Inducible nitric oxide synthase and cardiac dysfunction in salt-sensitive hypertension

Nitric oxide (NO) is generated from l-arginine by nitric oxide synthase (NOS), an enzyme that has three isoforms. Endothelial (NOS3) and neural (NOS1) isoforms are so labelled because they were initially identified in endothelial cells and neurones, although they have subsequently been identified in many different cell types. Both are expressed constitutively and activated by Ca2+. Inducible NOS (NOS2) was first identified in macrophages but its expression can be induced in almost any cell type. NOS1 and NOS3 are both found in the normal heart in a restricted distribution: (i) NOS1 primarily in nerves innervating the heart but also in the sarcoplasmic reticulum of cardiomyocytes; (ii) NOS3 in the endocardial and vascular endothelium as well as in cardiomyocytes, albeit at much lower concentrations. In contrast, NOS2 may be expressed in all cell types in response to an adequate stimulus. This complex distribution of NOS isoforms is matched by the variety and complexity of the cardiac effects attributed to NO. The response to NO appears to depend critically on the amount of NO formed, the context in which it is formed and even the precise subcellular localization in which it is formed. Thus, NO has been reported to be positively inotropic at low dose and negatively inotropic at high dose [1] with the inotropic response being altered by damage to the endocardial endothelium or by prior cholinergic stimulation. The importance of subcellular localization of NO formation in determining inotropic responses was recently highlighted by Barouch et al. [2]. NOS1 is colocalized with the ryanodine receptor in the sarcoplasmic reticulum where it acts to stimulate calcium release increasing contractility. In contrast, NOS3 localizes to caveolae together with β-receptors and l-type Ca2+ channels facilitating the inhibition by NO of the positive inotropic response to β-receptor activation. Diverse effects have also been reported as a consequence of NO formation following induction of NOS2. A number of cytokines induce NOS2 in the heart and elicit contractile dysfunction that can be prevented by NOS inhibition [3]. Conversely, the protection against post-ischaemic stunning afforded by basic fibroblast growth factor is absent in NOS2-deficient hearts and is abolished by selective NOS2 inhibition [4]. It is on this background of isoform-specific and context-dependent effects of NOS that the results reported by Kobayashi et al. [5] in this issue of the journal have to be interpreted. Aminoguanidine, a NOS2 inhibitor, was given to Dahl salt-sensitive (DS) rats receiving a high-salt diet. Aminoguanidine treatment attenuated the hypertrophy and prevented the decrease in contractility observed in vehicle-treated DS rats on a high-salt diet. Expression of NOS2 mRNA and protein were increased in vehicle-treated DS rats on a high salt diet and the increase was abolished by aminoguanidine treatment. The phosphorylation of p42 and p44 extracellular signal-regulated kinases (ERK) were increased in DS rats on a high-salt diet and this increase was attenuated by aminoguanidine treatment. The authors infer that NOS2 contributes to impaired ventricular function and vascular remodelling in DS rats receiving a high-salt diet and imply that this effect might be mediated by inhibition of the phosphorylation of p42 and p44ERK. The interpretation of this study is critically dependent on the selectivity of the NOS2 inhibitors used. While aminoguanidine is relatively selective for NOS2 in comparison to other NOS isoenzymes, it has other biological actions, including actions to prevent the accumulation of advanced glycation end-products and to prevent lipid peroxidation. Thus, it is reassuring that a second NOS2 inhibitor, S-methylisothiourea sulphate, also reduced left ventricular NOS2 mRNA and protein levels and prevented hypertrophy and contractile dysfunction. Although NOS2 inhibition did not alter blood pressure in this study, previous studies have reported that the NOS2 locus cosegregates with blood pressure in the Dahl salt-sensitive rat [6] and that NOS2 inhibition elevates blood pressure in Dahl salt-sensitive and salt-resistant rats on a high-salt diet [7]. The reduction in NOS2 mRNA and protein levels with the NOS2 inhibitors is not related to their mechanism of action [8] and is difficult to explain. It could indicate that induction of NOS2 by a high-salt diet includes a positive feedback mechanism so that NOS2 induction begets further NOS2 induction. The restoration of normal cardiac contractility by NOS2 inhibitors is striking and is accompanied by attenuation of cardiac hypertrophy. The activation (phosphorylation) of the p42/p44 ERK observed is likely to contribute to the cardiac hypertrophy observed in this model, as it is in other models of cardiac hypertrophy [9]. That NOS2 inhibition suppresses ERK phosphorylation while preventing hypertrophy suggests that it acts proximally to ERK activation in the pathways leading to cardiac hypertrophy. NOS2 inhibition may prove to be therapeutically useful if the effects observed in this model can be generalized to other forms of hypertrophy and cardiac dysfunction. The diverse effects of NO on the heart indicate that generalizability cannot be assured.