The vasoconstrictor peptide endothelin-1 (ET1) has only recently been characterized and its effects are at present largely speculative. It has been hypothesized that ET1 acts on mesangial cells to cause vasoactive changes which might ultimately contribute to the development of glomerulosclerosis. Opposite to ET1, nitric oxide (NO) inhibits mesangial cell contraction and proliferation. NO activates soluble guanylic acid cyclase and the final product, cyclic GMP (cGMP), has been recently used as a marker of NO action. Urinary levels of ET1 and cGMP were detected in 58 patients with biopsy-proven glomerulonephritis (GN), including 36 IgA nephropathy (IgAGN), 30 with normal and 6 with impaired renal function, 10 patients with non-IgA mesangial GN and 12 pts with membranous GN (MGN) with normal renal function. Compared to normal controls (0.019 +/- 0.006 ng/min), urine ET1 levels were significantly higher in patients with normal renal function having IgAGN (0.035 +/- 0.017, p < 0.01), MGN (0.028 +/- 0.013, p < 0.05), non-IgA mesangial GN (0.027 +/- 0.012, p < 0.05) and those with IgAGN and renal failure (0.032 +/- 0.011, p < 0.01). However no difference was found between MGN patients and normals by deleting MGN cases with mild to moderate mesangial proliferation. The mean value of urinary cGMP in IgAGN patients with renal failure (0.186 +/- 0.117 nmol/min) was lower (p < 0.05) than that of each group with normal renal function (IgAGN: 0.378 +/- 0.010 nM/min; MGN: 0.338 +/- 0.064 nmol/min, non-IgAGN: 0.436 +/- 0.168 nmol/min). The same significant differences were obtained by correcting cGMP values for creatinine urinary excretion. Urinary ET/cGMP ratio (assumed as an index of the relative balance between vasoconstrictor and vasorelaxing factors) was found to be higher than normal (0.570 +/- 0.010 ng/nmol) both in IgAGN patients with normal renal function (0.103 +/- 0.064 ng/mol, p < 0.05), and in those with renal failure (0.203 +/- 0.108 ng/nmol, p < 0.02). Urinary cGMP values were not related to plasma levels of atrial natriuretic peptide (ANP). These data show that hyperexcretion of ET1 occurs in a number of patients with mesangial proliferative GN. In some of them, mainly those with established glomerular damage, the local production of ET1 is not counter-balanced by adequate cGMP biosynthesis.