PREVENTION OF GLOMERULOSCLEROSIS BY EARLY CYCLOSPORINE TREATMENT OF EXPERIMENTAL LUPUS NEPHRITIS

The influence of cyclosporine A (CsA) treatment on the development of glomerulonephritis and glomerulosclerosis was investigated in chronic graft-versus-host disease (GVHD), a murine model for lupus nephritis. The renal disease is characterized by the formation of IgG-containing electron-dense deposits along the glomerular basement membrane (GBM) and in the mesangium, followed by the onset of proteinuria which starts, varying per individual mouse, about six weeks after the induction of the disease. Glomerular mRNA levels for matrix molecules were increased from week 4, preceding mesangial matrix expansion and GBM thickening which occurred from week 6. These initial events finally led to development of glomerulosclerosis, and end-stage renal failure. Groups of mice received three intraperitoneal (i.p.) injections per week with different doses of CsA, and treatment was started 2, 4, or 6 weeks after induction of the disease. Treatment with 10 or 50 mg CsA/kg/week did not influence the development of glomerulonephritis or glomerulosclerosis. Injection of 100 mg CsA/kg/week delayed the onset of proteinuria only when treatment was started in week 2. In week 6 some mice had already developed proteinuria whereas others had not. Treatment with 250 mg CsA/kg/week starting in week 6 abrogated glomerulonephritis and glomerulosclerosis only in those animals which were not yet proteinuric at that time. This, despite comparable increased autoantibody levels against DNA, GBM, and renal tubular epithelium (RTE) in both treated and untreated GVHD mice. Further increase in proteinuria and development of glomerulosclerosis could not be prevented if the mice already had developed proteinuria when CsA treatment was started. Dot blot analysis and in situ hybridization showed significantly decreased mRNA levels for alpha 1(I) and alpha 1(IV) collagen in kidneys of CsA-treated mice as compared to those of untreated mice 12 weeks after induction of the disease, if the highest dose of CsA was administered before the onset of proteinuria. No effect on these whole-kidney mRNA levels was observed in mice which had already developed proteinuria before CsA injections were started. Increased mRNA expression for matrix molecules in this group and in untreated GVHD mice was observed mainly in the interstitium. The kidneys of the treated GvHD mice and those of mice injected with 250 mg CsA/kg/week without induction of GvHD showed no morphological signs of CsA nephrotoxicity. We conclude that treatment with 250 mg CsA/kg/week prevents the development of glomerulonephritis and glomerulosclerosis in this model of lupus nephritis, if started before the onset of proteinuria. This might result from a direct effect of CsA on matrix synthesis, since autoimmunity was not affected as reflected by unaltered autoantibody levels, which increased to a similar extent in both treated and untreated animals. Furthermore, our results may indicate the possible diagnostic and prognostic value of extracellular matrix steady-state mRNA levels in early stages of renal disease.