Effect of 17 beta-estradiol on the growth of estrogen receptor-positive human melanoma in vitro and in athymic mice.

Physiochemical properties of an estrogen binding protein were characterized in three human melanoma cell lines, UISO-MEL-1, UISO-MEL-2, and UISO-MEL-4. Estrogen binding to melanoma cytosol was saturable, specific for estrogens, and represented by a single class of high-affinity, limited-capacity binding sites (Kd 5.5 x 10(-10) M, 2.7 +/- 0.5 fmol/mg of cytosol protein, UISO-MEL-2; 2.2 x 10(-10) M, 7.8 +/- 3.3, UISO-MEL-4) (SEM). UISO-MEL-1 cytosols did not bind estradiol. The binding protein in UISO-MEL-2 and -4 sedimented at 8.5S and 9.2S, respectively, in the presence of 10 mM sodium molybdate. Solid-phase radioimmunoassay with a monoclonal antibody specific for human estrogen receptor (H222 sp lambda) showed good correlation (r = 0.84) with a hydroxyapatite biochemical assay of identical melanoma cytosols. Exposure of UISO-MEL-2 to estradiol produced a time- and temperature-dependent increase in total nuclear receptor for estrogen in vitro. Estradiol treatment of athymic mice also significantly increased cytosol progesterone receptor content in UISO-MEL-2 and UISO-MEL-4 xenografts. Estradiol had no effect on the plating efficiency or growth of any melanoma cell line or normal melanocytes in vitro. Tamoxifen also had no effect on melanoma growth in vitro. In contrast, chronic exposure of athymic mice carrying estrogen receptor-positive UISO-MEL-2 to estradiol resulted in a sex-dependent increase in tumor latency and overall inhibition of tumor growth. Taken together, these observations suggest that a subset of human melanomas contains limited amounts of an estrogen binding protein similar to that observed in other estrogen-responsive tissues. The lack of effect of estradiol on melanocyte and melanoma growth in vitro, coupled with a decrease in tumor growth in athymic mice, suggests that, while inhibition may be receptor mediated, possible indirect actions of estradiol must also be considered.