Cyclosporine A (CSA) and tacrolimus (FK506) are powerful immunosuppressive agents that have proven useful for antirejection therapy in patients with solid organ transplants, including kidney. However, both drugs are nephrotoxic, each producing similar histological patterns of injury to renal tubules and preglomerular arterioles, and this toxicity is a major cause of renal allograft dysfunction. A renal transplant biopsy presently represents the most reliable means of diagnosing nephrotoxicity caused by CSA or tacrolimus and distinguishing it from acute rejection. Because CSA and tacrolimus nephrotoxicity often involve arteriolar smooth muscle, whereas vascular smooth muscle is rarely involved in acute rejection, we investigated if the appearance of a smooth muscle-specific isoform of alpha-actin (SMA) in the urine of renal transplant recipients about to undergo a biopsy for graft dysfunction correlated with biopsy evidence of CSA or tacrolimus toxicity. Eighty-nine urine samples from 61 patients, plus 6 samples from healthy control subjects, were analyzed in a blinded manner by enzyme-linked immunosorbent assay using a specific anti-SMA monoclonal antibody. For the patient samples, the results of these assays were then correlated with the biopsy findings. Those 40 cases in which the biopsy showed evidence of CSA or tacrolimus nephrotoxicity had a significantly (P < 0.01) greater SMA level in the corresponding urine samples (0.089 +/- 0.126 microgram/mL; mean +/- SD) than the 49 cases without toxicity (0.018 +/- 0.027 microgram/mL) or 6 control subjects (0.003 +/- 0.007 microgram/mL), although there was considerable overlap of SMA values among these groups. The greatest SMA levels were seen in patients with CSA or tacrolimus nephrotoxicity that was likely to be relatively acute, namely those with thrombotic microangiopathy and those without previous biopsy evidence of toxicity. SMA levels correlated significantly with the estimated severity of arteriolopathy on biopsy. In patients with tubular but not arteriolar lesions of CSA or tacrolimus toxicity, the mean SMA level was not significantly greater than that in patients without toxicity. Urine SMA levels in patients with a biopsy specimen showing acute rejection were not significantly different from those in patients without rejection, and there was no correlation between urine SMA level and severity of rejection. Whereas the degree of overlap of SMA levels in patients with and without nephrotoxicity was far too great to consider this assay as a potential alternative to renal transplant biopsy for the diagnosis of nephrotoxicity, the assay may have potential as a marker for active arteriolar injury in renal transplant recipients and other patients receiving CSA or tacrolimus therapy.