Properties of epithelial cells cultured from human carcinomas and nonmalignant tissues

Human breast cancer cell lines as models of growth regulation and disease progression

The routine isolation and culture of human breast cancer cells from patient samples has been a goal of breast cancer cell biologists for over 30 years. Despite extensive work in this area and the development of many human breast cancer cell lines, the proportion of patient samples that give rise to immortalized breast cancer cell lines is still disappointingly low. The majority of human breast cancer cell lines that have been established were isolated many years ago and have been grown continuously under poorly defined culture conditions. These cell lines have been useful for studies of the estrogen receptor biology in human breast cancer cells, in identifying growth factors synthesized by breast cancer cells, and for the characterization of genetic alterations in oncogenes and tumor suppressor genes present in these cells. More recently, tissue culture methods have improved, resulting in the ability to culture routinely normal human mammary epithelial cells of specific lineages and this has resulted in the development of new human breast cancer cell lines. The ability to isolate and culture normal and neoplastic human mammary epithelial cells under similar culture conditions has improved these models dramatically and has resulted in the identification of altered cellular phenotypes of human breast cancer cells.

Hormonal regulation of plasminogen activator and peroxidase activities in 7,12-dimethylbenz(a)anthracene-induced rat mammary tumors and the rat uterus

Oophorectomy was found to decrease the plasminogen activator activity of rat mammary tumors induced by 7,12-dimethylbenz(a)anthracene (DMBA) to less than 7 per cent, while in vivo estradiol treatment restored its activity in a dose dependent fashion. The peroxidase activity was not changed either by oophorectomy or by the administration of estrogen. In the rat uterus, plasminogen activator activity was not changed by oophorectomy or by the administration of estrogen, however, its peroxidase activity decreased to less than 2 per cent following oophorectomy, while estrogen administration restored its activity. Estrogen regulated plasminogen activator activity in the DMBA-induced rat mammary tumors but not in the uterus and thus, the specific hormonal regulation of this enzyme may be an important factor for the hormonal dependent growth of such tumors.

Specificity of tumor necrosis factor toxicity for human mammary carcinomas relative to normal mammary epithelium and correlation with response to doxorubicin

By using a unique short-term culture system capable of growing both normal and malignant breast epithelial tissue, human recombinant tumor necrosis factor (TNF) showed preferential cytotoxicity to malignant cells as compared to the corresponding nonmalignant cells. Most of the malignant specimens were sensitive to TNF with 13 of 18 specimens showing 90% inhibition of clonal growth (ID90) by less than 500 units of TNF per ml of culture fluid. In contrast, all 13 nonmalignant specimens tested clustered at the resistant end of the TNF response spectrum, with ID90 values being greater than 5000 units of TNF per ml of culture fluid. This differential sensitivity to TNF was seen in three cases in which malignant and nonmalignant breast epithelial tissues from the same patient were studied. To investigate the mechanism of resistance to TNF by normal cells, the presence of receptors for TNF was determined. Five of six cultures showed specific binding of 125I-labeled TNF and there was no relationship between the degree of resistance and the degree of specific binding. Simultaneous comparison of tumor responsiveness to doxorubicin and TNF revealed a positive correlation in ID90 values; these results may have important implications for the clinical use of TNF in cancer patients heavily pretreated with doxorubicin.

Purification and properties of plasminogen activators from epithelial cells

Two epithelial plasminogen activators were purified from the serum-free conditioned medium of guinea pig keratocytes (GPK) and human breast epithelial (BEB) cells in culture. The cells were cultured on Cytodex 3 microcarrier beads in Eagles' minimum essential medium. The purification procedure was essentially as described by Rijken and Collen (1981) [J. Biol. Chem. 265, 7035-7041]. The specific activities of the purified GPK and BEB activators were 12500 and 6000 IU/mg. Unlike other tissue activators, both the epithelial activators had an isoelectric point of approximately 4.7 +/- 0.2. Pure enzymes were shown to be homogeneous by dodecyl sulphate/polyacrylamide gel electrophoresis with an apparent molecular mass of 62 +/- 2 kDa under reducing conditions. Immunological experiments have shown that both the activators are different from urokinase and do not cross react with anti-urokinase antibodies. Both GPK and BEB activators bound tightly to fibrin clots in vitro. Preliminary N-terminal sequence results indicate that both the epithelial activators appear to be similar to one another but different from melanoma and other tissue activators. These findings indicate that the plasminogen activators secreted by epithelial cells represent a unique and different class of tissue plasminogen activator.

Growth of diploid cells from breast cancers

Cell cultures were derived from normal and cancerous breast tissues and from metastases by methods that selected for relatively adherent epithelial aggregates. Karyotypic analyses of first or second passage cultures yielded predominantly normal diploid cells. Nonclonal aberrations were more common in tumor-derived than in normal cultures. Three of the cultures that originated from metastases were characterized by abnormal clones. These results support observations based on DNA content, which indicate that a considerable fraction of breast cancers are composed predominantly of diploid cells. They differ greatly from chromosomal findings in long-term cultures of tumor effusions and thus emphasize the karyotypic diversity that can be found in tumors from a single tissue of origin--the breast.

The biology of breast cancer at the cellular level

Two properties seem fundamental to cancer; heterogeneity and progression (Foulds (1975) Academic Press, New York; Heppner et al. (1979) Commentaries on Research in Breast Disease, Vol. 1 (Bulbrook, R. and Taylor, D.J., eds.), pp. 177-191, Plenum Press, New York). Relatively little is understood about the premalignant stages of human breast disease in vivo. When the disease manifests as invasive carcinoma, its behavior exhibits great diversity, sometimes metastasizing rapidly, while in other cases 10-30 years pass before metastases proliferate. Here we review various aspects of breast cancer in vivo and consider how they predict properties of breast cancer found in culture. All of the experiments are consistent with the hypothesis proposed by Nowell (1976) Science 194, 23-28, that a fundamental aspect of malignancy is an increased genetic instability and that many of the cells within tumors are nonviable results of genetic instability. We suggest that most of the viable cells within primary breast carcinomas are diploid and are not yet capable of aspects of metastatic spread. What these cells have attained is an increased propensity for genetic instability which enables them to generate randomly aneuploid but frequently lethal genetic configurations. Occasionally one of these altered genomes is associated with the ability to proliferate at a metastatic site. This hypothesis implies that metastases from various patients may have arisen by divergent pathways and may also be divergent in many other aspects of their physiology, unrelated to malignancy. Such extreme heterogeneity may hamper attempts to understand fundamental aspects of malignancy. Hence we suggest that the less anaplastic and less divergent diploid cells within the primary carcinomas might be an important resource to gain insights into the critical alterations that are responsible for initiating frankly malignant behavior.

Metabolism of benzo(a)pyrene by human epithelial and fibroblastic cells: metabolite patterns and DNA adduct formation

We demonstrate in cell culture that mammary epithelial cells from normal human breast specimens metabolize benzo(a)pyrene (BaP) and form adducts with the bases of their DNA more readily and at lower concentrations of BaP than do fibroblasts from the same specimens. BaP metabolism and adduct formation was determined in the same incubations with epithelial cells grown out in early passage from each of three specimens and with fibroblasts from one of these specimens. The metabolite pattern of the epithelial cells was indicative of preferential formation of 7, 8-dihydrodiol-9, 10-dihydroepoxybenzo(a)pyrene the ultimate carcinogen. In contrast, fibroblasts formed mainly mono- and dihydroxide derivatives of BaP. The metabolite pattern from epithelial cells was compatible with the ease in which adducts between DNA and the diolepoxide of benzo(a)pyrene were formed. These results provide evidence that chemical carcinogens should be considered as possible factors in the induction of breast cancer in women.

Use of an efficient method for culturing human mammary epithelial cells to study adriamycin sensitivity

Techniques are described for isolating, cryopreserving, and culturing human mammary epithelial cells of both normal and malignant origin. The cells can be grown either in mass culture or as a clonal assay suitable for quantitating drug sensitivity. With this clonal assay plating efficiencies of 6%-41% were routinely obtained. We examined the response to adriamycin of five different primary carcinoma cultures from patients without prior drug therapy. We were able to detect heterogeneity in response to adriamycin among the breast carcinoma cultures as well as heterogeneity among subpopulations within a single carcinoma. The differences in adriamycin sensitivity were unrelated to growth rates in culture.

Clonal growth of normal and malignant human breast epithelia

We have developed an efficient method for the growth of colonies from single breast epithelial cells derived from nonmalignant [6], primary [7], and metastatic [1] surgical specimens. The immediate sources of the cells were primary and secondary mass cultures. An enriched culture medium and fibroblast feeder layers were used. A linear relationship between the number of cells cultured and the number of colonies obtained was found. Efficiency of colony formation was higher for those derived from nonmalignant specimens (mean 24%) than those derived from malignant specimens (mean 11%). Fibroblast feeder layers increased the efficiency of colony formation and also supported extended growth of colonies.