MCF-7/VD(R): a new vitamin D resistant cell line.

Several in vitro and in vivo experiments have demonstrated potent cell regulatory effects of vitamin D compounds in cancer cells. Moreover, a promising phase I study with the vitamin D analogue Seocalcitol (EB 1089) in patients with advanced breast and colon cancer has already been carried out and more clinical trials evaluating the clinical effectiveness of EB 1089 in other cancer types are in progress (Mørk Hansen et al. [2000a]). However, only little is known about the mechanisms underlying the actions of vitamin D or about the possible development of drug resistance in the patients. Therefore, in an attempt to gain more insight into these aspects, we have developed the MCF-7/VD(R) cell line, a stable subclone of the human MCF-7 breast cancer cell line, which is resistant to the growth inhibitory and apoptosis inducing effects of 1alpha,25(OH)(2)D(3). Despite this characteristic, receptor studies on the VDR have clearly demonstrated that the MCF-7/VD(R) cells contain fully functional VDRs, although in a lower number than seen with the parental MCF-7 cells. The regulation of the 24-hydroxylase enzyme appeared to be intact in the MCF-7/VD(R) cells and no differences with regard to growth rate and morphological appearance between the MCF-7/VD(R) cells and the parental MCF-7 cells were observed. Interestingly, however, the sensitivity of the MCF-7/VD(R) cells to the pure anti-estrogen ICI 182,780 was found to be increased. The MCF-7/VD(R) cell line shows characteristics different from those of previously described vitamin D resistant breast cancer cell lines but also some similarities. Together such vitamin D resistant cell lines therefore serve as a useful tool for studying the exact mechanism of action of vitamin D and the development of vitamin D resistance.

Inhibitory effects of 1alpha,25-dihydroxyvitamin D(3) on the G(1)-S phase-controlling machinery.

The nuclear hormone 1alpha,25-dihydroxyvitamin D(3) induces cell cycle arrest, differentiation, or apoptosis depending on target cell type and state. Although the antiproliferative effect of 1alpha,25-dihydroxyvitamin D(3) has been known for years, the molecular basis of the cell cycle blockade by 1alpha,25-dihydroxyvitamin D(3) remains largely unknown. Here we have investigated the mechanisms underlying the G(1) arrest induced upon 1alpha,25-dihydroxyvitamin D(3) treatment of the human breast cancer cell line MCF-7. Twenty-four-hour exposure of exponentially growing MCF-7 cells to 1alpha,25-dihydroxyvitamin D(3) impeded proliferation by preventing S phase entry, an effect that correlated with appearance of the growth-suppressing, hypophosphorylated form of the retinoblastoma protein (pRb), and modulation of cyclin-dependent kinase (cdk) activities of cdk-4, -6, and -2. Time course immunochemical and biochemical analyses of the cellular and molecular effects of 1alpha,25-dihydroxyvitamin D(3) treatment for up to 6 d revealed a dynamic chain of events, preventing activation of cyclin D1/cdk4, and loss of cyclin D3, which collectively lead to repression of the E2F transcription factors and thus negatively affected cyclin A protein expression. While the observed 10-fold inhibition of cyclin D1/cdk 4-associated kinase activity appeared independent of cdk inhibitors, the activity of cdk 2 decreased about 20-fold, reflecting joint effects of the lower abundance of its cyclin partners and a significant increase of the cdk inhibitor p21(CIP1/WAF1), which blocked the remaining cyclin A(E)/cdk 2 complexes. Together with a rapid down-modulation of the c-Myc oncoprotein in response to 1alpha,25-dihydroxyvitamin D(3), these results demonstrate that 1alpha,25-dihydroxyvitamin D(3) inhibits cell proliferation by targeting several key regulators governing the G(1)/S transition.

Estradiol down regulates expression of vasoactive intestinal polypeptide receptor type-1 in breast cancer cell lines.

Three breast carcinoma cell lines were tested for 17beta-estradiol (E(2)) mediated regulation of vasoactive intestinal polypeptide receptor type-1 (VPAC(1)) expression. In all three, E(2) was found to down-regulate the mRNA level. We studied T47D cells in more details and found a 25 and 70% decrease in the VPAC(1) mRNA level upon 7 and 48 h of E(2) treatment, respectively. The number of vasoactive intestinal polypeptide (VIP) binding sites was reduced 66% upon treatment with E(2) for 72 h. After cycloheximide pretreatment, the E(2) mediated mRNA reduction was attenuated from 50% to 25% after 24 h suggesting the effect to be at least partly independent of protein synthesis. Experiments with the transcriptional inhibitor actinomycin D showed that E(2) did not influence the VPAC(1) mRNA half-life while nuclear run-on experiments indicated that E(2) decreased the VPAC(1) transcription rate. Two antiestrogens: ICI 182780 (ICI) and 4-hydroxy-tamoxifen (4-OHT) mediated a concentration dependent inhibition of E(2)'s effect on the mRNA level. Transient transfection with reporter-gene constructs containing various portions of the VPAC(1) 5'-flanking sequence revealed the most proximal 100 bp to be essential for the basal transcriptional activity. However, E(2) did not influence the expression of the reporter gene using up to 3250 bp of the VPAC(1) 5'-flariking region.

Downregulation of VPAC1R expression in breast cancer cell lines.

The breast carcinoma cell line T47D was tested for 17 beta-estradiol (E2) mediated regulation of vasoactive intestinal polypeptide receptor type-1 (VPAC1) expression. E2 was found to downregulate the mRNA level. The number of VIP binding sites was reduced 66% on treatment with E2 for 72 h. Experiments with cycloheximide suggested that the effect was independent (at least partly so) of protein synthesis. Experiments with the transcriptional inhibitor, actinomycin D, showed that E2 did not influence the VPAC1 mRNA halflife. Both of two antiestrogens, ICI 182,780 and 4-hydroxy-tamoxifen, mediated a concentration dependent inhibition of the effect of E2 on the mRNA level. Transient transfection with reporter-gene constructs containing various portions of the VPAC1 5'-flanking sequence revealed the most proximal 100 bp to be essential for the basal transcriptional activity. However, E2 did not influence the expression of the reporter gene using up to 3,250 bp of the VPAC1 5'-flanking region.

The estrogen receptor variant lacking exon 5 has dominant negative activity in the human breast epithelial cell line HMT-3522S1.

The estrogen receptor variant lacking exon 5 (ERdeltaE5) encodes a truncated protein that lacks the hormone-binding domain and has been suggested to be responsible for the estrogen-independent growth of human breast tumors and resistance to antiestrogen therapy. The biological function of the ERdeltaE5 in human breast epithelial cells has been studied by transient transfection of HMT-3522S1 cells with wild-type (wt) estrogen receptor (ER) and ERdeltaE5. A 10-fold higher expression of ERdeltaE5 mRNA compared to that of wt ER mRNA was found. However, the expression of ERdeltaE5 protein was significantly lower than the expression of wt ER protein. The ERdeltaE5 was unable to activate the transcription of an estrogen-responsive reporter gene in the absence as well as in the presence of estrogen. The ERdeltaE5 disclosed a dominant negative activity when expressed together with wt ER. These data indicate that the biological significance of ERdeltaE5 in human breast cancer is dubious.

Down-regulation of laminin-binding integrins by 1 alpha,25-dihydroxyvitamin D3 in human melanoma cells in vitro.

In the present investigation the effect of 1 alpha,25(OH)2D3 on the expression of the integrin laminin receptor on the melanoma cell line SK-MEL-28 has been examined. The SK-MEL-28 cells were shown to contain high-affinity receptors for 1 alpha,25(OH)2D3 and cell proliferation was found to be inhibited in a dose-dependent manner in response to the hormone. Using monoclonal antibodies against the alpha 6-sub-unit of the integrin laminin receptor, immunocytochemistry demonstrated that exposure of cells to 1 alpha,25(OH)2D3 for 5 days caused a reduced staining intensity. This observation was further confirmed by dot blot analysis, where a dose-dependent decline of alpha 6 expression was obtained after treatment of the cells with 1 alpha,25(OH)2D3 for 6 days. FACS-analysis was performed in order to quantify this decline, and it was found that the level of alpha 6-subunits on the cell surface was reduced by more than 40%. Additional investigations including Northern blot analyses of poly(A)+RNA extracts also showed a dose-dependent reduction of alpha 6 mRNA. Interestingly, the decrease of alpha 6 expression on the surface of SK-MEL-28 melanoma cells was accompanied by a reduced ability of the cells to adhere to an artificial basement membrane. In conclusion, the present investigation shows that besides having an antiproliferative effect on the SK-MEL-28 melanoma cells, 1 alpha,25(OH)2D3 is also able to inhibit the surface expression of the alpha 6-subunit of the integrin laminin receptor. Moreover, the results strongly indicate that 1 alpha,25(OH)2D3 exerts its regulatory effect on the alpha 6-subunit at the transcriptional level rather than at the protein level.

Resistance of human breast-cancer cells to the pure steroidal anti-estrogen ICI 182,780 is not associated with a general loss of estrogen-receptor expression or lack of estrogen responsiveness.

To elucidate the mechanisms responsible for the development of anti-estrogen resistance, we have cloned and established 3 stable ICI-182,780-resistant sub-lines, MCF-7/182R-1, MCF-7/182R-6 and MCF-7/182R-7 from the estrogen-receptor(ER)-positive and estrogen-responsive human breast-cancer MCF-7 cell line by long-term treatment with 10(-7) M ICI 182,780. The ICI-182,780-resistant MCF-7 sub-lines express ER, but compared with MCF-7 cells the level is significantly lower in all 3 sub-lines. In the MCF-7 cell line we find that ER expression is regulated by estrogen and anti-estrogens at the transcriptional and post-transcriptional level. This is in contrast to the ICI-182,780-resistant sub-lines, in which we find very little hormonal effects on the ER mRNA expression level. The resistant sub-lines also deviate from parent characteristics by the complete lack of expression of progesterone receptor even when grown in the presence of estradiol. All 3 resistant sub-lines have a lower basal expression of cathepsin-D mRNA comparable with the lower ER expression, but, in contrast, they have higher basal expression of the pS2 mRNA than the parent MCF-7 cell line. Although there are different basal expression levels of the pS2 and cathepsin-D genes, the resistant sub-lines behave like the parent MCF-7 cell line with respect to the hormonal regulation of both genes. The estrogen receptors in the resistant sub-lines have also maintained wild-type characteristics with respect to estrogen and anti-estrogen regulation of the estrogen-regulated proteins procathepsin D, alpha1-antitrypsin and a 42-kDa protein. The resistant cells require estrogen for growth in athymic nude mice. Our results clearly demonstrate that the ER in the resistant sub-lines have a normal function for most parameters investigated, supporting our earlier observation that only wild-type ER protein is expressed in these cells. The few observed differences in ER function between the parent MCF-7 cell line and the resistant sub-lines are not likely to be responsible for the ICI-182,780-resistant phenotype.

Molecular cytogenetic analysis of a nontumorigenic human breast epithelial cell line that eventually turns tumorigenic: validation of an analytical approach combining karyotyping, comparative genomic hybridization, chromosome painting, and single-locus fluorescence in situ hybridization.

The immortalized, nontumorigenic human breast epithelial cell line HMT-3522 has been used as a model for premalignant and, eventually, malignant development. During cultivation, the karyotype evolution was followed. At an early stage, a very long constant phase showed a near-diploid karyotype, with only five marker chromosomes. DNA from this phase was used for comparative genomic hybridization (CGH) analysis, confirming a previously known MYC amplification, and the integration sites were subsequently determined by single-locus fluorescence in situ hybridization (FISH). Furthermore, gains of 5q22-qter and 20q11-qter and deletion of most of chromosome 6 (6p23-qter) were detected by CGH. Because of uncertainty about some of the indicated changes, including a deletion of Ip35-pter, the CGH findings were investigated more closely by chromosome painting, leading to a revision of the karyotype: 45,XX,del(I)(p35),-6,dup(8)(pter-->qter::qter-->q24),der(12) t(6;12)(p23; p13),der(14)t(5;14)(q22;q32.3),der(17)t(8;17;20)(17pter-->17q25 ::8qter--> 8q23::8q24-->8qter::8q24-->8qter:: 8q23-->8q24.1::20q11-->20qter). Some karyotypic changes were confirmed by CGH; others had to be revised; and, in the Ip35 region, classical cytogenetics seems superior to CGH. However, CGH revealed a karyotypically unsuspected dup(20q) that might be of special relevance to breast tumor initiation or progression. Our study confirms that CGH is supplementary to current technologies, e.g., karyotyping and Southern analysis, but cannot replace them. In addition, our cell line turned out to be an excellent model for comparison among the different methods. The results imply that future cytogenetic analyses of complex karyotypes should be based on a combination of karyotyping, CGH, and FISH.

Molecular aberrations in the MHC class I-restricted pathway for antigen presentation in methylcholanthrene sarcomas from nude mice: discrepancies between MHC mRNA and surface protein.

In a previous study, we demonstrated that eight sarcomas induced by chemical carcinogenesis in nude mice were rejected by syngeneic immunocompetent recipients at a much higher rate than eight sarcomas induced with the same method in syngeneic immmunocompetent mice. In the present study, we investigated these 16 sarcomas for structural and quantitative aberrations in components of the MHC class I-restricted antigen-processing and -presentation pathway. Considerable discrepancies between mRNA levels and cell surface protein expression of MHC class I (Kd, Dd and Ld) molecules were observed almost exclusively in the tumours derived from nude mice. Several of the nude mouse-derived tumours also displayed incongruent levels of heavy chain mRNA and beta2-microglobulin mRNA. These findings are taken as indications of abnormal regulation of gene transcription in nude mouse tumours, and if this abnormal regulation extends to the entire genome, it may explain the pronounced immunogenicity of these tumours. Proteasome composition, heat shock protein expression, TAP-molecule inducibility and intercellular adhesion molecule-1 expression were investigated in the same tumours. We found no indications of structural defects or quantitative differences in these molecules between the two groups of tumours.

Estrogen receptor messenger RNA splice variants are not involved in antiestrogen resistance in sublines of MCF-7 human breast cancer cells.

Development of resistance to tamoxifen is a serious problem in treatment of breast cancer patients. Although the mechanisms for development of resistance are unclear, an altered expression of alternatively spliced estrogen receptor (ER) mRNA has been suggested to be involved. We have looked for differential expression of ER splice variants lacking exon 2 (ERdeltaE2), exon 3 (ERdeltaE3), exon 4 (ERdeltaE4), exon 5 (ERdeltaE5), exon 7 (ERdeltaE7), and exons 4 and 7 (ERdeltaE4, 7) in the human breast cancer cell line MCF-7 and 10 ER-positive MCF-7 sublines resistant to the antiestrogens tamoxifen, ICI 164,384 or ICI 182,780. No major differences in the expression were demonstrated between MCF-7 cells and resistant cells, indicating that ER splice variants are not involved in antiestrogen resistance in this model system. Furthermore, despite a high mRNA level of some of the ER splice variants, no corresponding proteins could be detected using Western blot analysis.

Characterization of a nontumorigenic human breast epithelial cell line stably transfected with the human estrogen receptor (ER) cDNA.

Estrogens play an important role in breast cancer and the effect of estrogen on growth of breast cancer cells has been extensively studied. However, only little information is available about the response of normal breast epithelial cells to estrogen, mainly due to the difficulties in establishing estrogen receptor (ER)-positive human breast epithelial cells in culture. We have stably transfected the human estrogen receptor (hER) wt cDNA into the ER-negative, spontaneously immortalized human breast epithelial cell line, HMT-3522S1, in order to develop a model for studying the effect of estrogen on nonmalignant human breast epithelial cells. Characterization of the transfected clone F9 confirmed incorporation of the estrogen receptor gene in the genome, expression of hER mRNA and hER protein. However, proliferation of F9 cells was inhibited by both estradiol (E2) and tamoxifen, whereas the pure antiestrogen ICI 182,780 had no effect on cell proliferation. This seems paradoxical since E2 stimulated the expression of the endogenous genes, TGF-alpha, cathepsin D, and alpha1-antitrypsin. In breast cancer cell lines, high expression of these genes is correlated to estrogen-stimulated cell proliferation. The spontaneously immortalized HMT-3522S1 cells transfected with wt ER cDNA behave similarly to cell lines from nonmalignant breast tissue immortalized by carcinogens and transfected with mutated ER cDNA as described by others. The discrepancy between growth inhibition and induction of positive growth factors by E2 indicates that either ER-positive nonmalignant breast epithelial cells are growth-inhibited by E2 in contrast to malignant cells or that introduction of the ER into ER-negative cells is not sufficient for restoring "normal' estrogen responsiveness.

Trisomy 7p and malignant transformation of human breast epithelial cells following epidermal growth factor withdrawal.

We have reported previously on the first spontaneously immortalized, nonmalignant human breast epithelial cell line, HMT-3522, which is entirely dependent on exogenous epidermal growth factor (EGF). In passage 118, cells were adapted to grow in medium without EGF and a new growth-transformed subline, HMT-3522/gt-1, was generated and propagated at high growth rate without exogenous EGF (Madsen et al., Cancer Res., 52: 1210-1217, 1992). Here we have used this subline and the continuum of the parent line, HMT-3522/wt, to pose the question whether a relevant change in a physiological parameter of the microenvironment will induce malignant transformation. The two cell lines were cultured under identical conditions with the only exception that EGF was omitted in the medium for gt-1. Initially, wt and gt-1 were identical in terms of karyotype as well as morphology, growth rate, and protein expression as revealed by two-dimensional gel electrophoresis. A highly dramatic shift to phenotype was observed in passage 238 when the gt-1 line became tumorigenic in nude mice. After two mouse-culture passages, the resulting malignantly transformed cell line (HMT-3522/mt-1) was refractory to the growth-modulating effect of EGF and presented an extra copy of a chromosome marker, 7q-, as the only cytogenetic difference from the gt-1. Our results suggest that microenvironmental cues are powerful factors in the induction of malignancy. A major role of EGF receptor in the malignant transformation is emphasized by loss of EGF sensitivity and acquisition of an extra chromosome 7p harboring the EGF receptor gene. We hypothesize that during premalignant hyperplasia, a population of EGF/transforming growth factor alpha autonomous epithelial cells in situ may develop as a consequence of local transforming growth factor alpha deprivation, a condition reflected in the culture model as autonomy after EGF withdrawal.

Differential expression of estrogen receptor mRNA splice variants in the tamoxifen resistant human breast cancer cell line, MCF-7/TAMR-1 compared to the parental MCF-7 cell line.

Breast cancer patients with an estrogen receptor (ER) positive tumor can be treated with the anti-estrogen tamoxifen, but development of anti-estrogen resistance is a serious problem. We have analyzed a tamoxifen resistant human breast cancer cell line MCF-7/TAMR-1 for alterations in ER which might explain the tamoxifen resistance. The MCF-7/TAMR-1 cells expressed both wild-type ER mRNA and protein, and by RT-PCR we were able to clone ER cDNAs corresponding to the following mRNA splice variants: ER delta E2, ER delta E4, ER delta E5, ER delta E7 and a new double splice variant lacking both exon 4 and 7 (ER delta E4,7) The existence of the ER delta E4,7 variant was confirmed by RNase protection assay. Semi-quantitative RT-PCR revealed that ER delta E2 mRNA was expressed at a higher level in MCF-7/TAMR-1 cells, whereas the ER delta E5 mRNA was expressed at a significantly lower level in MCF-7/TAMR-1 cells compared with MCF-7 cells. The differential expression of the two ER mRNA splice variants indicates that they may be involved in anti-estrogen resistance, although the present knowledge of their biological function does not provide us with an explanation.

Growth factor requirement, oncogene expression and TP53 mutations of a tumorigenic and a non-tumorigenic subline of the human breast carcinoma cell line, HMT-3909.

From a human breast carcinoma cell line, HMT-3909, a tumorigenic and a non-tumorigenic subline have previously been described. Cells of both sublines have been characterised as carcinoma cells. In the present work we examined whether differences in growth factor requirements or oncogene expression may explain the difference in tumorigenicity. We found that exogenous growth factor dependence discriminated between the two sublines. No alterations in oncogenes or tumour suppressor genes were demonstrated that could explain the differences in tumorigenicity. The lower growth factor requirement and the higher growth rate of the tumorigenic subline indicates that, in these cells, growth potential may determine the outcome of the tumorigenicity assay.

In vitro karyotype evolution and cytogenetic instability in the non-tumorigenic human breast epithelial cell line HMT-3522.

The "spontaneously" immortalized cell line HMT-3522, derived from a fibrocystic breast lesion, is used as a model for premalignant breast epithelium. During 205 passages the cytogenetic evolution was followed. The results were compared with our earlier results on oncogene expression and growth factor requirements. During in vitro growth, gain and loss of markers, loss of normal chromosomes, and duplication of the chromosome complement could be demonstrated. The variability increased during in vitro growth. This variability, probably created randomly, leads to cells with different growth capacities, from which sidelines may be selected and become stemlines. The karyotypic evolution (including polyploidization) demonstrated here may be a result of genetic instability and heterogeneity. Although tumorigenicity was not achieved, either due to lack of cancer-specific gene alterations or to lack of proper selection pressure, the results suggest an ongoing process towards malignancy.

Gradual selection of a cellular clone presenting a mutation at codon 179 of the p53 gene during establishment of the immortalized human breast epithelial cell line HMT-3522.

We studied the occurrence of a p53 mutation along passages stored as frozen vials during establishment of a nontumorigenic human mammary epithelial cell line HMT-3522. Mutations were identified by a PCR-SSCP approach using DNA as a template. The mutation, a nonconservative nucleotide substitution at codon 179 changing a histidine into an asparagine, appeared between passages 51 and 63 and was concommitant to a change in growth conditions. Cells were no longer grown on collagen coat and cell growth was not responsive to insulin, transferrin, or hydrocortisone anymore. To assess if the mutation was an early or a late event during cell line evolution we put a vial of cells frozen at passage 30 back into culture and tested for the appearance of a p53 mutation along newly produced passages. The same mutation (His to Asp at codon 179), as previously identified, reemerged between passages 48 and 52, thus indicating that the mutation was preexisting in passage 30 and gradually selected out because of the growth advantage it conferred. In order to gain in sensitivity we used a RFLP approach on PCR fragments which allowed us to detect the mutation as early as passage 44. Hence it took 14 passages (approx 50 cell doublings) for the mutated cells to become detectable and another 9 passages (33 generations) to overgrow the wild-type component of the population. We calculated that the mutated cells acquired a growth advantage which allowed them to cycle 1.2 +/- 0.05 faster than wild type. Computer simulations were consistent with the mutation appearing at passage 20.

Altered expression of estrogen-regulated genes in a tamoxifen-resistant and ICI 164,384 and ICI 182,780 sensitive human breast cancer cell line, MCF-7/TAMR-1.

A stable, tamoxifen-resistant subline, MCF-7/TAMR-1, of the human breast cancer cell line MCF-7 has been established in tissue culture after long-term treatment with 10(-6) M tamoxifen. The MCF-7/TAMR-1 cell line grows equally well in the presence and absence of tamoxifen, whereas the steroidal antiestrogens ICI 164,384 and ICI 182,780 exert profound inhibitory activity on cell proliferation, although higher concentrations are required to inhibit these cells compared to the parent cells. The MCF-7/TAMR-1 cells grown in tissue culture deviate from parent characteristics by the complete lack of expression of progesterone receptors even when grown with estradiol, by an altered tamoxifen regulation of M(r) 52,000 cathepsin D synthesis and secretion, and by lack of tamoxifen stimulation of an estradiol down-regulated M(r) 42,000 protein with presumed growth inhibitory function. MCF-7/TAMR-1 cells are estrogen receptor positive. The estrogen receptors have wild-type characteristics with respect to (a) binding of estradiol, tamoxifen, and ICI 164,384; (b) estrogen and antiestrogen regulation of the estradiol-regulated proteins pS2, M(r) 61,000 alpha 1-antitrypsin-like protein, M(r) 66,000 alpha 1-antichymotrypsin-like protein, and corresponding mRNAs; and (c) estrogen and antiestrogen regulation of a transiently transfected estrogen responsive reporter gene. We suggest that the lack of tamoxifen up-regulation of the M(r) 42,000 protein synthesis in MCF-7/TAMR-1 cells may at least partly explain the resistance to tamoxifen treatment. The sensitivity to the growth inhibitory activity of ICI 164,384 and ICI 182,780 may be ascribed to the maintenance of the pure antagonistic effect of these steroidal antiestrogens on MCF-7/TAMR-1 cells. Our results indicate that treatment with pure antiestrogens may be effective when patients become refractory to tamoxifen therapy.

Altered gene expression of c-myc, epidermal growth factor receptor, transforming growth factor-alpha, and c-erb-B2 in an immortalized human breast epithelial cell line, HMT-3522, is associated with decreased growth factor requirements.

Activation of protooncogenes and constitutive secretion of autocrine growth factors are thought to be involved in the uncontrolled growth of cancer cells. We have attempted to elucidate the role of oncogenes and growth factors in the premalignant progression of human breast epithelial cells by using an immortalized, nontumorigenic, near-diploid human mammary epithelial cell line, HMT-3522, derived from a fibrocystic lesion and established in our laboratory. During propagation in tissue culture, the growth factor requirements of the HMT-3522 cells decreased simultaneously with an amplification and overexpression of the c-myc protooncogene. Other protooncogenes related to human breast cancer were unaltered with regard to gene copy number and expression. In passage 118, in which the most important growth factor still was epidermal growth factor (EGF), we were able to isolate an EGF-independent subline (S2). The EGF independence of S2 was accompanied by an overexpression of the mRNAs for epidermal growth factor receptor (EGF-R), transforming growth factor-alpha, and c-erb-B2 as compared to the EGF-dependent subline (S1). Moreover, by application of a blocking anti-EGF-R antibody, growth of S2 cells in EGF-free medium was inhibited significantly, indicating that EGF-R was involved in an autocrine loop probably with transforming growth factor-alpha as ligand. Neither the late passages of S1 cells nor S2 cells were tumorigenic after subcutaneous transplantation to athymic mice. Our results indicate that c-myc amplification and overexpression are correlated with a decreased requirement for growth factors. Even when these alterations are combined with immortalization and EGF independence, they are insufficient for malignant transformation of these human breast epithelial cells.

Differential tumorigenicity of two autologous human breast carcinoma cell lines, HMT-3909S1 and HMT-3909S8, established in serum-free medium.

In a serum-free medium we have established two new human breast carcinoma cell lines from a single primary tumor. Cultures were maintained on chemically defined medium CDM3 or on minor modifications of this medium, Dulbecco's modified Eagle medium-Ham's F12 supplemented with epidermal growth factor, insulin, transferrin, estradiol, hydrocortisone, triiodothyronine, cyclic AMP, phosphoethanolamine, ethanolamine, fibronectin, fetuin, ascorbic acid, bovine serum albumin, and trace element salts including selenite (Petersen and van Deurs, Cancer Res., 47: 856-866, 1987). Primary cultures comprised both NADPH-neotetrazolium reductase-positive carcinoma cells and NADPH-neotetrazolium reductase-negative cells of stromal appearance, as well as normal epithelial cells (Petersen and van Deurs, Cancer Res., 46: 2013-2020, 1986). In subsequent passages the cells were monitored exclusively using the tumorigenicity assay on nude mice. Two cell lines, one nontumorigenic, HMT-3909S1, and one tumorigenic, HMT-3909S8, were selected from the primary cultures. Selection of S8 through subline S4 required transient supplementation of CDM3 with fetal calf serum. Permanent lines S1 and S8 were maintained on serum-free medium. Further characterization of the two cell lines in terms of normal breast gland differentiation (Petersen and van Deurs, Differentiation, 39: 197-215, 1988) was carried out using immunocytochemistry, immunochemistry, electron microscopy, and cytogenetics. S8 appeared to be identical with the NADPH-neotetrazolium reductase-positive carcinoma cells of the primary cultures, with a particular subpopulation of carcinoma cells in the tumor of origin, and with the tumorigenic cells of the nude mice. This subline was aneuploid, typically epithelial in morphology, and expressed keratins K8 and K18 and the glycoprotein MAM-6, typical of luminal epithelial cells in the normal breast gland. Subline S1 appeared more like the elongated cells in the primary cultures and like a second subpopulation of cells in the carcinoma of origin. However, S1 cells were in fact epithelial, since they expressed keratins. Also, S1 cells seemed to be a triploidation of a cell with close resemblance to S4, while only few cytogenetic differences were found between S4 and S8, suggesting an origin of S1 and S8 via S4 from a single hypothetical stem cell.