Asymmetric dimethylarginine (ADMA) in human blood: effects of extended haemodialysis in the critically ill patient with acute kidney injury, protein binding to human serum albumin and proteolysis by thermolysin.

Free, non-protein bound asymmetrically guanidine-dimethylated arginine (ADMA) is an endogenous inhibitor of nitric oxide (NO) synthesis. Human erythrocytic membrane comprises considerable amounts of large (>50 kDa) ADMA-containing proteins. Location in the erythrocyte membrane and identity and physiological functions of ADMA-containing proteins are unrevealed. In healthy subjects, the concentration of free ADMA in heparinised plasma is almost identical to that of serum. We hypothesised that the robustness of free ADMA concentration in human blood is due to a remarkable resistance of erythrocytic ADMA-containing proteins against proteases. In vivo, we investigated the course of the concentration of ADMA in serum and EDTA plasma of a critically ill patient with acute kidney injury during extended haemodialysis. In vitro, we studied the effects of thermolysin, a useful experimental proteolytic enzyme of erythrocyte membrane proteins, on erythrocytic ADMA. The protein binding (PB) of ADMA to human serum albumin (HSA) was also determined. In these studies, ADMA was measured by a previously reported, fully validated GC-MS/MS method. We measured almost identical ADMA concentrations in plasma and serum samples of the patient. During dialysis, the circulating ADMA concentration decreased slowly and moderately indicating removal of this substance, which was however much less than expected from its low molecular weight (202 Da) and high water solubility. After dialysis, circulating ADMA concentration increased again, a phenomenon called rebound, and ADMA reached higher levels compared to the baseline. The PB value of ADMA to HSA was about 30 %. This surprisingly high PB value of ADMA to HSA may be an explanation for the rather poor dialysance of ADMA. Washed human erythrocytes suspended in phosphate-buffered physiological saline were found not to release appreciable amounts of free and ADMA-containing proteins. The lack of effect of coagulation or anticoagulation on the concentration of circulating free ADMA in humans is likely to be due to a remarkable resistance of ADMA-containing proteins in the erythrocyte membrane against proteases in vivo in humans. Our study suggests that free ADMA is released in the circulating blood at relatively high rates. The considerable PB of ADMA to HSA is likely to add to the apparently poor dialysability of ADMA. Other contributing factors could be redistribution of free ADMA between plasma and erythrocytes in favour of plasma ADMA and parallel formation of free ADMA from erythrocytic ADMA-containing proteins during haemodialysis.