Heme as a Playmaker in the Regulation of the Nitric Oxide System

Abstract

We read with great interest the article by Tsai et al. 1In this article, the authors presented a laboratory investigation in which they showed that heme oxygenase-1 (HO-1) induction significantly inhibits type 2 cationic amino acid transporter expression and l-arginine transport in lipopolysaccharide-stimulated macrophages. The authors further suggested that this effect may be related to the activation of nuclear factor erythroid 2–related factor 2 and inhibition of nuclear factor κB. After we read this analysis, it occurred to us that some points may be added to the discussion.The authors showed that lipopolysaccharide treatment resulted in a significant increase in type 2 cationic amino acid transporter expression and this effect was reversed by concomitant treatment with hemin (fig. 1). However, there are no data indicating the effect of hemin treatment on nitric oxide formation. These set of experiments could have rendered the authors’ conclusions stronger; in fact, heme may act as a pro-oxidant molecule, thus leading to an increased expression of the inducible isoform of nitric oxide synthase, which in turn leads to increased nitric oxide production. In this case, hemin, although resulting in a significant decrease in type 2 cationic amino acid transporter expression and activity, may still induce the release of nitric oxide. In addition, heme serves as prosthetic group of inducible isoform of nitric oxide synthase, and thus heme treatment may result in an increased synthesis of the enzyme. Different HO-1 inducers, such as SnCl2or cobalt-protoporphyrin, could have added more information because they potently induce HO-1 without increasing intracellular heme levels. In this regard, we and other authors previously showed that HO-1 induction by using cobalt-protoporphyrin or gene targeting modulates intracellular heme level, thus regulating the synthesis of heme-dependent proteins such as nitric oxide synthases, cyclooxygenases, nicotinamide adenine dinucleotide phosphate oxidase, and cytochrome P-450.2,3These observations may be consistent with previous work performed by the same authors4showing that propofol treatment resulted in a concomitant reduction of both the inducible isoform of nitric oxide synthase and type 2 cationic amino acid transporter expression. In this regard, we also showed that propofol may act as an inducer of HO-1 via activation of the nuclear factor-κB pathway.5Another point that we believe needs to be raised is in regard to the authors’ choice of adding hemin immediately after lipopolysaccharide stimulation, thus not permitting a strong preinduction of HO-1 activity, which would have allowed increased carbon monoxide levels and a reduction of the intracellular heme pool. Interestingly, the authors also showed that tin protoporphyrin, a strong inhibitor of HO activity, results in a significant increase of HO-1 protein (even though in the Results section it was indicated that tin protoporphyrin did not increase protein expression) and partial reversion of hemin effects. The molecular mechanism underlying this effect is still unclear, and several hypotheses may be carried out. One is that HO activity inhibition after tin protoporphyrin treatment results in increased intracellular heme level after strong HO activity inhibition, thus leading to increased HO-1 expression. The second hypothesis is based on the fact that because tin protoporphyrin possesses a structure similar to that of heme, it may directly induce HO-1 expression, probably by binding to specific heme binding motifs and activating the gene transcription. Moreover, the authors’ observations further confirm that the activity of the protein rather than its expression or intracellular localization is required for its beneficial effects. We also agree with the authors regarding their conclusions about carbon monoxide. In fact, further studies using carbon monoxide–releasing molecules or carbon monoxide gas are required to confirm a possible interaction between carbon monoxide, nuclear factor erythroid 2–related factor 2, and nuclear factor κB under these experimental conditions. In this regard, further studies should also be performed to exclude a possible involvement of biliverdin, the other byproduct of HO activity, which has also been shown to impact on inducible nitric oxide synthase expression and activity.6,7In conclusion, the authors’ observations are novel and intriguing and add a new, important piece in the complicated puzzle of the interaction between the HO–carbon monoxide and nitric oxide–nitric oxide synthase systems.*University of Catania, Catania, Italy. livolti@unict.it