Arginine and arginase: endothelial NO synthase double crossed?

In a remarkable article published in this issue of Circulation Research, Ryoo et al1 propose endothelial arginase II as a novel target for the treatment of atherosclerosis. Actually, they already had done so 2 years ago, based on work mainly on cultured human aortic endothelial cells,2 which indeed appeared to provide an unexpected way to explain endothelial dysfunction in terms of NO production.3 This new road to endothelial dysfunction proposed that oxidized low-density lipoproteins (LDLs), which we all agree are a main culprit in the initiation of the atherosclerotic process, increase the activity of arginase II, which, in turn, decreases NO production presumably by shunting the common substrate arginine away from endothelial NO synthase (eNOS) (Figure). The present study demonstrates that the activity of endothelial arginase II is augmented in apolipoprotein E–null (ApoE / ) mice, as well as in wild-type mice, fed a high cholesterol diet. Inhibitors of arginases restored endothelial function to normal in the former and genetic deletion of arginase II did so in the latter. The study covers a wide array of convincing experiments, both in vitro and in vivo, performed by a large group of investigators (actually, there are almost as many authors on the report as mice reported in the results section) with established expertise in the field of arginases and endothelial function.2,4,5 One can always question the selectivity of pharmacological agents, and the authors recognize that part of the effect of the drugs used in their in vivo studies may be attributable to inhibition of arginase in the vascular smooth muscle cells.1 However, the combination of the effect of the inhibitors in ApoE / mice and the reduction of the impact of high cholesterol in Arg / animals is particularly convincing. Thus, one can only bow to the evidence and, after reading the article, accept the conclusion of the authors that arginase II plays a critical role in the pathophysiology of cholesterol-mediated endothelial dysfunction, at least in the mouse. This is not the first (or second) time that an increased arginase II activity has been involved in atherosclerotic endothelial dysfunction in the ApoE / mouse and that the enzyme has been proposed as a new therapeutic possibility for atherosclerosis,6 which raises the question: how novel is novel? Interestingly, this earlier6 and the present1 studies agree that the protein level of the enzyme is not increased in the ApoE / endothelium. Thus, it is all a matter of activity and access to eNOS. Ming et al6 propose that the augmented activity is attributable to an augmented RhoA protein level. Ryoo and colleagues1,2,4 propose that oxidized (Ox)LDL increases arginase II activity by a sequence of regulatory events that involve decreased association with microtubules and a later increase in transcription. The dissociation of the enzyme from the microtubule cytoskeleton facilitates its cytosolic localization and thus its homing to the eNOS sites where the 2 enzymes can compete for the common substrate, arginine. Obviously, those 2 mechanisms could reinforce each other. Arginases are present in 2 isoforms, I and II. Arginase I is the hepatic isoform, whereas arginase II has a mainly extrahepatic, mitochondrial localization. Both isoforms catalyze the hydrolysis of L-arginine to L-ornithine and urea, the final step of the urea cycle (see elsewhere3,7,8), and the enzymes have been involved in a number of pathologies.8 With regard to atherosclerosis specifically, although arginase I can contribute to endothelial dysfunction in aging and diabetic rats,5,9 cultured human endothelial cells do not appear to express this isoform, and it is not induced by stimuli that augment the activity of arginase II.2,6 Thus, if arginase were involved in human endothelial dysfunction, the II isoform would be the likely culprit, which reinforces the interest of the studies by Ming et al6 and Ryoo colleagues.1,2 However, before jumping to the conclusion of such involvement, a few specific and general questions need to be answered. First, specifically, how does this competition work? Shared substrate (L-arginine) availability for 1 enzyme (eNOS) may be influenced by the activity of the other (arginase II). Thus, it is proposed that increased arginase activity and L-arginine utilization limits the access of the common substrate to NO synthase, leading to uncoupling with reduced production of NO and augmented production of superoxide anions.1,2,10,11 In support of this interpretation are the observations that OxLDL reduce the production of NO metabolites and augment that of oxygen-derived free radicals (reactive oxygen species [ROS])2 and that arginase inhibitors have an opposite effect.1 A nonspecific antioxidant effect of the inhibitors supposedly was ruled out by showing no reduction of the ROS signal generated in vitro by the xanthine-xanthine oxidase reaction.1 However, in the rodent endothelium, the major enzymes generating free radicals contributing to endothelial dysfunction are NADPH oxidase and cyclooxygenase,3,12,13 and one can but wonder how apocynin and indomethacin would affect the results.13,14 Also, the mere presence of an augmented The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association. From the Department of Pharmacology, Li Ka Shing Faculty of Medicine, The University of Hong Kong. Correspondence to Paul M. Vanhoutte, Head, Department of Pharmacology, Li Ka Shing Faculty of Medicine, University of Hong Kong, 2/F Laboratory Block, 21 Sassoon Rd, Hong Kong. E-mail vanhoutt@ hkucc.hku.hk (Circ Res. 2008;102:866-868.) © 2008 American Heart Association, Inc.