A gas, an amino acid, and an imposter: The story of nitric oxide, l-arginine, and ADMA in portal hypertension


  • Daniel A. Langer,

    1. Gastroenterology Research Unit, Department of Medicine, Physiology and Tumor Biology Program, Mayo Clinic, Rochester, MN
    Search for more papers by this author
  • Vijay H. Shah

    Corresponding author
    1. Gastroenterology Research Unit, Department of Medicine, Physiology and Tumor Biology Program, Mayo Clinic, Rochester, MN
    • Gastroenterology Research Unit, Al 2-435, Mayo Clinic, 200 First St. SW, Rochester, MN 55905
    Search for more papers by this author
    • See Article on Page 1382

  • Potential conflict of interest: Nothing to report.

  • fax: 507-255-6318

Portal hypertension and its complications are major causes of morbidity and mortality in patients with cirrhosis. Therefore, studies aimed at understanding the genesis of portal hypertension are of great interest. In this regard, over the years it has become clear that increased portal pressure occurs secondary to increased intrahepatic resistance not only from mechanical factors, such as increased collagen deposition and development of regenerative nodules, but also through intrahepatic vasoconstriction, which occurs through signaling defects within the hepatic vascular endothelium,1 and which is the focus of the study in this issue of HEPATOLOGY by Laleman et al.2 Indeed, a number of novel therapies for portal hypertension are now focused on increased intrahepatic vascular resistance as a target, especially through the nitric oxide (NO) pathway.3, 4


NO, nitric oxide; eNOS, endothelial NO synthase; ADMA, asymmetric dimethylarginine; DDAH, dimethylarginine dimethylhydrolase; BDL, bile duct ligated; L-NAME, NG-nitro-L-arginine-methyl ester.

NO is a free radical gas generated by endothelial NO synthase (eNOS), whose signaling functions are a key component of endothelium-dependent vasodilation in liver through its paracrine actions on adjacent contractile cells.5–8 Furthermore, deficient NO from hepatic endothelial cells contributes to endothelial dysfunction, which is defined as an impaired endothelium-dependent relaxation and which is present in the liver with cirrhosis.9–12 Based on this important observation, significant efforts have focused on understanding why production of gaseous NO is deficient in liver endothelial cells with ensuing endothelial dysfunction.9, 11 These studies have focused on mechanisms that limit NO generation from endothelial cells in liver and include putative roles for caveolin, AKT dysfunction, and GRK2 (Fig. 1).6, 13, 14 Thus, multiple molecular defects likely conspire to limit hepatic NO production in cirrhosis.

Figure 1.

ADMA and NO signaling in hepatic sinusoids. NO is generated by the nitric oxide synthase (NOS).23 In the quiescent liver, endothelial NOS is the predominant isoform constitutively expressed by sinusoidal endothelial cells and is involved in maintaining sinusoidal tone and flow.6, 9 NO generated from sinusoidal endothelial cells acts on adjacent hepatic stellate cells partially through activation of soluble guanylyl cyclase signaling and culminates in stellate cell relaxation. Circulating ADMA competes with L-arginine for eNOS resulting in diminished NO production and impaired HSC relaxation. Other important regulators include Akt phosphorylation of eNOS, eNOS binding to the inhibitory protein caveolin, and impaired Akt dependent phosphorylation activation via Akt binding to GRK2. ADMA is metabolized by DDAH in endothelial cells24–26 but also probably in hepatocytes.27 ADMA, asymmetric dimethyl arginine; eNOS, endothelial nitric oxide synthase; VEGF, vascular endothelial growth factor; ACh, acetylcholine; Akt, protein kinase B; GRK2, G-protein–coupled receptor kinase-2; sGC, soluble guanylate cyclase; DDAH, dimethylarginine dimethylhydrolase.

Laleman and colleagues provide evidence for another player in this conspiracy by focusing on a circulating mimic of L-arginine. L-arginine is an amino acid that serves a very important function as the sole substrate for eNOS; in the absence of L-arginine NO cannot be generated. Indeed, many commonly used synthetic pharmacological inhibitors act by competing off L-arginine from NOS. So what is the nature of this arginine imposter? The molecule is one of a related structure, asymmetric dimethylarginine (ADMA). ADMA is generated from proteolysis of previously methylated arginine residues within proteins and then degraded by conversion to citrulline by dimethylarginine dimethylhydrolase (DDAH) (Fig. 1).15, 16 ADMA has been shown to compete with L-arginine for NOS, thereby limiting NO production.16 ADMA is not actually new in the endothelial dysfunction scene. Elevated ADMA levels have already been shown to be a independent predictor of cardiovascular disease and mortality in patients with end-stage renal disease17 and other syndromes associated with endothelial dysfunction.15 More recent studies in the liver arena demonstrate that elevated levels of ADMA are associated with decompensated alcoholic cirrhosis18 and also that hepatic clearance of ADMA is impaired in patients with acute alcoholic hepatitis.19

Laleman et al. investigated ADMA in a cholestatic model of portal hypertension, the bile duct–ligated (BDL) rat, in which hepatic NO production is diminished.2, 20 They demonstrate that in BDL rats there was a significant increase in the level of circulating ADMA. Using an ex vivo rat liver perfusion system they also showed that acetylcholine-mediated vasodilation, which assesses NO- and endothelium-dependent vasodilation, is blunted by addition of ADMA to the perfusate, an effect reminiscent to that of the more traditional synthetic NOS inhibitor, NG-nitro-L-arginine-methyl ester (L-NAME). Interestingly, however, the combination of both ADMA and L-NAME had an additive effect. Furthermore, perfused BDL rat livers demonstrated impaired vasorelaxation to acetylecholine that was significantly exacerbated by ADMA. Last, high-performance liquid chromatography analysis of liver perfusates demonstrated impaired clearance of ADMA in BDL rat livers but not from rat livers with toxin-mediated cirrhosis, suggesting that ADMA clearance by liver is impaired selectively in cholestatic cirrhosis. Indeed, recent evidence does support a role for hepatic clearance of ADMA by DDAH.21, 22 Thus, with increased cholestatic liver damage there appears to be diminished capacity for hepatic ADMA metabolism contributing to increased circulatory levels, thereby perpetuating increased hepatic vascular tone.

Although the ideas are novel and data are compelling and intriguing, what are the limitations of this study? One key issue that warrants further attention is the role of DDAH in the liver. As the liver, rather than the systemic circulation, is where NO generation is deficient, one would anticipate that DDAH expression or activity in cholestatic liver would be diminished, thereby accounting for increased levels of ADMA. This concept was suggested by the impaired clearance of ADMA in perfusates of BDL rats; however, direct biochemical proof of decreased DDAH expression/activity is lacking. Furthermore, one wonders why circulating ADMA does not inhibit NOS in the systemic endothelial cells that paradoxically suffer from excess NO generation, in contrast to the dysfunctional endothelial cells within the liver with cirrhosis. Last, the additive effects of ADMA and L-NAME on hepatic vascular function are surprising and may suggest additional potential mechanisms of action of ADMA beyond NOS inhibition alone. Thus, these points require further investigation. In summary, the study by Laleman and colleagues has unmasked a new imposter of the amino acid L-arginine, ADMA, that causes eNOS to limit its production of NO gas in cirrhosis, thereby contributing to endothelial dysfunction, increased intrahepatic resistance, and portal hypertension. With the increasing emphasis of therapies targeting increased intrahepatic resistance, ADMA may be an imposter that is worth pursuing.