Asymmetric dimethylarginine and kidney disease--marker or mediator?

A symmetric dimethylarginine (ADMA) is a naturally occurring amino acid that circulates in blood and is excreted in urine (1). It is also a competitive inhibitor of all three isoforms of nitric oxide synthase (NOS). ADMA competes with arginine for the binding site of NOS with an IC50 of about 10 M depending on the prevailing concentration of arginine (2). Combine these observations and it is possible to make the statement that ADMA is an endogenous inhibitor of NOS. Although correct, this implies only that endogenously generated ADMA has the capacity to act as an inhibitor of NOS, and does not necessarily mean that this is either a physiologic or pathophysiologic role of ADMA. Ever since the first description of ADMA as an endogenous inhibitor of NOS (1), two approaches have been taken to try to answer the question of its biologic importance. One approach has been to explore the relationship between the circulating concentration of ADMA and disease states, and the other has been to undertake experiments designed to test causal relationships. The first disease association established was in renal failure (1). The idea was simple: Because ADMA is excreted in the urine, it should accumulate as the kidneys fail, and, as a consequence, the inhibition of NOS would produce adverse effects in many different organ systems. Reassuringly, ADMA fulfils many of the characteristic features of a uremic toxin. It is a guanidino compound, a product of protein metabolism, accumulates in renal failure, is removed by dialysis, and has a clear mechanism of action (inhibition of NO generation) to produce pathophysiology. Studies have already indicated that plasma ADMA levels are predictive of cardiovascular morbidity and mortality in renal failure (3,4) and now, in this issue of the Journal of the American Society of Nephrology, two articles describe that ADMA levels also predict progression to renal failure (5,6). It might also be reasonable to expect other effects on NO systems to affect the gut and the immune and nervous systems (1). Despite the encouraging results, the literature on ADMA levels in disease states is not easy to interpret with any degree of certainty. Different laboratories have reported widely differing ranges for ADMA concentration in plasma (7,8), and the use of laborious HPLC techniques to measure ADMA has meant that samples sizes for clinical studies have usually been relatively small. The two studies reported here are of 177 and 131 patients with kidney failure. One assessed ADMA by liquid chromatography–mass spectrometry and the other used a newly described ELISA. Taking into account differences in age and degree of renal failure in the subjects, ADMA concentrations were similar between the two studies and both found that those patients with the highest ADMA had the greatest progression to end stage disease, and gave comparable estimates of risk. The predictive power of ADMA appears to be independent of other markers, and may be additive in predictive value when combined with brain natriuretic peptide and C-reactive protein (9). Even allowing for the relatively low event rates, small sample sizes, and the possibility of publication bias, it does now appear likely that ADMA is a risk marker for death, cardiovascular mortality, and renal progression in patients with kidney failure. A meta-analysis of the available studies may be useful to get a better estimate of risk, but would be hampered by the lack of standardization in ADMA measurement techniques. As well as noting the predictive power of ADMA, it is clear that markers of renal function also predict cardiovascular risk (10) and that the best predictor of ending up with no renal function is starting off with reduced renal function. In this context it is worth noting that the predictive power of cystatin C, creatinine, and symmetric dimethylarginine (SDMA) is probably better than that of ADMA (5,6,10). This is important because creatinine has not been proposed as a biologic mediator, and SDMA, unlike ADMA, does not inhibit NOS (1) and does not have a clear biologic rationale to act as a mediator of cardiovascular risk. Two questions follow: First, is the accumulation of ADMA simply due to failed renal clearance? Second, is ADMA more than just another marker of renal dysfunction? SDMA is not metabolized to any great extent and, as expected, its concentration in plasma correlates closely with creatinine. In contrast, ADMA is actively metabolized to dimethylamine and citrulline by the action of the dimethylarginine dimethylaminohydrolases (DDAH I and II) (11,12). Indeed, the major route of elimination of ADMA from the body is by metabolism rather than by urinary excretion (13). Therefore, as the kidneys fail ADMA does indeed accumulate, but the extent of accumulation is limited because metabolism continues. Thus, though the Published online ahead of print. Publication date available at www.jasn.org.