To examine the role of carbon monoxide (CO) as a putative neuronal messenger and regulator of cGMP level in vivo, we exploited an animal model to increase brain capability to generate CO. The sole source of CO in mammalian systems is the α-meso carbon bridge of the heme molecule cleaved by heme oxygenase isozymes, HO-1 and HO-2. In adult animals, the noninducible isozyme HO-2 is the predominant form in the brain. We chose to increase, rather than inhibit, brain heme oxygenase activity because synthetic metalloporphyrins, such as Zn-protoporphyrin, which are the only known effective inhibitors of the isozymes, are also potent inhibitors of soluble guanylate cyclase, the enzyme that generates cGMP. In newborn rats both heme oxygenase isozymes were found expressed at low levels, and in the cerebellum heme oxygenase activity could be induced by treatment of 2-day-old animals with a selective depletor of glutathione, buthionine-SR-sulfoximine. The increase in activity was accompanied by marked increases in HO-1 protein and the 1.8 kb HO-1 mRNA in the cerebellum. Despite a pronounced decrease in activity of the hemoprotein nitric oxide synthase, no change in cGMP level was observed. The decrease in the synthase could not be explained by an inhibited heme biosynthesis activity. This unchanged level of cGMP suggests that NO is not the only gaseous heme ligand that can activate guanylate cyclase resulting in the generation of cGMP, but rather that CO may also function in this capacity. Increased capability of select cerebellar cell populations to generate CO, as indicated by an increase in their HO-1 protein content, points to the active role of this isozyme in maintenance of cGMP level under stress conditions, when nitric oxide production is compromised. The cell populations expressing HO-1 protein included those in pia matter and glia, such as astrocytes.

ABSTRACT: The ability of synthetic metalloporphyrins to suppress heme oxygenase activity and bilirubin formation has recently become of considerable clinical and experimental interest for suppression of jaundice in humans, including neonatal hyperbilirubinemia. The present investigation compares the biochemical effects of Sn- and Znprotoporphyrins on the predominant heme oxygenase iso-zyme present in the brain (HO-2) at activity, protein, and transcript levels and describes the ability of Sn-protopor-phyrin to adversely affect this isozyme. Specifically, 6 h after a modest dose (50 μmol/kg, i.v.) of Sn-protoporphyrin, heme oxygenase activity in rat brain was nearly undetectable. In addition, as revealed by Western blot analysis, HO-2 protein level was decreased by 20% and the electrophoretic behavior of the protein in the microsomal membranes was altered. Moreover, the activity of NADPH-cytochrome P-450 reductase, which is required for the oxidation of heme molecule, was markedly decreased (60% of control). Western immunoblot analysis revealed also a pronounced decrease in the reductase protein level. The inducible form of heme oxygenase, HO-1, was not detectable by immunoblotting in brain microsomes of either control or Sn-protoporphyrin-treated animals. Northern blot analyses did not reveal decreases in the levels of the single HO-1 mRNA (1.8 kb) or the two HO-2 transcripts (1.3 and 1.9 kb), suggesting that Sn-protoporphyrin mediates its effects on heme oxygenase isozymes at the protein level. Zn-protoporphyrin, on the other hand, had no deleterious effect on brain parameters presently investigated. The data suggest that Sn-protoporphyrin exerts its inhibitory effects on heme oxygenase activity in the brain through multiple mechanisms involving direct effect on HO-2 protein and its cellular level, inhibiting NADPH-cytochrome P-450 reductase activity, decreasing the cellular content of the reductase, as well as inhibiting heme oxygenase activity by serving as a competitive inhibitor. We suggest that Zn-protoporphyrin may be the safer alternative if metalloporphyrins are to be used to control hyperbilirubinemia.