Gail Bolan

Human breast cancer responsive to androgen in long term tissue culture

ALTHOUGH a significant role has been established for oestrogen in the growth of some animal1,2 and human3,4 mammary carcinoma, evidence for androgen dependency of human breast cancer has remained equivocal. Although the existence of a mouse mammary carcinoma which shows modest androgen responsiveness5 has supported the notion that some human breast cancer might also be dependent on androgens, no direct proof has been available. We have described our work characterising the oestrogen responsiveness of MCF-7, a human breast cancer cell line maintained in tissue culture for over 3 yr (ref. 6). This cell line was shown to contain oestrogen receptor, to be killed by antioestrogens, and stimulated by physiological concentrations of oestradiol. We now describe further investigations of this cell line which reveal that in addition to oestrogen responsiveness, this cell line shows threefold enhancement of precursor incorporation into macromolecules by androgens using serum-free conditions which preclude stimulatory effects of other trophic hormones. Furthermore, this stimulation by androgens seems to be mediated by interaction with a cytoplasmic androgen receptor protein which is clearly differentiable from the oestrogen receptor also found in these cells. Aside from defining an interesting new system in which the action of androgen can be studied in a human cell line in tissue culture, the present study provides unequivocal evidence for the androgen responsiveness of some human breast cancer at least in vitro.

Model systems for the study of estrogen action in tissue culture

Abstract Physiologic concentrations of estrogen stimulate precursor incorporation and growth in several lines of human breast cancer cells in long term tissue culture. Antiestrogens inhibit precursor incorporation and eventually kill the cells. Estrogens reverse antiestrogen effects. Responsive cell lines contain high affinity specific cytoplasmic receptors for estrogen. When binding and stimulation are measured under similar conditions, it appears that only a small proportion of receptor sites need be occupied for maximal stimulation. Thymidine kinase specific activity is increased by estradiol and may, in part, explain enhanced thymidine incorporation. These cell lines should prove useful for the study of the mechanisms by which estrogens regulate growth in human breast cancer

Progesterone and glucocorticoid interactions with receptor in breast cancer cells in long-term tissue culture.

I t has long been appreciated that the clinical course of breast cancer can be altered by a variety of hormonal manipulations.’ Despite an awareness that a variety of ablative and additive hormonal therapies may be accompanied by objective tumor regressions in some cases, the exact mechanism whereby these therapeutic interventions cause tumor regression is unclear. While clinical studies of breast cancer and investigation of hormone-dependent breast cancer in animals have contributed much to our understanding, the inherent complexities of studying hormonal regulation of tumor growth in vivo led us to attempt to investigate this process in long-term tissue culture lines of human breast cancer. By using defined media in which single hormones could be added to well-characterized uniform populations of cells, we hoped to avoid some of the difficulties intrinsic in more complexly organized systems. Thus, effects of hormone mediated by alterations in levels or activity of other trophic hormones, or effects of hdrmone on supporting stromal elements or the immune system, could be eliminated from consideration. I n addition, some recent work has suggested that determination of progesterone receptor activity may be helpful in predicting which patients with breast cancer might benefit from hormonal therapy,2 much as the knowledge of estrogen receptor levels has also been shown to be of value.) We have previously reported that some lines of human breast cancer in long-term tissue culture contain glucocorticoid and progesterone receptor^,^.^ and this finding has been confirmed by others.6 Because of the biologic interactions of glucocorticoids and progesterone, the availability of new compounds, such as R 5020, for receptor quantification (see below), and the need to establish assays that would unambiguously measure each of these receptors independently, we undertook a series of experiments aimed at studying progesterone-glucocorticoid interactions with human breast cancer in long-term culture. To substantiate the likelihood that observations made on such tissue culture systems bear any relevance to breast cancer in vivo. it is important to review the criteria that suggest that these cells resemble those found in the tumor in vivo. Four lines of evidence suggest that these cells,* which were initially derived from breast cancer specimens, continue to manifest properties characteristic of the primary tumor. First, morphologically, all of the cell lines employed have epithelial characteristics. In some cell lines, such as the MCF-7,’ the tendency of