Introduction
My long-term interest has always been in development and characterization of the HMEC system - to make it easier and more appealing for studies to be done with human epithelial cells. I remain interested in all work done with these cells, and encourage everyone to send me preprints of publications, meetings abstracts, and experimental results on an ongoing basis. I welcome all suggestions on what you would find most useful in terms of newsletter/E-mail formats and content, and resource sharing.
The following are the categories that seemed to fall out from the material on hand, as well as some commentaries on my part.
Growth factors, hormones, and the cell cycle; Cellular senescence.
For many years, one of my goals has been to address questions about cell cycle function and control in normal finite lifespan human epithelial cells, and to compare normal (cycling and senescent) and immortal cells for their cell cycle behavior. To do this, I wanted to find a method of synchronizing the HMEC that did not require the use of metabolic inhibitors or general starvation (to minimize possible secondary effects). Our recent publication [see reference #1 below] details how a Go-like arrest can be achieved by blocking the EGF receptor signal transduction. This method appears to be non-stressful, at least in that the cells remain viable and the arrest is readily reversible even after 2 weeks. For all cell cycle enthusiasts, this is really a very simple method. The hybridoma producing blocking monoclonal antibodies to the EGF receptor can be obtained from ATCC.
I am particularly intrigued by the roles played by TGFa/EGF and TGFß in the normal physiology of HMEC. The EGF receptor offers a one point control for cell growth in these cells - in that absence of signal transduction through this receptor sends the cells into Go regardless of what other growth factors are present. TGFa/EGF is also involved in cell motility [3], presumably involving expression of various integrins and possibly cell-cell adhesion proteins. The pathway from signal transduction to observed cell behavior is obviously an area of active research. In addition to signaling via ras pathways, recent studies in other systems suggest that some of TGFa/EGF action could occur via the arachadonic acid pathway. The pathways by which TGFa/EGF influences cell cycle progression and cell-cell interactions in these HMEC, and how these pathways may change during the course of transformation, are now being studied by several groups.
One of the aspects of TGFß that continues to puzzle me is the plasticity in the growth inhibitory response (i.e., the rapidity and extent to which cells display variability in their growth inhibition) [7, & Hosobuchi & Stampfer, In Vitro 1989]. I welcome potential mechanistic explanations for this, suggested experiments, and anyone wishing to pursue this question. One of the few all-or-none differences that I can detect between the finite lifespan cells and the immortally transformed cell lines is that I have never seen a normal cell continue growth in the presence of TGFß, whereas the cell lines readily give rise to populations that can maintain growth in the presence of TGFß.
Tumor supressors and oncogenes.
Examination of 184A1 and 184B5 has thus far not shown any detectable abnormalities in p53 or Rb behavior or sequence [18,19,23]. However, the normal finite lifespan HMEC in culture as well as the established cell lines show p53 with a long half life relative to what has been reported in rodent cells, and a conformation recognized by antibodies to "mutant" p53. The absence of detectable p53 abnormalities in the immortal cell lines is unusual. Other human epithelial cell lines examined have shown some defects in p53, or were immortalized by use of oncogenes that compromise p53 function. Just because we haven't detected it yet doesn't mean that some aspect of p53, or the p53 functional pathway (e.g., some downstream activity) isn't abnormal in these cells. br>
Several groups have now shown that 184B5 transfected with and overexpressing erbB-2/HER-2/neu [20-22, our unpublished results] is not readily transformed to tumorigenicity, although they acquire anchorage independence. However, transfection with the mutated erbB-2 does render the cells tumorigenic. We [Blam & Stampfer, unpublished] were not able to detect any phenotypic change when normal specimen 48R was transfected with and overexpressed normal erbB-2.
An additional comment about transfection: In two cases, groups have transfected 184B5 with two separate genes, for the estrogen receptor [10,11] and erbB-2 [22] and seen high levels of protein expression from the transfected genes. However, in both cases, the transfected cells had properties that differed from the properties seen in cells which normally express those genes, e.g., 184B5 expressing the ER showed reduced growth when exposed to estrogen, and 184B5 overexpressing erbB-2 still produced heregulin (the erbB-2 ligand). These results suggest to me that some caution is needed in analyzing results of transfection studies. The addition of the specific gene can confer properties that directly relate to the presence of that gene (e.g., binding of the appropriate ligand and some downstream effects). Nonetheless, the transfected cell is not equivalent to the cell type which normally expresses that gene.
I would be interested in hearing from others what techniques they have been using to transfect these cells, and what efficiencies of transfection have been achieved. We know that electroporation and calcium phosphate are feasible, but the efficiencies are very low, particularly in the normal HMEC. We are just trying Lipofectamine (Life Technologies) now.
This is a recently identified protein with extensive homology to calmodulin that is expressed in human stratified and pseudostratified epithelium, and downregulated during transformation. Two main groups are working in this area - Paul Yaswen and myself at LBL and Emanuel Strehler and his group now at the Mayo Foundation [25-28]. The calmodulin like protein may possibly play a role in mammary cell differentiation and carcinogenesis.
Transformation and carcinogenesis.
It seems to me that the ideal model system for studying carcinogenesis in vitro has yet to emerge. On the one hand, several groups (including ourselves) have been able to take normal cells through all the stages to tumorigenicity. This model provides "syngeneic" cells for comparison [29-31, 33, 36], but the relationship of the in vitro tumorigenically transformed cells to real in vivo tumor cells is questionable, particularly since most of the oncogenes used (e.g., mutated ras, mutated erbB-2, SV40-T, papilloma virus E7) are not found in human breast cancers. The relationship of immortally transformed post-crisis, long-term cultured tumor cells lines to in vivo primary tumor cells can also be questioned. The ability to be more successful in growing bone fide tumor cells from tumor tissues would help this situation. Several groups are working on this, with some limited success [Ethier, Dairkee, Bergstraesser]. However, because of the heterogeneity of primary tumor populations, it can be hard to know definitively what's growing. Also, there may be "field " influences of the tumor on non-tumor cells. For example, many years ago we found that although early passage tumor-derived cells sometimes exhibited anchorage independent growth, while reduction mammoplasty derived cells did not, we could also detect anchorage independent growth in some non-tumor mastectomy derived populations [Stampfer et al. CSH Conf. Cell Prol. 9, 1982). I would suggest that karyology could be a good marker, since most tumors are aneuploid.
Mutagenesis, carcinogen metabolism, and toxicity.
Another of my early hopes for an HMEC system was that it could be used to detect environmental carcinogens and mutagens. While there's still a way to go on this (e.g., its hard to detect carcinogens without a reliable, quantitative assay for malignant transformation), a number of groups have been producing significant data relative to mutagenesis and toxicity [39-45].
A couple of reminders:
Please remember to use the proper names of the cells in publications. For example, the immortal cell lines are called 184A1 and 184B5. The specimen ID is listed on the mailing sheets; it is not the FDN! The FDN is just a short code to identify ampoules for the freezer and computer. It has no meaning as an identifier of a cell type. Feel free to call if you have a question about this.
ABSTRACTS
Growth Factors, Hormones, and the Cell Cycle
[1] Stampfer, M. R., Pan, C. H., Hosoda, J., Bartholomew, J., Mendelsohn, J., Yaswen, P., Blockage of EGF receptor signal transduction causes reversible arrest of normal and transformed human mammary epithelial cells with synchronous reentry into the cell cycle. Exp. Cell Res., 208: 175-188, 1993.
We demonstrate that blockage of EGF receptor signal transduction is sufficient by itself to cause a rapid, efficient, and reversible Go-like growth arrest of normal human mammary epithelial cells (HMEC) of finite lifespan as well as two immortally transformed cell lines derived from normal HMEC following in vitro transformation with benzo(a)pyrene. For normal HMEC, the significant level of endogenous production of TGFa requires utilization of blocking antibodies to the EGF receptor to achieve cessation of growth in mass culture, whereas removal of EGF is sufficient to arrest the immortal cell lines. In the growth arrested cells, protein synthesis remains depressed; re-exposure to EGF leads to a rapid increase in protein synthesis. Inhibition of DNA synthesis is not detectable until approximately 12 hours after removal of EGF/TGFa, and is pronounced by 24 hours. Re-exposure to EGF produces high levels of synthesis of the early response genes; c-myc, c-fos, c-jun, and MGSA, within one hour. DNA synthesis increases only after 10 hrs, with a sharp peak after 15-20 hours. Re-exposure of the growth arrested normal HMEC for 1 hour with EGF allows a majority of the cells capable of cycling to subsequently enter the S phase. Little is currently known about
cell cycle control in normal human epithelial cells. The efficient and gentle method of achieving reversible Go growth arrest reported here may facilitate studies on the cell cycle of this cell type. Additionally, results from normal HMEC can be compared with syngeneic immortalized cell populations to determine possible cell cycle parameters altered as a result of immortal transformation.
[2] Bates, S. E., Valverius, E., Ennis, B. W., Bronzert, D. A., Sheridan, J. P., Stampfer, M., Mendelsohn, J., Lippman, M. E., Dickson, R. B., Expression of the TGFa/EGF receptor pathway in normal human breast epithelial cells. Endocrin., 126: 596-607, 1990.
To better understand the possible roles and interactions of transforming growth factor-a (TGFa) and its receptor, the epidermal growth factor (EGF) receptor in human breast epithelium, we have studied the expression of TGFa and the EGF receptor in a series of normal human mammary epithelial cells derived from reduction mammoplasty before in vitro propagation, during short term proliferation in vitro, and after immortalization. Increased TGFa mRNA expression coincided with conversion of the cells to a proliferative state in vitro. After establishment, propagation, and proliferation in vitro, the cells expressed high levels of both TGFa and EGF receptor mRNAs. Addition of diverse growth inhibitory agents, including 12-O-tetradecanoylphorbol-l3-acetate (TPA), TGFa, and sodium butyrate, to one of these rapidly proliferating cell populations (no. 184) failed to reduce the expression of either TGFa or the EGF receptor. Likewise, cessation of growth associated with both senescence and confluence of the 184 cells did not result in reduced expression. However, regulation of TGFa mRNA could be demonstrated by withdrawal of EGF from the medium or by antibody-mediated blockade of the EGF receptor in 184 cells. Antibody-mediated EGF receptor blockade also results in inhibition of growth and [3H]thymidine labeling. An autoregulatory autocrine loop appears operant in proliferating breast epithelial cells. Both growth and levels of TGFa mRNA expression are controlled by binding of ligand to the EGF receptor. These studies suggest a role for the TGFa/EGF receptor pathway in normal breast cell physiology.
[3] Matthay, MA, Thiery, J-P, Lafont, F, Stampfer, MR, and Boyer, B, Transient effect of epidermal growth factor on the motility of an immortalized mammary epithelial cell line, J. Cell Sci. (in press)
The effects of growth factors on epithelial cell motility and dispersion have been examined on an immortalized human mammary epithelial cell line, the 184A1 nontumorigenic cell line. Among all the molecules tested, epidermal growth factor (EGF) and transforming growth factor-a (TGF-a) were demonstrated to stimulate an increase in mammary epithelial cell motility and wound closure that was associated with a morphological transformation of the cells and was accompanied by modifications in cell-cell and cell-substrate adhesion systems. The EGF-induced increase in cell motility and monolayer wound closure occurred over a 24 hour period and was not dependent on an increase in cell numbers. The effect of EGF was abolished by inhibiting a2 integrins with specific antibodies, indicating that part of the mechanism for the increase in cell motility and accelerated wound closure depends on a2 integrin functional expression. After 72 hours of exposure to EGF, the EGF-induced alterations in cell morphology, motility and cell adhesion systems underwent a spontaneous reversion that was correlated with a 10-fold reduction in the number of EGF receptors. The ability to regulate the scattering response induced by growth factors might be an important feature distinguishing normal epithelial cells from their tumoral counterparts.
[4] Li, S., Plowman, G.D, Buckley, S.D, Shipley, G D, Heparin inhibition of autonomous growth implicates amphiregulin as an autocrine growth factor for normal human mammary epithelial cells. J. Cell Physiol., 153: 103-111, 1992.
Normal human mammary epithelial cells (HMECs) proliferate in a serum-free defined growth medium in the absence of epidermal growth factor Li and Shipley, 1991). Amphiregulin (AR) is a heparin-regulated, EGF-like growth factor. Our observation that one strain of HMECs produce AR mRNA (Cook et al., 1991) stimulated us to determine whether AR expression was a common phenomenon in HMECs and whether AR could act as an autocrine growth factor to support the EGF-independent growth of these cells. In this study, we detected high levels of AR expression in four separate HMEC strains while one immortal mammary cell line (HBL-100) and six mammary tumor-derived cell lines had low to undetectable levels of AR. The EGF-independent growth of HMECs was blocked by the addition of heparin or a monoclonal anti-EGF receptor antibody to the culture medium, implicating AR as an autocrine growth mediator. This hypothesis is further supported by the fact that medium conditioned by HMECs contains secreted AR protein. A mammary tumor-derived cell line, Hs578T, which proliferates in an EGF-independent manner, does not express detectable levels of AR and is not growth inhibited by heparin. Examination of the same cell types for expression of transforming growth factor type-alpha (TGF-alpha) mRNA revealed coordinate expression of AR and TGF-alpha in these cells. These data suggest that both AR and TGF-alpha mRNA are produced in much greater abundance by normal HMECs than in tumor-derived cells in culture, and that AR is an important autostimulatory factor for the growth of normal HMECs.
Reports from other groups
[5] David S. Salomon, Ralf Brandt , Fortunato Ciardiello and Nicola Normanno.
Tumor Growth Factor Section, Laboratory of Tumor Immunology & Biology, National Cancer Institute, Bethesda, MD.; Oncologia Spermentale D, Instituto Nazionale per lo Studio e la Cura dei Tumori-Fondazione Pascale and Cattedra di Oncologia Medica, Facolta' di Medicina e Chirurgia, Universtita' degli Studi di Napoli Federico II, 80131 Napoli, Italy.
Epidermal Growth Factor-Related Peptides, Amphiregulin and Cripto-1, in Normal and Transformed Mammary Epithelial Cells
CR-1 can function as a transforming gene since overexpression of this protein in mouse NOG-8 or human MCF-10A mammary epithelial cells leads to their ability to form colonies in soft agar. Both CR-1 mRNA and protein are expressed at a high level in a majority of primary human breast carcinomas and in several human breast cancer cell lines (5,6). The expression has been determined by immunocytochemistry in 68 primary infiltrating ductal and infiltrating lobular breast carcinomas and in 23 adjacent noninvolved human mammary tissue samples. The expression of CR-1 mRNA has been detected in MCF-7, ZR-75-1, T47-D, MDA-MB-231 and MDA-MB-468 human breast cancer cells. The CR-1 protein could be immunocytochemically detected in the same breast cancer cell lines which were expressing CR-1 mRNA. Within the 68 breast carcinomas, 56 (82%) expressed the CR-1 protein. CR-1 has been detected in only 3 cases (13%) of adjacent, noninvolved breast tissue while TGFa and AR are expressed in 26% and 43% of these tissues, repec
tively. These data suggest that the differential expression of CR-1 in malignant breast epithelial cells may serve as a potential tumor marker for breast cancer.
REFERENCES
1. Lippman, M.E. and Dickson, R.B. 1989, Mechanisms of growth control in normal and malignant breast epithelium. Rec. Prog. Hormone Res. 45:383-440.
[6] Carlos Arteaga, Vanderbilt University
EGF Rs are expressed in high levels in some poor prognosis breast carcinomas. We have examined the cytotoxic effect of TGFa-Pseudomonas exotoxin 40 (PE40; Merck Sharp & Dohme) on a panel of normal and tumorigenic human breast epithelial cells in vitro and in vivo. This is a recombinant fusion protein in which TGFa, one of the ligands for the EGF R, is coupled to the catalytic domain of PE. The MDA-468, MDA-231, BT-20, and MCF-7ADR estrogen receptor (ER)-negative, EGF R-rich human breast cancer lines w
ere exquisitely sensitive in vitro to TGFa-PE40 with an IC50 <0.02 nM. Cytotoxicity was abrogated by coincubation with TGFa or EGF.
The ER-positive, low EGF R-expression MCF-7, ZR75-1 and T47D cell lines were relative resistant with IC50 >0.2 nM. The normal breast epithelial cells 184, 184A1, and 184B5, all from reduction mammoplasty tissue, were resistant to TGFa-PE40 inhibited despite exhibiting high levels of EGF Rs. When administered via an intraperitoneal pump over 7 days, TGFa-PE40 inhibited MDA-468, MDA-231, and BT-20 tumor growth in athymic mice without host tissue toxicity. Mixed MDA-468/MCF-7 tumors were established in nude mice after coinoculation of both cell types in estrogen-supplemented female nude mice. TGFa-PE40 killed the MDA-468 cells while sparing the adjacent low EGF R-expressing MCF-7 cells as examined by EGF R immunoblot and immunohistochemistry of the resulting xenografts. By these methods, EGF Rs were consistently more abundant in tumor tissue than in adjacent nontumor tissue from the same mastectomy specimen (n=9). Taken together these data suggest that human breast cancer cell EGF Rs can be selectively targeted for the delivery of antitumor agents. Further clinical trials with TGFa-PE40 or other similar chimerae will address that possibility.
Other groups that are working on areas involving EGF/TFGa include Steve Wiley, Kirk Lund, Steve Prescott, U. Utah Medical Center, Bob Weinberg, MIT, and John Mendelsohn, Sloan-Kettering Memorial Hospital.
[7] Stampfer, M. R., Yaswen, P., Alhadeff, M., Hosoda, J., TGFß induction of extracellular matrix associated proteins in normal and transformed human mammary epithelial cells in culture is independent of growth effects. J. Cell Physiol., 155: 210-221, 1993.
We have previously characterized a human mammary epithelial cell (HMEC) culture system for the effects of TGFß1 on cell growth. In the current report, the effects of TGFß1 on synthesis and secretion of proteins associated with the extracellular matrix and proteolysis were examined. In particular, we compared the TGFß responses of normal finite lifespan HMEC, which are growth inhibited by TGFß, to two immortally transformed cell lines derived from the normal HMEC. One of these lines maintains active gro
wth in the presence of TGFß and the other shows partial growth inhibition. In contrast to the differing effects of TGFß on cell growth, we found that all these cell types showed strong induction of most of the mRNA and protein species examined, including fibronectin, collagen IV, laminin, type IV collagenase, urokinase type plasminogen activator (uPA), and plasminogen activator inhibitor 1 (PAI-1). The profile of TGFß1 binding proteins was the same in HMEC that were, and were not growth suppressed by TGFß. Therefore, the effects of TGFß on cell growth could be dissociated from its effects on specialized responses, indicating that within this one cell type there must be at least two independent pathways for TGFß activity, one which leads to cessation of proliferation and one which induces a specific set of cellular responses. This cell system may be useful for examining of the pathway of TGFß induced growth inhibition using closely matched cells which vary in their growth response but retain similar specialized responses to TGFß.
[8] Bronzert, D. A., Bates, S. E., Sheridan, J. P., Lindsey, R., Valverius, E. M., Stampfer, M. R., Lippman, M. E., Dickson, R. B., Transforming growth factor-b induces platelet-derived growth factor (PDGF) secretion while inhibiting growth in norm
al human mammary epithelial cells. Mol. Endocrin., 4: 981-989, 1990.
Platelet-derived growth factor (PDGF) is a potent mitogen in human serum which specifically stimulates the proliferation of mesenchymal cells. We have now examined normal human mammary epithelial cells (HMEC) derived from reduction mammoplasties and grown in a serum-free defined medium. Medium conditioned by HMEC contained a PDGF-Iike activity that competed with [125I]PDGF for binding to PDGF receptors in normal human fibroblasts. When conditioned media were incubated with antiserum specific for either
PDGF-A or PDGF-B, only PDGF-A antiserum was capable of inhibiting binding of conditioned media to PDGF receptors. Using an RNase protection assay, mRNA from normal HMEC was probed for both the PDGF-A and PDGF-B chains. Little or no PDGF-B was found in HMEC strains, while a strong signal was seen with the PDGF-A probe. When HMEC were grown in the presence of transforming growth factor-b (TGFb) for 48 h, inhibition of growth was observed in association with a 20- to 40-fold stimulation of PDGF-B mRNA and a 2-fold stimulation of PDGF-A mRNA. This mRNA induction was extremely rapid (within 1 h), and secreted PDGF activity was induced 2- to 3-fold. Two other HMEC growth inhibitors and differentiating agents, sodium butyrate and phorbol ester 12-O-tetradecanoylphorbol-13-acetate, had no effect on PDGF mRNA regulation. The current study suggests that PDGF gene induction is an extremely rapid and specific indicator of TGFb function regardless of whether TGFb is acting in a growth stimulatory or inhibitory
manner. Any role of PDGF-B in TGFb modulation of differentiation of normal or malignant mammary gland remains to be determined.
[9] Slingerland, J.M., Hengst, L., Alexander, D., Stampfer, M.R., and Reed, S.I., A novel inhibitor of cyclin/Cdk activity detected in TGFb arrested epithelial cells, submitted to Mol. Cell Biol.
Transforming growth factor-beta (TGFb) is a potent inhibitor of epithelial cell growth. Cyclins E and A in association with Cdk2 have been shown to play a role in the G1-to-S phase transition in mammalian cells. We have studied the effects of TGFb mediated growth arrest on G1/S cyclins E and A. Inhibition of cyclin A-associated kinase by TGFb is primarily due to a decrease in cyclin A mRNA and protein. By contrast, while TGFb inhibits accumulation of cyclin E mRNA, the reduction in cyclin E protein is minimal. Instead, we find that phosphorylation of cyclin E and of its bound kinase partner Cdk2, that normally accompany activation of cyclin E-associated kinase at the G1-to-S phase transition, are inhibited. A novel inhibitor of cyclin/Cdk complexes
was detected in TGFb treated cell lysates. Inhibition is mediated by a heat stable protein(s) that target both Cdk2 and Cdc2 kinases. In Go arrested cells a similar inhibitor of Cdk2 kinase was detected. However, the activity of Cdc2 was not inhibited. These data suggest the existence of a family of inhibitors of cyclin dependent kinases induced under different conditions to mediate antiproliferative responses.
Other groups working on TGFß include Rik Dernyk, UC San Frabncisco, Mario Anzano and Michael Sporn, NCI.
[10] Zajchowski, D. A., Sager, R., Induction of estrogen-regulated genes differs in immortal and tumorigenic human mammary epithelial cells expressing a recombinant estrogen receptor. Mol. Endo., 5: 1613-1623, 1991.
Studies on estrogen receptor (ER) positive human breast cancer cell lines have shown that estrogen treatment positively modulates the expression of the genes encoding transforming growth factor a (TGFa), 52kd cathepsin D, and pS2. In order to determine whether these genes would be similarly regulated by estrogens in normal human mammary epithelial cells (HMEC), we stably transfected immortal, non-tumorigenic HMEC [184B5] with an ER-encoding expression vector. ER negative tumor cells were also transfected for comparison. Levels of TGFa and 52kd cathepsin D mRNA were enhanced by estrogen treatment of both ER-transfected immortal and tumorigenic cells, demonstrating that the ER, by itself, is sufficient to elicit estrogenic regulation of the expression of these genes.In contrast, expression of the pS2 gene was detected only in the ER-tranfected tumor cells. The ER in both cell lines is capable of recognizing the pS2 promoter, however, since estrogen enhanced the activity of an introduced pS2cat reporter plasmid in transient expression analyses. These and other experiments with somatic cell hybrids between the immortal cells and ER+/pS2+ MCF-7 tumor cells, where pS2 gene expression is extinguished, support the conclusion that the immortal, non-tumorigenic cells encode gene products which block endogenous pS2 expression. These results also imply that such repressors are not active in the tumor cells.
[11] Zajchowski, D. A., Sager, R., Webster, L., Estrogen inhibits the growth of estrogen receptor-negative, but not estrogen receptor-positive, human mammary epithelial cells expressing a recombinant estrogen receptor. Cancer Res., 53: 5004-5011, 1993.
Estrogen is essential for the growth of the normal mammary gland and most estrogen receptor (ER)-positive mammary carcinomas. To better understand the differences between the estrogen response pathways in normal and tumor cells, we have stably transfected ER-negative immortal, nontumorigenic human mammary epithelial cells [184B5] and ER-negative breast cancer cells with an ER-encoding expression vector. Unexpectedly, estro-gen treatment (1.0 nM) inhibited the proliferation of ER-transfected non-tumorigenic and tumor-derived cells. The control transfectants and pa-rental cells exhibited no response to estrogen concentrations as high as 1.0 mM. This inhibitory effect was attributed to a decreased growth rate and a perturbation of the cell cycle distribution by estrogen treatment of the ER transfectants. The inhibitory response was blocked by cotreatment with the antiestrogen ICI 164,384 as predicted for a pure antagonist of estrogen action. However, treatment with the antiestrogen hydroxytamoxifen caused growth inhibition, implying that hydroxytamoxifen acts as an ago-nist of estrogen action in ER-transfected cells. Since estrogen is a mitogenic and not a growth-inhibitory stimulus for ER-positive breast cancers and cell lines, we tested the effect of constitu-tive, high level expression of the ER in ER-positive tumor cells. Stable transfection of ER-positive MCF-7 and T47D cells with the ER expression vector yielded cells with varying amounts of ER. At ER levels comparable to those found in the ER-negative transfected cells, the MCF-7 and T47D ER transfectants were not inhibited by estrogen. These data suggest that ER-positive breast cancer cells can tolerate higher constitutive levels of ER expression than ER-negative cells. The mechanism by which this is ac-complished may be an essential step in the process which yields ER--positive tumors.
Other growth factors (e.g., insulin, FGFs, HGSFS)
[12] Milazzo, G., Giorgino, F., Damante, G., Sung, C., Stampfer, M. R., Vigneri, R., Goldfine, I. S., Belfiore, A., Insulin receptor expression and function in human breast cancer cell lines. Cancer Res., 52: 3924-3930, 1992
We have previously reported that insulin receptor expression is increased in human breast cancer specimens (V. Papa et al., J. Clin. Invest., 85:1503-1510, 1990). In the present study, in order to further understand the role of the insulin receptor in breast cancer, insulin receptor expression and function were characterized in three human breast cancer cell lines, MCF-7, ZR-75-1, and T-47D, and compared to a nonmalignant human breast epithelial cell line, 184B5. Insulin receptor content, measured by radioimmunoassay, was elevated 5- and 3-fold in MCF-7 and ZR-75-1 breast cancer cell lines, respectively, when compared to the nonmalignant cell line 184B5. In contrast, the insulin receptor content of T-47D cells was not increased. The increase in insulin receptor content in MCF-7 and ZR-75-1 cells was not due to amplification of the insulin receptor gene. Also, total insulin receptor mRNA content was not increased in breast cancer cells in respect to nonmalignantly transformed 184B5 breast epithelial cells. However, significant differences in the content of receptor mRNA species were observed. The insulin receptors in the breast cancer cell lines were functional: (a) In all 4 cell lines, high-affinity insulin-binding sites were detected, and, in concert with the insulin receptor radioimmunoassay data, binding capacity was highest in MCF-7 and then in ZR-75-1 cells. (b) In all cell lines, insulin stimulated insulin receptor tyrosine kinase activity. However, the effect of insulin was greater in breast cancer cell lines than in nonmalignant breast cells. (c) In all cell lines, insulin at concentrations of 1 nM or less stimulated [3H]thymidine incorporation. This effect of insulin was inhibited by 50% in MCF-7 cells and by 60% in 184B5 cells when alpha-IR3, a monoclonal antibody to the insulin-like growth factor I receptor, was present. In these cells, therefore, insulin was active via both its own receptor and the IGF-I receptor. In contrast, alpha-IR3 antibody was without effect in T-47D and ZR-75-1 cells, suggesting that in these cell lines insulin acted only via its receptor. In the breast cancer cells, MA-5, an agonist monoclonal antibody to the insulin receptor, stimulated [3H]thymidine incorporation. This present study indicates therefore that in breast cancer cell lines there are functional insulin receptors that regulate breast cancer cell growth.
[13] Li, S., Shipley, G. D., Expression of multiple species of basic fibroblast growth factor mRNA and protein in normal and tumor-derived mammary epithelial cells in culture. Cell Growth & Diff., 2: 195-202, 1991.
We examined the expression of the basic fibroblast growth factor (bFGF) gene in cultured normal and tumor-derived human mammary epithelial cells at both the transcriptional and translational level. Northern blot analysis revealed three bFGF mRNA transcripts of 7.5, 4.4, and 2.2 kilobases in all four strains (donors) of normal cells (HMECs) we examined and in the immortal mammary cell line HBL-100. Of the four mammary tumor-derived cell lines we examined (MCF-7, BT-474, T-47D, and Hs578T), only the Hs578T cells produced detectable levels of bFGF mRNA. Western blot analysis of cell lysates using an anti-bFGF monoclonal antibody revealed corresponding results. bFGF protein was detected in normal HMEC strains 161 and 48 (other normal strains not tested), in HBL-100 cells, and in Hs578T cells, but not in the other tumor cell lines. In each case, three distinct molecular weight species of bFGF protein were detected which migrated in sodium dodecyl sulfate-polyacrylamide gel at 18, 24, and 27 kDa. We also investigated the ability of bFGF to stimulate the proliferation of normal and tumor--derived mammary epithelial cells. Addition of bFGF to serum-free cultures of these cells had no effect on the proliferation of HMECs under a variety of conditions and was weakly mitogenic for Hs578T cells. Our results indicate that normal HMECs produce bFGF mRNA and protein(s), whereas only some mammary tumor-derived cells express this gene. Thus, our results do not support a general role for expression/overexpression of bFGF in the development of mammary tumors. However, bFGF could play a role in the normal development and homeostasis of the mammary gland.
Reports from other groups
[14] Ermanno Gherardi and Michael Stoker, ICRF, Cambridge
HGFSF is a multifunctional growth factor having mitogenic, morphogenetic and motogenetic effects. The mechanism of action of HGFSF is not understood, but one possible primary target of action is on gap junctional cell communication. To determine whether or not this is the case, we are looking at the effect of HGFSF on cell communication in a variety of epithelial cell types including MDCK cells in which HGFSF has a predominantly motogenic effect and mouse keratinocytes and 184B5 cells in which HGFSF also has a mitogenic effect. In initial studies, we are measuring cell communication via the extent of transfer of a low molecular weight fluorescent dye microinjected into the cells. In parallel experiments we are looking at the effect of HGFSF on connexin expression and phosphorylation status.
[15] Micsunica Platica, Mt. Sinai Medical Center
Our laboratory is involved in studying the effect of a novel pituitary differentiating factor (PDF) on breast cancer cells. Recently we reported the effect of PDF on the rat mammary tumor cell lines MTW9/Pl. Within 24 hours, MTW9/Pl cells, which normally grow in suspension, began to aggregate forming spheroid-like structures, and to adhere to the plastic dish. Spheroid formation was followed in 48 hours by electron microscopic evidence of polarization, lumen formation and by enhanced expression of laminin and lactalbumin. The activity could not be replaced by a variety of hormones and growth factors (Platica, et al., Endocrinology 131:2573-2580, 1992).
The pituitary factor has a similar action on human breast cancer cells which express estrogen receptors (MCF-9) and T47D), but fails to affect human breast cancer cells devoid of estrogen receptors (MDA231 and HBL-100), normal epithelial cells (161 and 48R), acute lymphocytic leukemia cells (CEM-C7 C+, C-) and human myeloid leukemia cells HL-60. Morphological differentiation is accomplished by early increased expression of p53 and hsp90, and by late expression of casein, lactalbumin and uvomorulin. Basement membrane constituents may be involved in PDF-induced differentiation (manuscript in preparation). Studies are in progress to clarify if aggregation is a necessary step for differentiation and if increases in hsp90 and p53 are associated with PDF-induced differentiation.
Other groups working on HGF/scatter factor include Eliot Rosen, Long Island Jewish Medical Center.
Stampfer, M. R., Yaswen, P.: Factors influencing growth and differentiation of normal and transformed human mammary epithelial cells in cultures. In: G. E. Milo, B. C. Castro, C. F. Shuler (eds.), Transformation of Human Epithelial Cells: Molecular and Oncogenetic Mechanisms, 117-140. Boca Raton, FL, CRC Press, 1992. This is a good recent general review of our HMEC cell system. If you don't already have a copy, I can send you one.
[16] Porter, M. B., Pereira-Smith, O. M., Smith, J. R., Common senescent cell-specific antibody epitopes on fibronectin in species and cells of varied origin. J. Cell. Physiol., 150: 545-551, 1992.t
The phenomenon of in vitro cellular senescence has been demonstrated in cul-tured cells derived from humans and various other species. We have previously shown that monoclonal antibodies SEN-1, SEN-2, and SEN-3 react to epitopes on fibronectin that are exposed when human diploid fibroblasts become senescent. We here present results demonstrating that exposure of these epitopes is specific to senescence for a variety of human cells: epidermal keratinocytes, mammary epithelial cells, as well as fibroblasts. Fibronectin from 11 additional species was also analyzed by Western immunoblot for ability to bind the SEN antibodies. SEN-1 bound only human and gorilla fibronectin, whereas SEN-2 and SEN-3 bound fibronectin from those two species as well as the horse, cow, sheep, goat, dog, and chick. None of the antibodies reacted with fibronectin from the rabbit, rat, or mouse. These data indicated a correlation between the ability of the SEN antibodies to bind fibronectin from a particular species and the ability of cells from that species to exhibit a stable senescent phenotype in vitro. Therefore, exposure of this region of fibronectin may be important in the establishment and mainte-nance of cellular senescence. In addition, the ability of the SEN antibodies to react with fibronectin from a variety of senescent cells emphasizes their usefulness as markers for cellular senescence.
[17] Calvin Harley and Rich Allsop, Geron Corporation
We have done some research using the epithelial cells, although we have not published the data in a paper or abstract. Our lab is interested in determining the possible role telomere shortening may have on cell senescence and tumorigenesis. We have shown that telomeres shorten during in vitro aging of human fibroblasts. Preliminary data with the epithelial cells indicates that telomeres shorten as a function of population doublings in these cells as well, although further analysis is required to confirm this observation.
Other groups interested in cellular senescence include Anton Jetten and Carl Barrett, NIEHS
Tumor suppressors and oncogenes
[18] Lehman, T., Modali, R., Boukamp, P., Stanek, J., Bennett, W., Welsh, J., Metcalf, R., Stampfer, M., Fusenig, N., Rogan, E., Reddel, R., Harris, C., p53 mutations in human immortalized epithelial cell lines. Carcin., 14: 833-839, 1993.
Although rodent cells have been immortalized following transfection with a mutant p53 gene, the role of p53 in the immortalization of human cells is unknown. Therefore, human epithelial cell lines were examined for p53 mutations in exons 4-9 which include the evolutionarily conserved regions. A spontaneously immortalized skin keratinocyte cell line, HaCat, and three ras-transfected clones, have a p53 mutational spectrum that is typical of ultraviolet light induced mutations. A normal finite lifespan cell strain (184) and two benzo[a]pyrene immortalized mammary epithelial cell lines derived from 184 (184A1 and 184B5) contain wild type p53 sequences in exons 4-9, although elevated levels of nuclear p53 indicate an alteration in the stability of the normally transient protein. Wild type p53 was found in human bronchial, esophageal and hepatic epithelial cells immortalized by SV40 T antigen gene and human renal epithelial cells immortalized by adenovirus 5. BEAS-2B, an SV40 T antigen immortalized bronchial epithelial cell line and two subclones, have a germline polymorphism at codon 47. Inactivation of p53 by mechanisms such as mutation or complexing with proteins of DNA tumor viruses appears to be important in the immortalization of human epithelial cells.
[19] For other relevant studies on p53 in HMEC, see references from Vimla Band's lab, New England Medical Center: Delmolino, L., Band, H., Band, V., Expression and stability of p53 protein in normal human mammary epithelial cells. Carcinogeneisis, 14: 827-832, 1993.
Band, V., Dalal, S., Delmolino, L., Androphy, E. J., Enhanced degradation of p53 protein in HPV-6 and BPV-1 E6-immortlaized human mammary epithelial cells. EMBO J., 12: 1847-1852, 1993.
[20] Pierce, J. H., Arnstein, P., DiMarco, E., Artrip, J., Kraus, M. H., Lonardo, F., DiFiore, P. P., Aaronson, S. A., Oncogenic potential of erbB-2 in human mammary epithelial cells. Oncogene, 6: 1189-1194, 1991.
Introduction of the normal erbB-2 gene into immortal-ized human mammary epithelial cells (184B5) by trans-fection conferred a growth advantage to these cells both in vitro and in vivo. The 184B5 cells overexpressing erbB-2 formed colonies in semi-solid medium, frequently induced transient nodules in athymic mice and produced progressive tumors in vivo at a low frequency. Those tumors which did arise from erbB-2-transfected cells dis-played substantially higher levels of normal gp185erb-2 protein when compared to the original transfectants, con-sistent with their selection for increased erbB-2 expres-sion. Introduction of genes encoding genetically altered erbB-2 molecules into 184B5 cells increased their colony--forming efficiency and converted the cells to a tumorige-nic phenotype at a high frequency. When the biological and biochemical properties of human mammary carcin-oma cell lines known to overexpress erbB-2 were com-pared to the transfected 184B5 lines, they behaved most like those overexpressing the normal erbB-2 protein. Results indicate that overexpression of normal erbB-2 may directly contribute to the transformation of human mammary epithelium if sufficient levels of erbB-2 protein are expressed or if the erbB-2 gene is genetically altered.
[21] Zhai, Y.-F., Beittenmiller, H., Wang, B., Gould, M. N., Oakley, C., Esselman, W. J., Welsch, C. W., Increased expression of specific protein tyrosine phosphatases in human breast epithelial cells neoplastically transformed by the neu oncogene. Cancer Res., 53: 2272-2278, 1993.
Protein tyrosine phosphorylation/dephosphorylation is a fundamental mechanism in the regulation of cell proliferation and neoplastic transfor-mation; this metabolic process is modulated by the opposing activities of protein tyrosine kinases and protein tyrosine phosphatases (PTPases). While the role of protein tyrosine kinases has been examined extensively in human breast tumorigenesis, the role of PTPases in this process is virtually unknown. To address this issue, an activated neu oncogene was introduced into an immortalized nontumorigenic human breast epithelial cell line (184B5). This resulted in a substantial increase in P185neu expres-sion, which led to the formation of progressively growing carcinomas after such cells were inoculated into athymic nude mice. Importantly, a striking increase in the expression of specific PTPases, LAR and PTP1B, was observed in 3 independently neu transformed cell lines and their derived tumors. This elevation was verified at both the mRNA and protein levels. TC-PTP PTPase expression was only slightly increased in these neu trans-formed cells, and no expression of CD45 PTPase was observed. The level of neu expression, as well as the differential expression between P185neu and LAR/PTP1B, directly correlated with tumorigenicity. Furthermore, rat mammary carcinomas with elevated neu expression (neu-induced) also had sharply elevated LAR-PTPase expression when compared to rat mammary carcinomas with little or no neu expression (7,12-dimethylben-zanthracene induced); the level of expression of LAR PTPase was directly correlated with the level of neu expression. Thus, our results provide the first evidence that, in human breast carcinoma cells and in rat mammary carcinomas that have an induced increase in neu expression, a consistent and substantial increase in the expression of specific PTPases occurs. The relationship between Pl85neu-protein tyrosine kinase expression and spe-cific PTPase expression may play a critical role in human breast tumor-igenesis.
[22] Park, JW, Slamon, DJ, and Shepard, HM, Loss of heregulin expression is associated with malignancy, AACR Proceedings, p.521, 1993
Heregulin (HRG), a ligand for the p185HER-2 receptor, is expressed in a number of human tissues, including breast, colon, and lung. HRG mRNA expression was studied using PCR and Northern analysis. Primary cultures and cell lines derived from normal breast epithelium demonstrated relatively high levels of HRG expression. In contrast, 8 breast cancer cell lines examined showed loss of HRG expression, including all 5 high HER-2 overexpressing cell lines. 2 breast cancer cell lines, MDA-MB-231 and MDA-MB-175, did show HRG expression, but with a pattern of transcripts different from that of breast tissue or breast epithelial cell lines. In three separate cell lines in which p185HER-2 was caused to be overexpressed via transfection, HRG expression was retained, indicating that loss of HRG expression is not merely a result of transformation or of HER-2 overexpression. HRG expression was also examined in tumors from patients with lung and colon cancer. In most of these cases, HRG expression was seen in surrounding normal colon or lung tissue, but was absent in the tumor. These results indicate that HRG does not function as an autocrine factor in HER-2 overexpressing tumors, and that loss of HRG expression may be involved in tumorigenesis.
Other groups working on erbB2/HER2/neu include Ruth Lupu, Georgetown U., Gail Philips, Genentech Inc., Powel Brown, NCI, Shelley Blam, Berlex Biosciences, Briggs Morrison and Phil Leder, Harvard Medical School
Reports from other groups
[23] Yuen Kai Fung and Anne T'ang, USC School of Medicine
Rb suppresses the growth of 184A1 and 184B5.
[24] Robert Swift, Eli Lilly
I am using 184, 184A1 and 184B5 as sources of RNA to study the expression of genes in the region of BRCA1. I will probably use one or more of the cells as a transfection recipient for some of the cDNAs expressed in the BRCA1 region.
Other groups working on the area of BRCA1 are Ray White, U. Utah Medical Center, and Carl Barrett, NIEHS.
Calmodulin-like Protein (CLP) / NB-1 gene
[25] Yaswen, P., Smoll, A., Peehl, D. M., Trask, D. K., Sager, R., Stampfer, M. R., Down-regulation of a calmodulin-related gene during transformation of human mammary epithelial cells. Proc. Natl. Acad. Sci. USA, 87: 7360-7364, 1990.
A human cDNA library obtained from cultured normal human mammary epithelial cells (HMEC) was searched by subtractive hybridization for genes whose decrease in expression might be relevant to epithelial transformation. One clone identified by this procedure corresponded to a 1.4 kb mRNA, designated NB-1, whose expression was decreased more than 50-fold in HMEC tumorigenically transformed in vitro after exposure to benzo(a)pyrene and Kirsten sarcoma virus *[this was a subtraction between normal 184 and the 184B5KTu line]. Sequence analysis of NB-1 cDNA revealed an open reading frame with a high degree of homology to calmodulin. NB-1 expression could be demonstrated by PCR amplification in normal breast, prostate, cervix, and epidermal tissues. The presence of NB-1 transcripts was variable in primary breast carcinoma tissues, and undetectable in tumor derived cell lines of breast, prostate, or other origins. NB-1 mRNA expression could be downregulated in cultured HMEC by exposure to reconstituted extracellular matrix material, while exposure to TGFb increased its relative abundance. The protein encoded by NB-1 may have Ca+2 binding properties and perform functions similar to those of authentic calmodulin. Its possible roles in differentiation and/or suppression of tumorigenicity in epithelial tissues remain to be examined.
[26] Yaswen, P, Smoll, A, Hosoda, J, Parry, G, Stampfer, M.R, Protein product of a human intronless calmodulin-like gene shows tissue-specific expression and reduced abundance in transformed cells. Cell Growth & Diff., 3: 335-345, 1992.
The recently identified NB-1 mRNA is transcribed from a single intronless gene, previously thought to be an unexpressed calmodulin pseudogene [see Koller & Strehler 1988 below]. Although expression levels of the three known human calmodulin genes fluctuate only slightly in all cell types and tissues examined, NB-1 expression is limited to certain cells of pseudostratified and stratified epithelial tissues. Like calmodulin, the protein encoded by NB-1 is heat stable and binds to phenyl-Sepharose in a calcium dependent manner. Despite the shared identity of 85% of their 148 amino acids, however, calmodulin and NB-1 protein are easily distinguished electrophoretically and immunologically. Polyclonal antibodies prepared against recombinant NB-1 protein recognize a protein with an apparent molecular weight of 16 kilodaltons which is abundant in cultured normal human mammary epithelial cells, but which is absent or barely detectable in fibroblasts or tumor cell lines. The immunohistochemical distribution of NB-1 protein in histologically normal tissues suggests that expression of the gene is regulated during epithelial differentiation. The majority of a small number of malignant tissues examined had lowered or undetectable NB-1 protein expression relative to normal tissues. Given its restricted distribution, the NB-1 protein may be involved in the initiation or maintenance of certain differentiated functions. Its absence may be due to or necessary for the manifestation of the transformed phenotype in certain cell types.
[27] Yaswen, P., Stampfer, M. R., Ghosh, K., Cohen, J. S., Effects of sequence of thioated oligonucleotides on cultured human mammary epithelial cells. Antisense Res. Develop., 3: 67-77, 1993.
We have compared the effects of a number of different oligonucleotides on the growth and morphology of normal finite life span and immortally transformed human mammary epithelial cells. The oligonucleotide sequences chosen initially for study were based on that of the NB-1 gene, which encodes a calmodulin-like protein of unknown function. We found that certain thioated oligonucleotides 15 - 20 residues in length altered the morphology and decreased the growth rate of the normal cells in a concentration dependent manner. These effects were rapid, occurring within 24-48 hrs of oligo addition. The effects, which occurred without an accompanying detectable decrease in the levels of NB-1 mRNA or protein, were most pronounced in the normal epithelial cells, less apparent in the immortalized epithelial cells, and unobserved in normal breast fibroblasts. Identical sequences having mixed phosphodiester and phosphorothioate backbones, or phosphodiester backbones alone, had little or no effect on normal epithelial cell morphology or growth. Two out of seven additional thioated oligonucleotides which were not complementary to NB-1 mRNA also affected normal epithelial cell morphology and growth, when used at similar concentrations (10mM). Taken together, the observed effects on normal epithelial cells indicate that certain thioated oligonucleotides may have pharmacological consequences which do not depend on strict complementarity of their sequences to known mRNAs.
[28] Current direction and report from Emanuel Strehler, Mayo Foundation
One of the research projects of my laboratory deals with structural and functional aspects of a relatively recently discovered human calmodulin-like protein (CLP). This protein displays a tissue-/cell-specific pattern of expression, being particularly abundant in primary breast epithelial cells, and is strongly down-regulated in transformed cells. Although several of its biochemical and physico-chemical properties have been elucidated (on the bacterially expressed protein), details of its structure, and its in vivo function and regulation, remain unknown. HMEC and transformed epithelial cells may be useful for a number of studies on CLP, as listed below: Koller, M., Strehler, E.E., Characterization of an intronless human calmodulin-like pseudogene. FEBS Lett., 239: 121-128, 1988.
Transformation and Carcinogenesis
[29] Sanford, K. K., Price, F. M., Rhim, J. S., Stampfer, M. R., Parshad, R., Role of DNA repair in malignant neoplastic transformation of human mammary epithelial cells in culture. Carcinogenesis, 13: 137-1141, 1992.
Epithelial cells derived from normal human mammary tissue were examined for capacity to repair radiation-induced chromatin DNA damage. Repair capacity was estimated by quantifying chromatid aberrations in metaphase cells arrested 0.5-1.5 h after X-irradiation during G2. The parental cells at passage 12 had 19 chromatid breaks and 16 gaps per 100 metaphase cells, representing efficient repair. Of two continuous cell lines, derived after benzo[a]pyrene treatment, 184A1 maintained the efficient repair phenotype through passage 50, while a subline of 184A1 [184A1N4] developed the repair-deficient phenotype characterized by a 3- to 5-fold higher frequency of chromatid breaks or gaps. This line was transformed to tumorigenic cells by HaMSV and SV40 T antigen. The second continuous line 184B5 and derivatives had 102-165 chromatid breaks and 87-134 gaps per 100 metaphases (deficient repair phenotype). This line was transformed to tumorigenic cells by KiMSV. As reported previously for human epidermal keratinocytes, acquisition of this repair-deficient phenotype appears to be an early requisite step in the malignant neoplastic transformation of human cells in culture.
[30] Thompson, E., Torri, J., Sabol, M., Sommers, C., Byers, S., Paik, S, Martin, G., Lippman, M., Valverius, E., Stampfer, M., Dickson, R., Oncogene-induced basement membrane invasiveness in human mammary epithelial cells. 1993, in press.
Expression of the intermediate filament protein vimentin, and loss of the cellular adhesion protein uvomorulin (E-cadherin) have been associated with increased invasiveness of established human breast cancer cell lines in vitro and in vivo (Cell growth and Differentiation, 2:365-372, 1991; J. Cell. Physiol, 150:534-544, 1992). In the current study, we have examined these relationship in oncogenically transformed human mammary epithelial cells. A normal human mammary epithelial strain termed 184 was previously immortalized with benzo[a]pyrene and then either of two sublines infected with retroviral vectors containing a single or two oncogenes of the nuclear, cytoplasmic, and plasma membrane-associated type (v-rasH, v-rasKi, v-mos, SV40T, and c-myc). Major effects were observed after infection of the cells with two oncogenes (v-rasH + SV40T and v-rasH + v-mos) on in vitro invasiveness of Matrigel, a reconstituted basement membrane extract, while effects of single oncogenes were only modest or negligible. All oncogenic infectants demonstrated increased attachment to laminin, but altered secretion of the 72kDa and 92kDa gelatinases was not associated with malignant progression. Both highly invasive double transformants were vimentin-positive and uvomorulin-negative, a phenotype indicative of the epithelial-mesenchymal transition (EMT) previously associated with increased invasiveness of established human breast cancer cell lines. Parental populations or single oncogene transformants which were positive for both vimentin and uvomorulin were only weakly invasive. These data demonstrate that oncogene combinations can induce the invasive phenotype in mammary epithelial cells, and that this induction may involve an EMT-like event.
[31] Guthridge, C.J., Stampfer, M.R., Clark, M.A., and Steiner, M.R., Characterization and comparison of phospholipase A2 in ras-transformed and immortalized human mammary epithelial cells, submitted to Biochem. Biophys. Acta.
Phospholipase A2 (PLA2) activities are characterized and compared in nontumorigenic and tumorigenic human mammary epithelial cells, namely 184B5 cells (immortalized cell line from a reduction mammoplasty) and B5KTu cells (cells from a tumor induced by ras-transformed 184B5 cells). The major PLA2 activity observed in these cells has the characteristics of the high molecular weight cytosolic PLA2, including a requirement for free Ca2+ for activity, preferential hydrolysis of arachidonyl containing phosphotidylcholine and phosphotidylethanolamine and Ca2+ modulated subcellular localization of the activity. This activity is maintained upon anion exchange f.p.l.c. The cytosol also contains a Ca2+ independent PLA2 activity; the two cell types have similar levels of this activity. The particulate fractions have a low level of PLA2 activity which does not appear to be a Group II PLA2 activity. Specifically, Group II PLA2 is not detected in human mammary cells upon ammonium sulfate extraction of particulate fractions followed by cation exchange f.p.l.c. of the soluble extracts whereas this PLA2 activity is readily detected in ras-transformed fibroblasts analyzed in a like manner. Thus, the high molecular mass cytosolic PLA2 activity is selectively elevated in tumorigenic versus nontumorigenic human mammary epithelial cells. This PLA2 is implicated in signal transduction, including regulation of the release of arachidonic acid. Hence, the increase in this PLA2 activity in tumorigenic mammary cells may lead to increased generation of lipid biomediators in these cells.
[32] Zajchowski, D., Band, V., Trask, D., Kling, D., Connolly, J., Sager, R., Suppression of tumor-forming ability and related traits in MCF-7 human breast cancer cells by fusion with immortal mammary epithelial cells. Proc. Natl. Acad. Sci. USA, 8
7: 2314-2318, 1990.
Somatic cell hybrids between MCF-7 human breast cancer cells and normal immortalized human mammary epithelial cells have been obtained by polyethylene glycol-mediated cell fusion. The hybrid cells are suppressed in their ability to form tumors in nude mice, as well as in traits specific to the tumorigenic MCF-7 parent: growth factor independence, tumor necrosis factor sensitivity, and pS2 gene expression. In addition, they display other characteristics of the "normal" parent, including increased expression relative to the MCF-7 cells of the genes for the extracellular matrix component fibronectin, the intermediate filament keratin 5, and the angiogenesis inhibitor thrombospondin. The levels of keratins 8 and 18 also resemble those of the nontumorigenic pa
rent. These results provide evidence for the existence of tumor suppressor gene products in immortal mammary epithelial cells. We propose a characteristic "suppressed" tumor cell phenotype, which encompasses altered cytoarchitecture, angiogenesis capabilities, and growth factor requirements.
[33] Valverius, E. M., Ciardiello, F., Heldin, N., Blondel, B., Merlo, G., Smith, G., Stampfer, M. R., Lippman, M. E., Dickson, R. B., Salomon, D. S., Stromal influences on transformation of human mammary epithelial cells overexpressing c-myc and S
V40T. J. Cell. Physiol., 145: 207-216, 1990.
The proto-oncogene c-myc and the oncogene SV40T, both of which have been implicated in the process of cellular immortalization in vitro, have been introduced via amphotropic retroviral expression vectors into the human mammary epithelial cell (HMEC) line 184A1N4 (A1N4). Two stable cell lines were established by growth in selective medium and were found to overexpress either c-myc (A1N4-myc) or SV40T antigen (A1N4-T). Neither the A1N4, A1N4-myc, or A1N4-T cells will grow in soft agar or form tumors in nu
de mice. However, A1N4-T or A1N4-myc cells, but not the parental A1N4 cells, form colonies in soft agar in response to either epidermal growth factor (EGF), transforming growth factor a (TGFa), or basic fibroblast growth factor (bFGF). Like EGF and TGFa, bFGF is moderately mitogenic for the anchorage-dependent growth (ADG) of all three cell lines. Further, co-cultivation of A1N4-T or A1N4-myc cells with primary diploid mammary fibroblasts can also induce the anchorage-independent growth (AIG) and stimulate the ADG of A1N4-T or A1N4-myc. In addition, conditioned medium obtained from these mammary fibroblasts also stimulated the AIG of the A1N4-T and A1N4-myc cells and was found to contain immunoreactive TGFa and bioactive FGF. The mammary fibroblasts express specific mRNA transcripts for bFGF and acidic FGF (aFGF). These results suggest that growth factors such as TFGa or FGF, which may be derived from the adjacent mammary stroma, might influence in a paracrine manner the phenotypic characteristics of a population of human mammary epithelial cells toward transformation.
[34] Bhattacharya, B., Prasad, G. L., Valverius, E. M., Salomon, D. S., Cooper, H. L., Tropomyosins of human mammary epithelial cells: consistent defects of expression in mammary carcinoma cell lines. Cancer Res., 50: 2105-2112, 1990.
Suppression of synthesis of specific tropomyosin (TM) isoforms occurs commonly in human, murine, and arian fibroblasts transformed by retroviral oncogenes or other modalities. The resulting deficiency or altered distribution of TMs may predispose the cells to microfilament instability and contribute to expression of the transformed phenotype. In this study we have asked whether defects in TM expression had relevance to human neoplasia, which arises most often in cells of the epithelial lineage rather than in fibroblasts and often is unrelated to demonstrable expression of oncogenes. TMs were characterized in normal primary human mammary epithelial cells (HMEC) and in an immortalized nontumorigenic cell line derived from them. Seven TM isoforms were identified in primary HMEC, two of which may be unique to epithelial cells. Immortalized nontumorigenic HMEC expressed the same array of isoforms. Of six established human breast carcinoma cell lines studied, all failed to express the Mr 39,000 TM isoform and five of six also lacked expression of either the Mr 38,000 or 35,000 isoform. Northern blot analysis with probes specific for the 1.1-kilobase mRNA of fibroblast TMI detected a mRNA of this size in normal HMEC. This mRNA, which probably encodes the Mr 39,000 TM missing from all the carcinoma lines, was absent from five of the six breast cancer cell lines. These results indicate that abnormalities in TM expression in neoplastic cells are not limited to fibroblasts. The high frequency and consistent nature of such abnormalities among cell lines derived from human breast cancer raises the possibility that such abnormalities in expression of a major cytoskeletal protein may play a role in human neoplasia.
[35] Prasad, G. L., Valverius, E. M., McDuffie, E., Cooper, H. L., Complementary DNA cloning of a novel epithelial cell marker protein, HME1, that may be down-regulated in neoplastic mammary cells. Cell Growth and Diff., 3: 507-513, 1992.
A full-length complementary DNA clone from a normal human mammary epithelial cell (strain 184) encoding a 25-kilodalton protein, HME1, was isolated. Expression of HME1 RNA appears to be limited to epithelial cells. The HME1 sequence has extensive sequence homology with bovine 14-3-3 protein, which is an activator of tyrosine and tryptophan hydroxylase. However, the tissue distribution, arrangement of charged amino acids, and location of potential phosphorylation sites of HME1 differ from those of 14-3-3. Compared with normal mammary epithelial cells, expression of HME1 RNA was dramatically low in two cell lines derived from human mammary carcinoma that were examined, and in a line of normal mammary epithelial cells transformed by oncogenes. HME1 therefore appears to be a cellular differentiation marker that may be downregulated during neoplastic transformation.
[36] Gaddamanugu L. Prasad, Eva M. Valverius, Bruno Blondel, Giorgio Merlo, David S. Salomon, Robert H.
We have examined the effect of expression of a variety of transforming oncogenes on tropomyosin (TN) production in a human mammary epithelial cell line, 184A1N4, in order to determine (a) whether oncogene expression in epithelial cells is associated with derangement of TM expression, as has been well documented in fibroblasts, and (b) whether the abnormalities of TM synthesis we have previously reported in mammary carcinoma cell lines might be due to the action of oncogenes. Lines used in this study expressed v-Ha-ras, SV40-T, v-mos, SV40-T plus v-Ha-ras, c-neu, point-mutated c-neu, v-bas or c-myc. TM 38 synthesis was suppressed in mammary epithelial cells expressing SV40-T, while lines expressing either SV40-T or point-mutated c-neu showed enhanced synthesis of TM 35. Expression of v-Ha-ras was associated with enhanced expression of the mRNA for TM 39 but had no apparent effect on synthesis of TM 39 protein. This is the opposite change from that found in fibroblasts expressing v-Ha-ras, where the relevant mRNA and its encoded protein are both suppressed. The line expressing both SV40-T and v-Ha-ras, which is the only line among the group studies that was frankly transformed, showed strongly enhanced expression of TM 35 together with suppression of TM 38. The other oncogenes studied had no detectable effects on TM expression in epithelial cells. None of the oncogene-expressing lines showed suppression of TM 39, the common feature found in breast cancer cell lines. This study shows that TM expression in mammary epithelial cells can be altered by some oncogenes, although this is less common than in fibroblasts. None of the oncogenes studied here were able to induce the changes in TM expression found previously in human breast carcinoma cell lines.
[37] Carl Barrett and Gloria Jahnke, NIEHS
Question: Has anyone tried growing your normal or immortal HMEC in the nude mouse fat pad? What were the results? A group in England has successfully transplanted organoids into the cleared fat pad of the nude mouse (Gusterson et al. Virchows Arch. 404: 325-333, 1984 and Stewart H. et al Virchows Arch. 410:495-500, 1987). Any suggestions from readers on this would be welcome.
[38] Tracy Yang, NASA
Using repeated irradiation of heavy ions, we have successfully obtained morphological transformed cells from a growth variant. From different loci found in the dish, transformed cells were cloned and tested in soft agar. These transformed cells showed anchorage independent growth and appeared to have a shorted doubling time.
Mutagenesis, Carcinogen Metabolism, and Cellular Toxicity
[39] Eldridge, S. R., Gould, M. N., Specific locus mutagenesis of human mammary epithelial cells by ultraviolet radiation. Int. J. Radiat. Biol., 59: 807-814, 1991.
Tissue and locus specificity of mutation induction was studied in human mammary epithelial cells (HMEC). Primary HMEC from normal tissue, as well as immortalized HMEC (184B5) derived from normal HMEC, were cultured under identical conditions and exposed to 10 J/m2 ultraviolet (UV) radiation (254 nm peak wavelength), which produced approximately 50% mean survival in all cell strains and lines tested. UV radiation was found to induce mutations at the Na(+)-K+ ATPase locus as determined by ouabain-resistance in both normal and immortalized HMEC. Mutation frequencies measured in these cells following UV exposure were similar to those reported for human diploid fibroblasts. In addition, mutation induction was investigated at the hypoxanthine-guanine phosphoribosyltransferase (HPRT) locus in normal and immortalized HMEC. Induced mutations at the HPRT locus as determined by 6-thioguanine resistance in normal primary HMEC were not observed following UV radiation. In contrast, mutation induction was observed at this locus in UV-exposed immortalized HMEC.
[40] Eldridge, S. R., Gould, M. N., Comparison of spontaneous mutagenesis in early-passage human mammary cells from normal and malignant tissues. Int. J. Cancer, 50: 321-324, 1992.
Spontaneous mutant frequencies were determined in early-passage epithelial-cell strains derived from either normal or malignant human breast tissues, as well as a non-tumorigenic immortalized human mammary epithelial cell line (184B5) derived from normal cells. Mutations at the hypoxanthine-guanine phosphoribosyltransferase (HPRT) locus were quantified by determining the 6-thioguanine resistance. Mutant frequencies in human mammary epithelial cells (HMEC) from 4 normal and 5 carcinoma tissue samples did not differ significantly. In contrast, the mutant frequency in the immortalized HMEC was approximately 10 times higher than in average normal HMEC and normal non-immortalized cells from early-passage cultures of the same cell lineage. Our data suggest that the progression of normal breast cells to invasive carcinoma cells does not necessarily involve the establishment of a general genetic instability during this progression.
[41] Eldridge, S. R., Gould, M. N., Butterworth, B. E., Genotoxicity of environmental agents in human mammary epithelial cells. Cancer Res., 52: 5617-5621, 1992.
Despite an increasing incidence of human breast cancer, its etiology remains unknown. Since some environmental chemicals are stored in human breast fat and are rodent mammary carcinogens, determining the genotoxic potential of environmental agents in this key target tissue is important. An assay was developed for detecting genotoxic activity, as unscheduled DNA synthesis (UDS), induced by chemicals and UV radiation in early passage cultures of normal human mammary epithelial cells (HMEC) derived from 5 different women. In order to measure UDS in culture, reduction in the percentage of cells in S-phase was accomplished either by depriving the cells of epidermal growth factor and bovine pituitary extract or by contact inhibition of growth. Cultures were incubated with test chemicals for 24 h in the presence of [3H]-thymidine. UDS was quantitated autoradiographically as net grains per nucleus (nuclear grains minus cytoplasmic background, population average) with > or = 6 net nuclear grains considered in repair for any individual cell. A positive response was observed with UV radiation, benzo(a)-pyrene, aflatoxin B1, ethylmethanesulfonate, 1,6-dinitropyrene, 2-acetylaminofluorene, and tobacco smoke condensate but not 7,12-dimethylbenz(a)anthracene or 2,3,7,8-tetrachlorodibenzo-p-dioxin. These results demonstrate that HMEC from all 5 women examined have the ability to metabolize a variety of environmental chemicals to DNA-reactive forms. Furthermore, some chemicals known either to cause mammary cancer in rodents or to be contaminants in human breast tissue are genotoxic in HMEC. A positive response in passage 9 cultures was observed only with direct acting agents, suggesting that HMEC may lose their metabolic capabilities in longer-term cultures. The HMEC UDS assay may be used to address the role of environmental agents in human breast cancer by determining whether chemicals are DNA reactive or metabolized to DNA reactive species in this critical target tissue.
[42] T.R. Sutter, S.R. Eldridge, B.E. Butterworth, and W.F. Greenlee, CIIT, Research Triangle Park, NC 27709.
Identification of a battery of genes responsive to 2,3,7,8-tetrachlorodibenzo-p-dioxin in the human mammary epithelial cell line 184B5. 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD), a contaminant in certain industrial emissions and commercial products, serves as the prototype for the halogenated aromatic compounds (HACs), a large group of environmental toxicants that include the dibenzo-p-dioxins, dibenzofurans, biphenyls, and azo- and azoxybenzenes. Several studies indicate that substantial concentrations of these chemicals are present in human breast milk and fat; whether these chemicals contribute to the etiology of breast cancer is unknown. TCDD is a very potent carcinogen in rodents. Furthermore, the tumor-promoting activity for TCDD has been demonstrated in several initiation-promotion model systems. TCDD promotes the formation of hepatocellular carcinomas in rats, papillomas in mouse skin and transformed foci in C3H10T1/2 cells. Comparison of the promoting activities of TCDD and 12-O-tetradecanoylphorbol-13-acetate (TPA) in mouse skin and C3H10T1/2 cells indicates that TCDD is approximately two to four orders of magnitude more potent than TPA. Although the mechanisms of action of TCDD remain unknown, it has been shown that TCDD binds to an intracellular receptor protein )Ah-R). This TCDD::Ah-R complex affects gene transcription, e.g., the regulation of the cytochrome P450IA1 gene, by interacting with specific DNA recognition sequences. To elucidate the molecular actions of TCDD, we have used the technique of differential hybridization to isolate a set of novel TCDD-responsive genes from a cDNA library constructed from a human keratinocyte cell line (SCC-12F/c12c2) that was previously shown to express a variety of Ah-R dependent responses when treated with TCDD. By Northern analysis, we show that levels of these RNAs are increased, time-dependently, in response to TCDD and, in some cases, also to growth factors and calcium. The human mammary epithelial cell lines 184B5 (HMEC) was treated for 24h with either 10 nM TCDD or the vehicle control (0.1% DMSO). Total RNA was isolated and analyzed by Northern blot using the keratinocyte-derived TCDD-responsive cDNA clones. Many of these genes, including P450IA1, were also expressed in response to TCDD in the HMEC. As such, this gene battery should provide a useful indicator of the biological (Ah-R-dependent gene regulation) potency of accumulated HACs in human breast fat.
B.E. Butterworth,S.R. Eldridge, CS Sprankle, PK Working, KS Bentley, ME Hurtt. Tissue-specific genotoxic effects of acrylamide and acrylonitrile, Env. Molec. Carcin. 20:148-155, 1992.
[43] Allan Tischler, Gerri Levine and Jack Bartley, Metabolism of Benzo[a]pyrene and Benzo[a]pyrene-7,8-dihydriol in Human Mammary Epithelial Cells: Feedback Inhibition by 7-Hyrdoxybenzo[a]pyrene. Carcinogenesis, 1991, 12 (9)
The metabolism of benzo[a]pyrene and (-)-trans-benzo[a]pyrene-7,8-dihydrodiol was compared in human mammary epithelial cells grown in serum-free medium, MCDB-170. Conversion of (-)-trans-benzo[a]pyrene-7,8-dihydrodiol to the carcinogen (+)-trans-benzo[a] pyrene-7,8-dihydroxy-9,10-epoxide, as measured by analysis of their tetraol hydrolysis products, occurred much more efficiently in cultures incubated with 3H-(-)-trans-benzo[a]pyrene-7,8-dihydrodiol than in cultures incubated with 3H-benzo[a]pyrene. In cultures pretreated with unlabeled benzo[a]pyrene (24 hrs., 400 nM), the conversion of 3H-(-)-trans-benzo[a]pyrene-7,8-dihydrodiol to 3H-tetraols is inhibited 49%, while the conversion of 3H-benzo[a]pyrene to 3H-(-)-trans-benzo[a]pyrene-7,8-dihydrodiol is not affected. These observations led to the identification of a major benzo[a]pyrene-derived metabolite as 7-hydroxybenzo[a]pyrene, which was found to be an extremely potent and selective inhibitor of the conversion of (-)-trans-benzo[a]pyrene-7,8-dihydrodiol to (+)- trans-benzo[a]pyrene-7,8-dihydroxy-9,10-epoxide, with a Ki estimated at 3 - 12nM. Thuse benzo[a]pyrene activation in human mammary epithelial cells appears to be significantly limited by a feedback inhibition pathway induced by 7-hydroxybenzo[a]pyrene. The potency and selctivity of the 7-hydroxybenso[a]pyrene-induced inhibition suggests that the dio to diolepoxide conversion is affected by a selective oxygenase in human mammary epithelial cells, rather than a non-enzymatic, peroxy radical-induced mechanism. 7-Hydroxybenzo[a]pyrene should prove to be a valuable tool in the study of benzo[a]pyrene carcinogenesis.
[44] Tony Leadon, U. North Carolina
We have been carrying out extensive studies on the metabolic activation of complex mixtures of polycyclic aromatic hydrocarbons using HMEC derived from specimen no. 184. In these studies, HMEC are exposed to either benzo(a)pyrene (B(a)P) or residues from eight different manufactured gas plant sites. The DNA is then extracted and the presence of direct DNA adducts and indirect oxidative DNA damage is measured. The direct adducts are quantitated using a 32P-postlabeling assay coupled to a high performance liquid chromatography, while the oxidative DNA damage is detected using a monoclonal antibody against thymine glycol. We find that the HMEC metabolize the complex residues to produce a variety of DNA adducts and oxidative DNA damage. While each site produces multiple DNA adducts, they all produce an adduct that co-chromatographs with B(a)P-DNA adduct standard. The relative amounts of direct adducts and oxidative damage varies between the sites, which most likely reflects the different compositions of the site residues.
We are also examining the mechanism by which metabolic activation of benzo(a)pyrene leads to the production of oxidative DNA damage. In collaboration with Dr. Allan Tischler at LBL, we found that the majority of the oxidative damage is produced by activation of the arachadonic acid cascade. Using inhibitors of two major enzymes involved in the metabolism of arachadonic acid, cyclooxygenase and 5-lipoxygenase, we found that the levels of oxidative damage induced by B(a)P was significantly reduced. We are currently measuring the levels of arachadonic acid metabolites, such as prostaglandins and leukotrienes, in HMEC in order to verify that the inhibitors are actually having an effect on this cascade.
Questions or comments for the newsletter: Have any labs had difficulty re-establishing HMEC from frozen stocks? Any consensus on the best way to store the cells in liquid nitrogen (vapor phase vs. total immersion in liquid)?
[45] Regine Goth-Goldstein, Lawrence Berkeley Laboratory, in collaboration with Allan Tischler and Tony Leadon
The effect of compounds contained in tobacco smoke on BaP metabolism was studied in HMEC. Various compounds, in particular catechol and the tobacco-specific nitrosamine NNK, induced the metabolism of BaP, as did a condensate of tobacco smoke. Therefore, the compounds exhibit a cocarcinogenic effect, since they enhance the activation of BaP to an ultimate carcinogen.
Besides the cocarcinogenic effects of tobacco smoke, the direct effects on DNA of HMEC werre investigated, i.e., formation of DNA adducts and oxidative damage. Main strean ands side stream tobacco smoke condensates were prepared. Treatment of cells with each fraction resulted in formation of bulky DNA adducts and thymine glycols at similar levels, but side stream was found more potent.
The effect of another environmental toxin, coal tar, on BaP metabolism has been tested. It was found that coal tar is a potnt inducer of BaP metabolinm and that it is equally effective as pure BaP, even though BaP constitutes less than 1% of coal tar.
Fututre experiments: The studies with tobacco smoke condensates will be continued and the contributions of particle phase and vapor phase of tobacco smoke will be examined.
The coal tar studies will be expanded using induction of BAP metablism in HMEC as a biomarker for the health risk of these residues.
We want to characterize the various p450 enzymes expressed in human mammary tissue.
Some other published papers using HMEC from our lab:
Imam, S. A., Esteban, E. F., Chen, R. S., Cardiff, R. D., Taylor, C. R., Identification of a cell-surface antigen (LEA.135) associated with favorable prognosis in human breast cancer. Cancer Res., 53: 3233-3236, 1993.
Campbell, T., Skilton, R. A., Coombes, R. C., Shousha, S., Graham, M. D., Luqmani, Y. A., Isolation of a lactoferrin cDNA clone and its expression in human breast cancer. Br. J. Cancer, 65: 19-26, 1992.
Cook, P. W., Mattox, P. A., Keeble, W. W., Pittelkow, M. R., Plowman, G. D., Shoyab, M., Adelman, J. P., Shipley, G. D., A heparin sulfate-regulated human keratinocyte autocrine factor is similar or identical to amphiregulin. Mol. & Cell. Bio., 11: 2547-2
557, 1991.
Rubin, J. S., Osada, H., Finch, P. W., Taylor, W. G., Rduikoff, S., Aaronson, S. A., Purification and characterization of a newly identified growth factor specific for epithelial cells. Proc. Natl. Acad. Sci. USA, 86: 802-806, 1989.
Kraus, M. H., Issing, W., Miki, T., Popescu, N. C., Aaronson, S. A., Isolation and characterization of ERBB3, a third member of the ERBB/epidermal growth factor receptor family: evidence for overexpression in a subset of human mammary tumors. Proc. Natl.
Acad. Sci. USA, 86: 9193-9197, 1989.
Bottaro, DP., Rubin, JS, Faletto, DL, Chan, A.M, Kmiecik, TE, Vande Woude, GF, Aaronson, SA, Identification of the hepatocyte growth factor receptor as the c-met proto-oncogene product. Science, 251: 802-804, 1991.
Chan, A. M., Rubin, J. S., Bottaro, D. P., Hirschfield, D. W., Chedid, M., Aaronson, S. A., Identification of a competitive HGF antagonist encoded by an alternative transcript. Science, 254: 1382-1385, 1991.
Rubin, J. S., Chan, A. M.-L., Bottaro, D. P., Burgess, W. H., Taylor, W. G., Cech, A. C., Hirschfield, D. W., Wong, J., Miki, T., Finch, P. W., Aaronson, S. A., A broad-spectrum human lung fibroblast-derived mitogen is a variant of hepatocyte growth facto
r. Proc. Natl. Acad. Sci. USA, 88: 415-419, 1991.
There are a number of other groups with human mammary cell resources. If we develop an E-mail network in the future, I hope these resouces can be listed. In the meantime, I would like to bring attention to at least a few other
s.
Steve Ethier, at the University of Michigan, has been working on methods to grow the tumor cells from primary and metastatic human breast tumors. He is open to assisting others.
Ethier, S., Mahacek, M., Gullick, W., Frank, T., Weber, B., Differential isolation of normal luminal mammary epithelial cells and breast cancer cells from primary and metastatic sites using selective media. Cancer Res., 53: 627-635, 1993.
Vimla Band, at the New England Medical Center, has been working on immortalizing HMEC by a variety of methods. She is open to sharing her papilloma virus transformed cells. (see references also under p53)
Joyce Taylor-Papadimitriou, at the ICRF in London, has been a pioneer in growth, characterization, and transformation of HMEC. Some of her recent references are listed below.
Bartek, J., Bartkova, J., Kyprianou, N., Lalani, E.-N., Staskova, Z., Shearer, M., Chang, S., Taylor-Papadimitriou, J., Efficient immortalization of luminal epithelial cells from human mammary gland by introduction of simian virus 40 large tumor antigen with a recombinant retrovirus. Proc. Natl. Acad. Sci. USA, 88: 3520-3524, 1991.
Berdechevsky, F., Gilbert, C., Shearer, M., Taylor-Papadimitriou, J., Collagen-induced rapid morphogenesis of human mammary epithelial cells: The role of the alpha-2/ beta-1 integrin. J. Cell Science, 102: 437-446, 1992.
D'Souza, B., Berdichevsky, F., Kyprianou, N., Taylor-Papadimitriou, J., Collagen-induced morphogenesis and expression of the alpha-2 integrin subunit is inhibited in c-erbB2-transfected human mammary epithelial cells. Oncogene, 8: 1797-1806, 1993.
Kovarik, A., Peat, N., Wilson, D., Gendler, S. J., Taylor-Papadimitriou, J., Analysis of the tissue-specific promoter of the MUC1 gene. J. Biol. Chem., 268: 9917-9926, 1993.
Shearer, M., Bartkova, J., Bartek, J., Berdichevsky, F., Barnes, D., Millis, R., Taylor-Papadimitriou, J., Studies of clonal cell lines developed from primary breast cancers indicate that the ability to undergo morphogenesis in vitro is lost early in mali
gnancy. Int. J. Cancer, 51: 602-612, 1992.
After a long pause, these mammary newsletters continue. My main personal goal in the original newsletters was to provide what I considered some basic information on the cell system for the users of HMEC that I distributed. The newsletters served this function well - the old editions are part of the "care package" of information I send to investigators starting to work with the HMEC. The main secondary goal of the newsletter has been to facilitate communication among investigators using HMEC. I suspect the older editions were not terribly successful at this goal, and it is this goal (along with sharing of resources) that I now would like to encourage. As I said in the cover letter, I think the optimal method to achieve this goal for the future would be an E-mail network. In the meantime, I present in this newsletter some summaries from the past four years - consisting of abstracts from papers and meetings, and reports from investigators. For the future, I think it would work best to have some define
d categories (and subcategories) for listing and finding information. I have tried to organize the material in this newsletter by categories, but without keywords to permit cross referencing, it can be somewhat arbitrary under what category to place information.
Maintaining proper pH is essential for the good health of these cells. The Hepes buffered medium is much better at this than bicarbonate buffered medium. I highly recommend its use if you have the incubator space.
Normal and malignant rodent and human mammary epithelial cells are able to synthesize and to respond to a number of different peptide growth factors and growth inhibitors (1). Among these different families of growth factors, the epidermal growth factor (EGF)-related peptides perform an important role in the the normal development of the mammary gland and in the pathogenesis of breast cancer in rodents and humans (2,3). The EGF superfamily includes a number of proteins of diverse function (4). The peptide growth factors that are members of this familty of proteins are EGF, heparin-binding EGF (HB-EGF), transforming growth factor a (TGFa), amphiregulin (AR), betacellulin, heregulin a and b (HRG), cripto-1 (CR-1) and a series of DNA pox virus-derived pepides that includes vaccinia virus growth factor (VGF), Shope fibroma growth factor (SFGF) and myxoma virus growth factor (MGF). Several of these growth factors such as EGF, HB-EGF, TGFa, AR, HRGa and AR are synthesized as glycosylated, membrane-associated precursors that are biologically active and can function as such through a juxtacrine pathway. In addition, they can also serve as potential cell-cell adhesion molecules. Some of these peptides such as HB-EGF, AR and HRGa can bind to heparin and can probably be sequestered in the ECM where they would be associated with heparin sulfate-containing proteoglycans. AR and HB-EGF have the additional property that they possess concensus nuclear targeting sequences of basic amino acids in their hydrophilic NH2-terminal regions that resembles the nuclear localization sequences found in the SV40 large T antigen, c-myc and the steroid hormone receptors. Nuclear and occasionally nucleolar localization of AR has been observed in normal HMEC and in the immortalized, no
ntransformed human mammary epithelial cell lines 184A1N4 (A1N4) and MCF-10A. AR expression is substantially enhanced in primary human breast carcinoma cells (5,6). All of these growth factors with the exception of the HRGs and members of the CR-1 subfamily bind to and activate the 170 kDa EGF receptor (EGF-R, c-erb B) which itself is a member of superfamily of structurally and functionally related cell surface receptors that have an associated intracellular tyrosine kinase activity. Other members of this family include c-erb B-2/ HER-2, c-erb B-3/ HER-3 and c-erb B-4/ HER-4. AR is as potent as EGF in stimulating the anchorage-dependent proliferation of either A1N4 or MCF-10A cells (7,8). In addition, AR and TGFa can function as cooperating autocrine growth factors in regulating the proliferation of primary cultures of normal human mammary epithelial cells and of A1N4 cells. Treatment of A1N4 cells with 1-10 mM of a 20-mer antisense phosphorothioate oligodeoxynucleotide (S-oliqo) directed against the 5' end of the human AR mRNA but not with a missense S-oligo or infection of these cells with a replication defective, recombinant amphotropic TGFa antisense mRNA retroviral expression vector was able to significantly inhibit the growth of A1N4 cells. (7) AR is expressed in a number of human breast cancer cell lines (6). ER positive cell lines generally express higher levels of AR when compared to the ER negative cell lines. In addition, estrogen treatment induces a 15- to 20-fold increase in the expression of AR mRNA in MCF-7 breast cancer cells. AR is overexpressed in both ras and c-erb B2 transformed MCF-10A cells (8). AR mRNA expression is induced by EGF in both parental and oncogene transformed MCF-10A cells with the transformed cells showing levels of AR mRNA and protein 20- to 30-fold higher than the parental cells. Inhibition of both anchorage-dependent and anchorage independent growth of the oncogene transformed MCF-10A cells could be achieved by using an AR antisense S-oligo. AR protein is expressed in approximately 80% of human primary breast carcinomas as detected by immunocytochemical analysis and in 43% of non-involved mammary epithelium adjacent to carcinoma but at much lower levels when compared to the carcinomas (5). The pattern of stainings mostly cytoplasmic, but nuclear and occassionally nucleolar staining was observed.
2. Vonderhaar, B.K. 1988, Regulation of development of the normal mammary gland by hormones and growth factors. Cancer Treat. Res. 40:252-266.
3. Salomon, D.S., Dickson, R.B., Normanno, N., Saeki, T., Kim, N., Kenney, N. and Ciardiello, F. 1992. Interaction of oncogenes and growth factors in colon and breast cancer. In: Current Perspectives on Molecular & Cellular Oncology, Vol 1, Part B, pages
211-260, ed. D.A. Spandidos. JAI Press Ltd., London.
4. Salomon, D.S., Kim, N., Saeki, T, and Ciardiello, F. 1990, Transforming growth factor-a: An oncodevelopmental growth factor. Cancer Cells. 2:389-397.
5. Qi, C.-F., Liscia, D.S., Merlo, G., Normanno, N., Johnson, G, Gullick, W.J.,Ciardiello,F., Brandt, R., Kim, N., Kenney, N., and Salomon, D.S. Transforming growth factor a, amphiregulin, and cripto-1 in human breast carcinomas. Br. J. Cancer (submitted)
.
6. Normanno, N., Qi, C.-F., Gullick, W.J., Persico, G., Yarden, Y., Wen, D., Plowman, G.,Kenney, N., Johnson, G., Kim, N., Brandt, R., Martinez-Lacaci, I., Dickson, R.B., and Salomon, D.S. 1993. Expression of amphiregulin, cripto-1, and heregulin a in hu
man breast cancer cells. Int. J. Oncol. 2: 903-911.
7. Kenney, N., Johnson, G., Selvam, M.P., Kim, N., Qi, F.-F., Saeki, T., Brandt, R.,Jones, B.-W., Ciardiello, F., Shoyab, M., Plowman, G., Day, A., Salomon, D.S., and Normanno, N.: Transforming growth factor a (TGFa) and amphiregulin (AR) as autocrine gr
owth factors in nontransformed immortalized 184A1N4 human mammary epithelial cells. Mol. Cell. Diff. (in press).
8. Normanno, N., Selvan, M.P., Saeki, T., Johnson, G., Brandt, R., Kim, N., Ciardiello, F., Shoyab, M., Plowman, G., Todaro, G., and Salomon, D.S.: Amphiregulin is an autocrine growth factor for c-Ha-ras and c-erb B-2 transformed human mammary epithelial
cells. Proc. Natl. Acad. Sci. USA (submitted).
9. Ciardiello, F., Dono, R., Kim, N., Persico, M.G., and Salomon, D.S. 1991. Expression of cripto, a novel gene of the epidermal growth factor gene family, leads to in vitro transformation of a mouse mammary epithelial cell line. Cancer Res. 51: 1051-1054
.
We have recently isolated several genes that are structurally homologous to Rb or SV40 large T. We are testing the effect of transfecting these genes into the A1N4, A1N4-Ras cells on their growth behavior.
Fung, Y.-K., T'Ang, A., Murphree, A. L., Zhang, F.-H., Qiu, W.-R., Wang, S.-W., Shi, X.-H., Lee, L., Driscoll, B., Wu, K.-J., The Rb gene suppresses the growth of normal cells. Oncogene, 8: 2659-2672, 1993.
1. Post-translational modifications of CLP: We plan to use primary HMEC cultures to isolate sufficient quantities of the "real" (not bacterially expressed) CLP for studies (e.g., by mass spectrometry) of its true sequence and potential post-translational modifications (N-acetylation, lysine methylation, phosphorylation). We hope to be able to obtain microgram quantities of sufficiently pure protein by applying a one-step phenylsepharose purification procedure to soluble extracts from about 5 ¥ 107 cells.
2. Regulation of CLP gene expression: We are in the process of assembling reporter gene constructs containing various fragments of the CLP gene upstream flanking DNA in front of a reporter (growth hormone, luciferase) gene. These constructs will be used in transient expression assays in both HMEC and transformed cells to obtain insights into the potentially tissue-/cell-specific promoter elements governing CLP gene expression.
3. Studies on the function of CLP: We plan to re-introduce CLP into transformed epithelial cells by means of suitable eukaryotic expression vectors and will try to "knock out" CLP expression in normal HMEC cultures by antisense strategies. Analyzing the effects on cell growth characteristics of CLP overexpression in cells that do not normally produce this protein, or of reducing CLP levels in cells that normally do contain CLP may yield information on the possible role(s) played by this protein in vivo. Extracts prepared from normal HMEC cultures will also be used to search for target proteins that may interact in a Ca2+-dependent way with CLP.
Koller, M., Schnyder, B., Strehler, E. E., Structural organization of the human CaMIII calmodulin gene. Bioch. et Biophys. Acta, 1087: 180-189, 1990.
Koller, M., Baumer, A., Strehler, E. E., Characterization of two novel human tetropseudogenes related to the calmodulin-encoding gene, CaMII. Gene, 97: 245-251, 1991.
Rhyner, J. A., Koller, M., Durussel-Gerber, I., Cox, J. A., Strehler, E. E., Characterization of the human calmodulin-like protein expressed in Escherichia coli. Biochemistry, 31: 12826-12832, 1992.
Durussel, I., Rhyner, J. A., Strehler, E. E., Cox, J. A., Cation Binding and conformation of human calmodulin-like protein. Biochemistry, 32: 6089-6094, 1993.
Berchtold, M. W., Koller, M., Egli, R., Hameister, H., Strehler, E. E., Localization of the intronless gene coding for calmodulin-like protein CLP to human chromosome 10p13-ter. Hum. Genet., 90: 496-500, 1993.
Koller, M., Strehler, E. E., Functional analysis of the promoters of the human CaMIII calmodulin gene and of the intronless gene coding for calmodulin-like protein. Biochim. Biophys. Acta, 1163: 1-9, 1993.
Ed response: Rafe Guzman in Ranu Nandi's lab did try growing our normal and immortal cell lines in the cleared fat pad of nude mice, with and without the presence of mammary fibroblasts. They did not grow. I'd be happy to hear from anyone else who may have tried variations on this.
The karyotype of a growth variant, obtained from heavy-ion irradiated 184B5 cells, has been done with G-banding technique; and, chromosome markers of 184B5 were also found in the growth variants. Karyotyping of other transformed cells cloned from growth variants is in progress. In preliminary studies, we found that both Rb and p53 genes were present in radiation transformed cells. Additional studies are planned to detect possible small changes in these tumor suppressor genes of transformants. The tumorigenicity of transformed cells, which were cloned from foci, will be tested in athymic nude mice.
editor comment: we use vapor phase and have not had problems with viability with cells stored for over 10 years.
Have any labs found alternate (and less expensive) sources for the growth media besides Clonetics?
editor comment: MCDB170 is available from the UCSF Cell Culture facility, 415 476-1450. It requires more effort to put together than the ready made Clonetics media, and is not tested on the cells. I believe other companies may sell MCDB170 but I do not know about their quality control.
Email:mrgs@mh1.lbl.gov__|__
index__|__
Back