Myofibroblasts and the progression of crescentic glomerulonephritis.

BACKGROUND

The cellular and humoral factors involved in the pathogenesis of glomerulosclerosis and renal fibrosis following a crescentic glomerulonephritis have not been fully elucidated. Myofibroblasts and transforming growth factor-beta (TGF-beta) have been implicated in the development of experimental and clinical renal fibrosis. We have attempted to identify these mediators in crescentic glomerulonephritis and determine their role in the progression of the disease.

PATIENTS AND METHODS

We studied retrospectively 21 patients with crescentic and necrotizing glomerulonephritis (CNG) with emphasis on the renal expression (detected by immunohistochemistry) of myofibroblasts (alpha-smooth muscle actin+ cells), TGF-beta and collagen (III and IV) as well as their relationship with the clinical outcome of these patients. In situ hybridization histochemistry was applied to determine the site of synthesis of TGF-beta1 and collagen III. All the patients were treated by immunosuppression and followed up for a median period of 14 months.

RESULTS

Myofibroblasts and TGF-beta were detected in the crescents as well as in the periglomerular and tubulointerstitial areas in CNG biopsies. TGF-beta1 was also detected within renal tubular cells. The percentage of glomeruli with fibrotic and fibrocellular crescents was positively correlated with the severity of Bowman's capsule disruption (r = 0.631, P < 0.01) and with the intensity of myofibroblast expression in the interstitium (r = 0.504, P < 0.05). Strong interstitial immunostain for myofibroblasts and TGF-beta was also noted in association with interstitial fibrosis. In situ hybridization revealed the site of synthesis of TGF-beta1 to be the renal tubular cells of patients with CNG. By contrast, the site of synthesis of collagen III appeared to be confined to interstitial cells surrounding vessels, tubules and the glomeruli in a distribution identical to that of myofibroblasts. There was a significant positive correlation between the number of interstitial alpha-SMA+ cells and both interstitial TGF-beta (r = 0.591, P < 0.01) and interstitial collagen IV (r = 0.588, P < 0.01). In addition, the number of interstitial alpha-SMA+ cells and the extent of immunostain for collagen IV were positively correlated with the final serum creatinine (r = 0.517, P < 0.05 and r = 0.612, P < 0.01 respectively) and partially predicted functional outcome (R2 = 26.7% and 37.5% respectively) as well as the response to treatment. An association was observed between periglomerular myofibroblasts and the generation of fibrotic and fibrocellular crescents.

CONCLUSION

These observations suggest a causal link between myofibroblasts and fibrotic crescent formation. We also believe that interstitial myofibroblasts are actively involved in the pathogenesis of interstitial fibrosis in CNG.