Nephrosis induced by doxorubicin (adriamycin) is an experimental model of glomerulosclerosis with relative stable proteinuria which is commonly used for pharmacological intervention studies. It is induced by a single or a double dose of doxorubicin, with doses that vary considerably among investigators from 2 to 7.5 mg/kg. Intervention studies with ACE-inhibitors in this model have provided conflicting results. We hypothesized that these discrepancies might be due to different properties of the doxorubicin model, related to the dose of doxorubicin used to induce proteinuria. We tested this hypothesis by inducing doxorubicin nephrosis with 1, 2 and 3 mg/kg, and evaluating the response to intervention with lisinopril. The 1-mg/kg doxorubicin dose did not induce significant proteinuria. The 2- and the 3-mg/ kg dose resulted in a proteinuria of 684+/-215 mg/24 h and 736+/-277 mg/24 h 6 weeks after induction, respectively (Mean+/-SD). Treatment with lisinopril 2 mg/kg/day reduced proteinuria to 160+/-170 mg/24 h(p<0.01) in the 2-mg/kg doxorubicin group, whereas in the 3-mg/kg doxorubicin group, proteinuria did not respond to lisinopril (529+/-264 mg/24 h). In time control rats, proteinuria remained stable. Renal damage developed in both time control groups, with a glomerulosclerosis score of 29+/-22 in the 2-mg/kg group and 84+/-41 in the 3-mg/kg doxorubicin group. Lisinopril resulted in a significantly lower glomerulosclerosis score in the 2-mg/kg doxorubicin group only (16+/-15, p<0.05), whereas the 3-mg/kg group showed no significant reduction (56+/-29, NS). In conclusion, the dose of doxorubicin used to induce nephrosis is an important determinant not only of the severity of the ensuring renal damage, but also of the response to intervention by ACE-inhibition. These findings have an impact on the interpretation of intervention studies in this model.