RBCs), or RBCs with DFX; neuronal damage, TNF-α concentration, and microglial HO-1 expression were measured. HO-1 conditional knockouts were used to study myeloid, neuronal, and astrocyte HO-1 involvement in DFX-induced neuroprotection and cognitive recovery. Results: DFX treatment after SAH decreased cortical damage and improved cognitive outcome after SAH yet had no effect on vasospasm; ICV DFX was most neuroprotective. ICV DFX treatment after SAH decreased cerebral IL-6 concentration and trended towards decreased mitochondrial superoxide anion production. ICV DFX treatment after SAH effected an increase in HO-1 co-localization to microglia. DFX treatment of WT microglia with RBCs in the trans-wells showed decreased neuronal damage; this effect was abolished in HO-1-/- microglia. ICV DFX after SAH decreased neuronal damage and improved cognition in Hmox1 fl/fl control and Nes Cre :Hmox1 fl/fl mice, but not LyzM Cre :Hmox1 fl/fl mice. Conclusions: DFX neuroprotection is independent of vasospasm. ICV DFX treatment provides superior neuroprotection in a mouse model of SAH. Mechanisms of DFX neuroprotection after SAH may involve microglial/macrophage HO-1 expression. Monitoring patient HO-1 expression during DFX treatment for hemorrhagic stroke may help clinicians identify patients that are more likely to respond to treatment.