Carbon monoxide: vasoconstrictor or vasodilator? That's the question.

A STUDY BY BOTROS ET AL. (3) provides evidence that induction of renal heme oxygenase-1 (HO-1) attenuates afferent arterioles (Aff-Art) constrictor responses to increased renal perfusion pressure (RPP), thus suggesting that increased HO-1 modulates renal autoregulation in pathophysiological conditions. The authors elegantly made their conclusions by first showing the HO-1 pattern distribution following heme or SnCl2 treatment. In particular, the authors demonstrated that in hemin-treated rats HO-1 was expressed in tubular epithelial cells with prominent staining of proximal tubules, whereas in SnCl2-treated rats HO-1 was detected in medullary tubular epithelial cells and few proximal tubular cells. Interestingly, HO-1 was not detected in interlobular arteries or Aff-Art following heme or SnCl2 treatment (Fig. 1). This first set of experiments is consistent with previous studies suggesting the possibility of pharmacologically regulating HO-1 expression and HO activity in a site-specific manner along different nephron segments (1, 18). The authors then examined the physiological significance of such upregulation on Aff-Art constrictor responses to increased RPP. This evidence is of great physiological significance since this strategy has been employed to regulate vascular response in different pathological models such as hypertension (4) or diabetes (6). Additional evidence was provided by more sophisticated experiments employing HO-1 gene transfer by viral agents driven by specific promoters (15). The authors further confirmed this hypothesis showing that HO inhibition by its competitive inhibitor CrMP reversed the effects of heme and SnCl2 (Fig. 1). This observation led the authors to hypothesize that pharmacological upregulation of HO activity could somehow be related to Aff-Art constrictor responses. To solve this issue, preparations were obtained following treatment with the two by-products of HO activity: carbon monoxide (CO) and biliverdin. In this set of experiments, Botros et al. (3) showed that biliverdin failed to modify Aff-Art autoregulatory responses, whereas superfusing with 1 mmol/l CO results in a significant attenuation of autoregulatory responses to RPP increase (Fig. 1). To better comprehend the authors’ conclusions and point out the importance of the present work, one must examine the heme-HO system and its role in the regulation of key elements involved in vascular tone regulation (Fig. 2). Heme is the prosthetic group of numerous enzymes involved in this function, such as soluble guanylate cyclase (sGC) (5), nitric oxide synthase (NOS) (21), cytochrome P-450 (2) and cyclooxygenases (COX) (12). HO is the rate-limiting step in heme degradation, which leads to the formation of CO, biliverdin, and iron. HO exists as the isoenzymes HO-1 (inducible form) and HO-2 (constitutive form) (17). It is well documented that HO-1 is markedly induced by its substrate heme and by heavy metals (17). This fact, in conjunction with the robust ability of HO-1 to regulate blood pressure (18), has stimulated research on the mechanism(s) related to this physiological effect. The formed biliverdin is rapidly and stereospecifically reduced to bilirubin by biliverdin reductase (17). Both biliverdin and bilirubin are antioxidants and do not seem to be involved in vascular tone regulation, as also confirmed by Botros et al. (3). It should be noted that pharmacological induction of HO-1 does not decrease the constitutive heme proteins, such as COX-1 (12), eNOS (13), or mitochondrial proteins, but it decreases heme proteins with a rapid turnover, i.e., inducible enzymes such as iNOS or COX-2. Therefore, heme or SnCl2 increases HO-1, while contextually decreasing iNOS (13) and COX-2 (12) and increasing CO, and bilirubin. Controversial hypotheses have been postulated regarding the physiological effects of CO on vascular tone regulation. CO arising from HO-metabolized heme exerts a vasodilator effect (9, 11), stimulating sGC (14) and calcium-activated potassium channels (20) in vascular smooth muscle, thus causing vasorelaxation, and, conversely, inhibition of HO causes vascular constriction (10). In addition, Johnson et al. (8) showed that inhibition of HO with CrMP vasoconstricted isolated firstorder gracilis muscle arterioles denuded of endothelium. Consistent with these observations, inhibition of HO with SnMP significantly magnified the renal vasoconstrictor effect of angiotensin II, enhanced pressure-induced constriction of isolated pressurized renal interlobular arteries (16), reduced renal blood flow, and increased renal vascular resistance, effects that were magnified in the absence of nitric oxide (NO) (16). On the other hand, other investigators suggested that CO promotes endothelium-dependent vasoconstriction by inhibiting endothelial NO formation (7). However, most of our knowledge about the effect of the HO-CO system on renal vascular tone is derived from experiments performed using of perfused vessels of various origins, which may act as different resistance systems. Finally, it should be considered that the effects of CO may be dependent on other variables such as an animal’s age (19). Botros et al. (3) provide an important initial step in the elucidation of the HO system in the regulation of the afferent arteriole system. Still, there remain numerous other potential avenues of exploration, including interaction of CO with 20hydroxyeicosatetraenoic acid and prostanoids. The article by Botros et al. offers further understanding of the relationship between CO and HO-1 induction and the mechanisms that are involved in afferent arterioles tone regulation and opens up new strategic approaches for the effective comprehension of the physiological significance of the HO system in renal physiology. Address for reprint requests and other correspondence: G. Li Volti, Dept. of Biological Chemistry, Medical Chemistry, and Molecular Biology, Univ. of Catania, Viale Andrea Doria 6, 95125 Catania, Italy (e-mail: [email protected]). Am J Physiol Renal Physiol 295: F901–F903, 2008; doi:10.1152/ajprenal.90441.2008.