The Effect of Streptozotocin-Induced Diabetes Mellitus on the Ability of Rat Serum to Stimulate Human Fibroblast Proliferation

Abstract

Impaired tissue repair, tumor growth, and diminished uterine response to β-estradiol are characteristics of experimental diabetes mellitus; this may be due to a specific defect in cellular proliferation. The aim of this study was to determine whether circulating growth factors other than the somatomedins are deficient in diabetic rat serum. We examined the effect of streptozotocin-induced diabetes mellitus on the ability of rat serum to stimulate the growth of cultured human diploid skin fibroblasts, a cell line that has been shown to proliferate in somatomedin-depleted serum. Normal rat serum resulted in a dose-dependent increase in the number of human skin fibroblasts [HES; 2.78 ± 0.15 × 10⁵ (±SEM) for 5% serum vs. 0.21 ± 0.02 ± 10⁵ for no serum; P < 0.0005]. In contrast, serum from untreated diabetic rats resulted in significantly less proliferation than normal serum (0.35 ± 0.01 × 10⁵vs. 2.78 ± 0.15 × 10⁵; P < 0.0005 for 5% diabetic and normal rat serum, respectively). Similarly, serum from untreated diabetic rats was significantly less potent in stimulating DNA synthesis (measured as tritiated thymidine incorporation into DNA) in quiescent growth-arrested fibroblasts (12,698 ± 612 vs. 7,242.8 ± 378.2 cpm/well; P < 0.0005 for 2.5% serum). Serum from insulin-treated diabetic rats was significantly more potent than serum from untreated diabetic rats in stimulating human skin fibroblast proliferation (1.66 ± 0.10 × 10⁵vs. 0.39 ± 0.01 × 10⁵ cells/well; P < 0.0005 for 2.5% serum) and DNA synthesis (11,475.6 ± 1,325.6 vs. 7,242 ± 378.2 cpm/well; P < 0.005 for 2.5% serum), but was not significantly different from normal rat serum despite the relatively poor metabolic control achieved with insulin treatment. Similar results were obtained when sera from untreated and insulin-treated diabetic rats were tested on other human fibroblast cell lines, including embryonic lung and adult skin fibroblasts. In some fibroblast cell lines, serum from insulin-treated diabetic rats was significantly less potent than normal rat serum. Impaired stimulation of fibroblast DNA synthesis by diabetic serum was not due to small molecules such as ketones, as dialyzed serum from untreated diabetic rats was significantly less potent than dialyzed normal rat serum (5,960 ± 1,400 vs. 10,800 ± 810 cpm/well; P < 0.0005 for 2.5% serum). The addition of normal rat serum (10%, vol/vol) and somatomedin-depleted (hypophysectomised) rat serum (10%), but not insulin alone, to diabetic rat serum resulted in the same degree of stimulation of DNA synthesis as that seen with normal rat serum. No evidence of an inhibitor of fibroblast DNA synthesis in untreated diabetic rat serum could be detected in this assay system. On the basis of these studies, we suggest that the defect in cellular proliferation in diabetes mellitus is due to depletion of circulating growth factors other than insulin and somatomedins.