Elevated S100A8 protein expression in breast cancer cells and breast tumor stroma is prognostic of poor disease outcome.

PURPOSE: Elevated S100A8 expression has been observed in cancers of the bladder, esophagus, colon, ovary, and breast. S100A8 is expressed by breast cancer cells as well as by infiltrating immune and myeloid cells. Here we investigate the association of elevated S100A8 protein expression in breast cancer cells and in breast tumor stroma with survival outcomes in a cohort of breast cancer patients. PATIENTS AND METHODS: Tissue microarrays (TMA) were constructed from breast cancer specimens from 417 patients with stage I-III breast cancer treated at the University of Michigan Comprehensive Cancer Center between 2004 and 2006. Representative regions of non-necrotic tumor and distant normal tissue from each patient were used to construct the TMA. Automated quantitative immunofluorescence (AQUA) was used to measure S100A8 protein expression, and samples were scored for breast cancer cell and stromal S100A8 expression. S100A8 staining intensity was assessed as a continuous value and by exploratory dichotomous cutoffs. Associations between breast cancer cell and stromal S100A8 expression with disease-free survival and overall survival were determined using the Kaplan-Meier method and Cox proportional hazard models. RESULTS: High breast cancer cell S100A8 protein expression (as indicated by AQUA scores), as a continuous measure, was a significant prognostic factor for OS [univariable hazard ratio (HR) 1.24, 95% confidence interval (CI) 1.00-1.55, p = 0.05] in this patient cohort. Exploratory analyses identified optimal S100A8 AQUA score cutoffs within the breast cancer cell and stromal compartments that significantly separated survival curves for the complete cohort. Elevated breast cancer cell and stromal S100A8 expression, indicated by higher S100A8 AQUA scores, significantly associates with poorer breast cancer outcomes, regardless of estrogen receptor status. CONCLUSIONS: Elevated breast cancer cell and stromal S1008 protein expression are significant indicators of poorer outcomes in early stage breast cancer patients. Evaluation of S100A8 protein expression may provide additional prognostic information beyond traditional breast cancer prognostic biomarkers.

Correlation of GREB1 mRNA with protein expression in breast cancer: validation of a novel GREB1 monoclonal antibody.

Studies of gene regulated by estrogen in breast cancer 1 (GREB1) have focused on mRNA levels with limited evidence about GREB1 protein expression in normal and breast cancer cells. A monoclonal antibody that recognizes GREB1 protein in breast tissues could be applied to correlate protein expression with established mRNA expression data. A hybridoma expressing a murine monoclonal antibody targeting a 119 amino acid peptide specific to human GREB1 was generated. The novel monoclonal GREB1 antibody (GREB1ab) was validated for use in Western blotting as well as immunohistochemical (IHC) applications. GREB1ab detects a 216 kDa protein corresponding to GREB1 in estrogen receptor alpha (ERalpha+) breast cancer cells as well as ERalpha- breast cancer cells transduced with a GREB1 expression vector. GREB1ab specificity was verified using an ERalpha antagonist to prevent GREB1 induction as well as a silencing siRNA targeting GREB1 mRNA. GREB1ab was further validated for detection of GREB1 by IHC in breast cancer cell lines and breast tissue microarrays (TMA). ERalpha+ cell lines were observed to express GREB1 while ERalpha- cell lines did not express detectable levels of the protein. Using breast cancer tissue whole sections, IHC with the GREB1ab identified protein expression in ERalpha+ breast cancer tissue as well as normal breast tissue, with little GREB1 expression in ERalpha- breast cancer tissue. Furthermore, these data indicate that GREB1 mRNA expression correlates well with protein expression. The novel monoclonal GREB1ab is specific for GREB1 protein. This antibody will serve as a tool for investigations focused on the expression, distribution, and function of GREB1 in normal breast and breast cancer tissues.

VEGFA links self-renewal and metastasis by inducing Sox2 to repress miR-452, driving Slug.

Cancer stem cells (CSC) appear to have increased metastatic potential, but mechanisms underlying this are poorly defined. Here we show that VEGFA induction of Sox2 promotes EMT and tumor metastasis. In breast lines and primary cancer culture, VEGFA rapidly upregulates SOX2 expression, leading to SNAI2 induction, EMT, increased invasion and metastasis. We show Sox2 downregulates miR-452, which acts as a novel metastasis suppressor to directly target the SNAI2 3'-untranslated region (3'-UTR). VEGFA stimulates Sox2- and Slug-dependent cell invasion. VEGFA increases lung metastasis in vivo, and this is abrogated by miR-452 overexpression. Furthermore, SNAI2 transduction rescues metastasis suppression by miR-452. Thus, in addition to its angiogenic action, VEGFA upregulates Sox2 to drive stem cell expansion, together with miR-452 loss and Slug upregulation, providing a novel mechanism whereby cancer stem cells acquire metastatic potential. Prior work showed EMT transcription factor overexpression upregulates CSC. Present work indicates that stemness and metastasis are a two-way street: Sox2, a major mediator of CSC self-renewal, also governs the metastatic process.

Targeting of RAGE-ligand signaling impairs breast cancer cell invasion and metastasis.

The receptor for advanced glycation end products (RAGE) is highly expressed in various cancers and is correlated with poorer outcome in breast and other cancers. Here we tested the role of targeting RAGE by multiple approaches in the tumor and tumor microenvironment, to inhibit the metastatic process. We first tested how RAGE impacts tumor cell-intrinsic mechanisms using either RAGE overexpression or knockdown with short hairpin RNAs (shRNAs). RAGE ectopic overexpression in breast cancer cells increased MEK-EMT (MEK-epithelial-to-mesenchymal transition) signaling, transwell invasion and soft agar colony formation, and in vivo promoted lung metastasis independent of tumor growth. RAGE knockdown with multiple independent shRNAs in breast cancer cells led to decreased transwell invasion and soft agar colony formation, without affecting proliferation. In vivo, targeting RAGE shRNA knockdown in human and mouse breast cancer cells, decreased orthotopic tumor growth, reduced tumor angiogenesis and recruitment of inflammatory cells, and markedly decreased metastasis to the lung and liver in multiple xenograft and syngeneic mouse models. To test the non-tumor cell microenvironment role of RAGE, we performed syngeneic studies with orthotopically injected breast cancer cells in wild-type and RAGE-knockout C57BL6 mice. RAGE-knockout mice displayed striking impairment of tumor cell growth compared with wild-type mice, along with decreased mitogen-activated protein kinase signaling, tumor angiogenesis and inflammatory cell recruitment. To test the combined inhibition of RAGE in both tumor cell-intrinsic and non-tumor cells of the microenvironment, we performed in vivo treatment of xenografted tumors with FPS-ZM1 (1 mg/kg, two times per week). Compared with vehicle, FPS-ZM1 inhibited primary tumor growth, inhibited tumor angiogenesis and inflammatory cell recruitment and, most importantly, prevented metastasis to the lung and liver. These data demonstrate that RAGE drives tumor progression and metastasis through distinct tumor cell-intrinsic and -extrinsic mechanisms, and may represent a novel and therapeutically viable approach for treating metastatic cancers.

Constitutive Raf-1 kinase activity in breast cancer cells induces both estrogen-independent growth and apoptosis.

Overexpression of many growth factor receptors, as well as growth factors, has been shown to confer varying degrees of estrogen-independent growth on estrogen receptor (ER) positive breast cancer cells. The proto-oncogene Raf-1 is a key intermediate in the signal transduction pathway of many of these growth factor receptors, and when constitutively activated in fibroblasts is transforming. To examine the effects of Raf-1 kinase activity on the estrogen-dependent growth of human breast cancer cells, ER + MCF-7 breast cancer cells were stably transfected with an expression construct directing the expression of an amino-truncated protein having constitutive kinase activity. Expression of constitutively activated Raf in MCF-7 cells is incompatible with growth in the presence of estrogen; that is, cells down-regulate expression of the transfected Raf. Constitutive Raf activity does allow for growth of the cells in the absence of estrogen, suggesting that activation of growth factor signaling pathways through Raf may confer a selective advantage for growth of breast cancer cells under estrogen-deprived conditions. In addition, the high levels of Raf activity induce apoptosis in cells grown under either condition. This is a novel activity for Raf, and may occur because the levels of the constitutive Raf are extremely high in these cells.

Expression of DNA methyl-transferase (DMT) and the cell cycle in human breast cancer cells.

Estrogen receptor (ER)-negative breast cancer cells display extensive methylation of the ER gene CpG island and elevated DNA methyltransferase (DMT) expression compared to ER-positive cells. The present study demonstrates that DMT protein levels tightly correlate with S phase fraction in ER-positive cells, whereas ER-negative cells express DMT throughout the cell cycle. In addition, levels of p21CIP1, which disrupts DMT binding to PCNA, are inversely correlated with DMT levels. Therefore increased DMT expression in ER-negative cells is not simply due to elevated S-phase fraction, but rather to more complex changes that allow cells to escape normal cell cycle-dependent controls on DMT expression. Because ER-negative breast tumors often have activated growth factor pathways, the impact of these pathways on DMT expression was examined in ER-positive cells. Stable transfection with fibroblast growth factors (FGFs) 1 and 4 led to increased DMT expression that could not be accounted for by a shift in S phase fraction. Elevated DMT protein expression in FGF-transfectants was accompanied by a significant decrease in p21, again suggesting a reciprocal relationship between these two proteins. However, acquisition of an estrogen-independent phenotype, even in conjunction with elevated DMT levels, was not sufficient to promote ER gene silencing via methylation. These results indicate that multiple steps are required for de novo methylation of the ER CpG island.

Tamoxifen-resistant fibroblast growth factor-transfected MCF-7 cells are cross-resistant in vivo to the antiestrogen ICI 182,780 and two aromatase inhibitors.

Although the antiestrogen tamoxifen has been the mainstay of therapy for estrogen receptor (ER)-positive breast cancer, successful treatment of responsive tumors is often followed by the acquisition of tamoxifen resistance. Subsequently, only 30-40% of patients have a positive response to second hormonal therapies. This lack of response might be explained by mechanisms for tamoxifen resistance that sensitize ER pathways to small amounts of estrogenic activity present in tamoxifen or that bypass ER pathways completely. To elucidate one possible mechanism of tamoxifen resistance, we treated ovariectomized tumor-bearing mice injected with fibroblast growth factor (FGF)-transfected MCF-7 breast carcinoma cells with the steroidal antiestrogen ICI 182,780 or one of two aromatase inhibitors, 4-OHA or letrozole. These treatments did not slow estrogen-independent growth or prevent metastasis of tumors produced by FGF-transfected MCF-7 cells in ovariectomized nude mice. FGF-transfected cells had diminished responses to ICI 182,780 in vitro, suggesting that autocrine activity of the transfected FGF may be replacing estrogen as a mitogenic stimulus for tumor growth. ER levels in FGF transfectants were not down-regulated, and basal levels of transcripts for estrogen-induced genes or of ER-mediated transcription of estrogen response element (ERE) luciferase reporter constructs in the FGF expressing cells were not higher than parental cells, implying that altered hormonal responses are not due to down-regulation of ER or to FGF-mediated activation of ER. These studies indicate that estrogen independence may be achieved through FGF signaling pathways independent of ER pathways. If so, therapies directed at the operative mechanism might produce a therapeutic response or allow a response to a second course of antiestrogen treatment.

Human progesterone receptor complexed with the antagonist RU 486 binds to hormone response elements in a structurally altered form.

Structural and functional properties of human progesterone receptors (PR) bound with the antiprogestin, RU 486, and the progestin agonist, R5020, were compared in order to identify receptor mechanisms responsible for the inability of RU 486 to activate the transcriptional capacity of receptors. RU 486 interaction with human PR did not inhibit receptor transformation as assessed by dissociation of nontransformed 8-10S oligomeric receptors (in vitro and in vivo) and by tight binding of PR to nuclei/chromatin in whole cells. Assays based on immunoprecipitation of PR-DNA complexes with an antibody to human PR and gel retardation were used to analyze the effect of RU 486 on receptor binding to the hormone response element (HRE) of the mouse mammary tumor virus (MMTV). RU 486 did not impair PR recognition of the MMTV HRE. Quantitative affinity constants and kinetic parameters of PR binding to these specific DNA sites were similar for receptors complexed with either agonist or antagonist. However, PR-RU 486 complexes exhibited an altered sedimentation rate on sucrose gradients and a faster mobility when bound to the MMTV HRE as assessed by gel retardation. These results indicate that human PR transformed by RU 486 exhibit no impairment in binding to specific DNA sites of target genes, but when bound to DNA assumes a structural form different from that of the receptor-agonist complexes to activate transcription results from this structural alteration in PR, which does not permit protein-protein interactions required for receptor-mediated induction of gene transcription.

Hyperactivation of MAPK induces loss of ERalpha expression in breast cancer cells.

ERalpha-negative breast tumors tend to overexpress growth factor receptors such as epidermal growth factor receptor or c-erbB-2. Raf-1 is a key intermediate in the signal transduction pathways of these receptors. High levels of constitutive Raf kinase (Deltaraf) activity imparts ERalpha- positive MCF-7 breast cancer cells with the ability to grow in the absence of estrogen. Deltaraf transfectants maintained in estrogen-depleted media showed greatly diminished responses to 17beta-estradiol or the pure antiestrogen ICI 182,780. Western blotting, ligand binding, and immunohistochemistry assays revealed a loss of ERalpha protein expression, and ribonuclease protection assays indicated that this correlated with loss of ERalpha message. In examining the basal expression of estrogen-induced genes in the stable transfectants or in transient cotransfection assays with an estrogen-response element- reporter construct and Deltaraf or constitutively active MAPK kinase (DeltaMEK), no ligand- independent activation of ERalpha was observed. Transient expression of Deltaraf and double-label immunostaining showed ERalpha was lost in those cells that transiently expressed Deltaraf. Abrogation of Raf signaling via treatment with the MEK inhibitors PD 098059 or U0126 resulted in reexpression of ERalpha. Similar studies performed with MCF-7 cells overexpressing epidermal growth factor receptor or c-erbB-2 confirmed that hyperactivation of MAPK resulted in down-regulation of ERalpha that was reversible by MEK inhibition or transfection with dominant negative ERK1 and ERK2 constructs. These data suggest that the hyperactivation of MAPK in epidermal growth factor receptor- or c-erbB-2-overexpressing breast cancer cells is directly responsible for generation of an ERalpha-negative phenotype and, more importantly, that this process may be abrogated by inhibiting these pathways, thus restoring ERalpha expression.

VEGF drives cancer-initiating stem cells through VEGFR-2/Stat3 signaling to upregulate Myc and Sox2.

Vascular endothelial growth factor-A (VEGF), a potent angiogenic factor, is also implicated in self-renewal in several normal tissue types. VEGF has been shown to drive malignant stem cells but mechanisms thereof and tumor types affected are not fully characterized. Here, we show VEGF promotes breast and lung cancer stem cell (CSC) self-renewal via VEGF receptor-2 (VEGFR-2)/STAT3-mediated upregulation of Myc and Sox2. VEGF increased tumor spheres and aldehyde dehydrogenase activity, both proxies for stem cell function in vitro, in triple-negative breast cancer (TNBC) lines and dissociated primary cancers, and in lung cancer lines. VEGF exposure before injection increased breast cancer-initiating cell abundance in vivo yielding increased orthotopic tumors, and increased metastasis from orthotopic primaries and following tail vein injection without further VEGF treatment. VEGF rapidly stimulated VEGFR-2/JAK2/STAT3 binding and activated STAT3 to bind MYC and SOX2 promoters and induce their expression. VEGFR-2 knockdown or inhibition abrogated VEGF-mediated STAT3 activation, MYC and SOX2 induction and sphere formation. Notably, knockdown of either STAT3, MYC or SOX2 impaired VEGF-upregulation of pSTAT3, MYC and SOX2 expression and sphere formation. Each transcription factor, once upregulated, appears to promote sustained activation of the others, creating a feed-forward loop to drive self-renewal. Thus, in addition to angiogenic effects, VEGF promotes tumor-initiating cell self-renewal through VEGFR-2/STAT3 signaling. Analysis of primary breast and lung cancers (>1300 each) showed high VEGF expression, was prognostic of poor outcome and strongly associated with STAT3 and MYC expression, supporting the link between VEGF and CSC self-renewal. High-VEGF tumors may be most likely to escape anti-angiogenics by upregulating VEGF, driving CSC self-renewal to re-populate post-treatment. Our work highlights the need to better define VEGF-driven cancer subsets and supports further investigation of combined therapeutic blockade of VEGF or VEGFR-2 and JAK2/STAT3.

Bimodal regulation of epidermal growth factor receptor by estrogen in breast cancer cells.

In breast cancer, epidermal growth factor (EGF) receptor (EGFR) expression is inversely correlated with expression of estrogen receptor (ER) and predicts the prognosis and failure of endocrine therapy. We report here, for the first time, that in ER-positive breast cancer cell lines, MCF-7, T47D, and BT474, 17 beta-estradiol (E2) transiently induced EGFR messenger RNA (mRNA) levels 2- to 3-fold; this induction was prevented by the presence of the antiestrogen ICI 164,384 and was also reflected in the level of EGFR protein. Up-regulation of EGFR mRNA is most likely due to a direct effect of ER on the EGFR gene, with no involvement of protein synthesis, as it was not inhibited in the presence of cycloheximide; however, the subsequent down-regulation of EGFR required de novo protein synthesis. E2 had no effect on EGFR mRNA stability, and EGFR transcript levels were found to parallel EGFR mRNA levels, further supporting a direct transcriptional mechanism in the regulation of EGFR expression by estrogens. Additionally, sequencing of the EGFR promoter revealed putative imperfect estrogen-responsive elements that were capable of binding human ER. The transient nature of EGFR induction by E2, with a rapid return to a basal level that is dependent on protein synthesis, suggests that breast cancer cells possess active mechanisms to maintain low levels of EGFR expression in the presence of estrogen and a functional ER.

Estrogen induction of TGF-alpha is mediated by an estrogen response element composed of two imperfect palindromes.

To investigate the molecular mechanisms underlying the two- to three-fold induction of human transforming growth factor-alpha (hTGF-alpha) mRNA and two- to five-fold induction of hTGF-alpha protein observed following estrogen treatment of hormone-responsive human breast cancer cell lines, the hTGF-alpha promoter was assayed for ERE-like sequences able to mediate estrogen induction of a heterologous gene. Transient co-transfection of a chloramphenicol acetyl transferase (CAT) construct consisting of either 1100 bp or 330 bp of hTGF-alpha promoter sequence and an estrogen receptor expression vector into either COS-7 cells or hormonally responsive MCF-7 human breast cancer cells resulted in a two- to five-fold induction of CAT activity by estrogen. Although no consensus estrogen response element (ERE) exists in the hTGF-alpha promoter, a sequence consisting of two imperfect ERE palindromes separated by 20 bp is located at -200 to -252. This sequence was inserted into a mouse mammary tumor virus (MMTV) based CAT construct and assayed for its ability to confer estrogen regulation of CAT expression to a heterologous promoter. Transient co-transfection of this construct with an estrogen receptor expression vector into either COS-7 cells or MCF-7 cells resulted in an average 30-fold estrogen induction of CAT activity. Gel shift assays with human recombinant estrogen receptor (ER) and 32P-labelled fragments revealed that the ER could specifically bind to this sequence. These results indicate that this 53 bp sequence can function as an ERE, and is likely to be responsible for the observed induction of TGF-alpha message and protein in response to estrogen. These data also indicate that the level of estrogen inducibility mediated by this element may be positively or negatively modulated by interaction or competition with other transcription factors.

Use of ERE and reporter gene constructs to assess putative estrogenic activity.

Estrogens primarily function through the activation of their receptors, which subsequently function as nuclear transcription factors. There are two estrogen receptor (ER) genes, now designated ERa (the classic ER gene) and ER/3. The key consequence of the activation of either gene product is the regulation of gene transcription. The extent and nature of transcription appear to be regulated by a series of coregulator proteins. One of the most sensitive assays for detection of potential estrogenic activity is measurement of the ability of a test compound to influence the transcription of reporter genes. In this regard, many investigators use promoter-reporter constructs. To assess putative estrogenic activity, an estrogen-responsive promoter is generally placed upstream of a reporter gene and transiently transfected into a target cell. When exposed to an estrogenic compound, expression of the reporter gene would normally be induced. We briefly discuss several issues pertinent to the use of these assays and the interpretation of resulting data, including estrogen-responsive, promoter-reporter constructs, reporter genes and measurements of activity, choice of target cell or cell line, transient introduction of promoter-reporter constructs into cells, basic statistical approaches to data analysis, and definitions of agonist, partial agonist, and antagonist.

Which patients with primary biliary cirrhosis or primary sclerosing cholangitis should undergo endoscopic screening for oesophageal varices detection?

BACKGROUND: Recent guidelines from an AASLD Single Topic Symposium suggest that patients with cirrhosis, including those with primary biliary cirrhosis (PBC) or primary sclerosing cholangitis (PSC), should be screened for oesophageal varices when the platelet count is <140,000/mm3. AIM: To determine the validity of these guidelines in clinical practice in patients with PBC or PSC. METHODS: Retrospective review of individuals undergoing screening upper endoscopy for oesophageal varices at a single centre. Oesophageal varices were reported as being present or absent. RESULTS: A total of 235 patients with chronic liver disease, including 86 patients with PBC (n=79) or PSC (n=7), 104 patients with chronic viral hepatitis, and 45 with non-alcoholic cirrhosis of differing aetiologies, underwent a single screening endoscopy between 1996 and 2001. Oesophageal varices were detected in 26 (30%) of the PBC/PSC group, 38 (37%) of the viral hepatitis group, and 21 (47%) of the "other" group. Applying multiple logistic regression analysis to the data in the group with PBC/PSC, platelets <200,000/mm3 (odds ratio (OR) 5.85 (95% confidence interval (CI) 1.79-19.23)), albumin <40 g/l (OR 6.02 (95% CI 1.78-20.41)), and serum bilirubin >20 micromol/l (OR 3.66 (95% CI 1.07-12.47)) were shown to be independent risk factors for oesophageal varices. Prothrombin time was unhelpful. The values at these cut offs were not useful in predicting oesophageal varices in the other groups. CONCLUSION: We conclude that current guidelines recommended by the AASLD Single Topic symposium are invalid in our cohort of patients with PBC and PSC. Patients with a platelet count <200,000/mm3, an albumin level <40 g/l, and a bilirubin level >20 micromol/l should be screened for oesophageal varices.

Molecular biology of breast carcinoma.

During the last several years basic research has resulted in the identification of many of the factors involved in signal transduction pathways, leading us to a greater understanding of the mechanisms of growth control in breast cancer cells. Many of these factors are the products of proto-oncogenes or suppressor genes. This review describes the role of some of these factors in breast cancer development, progression, and metastasis and discusses implications for future directions.

Mapping contacts between unpurified human progesterone receptor and the hormone response element of mouse mammary tumor virus.

Binding of steroid hormone receptors to specific recognition sites of hormone-inducible genes is one of the events required for hormonal regulation of gene transcription. We have employed an immunoprecipitation assay to map the interaction between unpurified human progesterone receptors from crude nuclear extracts of T47D cells and the hormone response element of the mouse mammary tumor virus (MMTV). DNase I footprints and methylation interference patterns are similar to those reported with highly purified rabbit progesterone receptors, suggesting that both human and rabbit receptors recognize similar features in the hormone response element. More importantly, these patterns suggest that if other factors are associated with unpurified nuclear receptor, they do not alter the contacts made by receptor nor do they make contacts themselves with MMTV DNA in a manner detected by DNase I or methylation interference assays. The sites of interaction of receptors bound with the clinically important progestin antagonist, RU 486, are comparable to those observed with an agonist-receptor complex. These results suggest that the antagonist prevents receptor action at a step after its recognition and binding to specific sites on a hormone-responsive enhancer element.

MCF-7 breast cancer cells overexpressing transfected c-erbB-2 have an in vitro growth advantage in estrogen-depleted conditions and reduced estrogen-dependence and tamoxifen-sensitivity in vivo.

A c-erbB-2 expression vector was transfected into the estrogen receptor positive (ER+) MCF-7 human breast cancer cell line to determine if overexpression of this transmembrane tyrosine kinase could increase the malignant phenotype of this cell line. Loss of transfected c-erbB-2 expression was observed when cells were carried in medium containing estrogen. Homogeneous populations stably overexpressing levels of the 185 kDa c-erbB-2 observed in the SKBR-3 a breast cancer cell line which overexpresses c-erbB-2 as a result of gene amplification could be obtained by continually maintaining the transfected cell lines in estrogen-free conditions. Levels of constitutively activated c-erbB-2 varied among clonal isolates. Whereas some overexpressing lines did acquire the ability to form transient tumor nodules in ovariectomized nude mice without estrogen supplementation, as well as in mice that received the antiestrogen tamoxifen, one cell line that exhibited the highest levels of constitutively activated c-erbB-2 was able to form static tumors of a larger size under both conditions. This same cell line formed progressively growing tumors in estrogen-supplemented mice that were much larger than observed in mice injected with control cell lines, and also showed reduced sensitivity to antiestrogens in vitro, but it continued to have a low metastatic phenotype. These results suggest that signal transduction mediated by the c-erbB-2 tyrosine kinase can partially overcome the estrogen dependence of ER+breast cancer cells for growth and that c-erbB-2 overexpression confers a selective advantage to such cells in the absence of estrogen.

Emergence of MCF-7 cells overexpressing a transfected epidermal growth factor receptor (EGFR) under estrogen-depleted conditions: evidence for a role of EGFR in breast cancer growth and progression.

Overexpression of epidermal growth factor receptor (EGFR) is correlated with loss of estrogen receptor and poor prognosis in breast cancer. To investigate this phenomenon, we transfected a cytomegalovirus expression vector directing the expression of EGFR into estrogen receptor-positive MCF-7 breast cancer cells and into a clone of MCF-7 cells previously transfected with transforming growth factor alpha. Cells arising from single clones or pooled polyclonal populations maintained in charcoal-stripped calf serum, a medium devoid of estrogen, overexpressed EGFR. Switching these cells to a medium containing fetal calf serum or charcoal-stripped calf serum plus 17 beta-estradiol resulted in the emergence of a population expressing low EGFR levels. Loss of expression was not a consequence of nonspecific repression of the cytomegalovirus promoter, because expression of the fibroblast growth factor (FGF)-4 complementary DNA in a similar vector was not lost in fetal calf serum. While loss of EGFR overexpression in fetal calf serum was seen at both the protein and mRNA levels, Southern blotting shows that this was not due to loss of the transfected gene. Subclones of a cell population with low EGFR expression were capable of increasing expression upon estrogen withdrawal, demonstrating that the changes in EGFR expression were reversible and suggesting a growth advantage conferred by EGFR overexpression under these restrictive growth conditions. Overexpression of EGFR did not result in loss of ER expression. These results suggest a role for overexpression of EGFR in the growth of estrogen receptor-positive breast cancer cells in the absence of estrogen.