The effect of estrogen on aromatase and vascular endothelial growth factor messenger ribonucleic acid in the normal nonhuman primate mammary gland.

In the present study, the baboon was used as a model to investigate the effects of steroid hormones on vascular endothelial growth factor (VEG/PF) and aromatase expression and on proliferation of the normal mammary gland. Immunocytochemistry revealed that both aromatase and VEG/PF were expressed in the epithelial cells of the terminal ductal lobular units. Mammary tissue biopsies were obtained from female baboons during the follicular and luteal phases of the menstrual cycle, 4 weeks after ovariectomy (OVX), and after 2 weeks of treatment with estradiol benzoate (E2B; 500 microg/day, i.m.). Although there was little apparent difference in aromatase messenger ribonucleic acid (mRNA) in tissue from follicular and luteal phases or after ovariectomy, aromatase mRNA was decreased in tissue from ovariectomized (OVX) animals treated for 2 weeks with E2B. Furthermore, aromatase activity in tissue from these animals was markedly reduced compared to activity in tissue from the OVX animals before treatment (P < 0.001). In one animal in which mammary aromatase activity was measured sequentially during the follicular and luteal phases, aromatase activity was increased significantly after OVX and was reduced to the level in the intact animal by subsequent treatment with E2B. This effect on both aromatase activity and mRNA occurred rapidly 2 and 4 h after injection with E2B. In contrast to its effect on aromatase, E2B treatment of OVX animals stimulated VEG/PF mRNA 2 and 4 h after injection. In histoculture of mammary biopsies from these animals in the follicular and luteal phases of the menstrual cycle or after OVX, [3H]thymidine incorporation was increased significantly by incubation with testosterone (T) as well as estrogen (P < 0.01). The effect of T was blocked by aromatase inhibitor, 4-hydroxyandrostenedione, suggesting that the tissue is responsive to E produced by aromatization of T in the tissue. When mammary tissue from OVX animals was cultured with T, there was a significantly greater increase in [3H]thymidine incorporation than in histocultures of tissue from intact animals (P < 0.01). However, in histocultures of tissue from the OVX animals treated with E2B (500 microg) for 2 weeks, [3H]thymidine incorporation was similar to the level in tissue of intact animals incubated with T. No significant changes occurred in [3H]thymidine incorporation with the nonaromatizable androgen dihydrotestosterone or progesterone alone. These findings suggest that estrogens produced locally by aromatization of T have a functional role in mammary tissue. Aromatase expressed in the mammary gland could be important in maintaining local estrogen concentrations, particularly after menopause. Estrogen appears to regulate transcription of both aromatase and VEG/PF in the mammary gland, suggesting a regulatory loop by which local estrogens could stimulate VEG/PF production. Thus, paracrine/autocrine mechanisms that can enhance the proliferation of malignant cells and their metastatic spread already exist before transformation.