Alam, Jawed

The mechanism of heme oxygenase-1 gene ( ho-1) activation by heme in immortalized rat proximal tubular epithelial cells was examined. Analysis of the ho-1promoter identified the heme-responsive sequences as the stress-response element (StRE), multiple copies of which are present in two enhancer regions, E1 and E2. Electrophoretic mobility shift assays identified Nrf2, MafG, ATF3, and Jun and Fos family members as StRE-binding proteins; binding of Nrf2, MafG, and ATF3 was increased in response to heme. Dominant-negative mutants of Nrf2 and Maf, but not of c-Fos and c-Jun, inhibited basal and heme-induced expression of an E1-controlled luciferase gene. Heme did not affect the transcription activity of Nrf2, dimerization between Nrf2 and MafG, or the level of MafG, but did stimulate expression of Nrf2. Heme did not influence the level of Nrf2 mRNA but increased the half-life of Nrf2 protein from ∼10 min to nearly 110 min. These results indicate that heme promotes stabilization of Nrf2, leading to accumulation of Nrf2 · MafG dimers that bind to StREs to activate the ho-1 gene.

This study examined the effect of hemin on the expression of heme oxygenase-1 (HO-1) and monocyte chemoattractant protein-1 (MCP-1) in immortalized rat proximal tubular epithelial cells (IRPTCs). Hemin elicited a dose- and time-dependent induction of HO-1 and MCP-1 mRNA. HO activity contributed to MCP-1 mRNA expression at early time points (4–6 h) because inhibition of HO activity by zinc protoporphyrin (ZnPP) prevented hemin-induced expression of MCP-1 mRNA. Catalytically active intracellular iron was markedly increased in hemin-treated IRPTCs and contributed to the induction of HO-1 and MCP-1 mRNA because an iron chelator blocked hemin-induced upregulation of both genes, whereas a cell-permeant form of iron directly induced these genes. N-acetylcysteine completely blocked hemin-induced expression of HO-1 and MCP-1 mRNA, thereby providing added evidence for redox regulation of expression of these genes. The redox-sensitive transcription factor NF-κB was recruited in hemin-induced upregulation of MCP-1 because two different compounds that abrogate the activation of NF-κB (TPCK and BAY 11-7082) completely blocked hemin-induced upregulation of MCP-1 mRNA. In contrast to this HO-mediated induction of MCP-1 through redox-sensitive, iron-dependent, and NF-κB-involved pathways observed after 4–6 h, hemin also elicited a delayed induction of MCP-1 at 18 h through HO-independent pathways. We conclude that hemin is a potent inducer of MCP-1 in IRPTCs: HO-dependent, heme-degrading pathways lead to an early, robust, and self-remitting induction of MCP-1, whereas HO-independent mechanisms lead to a delayed expression of MCP-1.

The current study was designed to determine the effect of recombinant heme oxygenase-1 (HO-1) gene expression on endothelial function in cerebral arteries. Isolated canine basilar arteries were exposed ex vivo (30 minutes at 37°C) to an adenoviral vector (10 10 PFU/mL, total volume 300 μL) encoding either the HO-1 gene (AdCMVHO-1) or the β-galactosidase (β-Gal) reporter gene (AdCMVβ-Gal). Twenty-four hours after transduction, arterial rings were suspended in organ chamber for isometric force recording. Endothelium-dependent relaxations were obtained in response to bradykinin (10 −10 to 10 −6 mol/L) during contraction to uridine-5′-triphosphate (UTP; 3 × 10 −6 to 3 × 10 −5 mol/L). Certain rings were incubated with oxyhemoglobin (OxyHb; 10 −5 mol/L) overnight (16 to 18 hours of 24 hours). Expression and localization of recombinant protein were shown by Western blot analysis and immunohistochemistry. Endothelium-dependent relaxation to bradykinin and endothelium-independent relaxation to forskolin (10 −9 to 10 −5 mol/L) and DEA-NONOate (10 −10 to 10 −5 mol/L) were identical in β-Gal– and HO-1–transduced arteries. Exposure to OxyHb caused impairment of endothelium-dependent relaxation to bradykinin ( P < 0.01). In contrast, OxyHb did not affect endothelium-dependent relaxation in arteries expressing recombinant HO-1 ( P > 0.05). This protective effect of HO-1 was reversed by coincubation with tin protoporphyrin (SnPP9; 10 −5 mol/L), a selective inhibitor of HO-1 ( P < 0.01). Basal levels of 3′,5′-cyclic monophosphate (cGMP) in HO-1–transduced vessels were not significantly different from those in β-Gal–transduced vessels. Pretreatment with OxyHb significantly reduced cGMP level in β-Gal–transduced rings ( P < 0.01), whereas it had no effect in HO-1–transduced rings. These results demonstrate that HO-1 gene transfer does not affect endothelial and smooth muscle function of normal arteries, and that expression of recombinant HO-1 in cerebral arteries protects vasomotor function against OxyHb-induced injury.