An in vitro model of epithelial cell growth stimulation in the rodent mammary gland.

Mouse mammary epithelial cell cultures previously described bring about extensive proliferation and a cell population with the appropriate markers for luminal ductal epithelial cells, and also the ability to form normal tissue after implantation into mice. This success may result from a culture environment that resembles certain aspects of the environment in the mammary gland. Mouse mammary epithelial cells, whose proliferation is limited when plated alone, can be stimulated to multiply by contact with lethally irradiated cells of the LA7 rat mammary tumour line. Most of the proliferative stimulus is imparted by direct cell contact between LA7 and mouse mammary cells. Junctions, including adherens junctions, form among all cells in the culture, much as junctions form in the mammary gland. LA7 cells secrete TGFalpha and bFGF, factors found in the mammary gland, and factors to which mouse mammary cells respond in culture. Mouse mammary cells express keratins 8 and 18, markers for luminal cells of the mammary duct. LA7 cells express keratin 14 and vimentin, markers for myoepithelial cells. These facts, taken together, fit a model of cell replacement in an epithelial tissue and also imitate the relationship between luminal ductal cells and myoepithelial cells in the mammary gland. This method of culturing cells is useful, not only for in vitro-in vivo carcinogenesis studies, but also for the study of mechanisms by which growth signals are imparted from one cell to another.

Cellular expression of estrogen and progesterone receptors in mammary glands: regulation by hormones, development and aging.

At present, there is an extensive body of literature documenting the participation of estrogen receptors (ER) and progesterone receptors (PR) in mammary gene expression. Yet, the precise roles of these receptors in regulating mammary development, carcinogenesis and the growth of a subset of tumors still remain unclear. Mammary glands are composed of various cell types with different developmental potentials. Further, ultimately, that it is their mutual interactions which dictate the behavior of mammary epithelial cells. Therefore, to resolve the roles of ER and PR in normal mammary growth, differentiation and carcinogenesis, analyses for the expression of these receptors at the level of individual cell types is of paramount importance. Accordingly, in the present studies using immunolocalization techniques, we document the ontogeny and cellular distribution of ER and PR during mammary development and in response to ovarian hormones and aging. In addition, we discuss the potential biological significances of the expression patterns of ER and PR during various physiological states. We believe that the observations reported here should provide a conceptual framework(s) for elucidating the roles of ER and PR in normal and neoplastic mammary tissues.

Growth and characterization of N-methyl-N-nitrosourea-induced mammary tumors in intact and ovariectomized rats.

It is well established that 85-90% of chemically induced mammary tumors in rats will disappear or diminish significantly in size after the ovaries are removed from the animal. However, it is less well established whether a high percentage of these mammary tumors will grow back with prolonged time after ovariectomy. It is also not known what changes in gene expression take place in the tumors as they develop an independence from hormones for growth. This study was carried out to investigate this. Virgin, 50-day-old female Sprague-Dawley rats were injected with N-methyl-N-nitrosourea (MNU) at the dose of 50 mg MNU/kg body wt. When at least one mammary tumor had grown to 1.0-1.5 cm in one dimension, the animal was bilaterally ovariectomized and reduction and then re-growth of the tumors monitored. Control animals were treated identically except they were not ovariectomized when tumors appeared. Re-growths and new tumors and tumors that developed in the control rats were removed when they reached 1.0-1.5 cm in diameter and all animals were killed 25 weeks after the MNU injection. All the animals in the study (100%) developed mammary tumors after MNU injection with an average latency of 56.5 days. After ovariectomy, 93% of the tumors showed 50% or more reduction in size and 76% of the tumors could not be detected by palpation. However, in 96% of the animals where tumor reduction or disappearance occurred, a re-growth or new mammary tumor development took place with an average latency period of 52.8 days from the day of ovariectomy. Of these post-ovariectomy tumors, 36% occurred at a location where tumors had developed prior to ovariectomy, but 64% appeared at new locations. The circulating levels of 17beta-estradiol (E2) was undetectable in the ovariectomized (OVX) rats and significant reduction was seen in the serum concentrations of progesterone (P4), prolactin (PRL), growth hormone (GH) and insulin-like growth factor-I (IGF-I). The tumors from the OVX rats showed indications of progression as evident from loss of differentiation and invasive characteristics. Comparison between tumors from OVX and intact rats revealed a significantly increased expression of P450 aromatase and elevated activation of extracellular signal-regulated kinase 1 and 2, but reduced levels of the progesterone receptor and cyclin D1 in OVX rats. However, the estrogen receptor (ER) content remained similar in tumors from both groups, at least at the protein level, and so did the expression of IGF-I, IGF-II, insulin receptor substrate-1 (IRS1), IRS-2 and epidermal growth factor receptor. IGF-I receptor (IGF-IR) and ErbB-2 were expressed, respectively, in 50 and 70% of the tumors from the OVX animals, whereas these genes were expressed in 100% of the tumors from the intact rats. It is concluded that chemically induced rat mammary tumors may still depend on the ER and local syntheses of E2 and growth factors for growth initially after ovariectomy. However, as these tumors progress, they develop a more aggressive phenotype and lose their dependency on the ER and possibly growth factors.

Identification of rat mammary tumor-1 gene (RMT-1), which is highly expressed in rat mammary tumors.

Full-term pregnancy early in life results in a permanent reduction in lifetime breast cancer risk in women. Parous rats and mice are also refractory to chemical carcinogenesis. Therefore, investigation of the differences between mammary glands from virgin and parous rats would provide valuable information regarding the protective effects of early full-term pregnancy. In this report, we examined the gene expression patterns in mammary glands from virgin and parous Lewis rats. Using differential display technology, a novel 4.2 kb cDNA, designated rat mammary tumor-1 (RMT-1) was isolated. Northern blot analysis of RMT-1 showed that RMT-1 expression was higher in the pre-pubertal and pubertal stages during rat mammary gland development while it was down-regulated in mammary glands from mature virgin and parous rats. RMT-1 expression was highest in rat mammary cancers compared with either the mammary glands of virgin or parous rats. At the Northern blot sensitivity level, RMT-1 expression was found only in the mammary gland. Northern blot analysis also showed that the expression of this gene was found in 74% of N-methyl-nitrosourea (MNU)-induced mammary cancers while it was not found in MNU-induced cancers from other organs. The examination of the RMT-1 gene structure revealed that it consists of five exons spanning 5.9 kb. Using fluorescence in situ hybridization, the gene was localized on rat chromosome 1 band q 43-51. The present data show that there is a correlation between high RMT-1 expression and rat mammary carcinogenesis or decreased RMT-1 expression and parity associated refractoriness to chemically induced mammary carcinogenesis. However, whether or not RMT-1 gene has a functional role in these processes remains to be investigated.

Short-term exposure to pregnancy levels of estrogen prevents mammary carcinogenesis.

It is well established that pregnancy early in life reduces the risk of breast cancer in women and that this effect is universal. This phenomenon of parity protection against mammary cancer is also observed in rodents. Earlier studies have demonstrated that short-term administration of estradiol (E) in combination with progesterone mimics the protective effect of parity in rats. In this study, the lowest effective E dosage for preventing mammary cancer was determined. Rats were injected with N-methyl-N-nitrosourea at 7 weeks of age; 2 weeks later, the rats were subjected to sustained treatment with 20 microg, 100 microg, 200 microg, or 30 mg of E in silastic capsules for 3 weeks. Treatments with 100 microg, 200 microg, and 30 mg of E resulted in serum levels of E equivalent to those of pregnancy and were highly effective in preventing mammary cancer. E treatment (20 microg) did not result in pregnancy levels of E and was not effective in reducing the mammary cancer incidence. In another set of experiments, we determined the effect of different durations of E with or without progesterone treatments on mammary carcinogenesis. These experiments indicate that a period as short as one-third the period of gestation is sufficient to induce protection against mammary carcinogenesis. The pioneering aspect of our study in contrast to long-term estrogen exposure, which is thought to increase the risk of breast cancer, is that short-term sustained treatments with pregnancy levels of E can induce protection against frank mammary cancer.

Parous rats regain high susceptibility to chemically induced mammary cancer after treatment with various mammotropic hormones.

Parity in humans and rats provides significant protection against mammary tumor development. This study was carried out to investigate whether treatment of parous rats with mammotropic hormones would affect methyl-nitrosourea (MNU)-induced mammary carcinogenesis. Parous rats were treated with 17beta-estradiol (E2), progesterone (P4) and thyroxine (T4) alone or in combination. E2 (20 microg/60 days) and P4 (20 mg/60 days) were administered by silastic tubing and T4 in the drinking water (3 microg T4/ml). Hormonal treatments commenced 7 days before MNU injection and continued for 33 weeks. Animals were palpated weekly for tumor detection. The effects of the hormonal treatments on the circulating concentrations of E2, P4, growth hormone (GH), prolactin (PRL), T4 and insulin-like growth factor-I (IGF-I) after 7 days of treatment, the time of MNU injection, was assessed. Animals treated with E2 had significantly elevated circulation concentrations of GH, PRL and P4, and serum levels of E2 were more consistent in this group than in the other animal groups. P4 treatment caused elevation in P4 concentration in serum but did not affect the circulating levels of other hormones. The proliferation of the mammary gland at the time of MNU injection was elevated in animal groups treated with E2 either alone or with P4 and T4 and in animals treated with P4 alone, but the mammary gland was most differentiated in untreated parous rats and least in animals treated with E2 either alone or with P4 and T4. Mammary tumor incidence was 10% in parous rats that did not receive any hormonal treatment. Treatments with E2 or P4 alone significantly increased the susceptibility of parous animals to 67 and 50.0%, respectively; a tumor incidence similar to that of untreated AMV rats (64%). Parous rats treated with E2 plus P4 had tumor incidence higher than 90%. T4 administered did not affect mammary carcinogenesis.

Expression of MAT1/PEA-15 mRNA isoforms during physiological and neoplastic changes in the mouse mammary gland.

MAT1 is a novel transforming gene which was cloned from a mouse mammary tumor induced by N-methyl-N-nitrosourea in vitro in the presence of lithium as a mitogen. Later, it was found to be identical to the 3' untranslated region (UTR) of the 2.5 kb isoform of PEA-15 (phosphoprotein enriched in astrocytes-15 kDa). We re-cloned MAT1/PEA-15 cDNAs and showed 2.5, 2.0 and 1.8 kb isoforms and confirmed MAT1 localization as reported. The 2.0 and 1.8 kb isoforms were produced by alternative splicing and alternative polyadenylation at the 3' UTR, respectively. To analyze the role of MAT1/PEA-15, we examined the expression of MAT1/PEA-15 mRNA in normal mammary tissues and in mammary tumors. The mammary gland during pregnancy, lactation and weaning showed weak but stable expression. Compared with normal mammary gland, mammary tumors showed stronger expression. Aberrant expression of MAT1/PEA-15 isoforms was found in mouse mammary epithelial cell lines, FSK7 and TM6, which lost the 2.5/2.0 and 2.5 kb isoforms, respectively. In contrast to other oncogenes like c-myc, MAT1/PEA-15 mRNA was extremely stable after actinomycin D and cycloheximide treatments suggesting that other protein expression is prerequisite for degradation of MAT1/PEA-15 mRNA. It evoked the possibility of the 3' UTR of MAT1/PEA-15 (designated as MAT1-T) as a riboregulator in mammary tumorigenesis and necessity for further analysis of human breast cancers as well as mouse mammary tumors.

Isolation of oncogenes from rat mammary tumors by a highly efficient retrovirus expression cloning system.

A majority of mammary tumors induced with N-methyl-N-nitrosourea in rats contain G to A transitional mutation of c-Ha-ras at the 12th codon. Additional oncogene activation is known to be necessary for further tumor progression. To isolate novel oncogenes, we used an expression cloning system utilizing the pMX retroviral vector in combination with BOSC23 packaging cells. First, we elucidated the sensitivity of this system in the NIH 3T3 focus assay; foci were detectable even after 10(-6) dilution using v-Ha-ras, neuT, and beta-galactosidase constructs in pMX vector. This system is sensitive enough to detect low copy number cDNAs. We used the pMX/BOSC23 expression cloning system to clone novel oncogenes from rat mammary tumors harboring an activated c-Ha-ras and isolated several candidate oncogenes that caused transformation of NIH 3T3 cells and/or generated tumors when transplanted to nude mice.

Induction of mammary carcinomas by N-methyl-N-nitrosourea in ovariectomized rats treated with epidermal growth factor.

The importance of epidermal growth factor (EGF) in both normal and malignant mammary gland development are presented in these studies. Initial findings demonstrated that in the absence of ovarian hormones, EGF had a significant proliferative effect on mammary epithelial cells. To determine whether mammary epithelial cells grown with EGF, in the absence of ovarian hormones, could be transformed by N-methyl-N-nitrosourea (MNU), female ovariectomized Lewis rats were implanted with pellets containing EGF for 1 week and then treated with MNU for initiation. Two days after MNU treatment, ovaries were implanted and EGF pellets were removed from all ovariectomized groups in order to promote carcinogenesis. The mammary carcinoma incidence of the EGF-stimulated group (90%) was not significantly different from the intact group (100%). The mammary cancer morphology of EGF-treated carcinomas was either ductal carcinoma or cribriform adenocarcinoma, whereas intact animals developed mainly papillary and occasional cribriform carcinomas. Fifty-eight percent of the carcinomas from the EGF group were ovarian hormone-independent compared with 10% of carcinomas from the intact group. These results demonstrate that EGF-induced proliferation during initiation with MNU was sufficient to induce the transformation of mammary carcinomas in the absence of ovarian hormones. The hormonal dependency of these EGF-induced carcinomas were different compared with MNU-initiated mammary carcinomas in intact rats.

Hormonal prevention of breast cancer: mimicking the protective effect of pregnancy.

Full term pregnancy early in life is the most effective natural protection against breast cancer in women. Rats treated with chemical carcinogen are similarly protected by a previous pregnancy from mammary carcinogenesis. Proliferation and differentiation of the mammary gland does not explain this phenomenon, as shown by the relative ineffectiveness of perphenazine, a potent mitogenic and differentiating agent. Here, we show that short term treatment of nulliparous rats with pregnancy levels of estradiol 17beta and progesterone has high efficacy in protecting them from chemical carcinogen induced mammary cancers. Because the mammary gland is exposed to the highest physiological concentrations of estradiol and progesterone during full term pregnancy, it is these elevated levels of hormones that likely induce protection from mammary cancer. Thus, it appears possible to mimic the protective effects of pregnancy against breast cancer in nulliparous rats by short term specific hormonal intervention.

Defective retrovirus insertion activates c-Ha-ras protooncogene in an MNU-induced rat mammary carcinoma.

Endogenous retrovirus sequences are present in the genome of a wide variety of animal species. The activation of the proto-oncogenes of the ras family, particularly c-Ha-ras, by either point mutation or overexpression, has been shown to be associated with a vast number, of different cancers. here we report that the insertion of a defective retrovirus in the -1 intron of rat c-Ha-ras is responsible for the activation of the gene by over 10-fold overexpression in an MNU-induced rat mammary cancer. A portion of the 3' end of the retroviral sequence is expressed as a part of the c-Ha-ras transcript in the carcinoma tissue, indicating the direct involvement of this element in the transcription of the c-Ha-ras gene. The c-Ha-ras structural gene transcribed by the promoter of the defective retroviral element can neoplastically transform the NIH 3T3 cell line upon transfection.

Changes in the parous rat mammary gland environment are involved in parity-associated protection against mammary carcinogenesis.

Parity in rats results in protection from methylnitrosourea (MNU)-induced mammary cancer. Our goal was to determine if systemic alterations in the mammary gland environment after a full-term pregnancy rendered the parous rat an inadequate host for promotion of initiated mammary epithelial cells to become cancerous. Lewis rat MNU-treated mammary epithelial cells were transplanted into uniparous (UP), age-matched virgin (AMV) (both 130-150 d), or young virgin (YV) (50-60 d) syngeneic hosts to examine if differences in the systemic environments of the three hosts had an effect on hyperplasia and cancer formation. More transplants in YV and AMV hosts contained hyperplasias and adenocarcinomas as compared to transplants in UP hosts. In addition, UP host transplants had significantly fewer numbers of hyperplastic lesions than transplants from the virgin hosts. The evidence presented here shows that the uniparous host environment is less supportive than that of the virgin host for hyperplasia and cancer development.

Cell proliferation and apoptosis during mammary carcinogenesis in pituitary isografted mice.

In the present study, pituitary isografted animals serve as a model for evaluating the changes in differentiation, cell proliferation and programmed cell death (apoptosis) in mammary epithelial cells during carcinogenesis. The percentage of bromodeoxyuridine (BrdU)-labeled ductal and alveolar cells was significantly higher in pituitary isografted animals than in non-isografted control animals. BrdU-labeled cells increased in lobular hyperplastic nodules, keratinized nodules and mammary carcinomas; similar changes were observed with apoptotic cells, which were rare in mammary glands of adult non-isografted animals (one to three apoptotic cells per 2000 mammary epithelial cells), but their number increased in hyperplastic lesions and mammary carcinomas. Among hyperplastic nodular lesions, variants with high, moderate and low proliferative activity and/or apoptotic cell death were identified, which suggests that they may have different growth potentials and different propensities for malignant transformation. After removing pituitary isografts, apoptosis occurs in hyperplastic lesions but not in mammary carcinomas-implying that malignant tumors are hormone-independent. The dynamics of the changes in apoptotic cell death among various hyperplastic lesions after removal of pituitary isografts suggests that these lesions are composed of heterogeneous cell populations, as far as the initiation of apoptosis is concerned. Our data indicate that apoptosis can be used together with cell proliferation as a potential marker in characterizing the growth potential and phenotypic diversity of hyperplastic, premalignant and malignant mammary gland lesions.

N-Ethyl-N-nitrosourea induces mammary cancers in the pituitary-isografted mouse which are histologically and genotypically distinct from those induced by N-methyl-N-nitrosourea.

N-Ethyl-N-nitrosourea (ENU) and N-methyl-N-nitrosourea (MNU) are alkylating agents which respectively ethylate or methylate nucleophilic centers in the cell such as DNA. In vitro studies with naked DNA and bacterial mutagenesis assays suggest that these two compounds induce different spectra of genetic lesions. In addition, the ethyl-DNA adducts induced by ENU persist longer than the methyl-DNA adducts induced by MNU. Since MNU is a known mammary carcinogen in the pituitary-isografted mouse, these data suggest that ENU may be an even more potent carcinogen than MNU. The purpose of this study was to determine whether ENU was a mammary carcinogen in the pituitary-isografted mouse and if so, to compare the genotype and phenotype of ENU-induced mammary tumors with those induced by MNU. Fifteen adult female virgin BALB/c mice were isografted with two pituitaries and subsequently treated with a single intravenous injection of ENU (50 micrograms/g body weight). Mammary adenocarcinomas arose in all of the survivors (n=12) with a median latency of 27 weeks and a mean frequency of 1.4 cancers per mouse. When tumor DNA was analyzed for mutations in the 12th and 61st codons of c-Ki-ras or c-Ha-ras protooncogenes, only wild type sequences were found. This is in contrast to MNU which causes a G to A transition mutation in the 12th codon of the c-Ha-ras proto-oncogene in about one of five mammary cancers induced in pituitary-isografted mice. Furthermore, the ENU-induced tumors were solid viable papillary adenocarcinomas, whereas MNU induced tumors are highly necrotic adenocarcinomas with squamous metaplasia. These results demonstrate that, in the pituitary-isografted mouse, ENU is as potent a mammary carcinogen as MNU and suggest that oncogenes other than c-Ki-ras or c-Ha-ras may be involved in ENU-induced mammary cancers.

Hormone dependent and independent mammary tumor development form N-methyl-N-nitrosourea-treated rat mammary epithelial cell xenografts in the nude mouse: multiple pathways and H-ras activation.

A nude mouse mammary fat pad xenograft system was developed to examine hormone dependent and independent mammary tumorigenesis and progression from N-methyl-N-nitrosourea (MNU)-induced hyperplastic lesions. Ninety-one percent of transplanted mammary tumors grew, with an orthotopic preference, and maintained their hormone dependence, histopathology, and H-ras mutation frequency. Grafted mammary epithelial cells, from MNU-treated rats, developed normal; and hyperplastic outgrowths, representative of those found in the rat mammary gland after MNU-treatment. Hyperplasias developed into neoplasias that were both hormone dependent and independent. We demonstrate that hormone independent tumors can develop directly either from lobuloalveolar or ductal hyperplasias or from hormone dependent tumors. H-ras mutation was detected in mammary preneoplasias (4 lines) before they developed into tumors and was associated with an elevated tumorigenic potential. Our observations suggest that there are multiple histopathogenic pathways in the development and progression to hormone independent rat mammary tumors.

Adenoviral-mediated gene transfer into primary human and mouse mammary epithelial cells in vitro and in vivo.

Primary mammary epithelial cells from both the human and mouse mammary glands can be genetically altered under a variety of situations using the replication-defective adenoviral vector containing a marker gene encoding the E. coli beta-galactosidase. Primary human and mouse mammary epithelial cells in monolayer culture and in three-dimensional collagen gel culture systems were transduced by adenovector at high efficiency. Successful gene transfer was also accomplished in situ and in vivo. In the mouse mammary gland, anatomically restricted gene transfer and expression was demonstrated by micro-injection of adenoviral vector directly into the main duct of the mammary gland. Injection of adenoviral vector directly into the human mammary tissues from reduction mammoplasty specimens, into the mouse mammary gland-free fat pad containing the previously transplanted dissociated human mammary epithelial cells, and intratumorally into the human breast cancer xenografts in nude mice, all resulted in successful gene transfer to human mammary epithelial cells. High efficiency introduction of genetic material into primary mammary epithelial cells is important in the study of mammary carcinogenesis and potentially for gene therapy of human breast cancer.

Refractoriness to mammary tumorigenesis in parous rats: is it caused by persistent changes in the hormonal environment or permanent biochemical alterations in the mammary epithelia?

Administration of a single i.v. injection of 50 mg N-methyl-N-nitrosourea (MNU)/kg body wt to 50- to 60-day old virgin rats, 120-day-old virgin rats, and 120-day-old parous rats (Sprague-Dawley; n = 18-37) resulted in a high incidence of mammary carcinomas in the virgin animals (97.3% in 50- to 60-day-old virgin rats; 75.0% in 120-day-old virgin rats), but mammary carcinomas did not develop in the parous rats. The concentrations in serum of various mammotropic hormones were measured in identical groups of rats at the time of MNU treatment. Growth hormone (GH) concentration was significantly reduced in parous rats, as compared with young or age-matched virgin rats. The concentrations of prolactin, 17 beta-estradiol, progesterone, corticosterone and thyroxine were not significantly altered in the parous rats compared to the two groups of virgin animals. Histological examination of the mammary glands from the three groups of rats showed that the epithelia of the parous animals were in a stage of regression, whereas the mammae of the young virgin rats showed the highest degree of lobulo-alveolar development. The levels of estrogen receptor (ER), epidermal growth factor (EGF) receptor (EGF-R) and GH receptor (GHR) in the mammary glands of the animals were also measured. We found a reduction in the receptor levels for both estrogen and EGF in mammary tissues from parous animals. Receptors for GH were present in normal mammary tissues from both virgin and parous rats. We hypothesize that the reduction in the circulating concentration of GH caused the reduced susceptibility of parous rats to mammary carcinogenesis possibly by decreasing the levels of ER and/or EGF-R in the mammary gland.

Reconstituted human normal breast in nude mice using collagen gel or Matrigel.

Human breast epithelial cells dissociated from reduction mammoplasty specimens were embedded in two commonly used extracellular matrices, type I collagen gel or Matrigel, and subsequently transplanted subcutaneously into athymic nude mice. Histological sections from both types of recovered gels showed epithelial structures arranged as short tubules with some branching as well as preservation of epithelial cell polarity. Proliferation was studied in vivo by 5-bromo-2'-deoxy-uridine labeling followed by immunostaining of sections from recovered gels. Human breast epithelia embedded in collagen gel or Matrigel had similar proliferative activity. Cholera toxin, 17 beta-estradiol, and epidermal growth factor, when tested singly, were growth promoting, and in combination 17,beta-estradiol and cholera toxin had an additive effect but 17,beta-estradiol and epidermal growth factor were not additive. Our model system provides a means to study the endocrine control of normal human breast development.