Search

The WHSC1L1 gene is amplified and overexpressed in ∼12% of breast cancers, and is thought to be a driving oncogene when amplified and overexpressed, however the mechanism of transformation is not known. SUM44 cells harbor an amplification of WHSC1L1 as part of the 8p11-p12 amplicon, and express high levels of both the long and short isoforms of the gene. The short isoform, containing only a PWWP domain and no catalytic SET domain, is the major species present in SUM44 cells. To create a model to study the effect of WHSC1L1 on gene regulation, SUM44 cells were lentivirally transduced with either a control shRNA for LacZ (shLacZ), an shRNA specific for both isoforms of WHSC1L1 (shWHtotal), or a short isoform-specific shRNA (shWHshort). Cell growth was significantly reduced in shWHtotal cells compared to shLacZ, however, the greatest decrease in growth was observed in cells transduced with the short isoform-specific shRNA. To identify genes transcriptionally regulated by WHSC1L1, we performed genome-wide expression profiling of SUM44 breast cancer cells transduced with either shWHtotal or shLacZ. A total of 5185 genes were found to be differentially expressed in shWHtotal SUM44 cells compared to shLacZ control. Among the 5185 differentially expressed genes, we found ESR1 to be significantly downregulated (-7.3-fold, p < 0.0001) in the shWHtotal cells compared to shLacZ controls. In addition MYCN, which is amplified in SUM-44 cells was reduced by 7.43-fold, (p < 0.004), ERBB3 was reduced by 2.5-fold, (p < 0.0009), and ERBB4, which is highly expressed in SUM-44 cells was reduced by 2.12-fold, (p < 0.00001). Immunoblotting confirmed downregulation of estrogen receptor alpha (ERα) in the shWHtotal compared to shLacZ control. Interestingly, knockdown of WHshort resulted in an even greater decrease of ERα than the shWHtotal knockdown. To determine whether downregulation of ERα affects sensitivity to Tamoxifen, we measured the Tamoxifen LD50 in SUM44 shLacZ and SUM44 shWHtotal cells, and found a 10-fold decrease in Tamoxifen LD50 compared to shLacZ control (3.8nM vs. 35nM). Tamoxifen affected SUM-44 cells in a dose-dependent manner, however 500nM Tamoxifen resulted in complete growth arrest in SUM44 shWHtotal cells, while the cells infected with the shLacZ virus cells grew at the same dose, albeit slowly. Our data indicate that WHSC1L1 regulates ERα expression in SUM44 cells, and WHSC1L1 knockdown increased the cells' sensitivity to Tamoxifen. The decrease in both cell growth and ERα expression in the SUM44 WHshort-specific knockdown cells suggests that the catalytically inactive short isoform of WHSC1L1 may be the isoform driving the oncogenic activity of WHSC1L1 amplification, perhaps through a dominant-negative interaction with the long isoform of WHSC1L1. Data showing increases in global H3K36 methylation levels following knock down of WHSC1L1 are consistent with this hypothesis. Citation Format: Jonathan Curtis Irish, Alexandria Rutkovsky, Stephen P. Ethier. Overexpression of histone methyltransferase WHSC1L1 regulates expression of ER-alpha in SUM44 breast cancer cells. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 5148. doi:10.1158/1538-7445.AM2014-5148

Large-scale efforts such as The Cancer Genome Atlas (TCGA) have identified the landscape of genetic, epigenetic and proteomic alterations in breast cancer through analysis of ∼1000 primary breast tumors. Importantly, all of the samples analyzed by TCGA also come with long-term clinical follow-up data that allows for examination of how identified alterations affect patient outcomes. Although powerful, the sheer number of alterations identified by TCGA and similar efforts makes it difficult to separate alterations that drive cancer from those that are merely passengers. We have been combining data from TCGA with data from genome-wide, RNAi-based, functional genetic screens that we have been performing on a panel of breast cancer cell lines that were developed in our lab. This combination of functional genetic data with data from the TCGA has resulted in the identification of genes that are responsible for driving breast cancer cell growth and survival that also affect patient outcomes. Using this approach, we have identified the TRiC chaperonin protein complex as playing a critical role in driving breast cancer and affecting patient outcomes. In the SUM-52 breast cancer cell line, a member of the TRiC complex known as TCP-1 was a hit in the genome wide, RNAi-based screen and examination of TCP1 in the TCGA database revealed that it is altered in ∼9% (85 of 929 cases) of breast cancers and patients with TCP1 alterations have significantly reduced long-term survival (pval Citation Format: Stephen Guest, Jon Irish, Zachary Kratche, Elizabeth Garrett-Mayer, Robert Wilson, Stephen Ethier. The TRiC chaperonin in breast cancer cell growth, survival and response to taxane therapy. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 440. doi:10.1158/1538-7445.AM2014-440

Nonporous (NPS) RP-HPLC has been used to rapidly separate proteins from whole cell lysates of human breast cell lines. The nonporous separation involves the use of hard-sphere silica beads of 1.5-microm diameter coated with C18, which can be used to separate proteins ranging from 5 to 90 kDa. Using only 30-40 microg of total protein, the protein molecular weights are detectable on-line using an ESI-oaTOF MS. Of hundreds of proteins detected in this mass range, approxinately 75-80 are more highly expressed. The molecular weight profiles can be displayed as a mass map analogous to a virtual "1-D gel" and differentially expressed proteins can be compared by image analysis. The separated proteins can also be detected by UV absorption and differentially expressed proteins quantified. The eluting proteins can be collected in the liquid phase and the molecular weight and peptide maps determined by MALDI-TOF MS for identification. It is demonstrated that the expressed protein profiles change during neoplastic progression and that many oncoproteins are readily detected. It is also shown that the response of premalignant cancer cells to estradiol can be rapidly screened by this method, demonstrating significant changes in response to an external agent. Ultimately, the proteins can be studied by peptide mapping to search for posttranslational modifications of the oncoproteins accompanying progression.

The underlying causes for different apoptotic responses in neoplastic cells are still not fully understood. We demonstrate here that a human breast cancer cell line, MDA-MB-468, which lacks the retinoblastoma protein (RB), is particularly sensitive to low doses of ultraviolet (UV) radiation. These cells are 15-20-fold more sensitive to UV radiation than RB-positive cell lines, as measured by both apoptosis and clonogenic assays. In addition, a prostate cancer cell line that lacks functional RB, DU-145, was found to have a similar apoptotic response to low doses of UV radiation. Based on these data, we hypothesized that the lack of RB is responsible for the extreme sensitivity of these cells to UV-radiation-induced apoptosis. To further examine the role of RB in apoptosis, cells of RB-positive human breast cancer and normal cell lines were infected with the human papilloma virus type 16 (HPV-16) E7 and assessed for UV-radiation sensitivity. The HPV-16 E7 protein is known to decrease levels of free RB in cells. Infection of RB-positive human breast cancer or normal cells with E7 resulted in a 4-5-fold increase in sensitivity to UV radiation compared to controls. The above data suggest a role for the RB protein in protecting cells from undergoing apoptosis in response to UV radiation.

Over the years, a number of groups have developed culture media and conditions that support the growth of normal human mammary epithelial cells from a variety of lineages, and these cell systems have increased our understanding of human mammary gland biology. In contrast, it remains difficult to isolate and culture human breast cancer cells from primary tumors and metastatic specimens. Our laboratory has now developed 13 human breast cancer cell lines. Some of these cell lines originated from primary tumor specimens, others were derived from chest wall recurrences in patients who had prior surgery, and some of the cell lines were derived from pleural effusion metastases. The cell lines that we have obtained exhibit an array of genetic alterations representative of the genomic diversity of uncultured human breast cancers. In addition, these cells are cultured under well-defined conditions (most cell lines are grown in serum-free media) that allow detained cellular studies to be carried out that complement the genetic analyses of these cell lines. This chapter outlines the culture methods and approaches that we have used to develop this panel of cell lines. Our methods emphasize certain key features that we feel are essential for the isolation and culture of breast cancer cells. These features are all focused on developing culture conditions that allow for the slow emergence of cancer cells in the relative absence of normal cells, which proliferate rapidly under highly-growth-factor-enriched conditions. Thus, selective growth media and methods for enriching cell populations for cancer cells while minimizing the presence of normal epithelial and stromal elements are important for the expansion of breast cancer cells in vitro. It is still not possible to develop cell lines from every breast cancer specimen obtained in the laboratory. Thus, whereas the cell lines we have developed have resulted in improvements in our understanding of breast cancer cell biology, much work remains to be done to understand why some breast cancer cells fail to grow in vitro while growing robustly in the patient. Improving our knowledge of the factors that influence breast cancer cell viability and proliferation will be important for development of novel therapeutic strategies for breast cancer.

Since its inception, Breast Cancer Research and Treatment (BCRT) has been a publication source for manuscripts that describe new laboratory, clinical, or epidemiologic data regarding breast cancer. In addition, the Editors have encouraged thoughtful state-of-the-art reviews and commentaries. During this time, the field has witnessed remarkable progress in our understanding of the genetics and molecular biology that drive the breast cancer phenotype and in the evaluation and treatment of patients with this disease. Indeed, these successes have led to a plethora of new diagnostic and treatment strategies that have been accompanied by a dramatic decline in breast cancer mortality in the Western world over the last 30 years [1, 2]. However, these new advances are not without cost. Concomitant with progress has come a concern over the financial implications of supplying these new treatments to patients. This concern has led to an entirely new field of research: analysis of cost effectiveness in medicine in general and oncology specifically [3]. BCRT has, like many journals with clinical content, included manuscripts in which the cost effectiveness of a new treatment has been compared to an existing one [4–18]. Recently, though, the Editors have noticed a disturbing trend in that the submitted manuscripts appear to have been mainly or totally prepared and written by pharmaceutical companies. In these manuscripts, a new therapeutic agent produced by the sponsor is usually compared to other treatments and, not surprisingly, usually with favorable results. After careful deliberation, the Editors have decided that although we remain interested in scholarly cost-effectiveness studies, the rigor of review of such manuscripts will be elevated. In the future, any such study will require a documented statement that either the study was not supported by a pharmaceutical company or if so, that the sponsor had no input into design or analysis of the cost-effective analysis, nor did they have final review. Rather, each author will be asked to document his or her role in the study, and in the future potential authors are encouraged to decline if they feel that their role is insufficient to merit their inclusion. The Editors of BCRT are not alone in our concern regarding apparent or real pharmaceutical influence over manuscript preparation. Recently, editors of The Oncologist expressed similar apprehension, concluding that they will ‘‘accept papers for review only if the article was written, endorsed, and proffered for publication by the authors identified in the byline’’ [19]. We hope that this new policy will enhance the quality of the journal without inhibiting submission of well-designed, and unbiased, manuscripts.

Previously, we demonstrated that human breast cancer cells with progressively elevated levels of constitutively tyrosine phosphorylated erbB-2 are independent of growth factors required by normal human mammary epithelial (HME) cells for proliferation in serum-free medium. To determine whether erbB-2 overexpression alone is sufficient to confer the growth factor-independence phenotype in HME cells, the spontaneously immortalized MCF-10A cell line and the HPV-16-immortalized H16N2 cell line were infected with the bicistronic retroviral vector pTPerbB-2 and tested for their ability to grow in the absence of specific factors. Selection of infected cells in G418-containing medium resulted in moderate levels of erbB-2 overexpression in approximately 40% of cells. The subpopulation of erbB-2 overexpressing cells could be selected for by culturing the cells in medium devoid of insulin. When MCF-10A or H16N2 cells were infected with pTPerbB-2 and directly selected in growth factor-deficient medium over long periods of time, populations of both cell lines emerged that expressed levels of erbB-2 protein equivalent to levels expressed by breast cancer cells with an erbB-2 gene amplification. Furthermore, overexpressed p185(erbB-2) was constitutively tyrosine phosphorylated in these cells. The levels of tyrosine phosphorylated p185(erbB-2) differed in the two recipient lines, with H16N2-erbB-2 cells having higher levels of activated receptor than MCF-10AerbB-2 cells. Furthermore, only the H16N2-erbB-2 cells were independent of both insulin and epidermal growth factor for growth in serum-free medium. Overexpression of erbB-2 also resulted in progressively increasing levels of tyrosine-phorphorylated erbB-3, without any significant changes in p180(erbB-3) levels. These studies demonstrate a direct relationship between the level of expression and activation of p185(erbB-2) and the requirements of HME cells for insulin-like and epidermal growth factor-like growth factors. The results also suggest that genetic alterations present in breast cancer cells, or mediated by HPV-16-induced alterations in pRb and p53, can influence the expression level and activation status of erbB-2 as well as erbB-3 and, in turn, their degree of growth factor independence.

Our laboratory has developed twelve human breast cancer cell lines from primary and metastatic sites. In this report we demonstrate that eight of eight breast cancer cell lines examined exhibit constitutively tyrosine phosphorylated and enzymatically active endogenous pp125fak when grown in monolayer. The activation status of pp125fak in breast cancer cells in monolayer is significantly elevated over that exhibited by normal mammary epithelial cells cultured under the same conditions. Constitutive activation of pp125fak is the only characteristic so far studied that all of these breast cancer cell lines have in common. In contrast to HBC cells, tyrosine phosphorylation of pp125fak in HME cells was low or absent in monolayer culture but was induced to high levels by culturing the cells in Matrigel. Thus tyrosine phosphorylation and activation of pp125fak is a regulated process in normal mammary epithelial cells, but is constitutive in breast cancer cells. Finally, analysis of the ability of normal human mammary epithelial cells and breast cancer cell lines to grow under anchorage-independent conditions indicated that normal human mammary epithelial cells rapidly and uniformly lost viability when not substrate-attached, whereas all of the breast cancer cell lines survived for a 3-week culture period. Furthermore, a subset of the breast cancer cell lines grew to form large colonies under anchorage-independent conditions. Interestingly, pp125fak activation decreased dramatically in HBC cells cultured for two weeks in suspension, suggesting that activation of this kinase is not necessary for long-term growth under anchorage-independent conditions. These results suggest that constitutive activation of pp125fak results in preferential survival of human breast cancer cells under anchorage-independent conditions but that activation of pp125fak is not the sole mediator of anchorage-independent colony formation.

Previous studies have shown that polycyclic aromatic hydrocarbons (PAHs) mobilize intracellular Ca2+ in human T cells by inositol trisphosphate-dependent mechanisms resulting from activation of phospholipase C-gamma by SRC-related protein tyrosine kinases, thereby mimicking antigen-receptor activation. Ca2+ appears to play an important second messenger role in growth factor control of cell proliferation in human mammary epithelial cells (HMEC), such as the epidermal growth factor receptor pathway. The purpose of the present studies was to determine if PAHs are able to increase intracellular Ca2+ in primary cultures of HMEC and increase cell proliferation. Two carcinogenic and two non-carcinogenic PAHs were tested for their ability to increase intracellular Ca2+ in HMEC. The carcinogenic PAHs dimethylbenz[a]anthracene (DMBA) and benzo[a] pyrene (BaP) were able to cause Ca2+ elevation in HMEC at early time points (2 h) and caused sustained alterations in Ca2+ homeostasis (18 h). DMBA showed maximal effects at early time points (2 h), while BaP showed maximal effects on sustained Ca2+ (18 h). 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD), a potent dioxin and tumor promoter, produced maximal Ca2+ elevation at 2 h, with a return to near baseline levels by 6 h. The non-carcinogenic PAHs benzo[e]pyrene and anthracene did not significantly alter intracellular Ca2+ at any time point. alpha-Naphthoflavone significantly reduced the Ca2+ response induced by BaP treatment, but not by DMBA or TCDD, suggesting that P450 1A or 1B metabolism of BaP may be important in the sustained Ca2+ elevating response. In evaluating the effects of BaP on HMEC proliferation, BaP was found to increase the number of cells recovered after 4 days in culture in the absence or presence of various concentrations of epidermal growth factor. These studies provide initial evidence that Ca2+ signaling may be associated with mitogenesis in HMEC, which may play a role in tumor promotion and progression produced by PAHs.

An amplified region of chromosome 8, the 8p11-p12 amplicon, is observed in ∼15% of breast cancers. This region contains 55 genes, of which 22 have been identified as candidate oncogenes based on higher than normal copy number and associated elevated expression levels in breast tumor cells compared to normal breast epithelial cells. Our lab previously identified the short isoform of one amplicon gene, Wolf Hirschhorn Syndrome Candidate 1-Like 1 (WHSC1L1), to be a potent inducer of transformation when over expressed in MCF10A breast epithelial cells in vitro. WHSC1L1 is a member of the NSD family of methyltransferases, and has been shown to methylate lysines 4, 27, and 36 of the histone H3 subunit, however, the potently transforming short isoform (WHshort) lacks the catalytic SET domain. Our lab has shown that over expression of WHshort in the immortalized human breast epithelial cell line MCF10A causes a significant increase in proliferation in insulin-free culture conditions, and the ability to form colonies in soft agar. In SUM44 cells, an ER+ breast cancer cell line harboring the 8p11-p12 amplicon, shRNA knockdown of WHSC1L1 resulted in growth arrest. The WHSC1L1 family member NSD1 has been shown to act as an oncogene in acute myelogenous leukemia by blocking differentiation in developing lymphocytes through H3K36me3-mediated inhibition of the polycomb repressor complex, resulting in constitutive activation of the HOXA gene family. While WHSC1L1 may work by a similar mechanism, little is known about the specific substrates of WHSC1L1, or even whether its oncogenic activity is caused by histone methylation changes. Knockdown by shRNA of WHSC1L1 in SUM44 cells resulted in a significant reduction in transcript levels of several genes important in breast cancer cell proliferation, including ESR1, MYCN (amplified in SUM-44), ERBB3, and ERBB4. Interestingly, Carroll et al. recently reported that ER+ breast tumors can be stratified by gene expression signature into poor outcome and good outcome groups, and that ER-alpha binding site patterns on chromatin were altered in the two groups. In this study, WHSC1L1 was part of the ER-regulated gene signature of the poor outcome group. Therefore, we hypothesize that WHSC1L1 may influence histone H3K36me3 marks at specific gene loci, leading to altered ER-alpha binding that is associated with the expression signatures of poor-outcome ER+ breast cancers. To investigate chromatin changes as a result of changes in WHSC1L1 expression, we have accomplished shRNA knockdown of WHSC1L1 in SUM44 cells using GIPZ and TRIPZ lentiviral vectors. Using these cell lines, we are investigating changes in H3K36me3 marks and ER-alpha binding patterns by chromatin immunoprecipitation followed by sequencing (ChIP-seq). These experiments will help to elucidate the interaction of WHSC1L1 and ER in regulated gene expression in breast cancer cells with the 8p11 amplicon. Citation Format: Jonathan C. Irish, Gang Liu, Stephen P. Ethier. Epigenetic mechanisms of transformation by WHSC1L1 in breast cancer cells with the 8p11-p12 amplicon. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 2964. doi:10.1158/1538-7445.AM2013-2964

Approximately half of all basal type breast cancers show evidence of autocrine activation of the epidermal growth factor receptor (EGFR) by amphiregulin (AREG). In addition, AREG expression is associated with aggressive and chemoresistant forms of breast cancer. Studies in our laboratory show that when AREG is the activating ligand, EGFR accumulates at the cell surface, resulting in altered receptor trafficking and downstream signaling. To investigate the effect of AREG-mediated activation of EGFR on the phenotype of AREG-expressing cancers, AREG expression was knocked-down in SUM149 cells, a highly tumorigenic and metastatic cell line isolated from a basal subtype inflammatory breast cancer, which expresses high levels of AREG and overexpresses EGFR. Microarray analysis of AREG knockdown cells revealed a number of relevant pathways affected by AREG-EGFR signaling. These pathways include Notch signaling, MAPK activation, IL-1 signaling and multiple components of the Wnt signaling pathway. In particular, knockdown of AREG or inhibition of EGFR signaling in AREG stimulated cells resulted in the upregulation of DKK-1, an antagonist of canonical Wnt signaling, and the expression of several WNT target genes. Flow cytometric analysis of these cells demonstrated that AREG knockdown resulted in a significant reduction in CD44+ cells while simultaneously increasing the non-stem cell fraction of CD44-/CD24+ cells. Because CD44hi/CD24lo is characteristic of undifferentiated cells that have the potential to initiate tumor formation, the tumorigenicity of these cells was investigated by transplanting parental and AREG knock-down SUM149 cells into the mammary fat pads of NOD-SCID mice. The tumorigenic activity of the AREG knock-down derivatives was significantly reduced compared to the parental SUM149 cells. In particular, the AREG knockdown clone that expressed the lowest level of AREG expression and the highest level of DKK-1 expression failed to form tumors in NOD-SCID mice. To investigate the link between AREG-mediated activation of EGFR, receptor trafficking, and downstream signaling pathways that are responsible for changes in expression of WNT pathway genes, we are investigating differences in EGFR phosphorylation and downstream signaling stimulated by AREG versus EGF. In conjunction with our previous findings that AREG-mediated EGFR signaling results in activation of NF-κβ and the upregulation of IL1α/β, our current results suggest a link between these pathways, WNT signaling, and tumorigenicity in SUM-149 cells. Citation Format: Christiana S. Kappler, Stephen P. Ethier. Inhibition of amphiregulin-mediated EGFR signaling suppresses tumorigenicity in breast cancer cells. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 530. doi:10.1158/1538-7445.AM2013-530

Data obtained using long-established human breast cancer cell lines have suggested that autocrine growth factors secreted by the cells are important for their growth in vitro. Such data alone cannot definitively establish a role for autocrine growth factors in human breast cancer cell proliferation in vivo, but they create a paradigm that can be tested by analysis of data obtained with primary breast cancer cells and tissues. This review is aimed at examining experimental data obtained using fresh human breast cancer cells and tissues to determine whether the results obtained are consistent with predictions made by autocrine models of human breast cancer cell proliferation. Conceptually, for autocrine loops to be of primary importance in human breast cancer cell proliferation, breast cancer cells in vivo must 1) synthesize biologically active growth factors that are available to growth factor receptors, 2) synthesize the cognate growth factor receptors, 3) require the specific factors for proliferation, and 4) express the autocrine loop as a pathologic, rather than a physiologic, process. Since the proportion of tumors that express growth factors of a given family is consistently higher than the proportion of tumors that express the cognate receptors, it is likely that growth factor synthesis has an important, nonautocrine role in breast cancer progression as well. Data obtained with primary human breast cancer specimens indicate that growth factors synthesized by breast cancer cells have an important role in tumor development and progression but that these factors act in a true autocrine fashion only in a subset of tumors.

Recently, a family of growth factors has been described that activates erbB-2 receptors. These factors, known as the neu differentiation factors (NDF) or heregulins (HRG), induce tyrosine phosphorylation of erbB-2 receptors as a result of their direct interaction with either erbB-3 or erbB-4 receptors. Although it is known that expression of erbB-2 receptors has relevance in human breast cancer progression, how erbB-2, -3 and -4 receptors regulate mammary epithelial cell proliferation is not known. Therefore, experiments were carried out to study the mitogenic activity of NDF/HRG on the human mammary epithelial cell line MCF-10A which can be cultured continuously under serum-free conditions. MCF-10A cells, like primary cultures of normal human mammary epithelial cells, express an absolute requirement for exogenous epidermal growth factor (EGF) and insulinlike growth factor I (IGF-I) for growth. The results of these experiments indicate that NDF/HRG can induce tyrosine phosphorylation of p185erbB-2 in MCF-10A cells and is mitogenic for these cells. This is consistent with the coexpression of erbB-2 and erbB-3 mRNA that we have observed in MCF-10A cells. In addition, we found that NDF/HRG can substitute for either EGF or IGF-I to stimulate proliferation of these cells. The ability to substitute for both EGF and IGF-I is a unique property of NDF/HRG and is not shared by other members of the EGF or IGF family of growth factors, nor by other factors that we have studied. A striking isoform specificity was also observed which indicated that the beta-isoforms of NDF/HRG were greater than ten times more mitogenic than the alpha-isoforms. We also examined the mitogenic activity of NDF/HRG on MCF-10A cells that overexpress the erbB-2 receptor as a result of infection with a retroviral vector containing the human c-erbB-2 gene (MCF-10AerbB-2 cells). These studies indicated that MCF-10AerbB-2 cells have increased sensitivity to the mitogenic effects of NDF/HRG and that these cells are responsive to the alpha-isoforms of NDF/HRG at physiological concentrations. Thus, NDF/HRG is a dual specificity growth factor for human mammary epithelial cells, and the responsiveness of the cells to NDF/HRG is influenced by the level of expression of erbB-2 receptors.

During the process of tumorigenesis and tumor progression cancer cells acquire an array of genetic alterations. The theory of oncogene addiction proposes that a small portion of these alterations generate driver oncogenes that are critical for maintaining the transformed phenotype. Strong support for this theory comes from the identification and subsequent therapeutic targeting of individual driver oncogenes that has resulted in significant clinical responses. However, in solid tumors that contain multiple driver oncogenes, these responses are not durable and it is therefore likely that multiple oncogenes will need to be identified and targeted in these tumors in order to reverse the transformed phenotype and effect durable clinical responses. Identifying multiple driver oncogenes among the large number of mutated/amplified genes that are present in most solid tumors is a nontrivial task that will require novel approaches. In this study, we have developed an approach that combines exome sequencing, array comparative genomic hybridization, expression analysis and genome-wide, RNAi-based functional screening to identify driver oncogenes that are functioning in individual cell lines derived from human tumors. In a pilot study using this approach we have shown that of the 583 amplified genes in the SUM225 breast cancer cell line only 45 are hits in the RNAi-based functional screen. Similarly, of the 806 genes that are mutated in this cell line only 26 are hits in the RNAi-based functional screen. Interestingly, the SMPD3 gene that was both amplified and a hit in the functional screen also contained a point mutation and this point mutation was present in another HER2 positive breast cancer cell line, SUM190. The identification of SMPD3 as a likely driver oncogene in the SUM225 and SUM190 cell lines highlights the power of an approach that combines sequence and copy number data with functional screening to identify driver oncogenes. Future analysis will incorporate expression data to further refine this approach. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5125. doi:1538-7445.AM2012-5125

Activated oncogenes are the genetic drivers of human cancer progression. Oncogenes become activated by structural DNA alterations including gene amplification and over expression, point mutations, and gene translocations/rearrangements. Data from existing targeted cancer treatments clearly show that the best response is observed when the activated oncogene is targeted. On the other hand, among the patients with activated oncogene who respond to a targeted treatment, the level and duration of response is highly variable. This could be explained by the genomic context of an activated oncogene that is unique to any given patient. Therefore, the ability to detect complex oncogene signatures in human cancer specimens is vital for better predictions of the efficacy of targeted agents, and to design therapeutic strategies with multiple targeted agents. In our laboratory, we have been working to identify candidate amplified and point mutated oncogenes in multiple human cancer specimens and cell lines. To obtain individual oncogene signatures, we have used high resolution aCGH coupled with a novel gene expression algorithm that allows for the identification of genes that are over expressed within amplicons in individual specimens. We have also used a high-throughput, mass spectrometry based method for detection of point mutations in oncogenes using the same specimens. These approaches have allowed us to identify tens to hundreds of candidate oncogenes in each specimen which were then refined by on-line computational tools such as Gene Tracker. Using this approach, we have elucidated complex oncogene signatures for over 50 cancer cell lines and specimens. Analysis of these data sets points to highly variable oncogene combinations in individual specimens. In other studies using a subset of breast cancers selected for the presence of the 8p11 amplicon, we found that even in these highly selected specimens, the oncogene signatures were unique to each patient. These signatures make clear predictions about combinations of targeted agents that could be considered. For example, we have observed activation of oncogenes in the 8p11 amplicon to be associated with other oncogenes such the EGFR, cdk4, MYST3, CRKL, and others. Additionally, the presence of PIK3CA mutations occurs variably in these specimens as well. Most of the breast cancer specimens and cell lines in our study have a well-defined signature of oncogenes, and understanding these signatures is important not only for the design of strategies using multiple targeted agents, but also for the proper interpretation of clinical trial results using single targeted agents. Thus, we have developed a high-throughput, cost effective molecular diagnostic platform that can accurately and rapidly identify the causal driving oncogenes in an individual cancer biopsy specimens and therefore can be adapted to analysis of large numbers of patient's specimens. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5099. doi:1538-7445.AM2012-5099

Epidermal growth factor receptor (EGFR) initiates signaling via several ligands that include but are not limited to EGF, amphiregulin (Areg) and epiregulin. These ligands are believed to differentially affect various physiological and cellular processes such as cell survival, proliferation and differentiation. This study is designed to delineate how amphiregulin alters the biology of both breast and ovarian cancer cells that overexpress amphiregulin and activate EGFR in an autocrine manner. AREG knock down drastically reduces the rate of proliferation of both a breast cancer cell line, SUM149 and an ovarian cancer cell line, OV1 that overexpress amphiregulin. Stable, clonal populations of SUM149 cells with AREG knockdown grow poorly in serum free media, but can be rescued by addition of exogenous AREG or other EGFR ligands to the culture medium. Clones with the most effective AREG knock down are completely unable to survive or proliferate in serum free media. Our Micro array analyses show that inhibiting amphiregulin in SUM149 cells dysregulates the expression of several important genes, including up-regulating DKK1, an antagonist of Wnt signaling. DKK1 up regulation upon AREG knockdown was confirmed by ELISA, with one of the clones, C2 showing a 16 fold increase in DKK1 secretion following AREG knock-down. Flow cytometetric analysis of SUM-149 cells and AREG knock-down derivatives demonstrated that 48% of control cells expressing a non-silencing shRNA construct were CD44+/CD24- but only a small population (7%) expressed CD24 alone. AREG knockdown induced significant reduction in the fraction of CD44 positive cells, down to ∼4% while, simultaneously increasing the non-stem cell fraction, CD44-/CD24+ to 60% of the total population. Since CD44hi/CD24lo is a characteristic of undifferentiated cells that have the potential to initiate tumor formation, we are currently examining the in vivo growth potential of SUM-149 cells and AREG knock-down derivatives. To determine the tumorigenicity of SUM149 cells three different cell innocula of unsorted parental SUM149 cells and SUM149 cells expressing a non-silencing shRNA construct (SUM149-NS) were transplanted into the mammary fat pad of NOD-SCID mice. As expected, the tumorigenic activity of the cells is proportional to the cell titer. In the parental SUM149 cells, the injection of 5x105 cells resulted in measurable tumors beginning at day 28 while SUM149-NS displays tumors at day 24, while transplantation of 2.5x105 cells per fat pad yielded tumor formation in 31 and 28 days respectively. Experiments to measure the in vivo growth potential of AREG knock-down SUM-149 cells is in progress. Thus, amphiregulin has a unique biological function to enhance cell survival and differentiation in vitro and in vivo. Future studies will focus on the role of AREG mediated WNT signaling in the malignant potential of SUM-149 cells. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 3248. doi:1538-7445.AM2012-3248

Interconnected genetic and epigenetic events control both the initiation and progression of cancer. Specifically, genetic alterations, such as amplification, and the subsequent over-expression of genes encoding ‘epigenetic modifying enzymes’ can directly lead to histone code changes that may be critical for cancer progression. KDM5A (lysine (K)-specific demethylase 5A, also known as RBP2 and JARID1A) was originally identified as the retinoblastoma-binding protein (RBP) implicated in the regulation of retinoblastoma target genes. Recently, several groups discovered that KDM5A is a member of a set of newly identified histone demethylases that control the chromatin-mediated regulation of gene expression. Specifically, KDM5A can function as a transcriptional repressor through the demethylation of tri- and dimethylated histone H3 at lysine 4 (H3K4) active marks. In this study, we observed a significant amplification and over-expression of the KDM5A gene in various tumors, including breast cancer. We found that breast cancer cells with KDM5A gene amplification had intrinsic drug resistance properties and knocking down KDM5A with shRNAs improved the efficacy of epidermal growth factor receptor (EGFR) inhibitors against these breast cancer cells. Further, up-regulation of KDM5A modified the histone methylation status and altered the expression of a subset of key genes, including the tumor suppressor p21 and the apoptotic effector BAK1, in breast cancer. Our findings suggest that alteration of the KDM5A gene may have a critical role in the pathogenesis of breast cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 2192. doi:1538-7445.AM2012-2192

PURPOSE Detection of occult carcinoma in patients with breast cancer may aid the establishment of prognosis and development of new therapeutic approaches. To improve on existing methods of detection, we have developed a reverse-transcriptase polymerase chain reaction (RT-PCR) assay for keratin 19 (K19) transcripts to identify mammary carcinoma cells in the peripheral blood and bone marrow of patients with breast cancer. PATIENTS AND METHODS Peripheral-blood or bone marrow samples obtained from 34 patients with stages I to IV breast cancer and 39 control subjects without breast cancer were screened for K19 mRNA by nested primer PCR. RESULTS In reconstitution experiments, K19 RT-PCR reliably detected 10 mammary carcinoma cells in 1 million normal peripheral-blood mononuclear (PBMN) cells. Four of 19 patients with stage IV breast cancer had detectable K19 transcript in peripheral blood. Five of six patients with histologically negative bone marrow biopsies following preablative chemotherapy and before autologous bone marrow transplant (BMT) were positive by this assay. Stem-cell apheresis harvests obtained from one of these patients and three additional patients immediately before BMT were all K19-negative. K19 RT-PCR analysis of CSF from a breast cancer patient with known carcinomatous meningitis was also positive. Thirty-eight of 39 non-breast cancer patients had negative K19 RT-PCR assays. The one exception was a patient with chronic myelogenous leukemia. CONCLUSION RT-PCR of K19 is a sensitive, specific, and rapid method for detection of occult mammary carcinoma cells in the peripheral blood and bone marrow of patients with breast cancer. The presence of residual breast cancer cells in histologically normal bone marrow aspirates but not in stem-cell apheresis harvests is a frequent finding. This assay may be useful in diagnosing metastatic disease, as well as in monitoring the effectiveness of systemic therapy.

Gene amplification is a major mechanism used by cancer cells to increase gene expression and subsequently gain a growth and survival advantage. Previously, we identified a novel amplicon at 9p23-24 region in human esophageal and breast cancer and originally cloned a novel gene GASC1 (Gene Amplified in Squamous Cell Carcinoma 1) from this amplicon. Accumulating evidence suggests common amplicons that occur in breast and other cancers contain multiple oncogenes that could play a role in cancer initiation and progression. In the current study, we aimed to comprehensively characterize the 9p23-24 amplified genes in human breast cancer. We performed array CGH analysis on a panel of cancer cell lines with the Agilent 244K microarray chip and narrowed the shortest region of overlapping (SRO) of the 9p23-24 amplicon to ∼2 Mb. Eleven genes within the SRO regions were examined for their mRNA expression by quantitative RT-PCR (qRT-PCR). Based on statistical association between copy number and expression, we confirmed GASC1 as a top candidate oncogene and identified three new potential oncogenes, UHRF2, KIAA1432 and C9orf123. Our more recent studies indicated that GASC1 is a member of a new class of oncogenes that are involved in the deregulation of histone methylation in cancer cells. We found that GASC1 induces the expression of core stemness transcription factors NANOG, SOX2 and OCT4, as well as classical oncogenes, including NOTCH1 and MYC, in breast cancer cells. On the other hand, UHRF2 encodes a nuclear protein that is involved in cell-cycle regulation. Knocking down UHRF2 slowed cell growth and inhibited colony formation of 9p23-24 amplified breast cancer cells. Furthermore, we demonstrated that expression levels of key tumor suppressing genes, including p16/INK4a, p21/WAF1 and p27/KIP1, were increased after knocking down UHRF2 in breast cancer cells. Collectively, our studies support the notion that the 9p23-24 amplicon contains multiple candidate oncogenes and may play an important role in human tumorigenesis. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4700. doi:10.1158/1538-7445.AM2011-4700

Utilizing a cross-species expression profiling strategy that combines microarray techniques and metastatic cell lines of human and mouse origins, 22 up-regulated and 12 down-regulated genes were identified as an essential genetic program in highly-metastatic cell lines. The correlation of all gene expression levels and metastasis potential was confirmed by RT-PCR in a set of human and mouse cancer cells. Among these genes, Forkhead-Box Q1 (Foxq1) is a transcription factor with a function in metastasis that has not yet been reported upon. We showed that ectopic expression of Foxq1 increased cell migration and invasion in vitro; significantly enhanced lung metastatic capabilities of mammary epithelial cells in vivo and resulted in a marked epithelial to mesenchymal transition (EMT). In contrast, knockdown of Foxq1 expression led to the opposite change in vitro and in vivo. Neither ectopic expression nor knockdown of Foxq1 had significant effects in cell proliferation and colony formation in vitro. Additionally, we showed that Foxq1 repressed E-cadherin expression by binding to the E-box in its promoter region. Finally, we found that the expression of Foxq1 is regulated by TGF-β1, and that knockdown of Foxq1 blocked EMT induced by TGF-β1 treatment at morphological and molecular levels. Our study highlights the feasibility of cross-species expression profiling as a strategy in identifying metastasis-related genes. It also reveals that the inducement of EMT is a potential mechanism for the novel metastasis-promoting function of Foxq1 in breast cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3441. doi:10.1158/1538-7445.AM2011-3441

The 8p11-12 genomic region is amplified in about 10-15% of human breast cancer and may be associated with poor prognosis. Earlier, we used genomic analysis of copy number and gene expression to perform a detailed analysis of the 8p11-12 amplicon for identifying candidate oncogenes in breast cancer. We identified phosphatidic acid phosphatase type 2 domain containing 1B (PPAPDC1B) as one of the candidate oncogenes based on statistical analysis of copy number increase and over expression. Recently, fine-mapping of this locus, and the updated human genomic database has shown there may be more than 3 splicing variants of PPAPDC1B. We measured the expression status of all 3 individual PPAPDC1B isoforms in breast cancer cell lines by real-time quantitative RT-PCR and western blot. Interestingly, the short splicing variant /isoform2 is the dominant over-expressed form in SUM44 and SUM52 breast cancer cell lines that harbor the 8p11-12 amplicon. The PPAPDC1B gene encodes a novel type of transmembrane phosphatidate phosphatase with an acid-PPc domain with broad lipid phosphate substrate specificity. This suggests that PPAPDC1B may play a role in cell signaling and lipid metabolism. The PPAPDC1B isoform 2 is predicted to have an inactive acid-PPc domain due to truncation of mRNA. To address the question of whether PPAPDC1B possesses transforming properties, and to functionally characterize the different isoforms, we established lentiviral expression constructs of full-length PPAPDC1B as well as the short isoform. MCF10A cells were infected with equalized viral titer of lentiviral vectors and assayed for alterations in growth rates, colony formation and growth factor-independent proliferation. Cell proliferation assays indicated that over expression of PPAPDC1B isoform 2 enhanced cell proliferation more dramatically than full-length PPAPDC1B. Furthermore, colony formation assay showed that MCF10A cells over expressing PPAPDC1B isoform 2 formed large highly proliferative colonies compared with control cells. All of the isoforms of PPAPDC1B could induce insulin-like growth factor-independence in MCF-10A cells, but not independence of EGF. Thus, our data suggest that PPAPDC1B isoform 2 is a candidate oncogene from the 8p11 region that can induce a proliferative advantage in subset of breast cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 1135. doi:10.1158/1538-7445.AM2011-1135

We recently described culture conditions that allow proliferation of metastatic human breast cancer cells from biopsy specimens of certain patient samples. These conditions resulted in the development of an immortalized cell strain designated SUM-44PE. These same culture conditions were used to isolate a human breast cancer cell strain from a metastatic lymph node of a separate breast cancer patient. The SUM-16LN human breast cancer cells isolated from this specimen were cultured either in serum-free medium or serum-containing medium supplemented with insulin and hydrocortisone. Unlike the SUM-44PE cells that have proliferated in culture continuously for over two years, SUM-16LN cells proliferated in culture for approximately 200 days and underwent 15 to 20 population doublings before undergoing cell senescence. No cells of this strain proliferated beyond passage 8. SUM-16LN cells were keratin-19 positive and had an aneuploid karyotype. These cells overexpressed p53 protein and had an amplified epidermal growth factor (EGF) receptor gene that resulted in high level expression of tyrosine phosphorylated EGF receptor protein. Despite the presence of high levels of tyrosine phosphorylated EGF receptor in these cells, they proliferated in serum-free, EGF-free medium and did not secrete detectable levels of EGF-like mitogenic growth factor. In addition, these cells were potently growth inhibited by all concentrations of exogenous EGF tested and by the neutralizing EGF receptor antibody Mab 425. These results suggest that the high level of tyrosine phosphorylated EGF receptor present in these cells is the direct result of receptor overexpression and not the result of the presence of a stimulatory ligand. Thus, SUM-16LN represents a human breast cancer cell strain that exhibited genetic and cellular characteristics of advanced human breast cancer cells. Nevertheless, these cells exhibited a finite proliferative lifespan in culture, suggesting that cellular immortalization is not a phenotype expressed by all human breast cancer cells.

Normal rat mammary epithelial cells have an absolute requirement for insulin (IN) or insulin-like growth factor-l (IGF-I) for proliferation in serum-free medium. By contrast, serially transplantable rat mammary carcinoma (RMC) cells are IGF-I-independent for continuous growth in vitro. Previously, we demonstrated that IGF-I independence is not mediated by an autocrine loop. Therefore, experiments were carried out to determine if IGF-I independence behaves as a dominant or recessive phenotype in somatic cell hyhridization experiments. IGF-I-independent I-9RMT cells were rendered hygromycin-resistant and IGF-I-dependent MCF-IOA cells were rendered G418 resistant by infection with retroviral expression vectors. Cells of each line were co-incubated in 60 mm dishes and fused with polyethylene glycol. Hybrid cells were selected with media containing hygromycin and G418 in the presence or absence of IN. In three experiments, approximately the same number of colonies emerged in double selection media in the presence or absence of IN. Furthermore, colonies that emerged in IN-containing media, when subcultured, grew equally well in the presence or absence of IN. Thus, fusion of IGF-I-independent RMC cells with IGF-I-dependent human mammary epithelial cells yields hybrids that are IGF-I-independent for growth in serum-free medium.

Activated oncogenes are the genetic drivers of human cancer progression. Oncogenes become activated by structural DNA alterations including gene amplification accompanied by gene over expression, point mutations, and gene translocations/rearrangements. Data from recent clinical trials using targeted agents strongly suggest that clinical efficacy of these drugs is optimal when the target is an activated oncogene. However, it is also clear that even among patients that respond to drugs that target activated oncogenes, the level and duration of response is highly variable. One possible reason for heterogeneity in response to oncogene-targeted agents is the genomic context of a particular activated oncogene is not the same in all patients. Thus, the genomic context of an oncogene-target can have a dramatic influence on drug efficacy. Therefore, the ability to detect and diagnose complex oncogene signatures in human cancer specimens is vital for better predictions of the efficacy of targeted agents, and to design personalized therapeutic strategies. In our Oncogenomics laboratory, we have been working to identify candidate amplified and point mutated oncogenes in human breast, ovarian and lung cancer specimens and cell lines. To obtain individual oncogene signatures, we have used high resolution array comparative genomic hybridization coupled with a novel gene expression algorithm that allows for the identification of genes that are over expressed within amplicons in individual specimens, and Sequenom analysis for identification of point mutated oncogenes. These approaches have allowed us to elucidate complex oncogene signatures for over 50 cancer cell lines and specimens. Analysis of these data sets suggests that oncogene combinations across specimens is nearly random. Thus, knowing the status of one oncogene such as HER-2 does not predict what other oncogenes may be activated in the specimen or cell line. Indeed, we have observed HER-2 amplifications in the context of amplifications of oncogenes in the 11q12 or 8p11 amplicons, among others, and in the presence of absence of PIK3CA mutations. Furthermore, analysis of the effects of small molecule inhibitors of HER-2 at the level of cell survival/proliferation and gene expression demonstrated a dramatic effect of oncogene context on response to a targeted agent. In particular, PIK3CA mutations in the presence of HER-2 amplification and over expression was associated with a greatly attenuated response of the cells to the HER-2 targeted agent. Collectively, these results point to a need for a molecular diagnostic platform that can accurately identify the causal driving oncogenes in an individual cancer biopsy specimen in order to optimize clinical trial design, and develop personalized therapeutic strategies.

Oncogenomic mutations found in NSCLC are potential therapeutic targets, as seen in the dramatic responses individuals with EGFR mutations have to tyrosine kinase inhibitors. We selected 149 patients with NSCLC from previous population-based studies of lung cancer which included 80 (54%) Caucasian samples and 69 African American (46%) samples. Tumor DNA was extracted from formalin fixed paraffin embedded tissue. We examined 238 mutations in 19 oncogenes using the OncoCarta Assay v1.0 (Sequenom). We identified a total of 52 mutations in tumors from 47 individuals; the remaining 102 tumors did not have mutations in the oncogenes we assayed. Mutations were identified in EGFR (n=20), KRAS (n=19), PIK3CA (n=6), PDGFRA (n=2), AKT2 (n=1), FLT3 (n=1) and CDK1 (n=1). Consistent with previous studies, EGFR mutation was more likely to occur in non-smokers (p=3.3E-08 Fisher's test) and the significance of this effect was more pronounced when analyzed against the number of pack years smoked (p=7.3E-08, Kruskal-Wallis test). Previously published reports have suggested that African American patients are less likely to exhibit EGFR mutations compared to Caucasian NSCLC patients and are therefore less likely to respond to tyrosine kinase inhibitors. We found no difference in the proportion of EGFR mutations in African American patients (n=8, 11.6%) compared to Caucasian (n=12, 15.0%, p-value=0.54). Furthermore, EGFR mutations in African American patients were exclusively in exon 19 (n=8, 100%), whereas Caucasian patients had mutations in exon 21 (n=3, 25%) and exon 18 (n=1, 8.3%) in addition to those in exon 19 (n=8, 66.7%). Interestingly, several reports have suggested a higher rate of response of NSCLC to erlotinib, an EGFR tyrosine kinase inhibitor, associated with exon 19 deletions relative to the exon 21 L858R mutation. Thus, our oncogenomic data demonstrates that African American NSCLC patients not only harbor a similar frequency of EGFR mutation as Caucasians, but that the specific nature of the mutation may result in a higher frequency of response to EGFR targeted therapeutics.

Endoplasmic reticulum (ER)-associated degradation (ERAD) components are highly induced in various tumors and are associated with cancer cell survival and resistance to anti-cancer treatments. The ERAD pathway is crucial for the cell to adapt to ER stress by degrading misfolded, unassembled and even some normal proteins in the ER lumen. Recent work in our group has led us to identify the ER lipid raft-associated 2 (ERLIN2, also known as SPFH2, C8ORF2) gene as one of the candidate oncogenes within the 8p11-12 amplicon in breast cancer based on statistical analysis of copy number increase and over expression. ERLIN2 protein, which localized specifically to the ER, is the key component of ERAD pathway. To elucidate the oncogenic properties of the ERAD factor ERLIN2 in human breast cancer, we established the MCF10A-ERLIN2 cell line, which stably over expresses ERLIN2 in human nontransformed mammary epithelial cells (MCF10A) using the pLenti6/V5-ERLIN2 construct. ERLIN2 over expressing cells and their respective parental cell lines were assayed for alterations in growth rates, growth factor-independent proliferation and other phenotypes expressed by cancer cells. From these experiments, we found that over expression of ERLIN2 in MCF10A cells results in growth factor-independent proliferation in insulin-like growth factor deficient media and formed large, highly-proliferative, and abnormally-structured colonies in three-dimensional morphogenesis assays. However, over expression of ERLIN2 in MCF10A cells did not result in cells with anchorage independent growth potential. To further elucidate the molecular basis that ERLIN2 is linked to the ER-stress response pathways in breast cancer cells, we knocked down the ER stress sensor IRE1α activity in the breast cancer cell lines in which ERLIN2 is amplified and over expressed. The knockdown of the IRE1α RNase activity, but not the kinase activity, significantly reduced protein levels of ERLIN2 in ERLIN2 amplified breast cancer cells. In addition, forced expression of wild-type IRE1α or spliced XBP1, the IRE1α target, led to increased expression of ERLIN2 in MCF10A cells. We also found that ERLIN2 protein was induced in response to the ER stress-inducing reagents Tunicamycin, Thapsigargin or growth factor depletion. Together, our findings indicate that ERLIN2 plays an unconventional oncogenic role through modulating ER stress response and ERAD pathways in a subset of breast cancer. Targeted inhibition of ERAD factor ERLIN2 in cancer could provide potential new avenues for therapeutic development. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 5022.

Oncogene over-expression due to DNA amplification is a frequent type of oncogene activation event that is responsible, in part, for tumorigenesis. The determination of which oncogenes in a tumor are genetically activated by DNA amplification can be predicted by the integration of aCGH and gene expression microarray data. Several methods have been previously implemented that combine aCGH and gene expression microarray data, but all of these methods focus on the gene expression changes in genomic regions with the most frequent DNA amplifications in a set of tumor samples. The implementation of personalized cancer therapy will require methods that predict which oncogenes are activated in a specific tumor sample by analyzing genes in all amplified regions. Here, we describe a method and R package for the Joint Analysis of Chromosomes and Karyotypes with Transcription On Microarrays (jackTom) that provides predictions on which genes are over-expressed due to DNA amplification in individual tumors by analyzing all amplicons. The analysis requires a collection of aCGH and gene expression arrays from a set of tumors. First, regions of genetic alteration are identified from aCGH microarrays with circular binary segmentation (DNAcopy package) using a copy number amplification minimum of 1.75. Additionally, regions lacking genetic amplification or genetic loss are also mapped using the same procedure. The expression of amplified genes in a specific tumor are compared to tumors without genetic alterations in the amplified region. The significance of the gene expression alterations in amplified regions is determined using either a Z-score or a modified ANOVA (limma package). The jackTom package was used to analyze the 8p11-p12 region in 10 human breast cancer (HBC) cell lines and 19 HBC tumor samples. This genetic region is amplified in approximately 15% of HBC. Results from jackTom were subsequently compared with a multiple regression analysis of the genes in this amplicon (performed by globalTest). In the 29 HBC samples analyzed, 9 samples exhibited chromosome 8p11-p12 amplification of at least 1.75 fold. Sixty one genes that were amplified in at least two samples spanned bases 31.01 Mb to 48.81 Mb. GlobalTest analysis demonstrated that 20 genes had expression changes that were significantly altered by the amplification event. Nevertheless, jackTom analysis of each sample demonstrated that MYST3 is significantly over-expressed in 3/8 tumors harboring an amplification (p less than or equal to 0.024, 0.010 and 0.0053, respectively) and DKK4 is significantly over-expressed in 1/6 tumors harboring an amplification (p less than or equal to 0.0074). Thus, jackTom analysis demonstrates that frequently amplified regions select for genetic activation of different candidate oncogenes in individual tumors. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 117.

Despite remarkable advances in research and clinical management, metastatic breast cancer remains incurable and causes most deaths in human breast cancer patients due to the unavailability of specific, targeted therapy. Therefore, identification of functional metastasis genes and further unveiling the molecular mechanisms underlying metastasis process remains the top priority in cancer research field. In recent years, much evidence supports the notion that epithelial-mesenchymal transition (EMT), a developmental program involved in a variety of tissue remodeling events, contributes to cancer cell plasticity and to the invasive potential of transformed epithelial cells. However, the legitimacy of EMT is still challenged by disagreements among pathologists who dismiss EMT as a fallacy. Utilizing a cross-species expression profiling strategy which combined microarray technique and model cell lines, we identified 22 up-regulated and 12 down-regulated genes as an essential genetic program in both human and mouse highly-metastatic model cell lines. About half of the gene candidates are novel and their roles in metastasis have not been reported. We randomly selected four genes and investigated their roles in metastatic characteristics including EMT program and cell migration and invasion. Overexpression of GCNT2 gene (an up-regulated gene candidate) not only led to a marked increase of cell migration and invasion, but also led to marked alterations in epithelial-mesenchymal transition (EMT) program. Ectopic expression of MAPK12 (another upregulated gene) only changed the EMT program but has no effect on cell migration and invasion. Overexpression of IRX5 (a down-regulated gene) not only led to a marked decrease of cell migration and invasion in vitro, but also a modulation of EMT program. However, expression of PHLDA3 (another down-regulated gene) only affect the cell migration and invasion, but has no effect on EMT program. These results highlight the feasibility of the cross-species expression profiling approach as a strategy to identify gene targets that are involved in breast cancer metastasis. The results of the functional studies for randomly selected genes further suggest that this multigenic program is consisted of different genes that play various roles in controlling EMT and/or Cell migration and invasion ability. Further studies are needed to investigate the effect of these novel gene candidates on metastasis in vivo and as therapeutic targets for treatment of breast cancer metastasis. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2301.

Amphiregulin (AREG) and epidermal growth factor receptor (EGFR) co-expression leads to aggressive growth phenotypes observed in some triple negative breast tumors. In SUM-149 cells, a triple negative breast cancer cell line with an AREG/EGFR autocrine loop, we have shown that these cells are more invasive and motile than MCF10A cells grown with EGF and that altered signaling occurs when AREG is the stimulating ligand. To provide additional insight into the role of AREG in the SUM-149 cells, we have successfully knocked down AREG expression in the SUM-149 cells. Q-RT-PCR analysis showed an approximately 40-fold knock-down of AREG mRNA following stable transduction of the most effective shRNA vector. In these knock-down cells, secreted AREG was undetectable by ELISA, and cellular AREG was undetectable by western blot. Knock-down of AREG in SUM-149 cells resulted in a dramatic decrease in the growth rate of these cells, resulting in a doubling time of about half that of the parental SUM149 cell line. Soft agar assays comparing the anchorage-independent growth potential of SUM-149 cells with AREG knock-down SUM-149 cells demonstrated an approximately 50% decrease in soft agar colony forming ability. To examine the migration and invasion capacity of the AREG knock-down cells relative to parental cells, Matrigel invasion assays were carried out and showed that knocking down AREG resulted in a 40-fold decrease in the cells ability to invade the matrix. Our lab has previously shown that in SUM-149 cells, NF-κB DNA binding activity is initiated as a result of AREG activation of EGFR. NF-κB activity in the AREG knock-down cells was altered suggesting that AREG affects cellular signaling by altering transcription factors. Addressing the continued proliferation of the AREG knock-down cells, we examined the requirement for EGFR activation for growth of these cells. Phosphotyrosine western blot analysis demonstrated detectable EGFR activation in the AREG knock-down cells. Growth assays where these cells were treated with an EGFR tyrosine kinase inhibitor or with a neutralizing EGFR antibody demonstrated the dependency of EGFR for growth of these cells. Q-RT-PCR analysis for expression of all seven EGFR family ligands suggested that epigen or EGF may be factors contributing to the continued growth of these cells, in the absence of AREG. In summary, knock-down of AREG expression in SUM-149 cells has a dramatic affect on the cells ability to proliferate, invade, and form colonies in soft agar, which links this growth factor with the aggressive properties associated with the SUM-149 cells. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 267.

Breast cancer remains a major health problem, despite advances in detection, diagnosis, and treatment. Four genomic subtypes of breast cancer have been identified: luminal A and B, HER2 positive, and basal type breast cancers. Three of these subtypes help to define the proper targeted therapeutic as breast cancers of luminal subtypes tend to be estrogen receptor positive and treatable with tamoxifen while breast cancers of HER2 positive subtype are treatable with trastuzumab. The fourth subtype, the basal type breast cancers, has a poor prognosis and has yet to have a molecular target identified. One molecule highly expressed in basal type breast cancers is epidermal growth factor receptor (EGFR). However, targeting EGFR kinase activity with small molecule kinase inhibitors has yet to show efficacy in the treatment of breast cancers. Recently, a panel of EGFR expressing breast cancer cell lines was characterized as sensitive or resistant to EGFR kinase inhibitors. One mechanism of resistance to EGFR kinase inhibitors was found to be the constitutive activation of Met, another molecule associated with basal type breast cancers. Here, the mechanism of constitutive Met activation in EGFR tyrosine kinase inhibitor resistant breast cancer cells is explored. Previously, constitutive Met activation was found to associate with acquired EGFR inhibitor resistance in lung cancer via genetic amplification of Met. In the EGFR TKI resistant breast cancer cells with constitutive Met phosphorylation, Met was not amplified or significantly overexpressed. In addition, constitutive Met phosphorylation was found to occur independent of ligand (hepatocyte growth factor (HGF)) stimulation, as HGF was not expressed in the breast cancer cell lines and inhibiting HGF association with Met via an inhibitory antibody failed to reduce Met phosphorylation. These data suggest a ligand-independent mechanism for Met constitutive phosphorylation. Interestingly, ligand-independent Met activation can be mediated by Met association with other membrane molecules including integrins, tetraspanins and receptor tyrosine kinases, including EGFR. Co-immunoprecipitation studies demonstrated that EGFR and Met co-associate and that this association occurs independent of EGFR or Met kinase activities. In addition, Met phosphorylation was reduced with EGFR TKIs as well as with downregulation of EGFR expression. Taken together, these data suggest that one potential mechanism of EGFR TKI resistance in breast cancer cells is the association and transphosphorylation of Met by EGFR. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 278.

Amplification of the 8p11-12 region has been found in about 10-15% of human breast cancer and may be associated with poor prognosis. Earlier, we used genomic analysis of copy number and gene expression to perform a detailed analysis of the 8p11-12 amplicon for identifying candidate oncogenes in breast cancer. We identified Wolf-Hirschhorn syndrome candidate 1-like 1 (WHSC1L1) as a candidate oncogene based on statistical analysis of copy number increase and over expression. The WHSC1L1 gene encodes a PWWP domain protein that regulates histone methylation marks, and regulates the transcription profile and the differentiated function of cells. To address the question of whether WHSC1L1 possesses transforming properties, we established lentiviral expression constructs containing the WHSC1L1 as well as other 20 candidate genes of 8p11-12 region. MCF10A cells were infected with these lentiviral vectors and assayed for alterations in growth rates, growth factor-independent proliferation, anchorage-independent growth and for three-dimensional morphogenesis assays in Matrigel culture. From these experiments, we found that the WHSC1L1 is the most potently transforming oncogene we tested from the 8p11-12 region. Furthermore, knockdown of this gene in WHSC1L1 amplified breast cancer cells resulted in profound loss of growth and survival potential in these cells. To identify genes that may be regulated in their expression by the WHSC1L1 when it is amplified and over expressed, we performed expression profiling of MCF10A cells, MCF-10AWHSC1L1 cells, and WHSC1L1 amplified SUM-44 breast cancer cells. Analysis of genes altered by over expression of WHSC1L1 in MCF10A-WHSC1L1 cells and in SUM-44 cells resulted in the identification of 184 WHSC1L1 candidate target genes. Among them was the iroquois homeobox 3 gene (IRX3), a member of the Iroquois homeobox transcription factor family. The regulation of IRX3 gene expression by WHSC1L1 was validated in SUM44 breast cancer cells using WHSC1L1 knock down. IRX3 mRNA was dramatically down-regulated in SUM-44 cells after WHSC1L1 shRNA infection. Intriguingly, in our array CGH study we found that SUM-44 cells also have a gene amplification of IRX3. Thus, our finding indicates that WHSC1L1 is the major driving oncogene in the 8p11-12 amplicon, and provides the first evidence that WHSC1L1 and IRX3 are genetically linked in subset of breast cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 3102.

A fundamental tenet of cancer biology is that cell transformation is driven by oncogenes, activated forms of proto-oncogenes. Yet, we have poor understanding of how oncogenes transform cells and induce phenotypes not expressed by normal cells. To examine signaling by the HER2 proto-oncogene and activated oncogene, we studied how HER2 signaling affects gene expression in a panel of well characterized isogenic breast cell lines: non-transformed MCF10A cells, HER2 over expressing and partially transformed MCF10HER2 cells, and fully transformed MCF10HER2/E7 cells. Cells were treated with small molecule kinase inhibitors to block HER2 signaling and RNA was collected every 3 hours for 45 hours (16 time points). Whole genome expression profiling was performed using Illumina microarrays. Analysis of time series expression data in the three cell lines identified genes specifically down and up-regulated by the proto-oncogene compared to the activated oncogene. Among the genes that were most highly up-regulated specifically by activated HER2 were chaperone proteins such as SCG5, and the heat shock proteins HSP6 and HSPA1A. Genes that were the most down regulated specifically by the activated oncogene included CITED2, the gene for a transcription factor that suppresses HIF1a expression, and DKK1, the gene for a repressor of Wnt signaling. Ingenuity Pathway Analysis identified several canonical pathways that were enriched for HER2 regulated genes in the most transformed cells compared to the nontransformed cells. For example, HER2 signaling in transformed cells upregulated more genes in pathways associated with cell motility and invasion than in MCF10A cells. Further, HER2 signaling in MCF10A cells did not influence the expression of any genes in the glucocorticoid, NRF2, Wnt, HIF1a, or aryl hydrocarbon signaling pathways. By contrast, signaling by the activated HER2 oncogene in MCF10HER2 cells up-regulated several genes in these pathways, including NFKBIB, PCK2, BAX, HSP40. HER2 oncogene signaling in fully transformed MCF10HER2/E7 cells stimulated upregulatation of still more genes associated with these pathways, including SOX4, RARA, HSP70, HSP90, MAPK6, SHC1, LDHA, MMP3, SLC2A1, HRAS. Similar experiments were performed with SUM-225 cells, a human breast cancer cell line with HER2 gene amplification. Activated HER2 regulated similar pathways in the SUM-225 cells as in the HER2 transformed MCF10 cells. In addition, we observed a significant enrichment of pathways unique to the HER2-upregulated genes in the SUM225 cells, such as the Nanog signaling pathway that has a role in stem cell pluripotency. This study identified many genes specifically regulated by the HER2 oncogene in HER2-transformed cells. Associating these oncogene-regulated genes with signaling pathways and biological networks helps to shed light on the mechanisms by which the HER2 oncogene regulates the expression of transformed phenotypes. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 114.

The positron-emitter-labeled glucose analogue 2-[fluorine-18]-fluoro-2-deoxy-D-glucose (FDG) accumulates into many cancers after intravenous injection, but the effect of serum glucose levels on FDG uptake in the tumor has not been extensively studied. In vitro, elevated media glucose levels markedly diminished FDG and FDG 6-phosphate uptake and retention in human adenocarcinoma cells, while insulin had no effect. Mammary cancers were established subcutaneously in 12 rats. Six control rats with mammary tumors were fasted overnight. Hyperglycemia was established in six rats by means of continuous glucose infusion (glucose clamp). All animals were then intravenously administered 50 microCi of FDG. Serum glucose levels were 87 mg/dL (4.83 mmol/L) in the control animals and more than 900 mg/dL (49.9 mmol/L) in the hyperglycemic animals. One hour after injection of FDG, mean F-18 uptake in the tumor, brain, small bowel, and ovaries was 2.7-9.7 times lower in the hyperglycemic animals (P less than .02). Mean F-18 activity in the kidneys tended to be somewhat higher in the hyperglycemic animals. FDG uptake in other tissues was comparable between the control and hyperglycemic groups. These data suggest that high serum glucose levels may substantially impair visceral tumor imaging with FDG positron emission tomography.

Abstract #6032 The goal of this study was to use a computational strategy to identify HER-2 regulated-gene expression networks in an isogenic series of cell lines in which HER-2 functions as a proto-oncogene, as an oncogene driving partial transformation, and as an oncogene driving full malignant transformation. The cells used for these experiments include non-transformed, immortalized MCF10A cells grown in media containing EGF and insulin; MCF10HER2 cells that over express HER-2 and are insulin-independent for growth; and MCF10HER2/E7 cells (also E7 transduced), which are insulin and EGF-independent for growth and fully malignant in vivo. We obtained whole-genome expression data every 3 hours for 45 hours while HER-2 signaling and cell proliferation were blocked by treatment with specific small molecule kinase inhibitors. A dynamic linear finite difference model was used to derive phenomenological gene expression networks from the time-series gene expression data. In these networks each gene is rank-ordered by its degree of connectivity, and hub genes are identified as those with the highest connectivity level. We then compared the hub genes regulated by HER-2 in each cell line, and used Ingenuity Pathway Analysis to identify the biological networks differentially regulated by HER-2. Results show that both normal and transformed gene expression networks have scale-free topology. Using the same connectivity value to define hubs, we found that HER-2 regulated the expression of more hub genes in transformed cells. These findings suggested that the HER-2 signal was integrated differently in progressively transformed cells and led us to map functional association networks for each cell-specific list of hubs. In MCF-10HER-2 cells, which are partially transformed and insulin-independent for growth, two hubs differentially regulated by HER-2 include VAMP8 and PHGDH. These genes influence cell surface expression of glucose transporters, and activate anabolic pathways from glycolysis intermediates, respectively. Thus, HER-2-mediated insulin independence is associated with constitutive glucose transport and the transition to aerobic glycolysis. HER-2/E7 cells are fully transformed, and in these cells, differentially regulated hub genes such as HSP1A1 and GPX4 are part of the coordinated regulation of the stress-protective heat shock response and enzymatic pathways of oxidative defense. In concert with these modular changes, the CITED2 hub gene, a negative regulator of HIF1α, is specifically down-regulated by HER-2 signaling in the MCF10A/HER2 and MCF10A/HER2/E7 cell lines, which results in enhanced HIF1α expression and transcriptional activity. In summary, our systems biology approach resulted in the identification of distinct gene expression and functional association networks that arise when HER-2 transitions from being a proto-oncogene to a fully transforming gene. These results provide evidence regarding the mechanisms by which HER-2 acts as a transforming gene when over expressed. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 6032.

In previous studies, we developed serum-free, bovine pituitary extract (BPE)-free culture conditions for the growth of normal and neoplastic rat mammary epithelial cells. The present studies were aimed at determining if these culture methods could be used to study the influence of specific growth factors on the proliferative potential of normal human mammary epithelial (HME) cells and cells derived from human breast cancer (HBC) specimens. Our results indicate that normal HME cells in primary culture express stringent requirements for insulin (IN), epidermal growth factor (EGF), and cholera toxin (CT). Of these factors, EGF is most important, with essentially no proliferation taking place in the absence of this factor. By contrast, when cells are grown in serum-free primary culture in the presence of a full complement of growth factors and then subcultured, growth in secondary culture is not influenced by the removal of individual growth factors. Growth in secondary culture in the absence of EGF is mediated by autocrine factors secreted by the cells. However, there is no evidence for autocrine activity that mediates growth in the absence of IN in secondary cultures. Primary culture of HBC cells in serum-free, BPE-free medium revealed two patterns of growth factor requirements. One set of HBC cells expressed identical requirements for IN and EGF in primary culture as normal cells. Likewise, these cells grew in secondary culture in the absence of either factor. The second set of tumors expressed independence of IN for growth in primary culture. These cells grew to confluence in primary culture in the absence of IN and could be subcultured in this medium. All tumor cells examined expressed a requirement for EGF for primary culture growth, whereas none of the HBC cells examined expressed a significant CT requirement. In many cases, growth in the absence of CT exceeded that observed in its presence. Thus, our culture system allows analysis of the growth factor requirements of HME and HBC cells in primary culture. Our results indicate significant differences between HME and HBC cells in this regard. However, the results of secondary culture experiments indicate that the growth factor milieu from which cells are taken can have a profound effect on the requirements for growth factors in culture.

Over the past several years our laboratory has been studying factors that regulate proliferation of normal human mammary epithelial (HME) cells in order to better understand the alterations in cellular growth control mechanisms that occur during breast cancer development. To perform these experiments, we have either modified or developed cell culture methods for the isolation and growth of normal and neoplastic HME cells obtained from patient biopsy specimens. From these studies we have found that normal HME cells of the luminal lineage (the lineage from which breast cancer arises) have strict requirements for specific growth factor combinations for in vitro growth. Furthermore, these cells have a finite proliferative lifespan in culture. By contrast, human breast cancer (HBC) cells isolated from primary and metastatic sites exhibit many growth phenotypes that distinguish them from normal cells. First, whereas normal HME cells proliferate in culture with doubling times of 24–36 hours, HBC cells obtained from patient samples proliferate with doubling times of 100–200 hours. These proliferation kinetics are consistent with the rate at which these cells proliferate in vivo. This observation indicates that there are fundamental differences in growth regulation between normal and neoplastic mammary epithelial cells. Second, the majority of HBC cells isolated from human samples exhibit an extended proliferative lifespan in culture. Whereas normal HME cells undergo cell senescence after 15–20 population doublings, HBC cells often give rise to cell lines with indefinite proliferative potential. Third, HBC cells become independent of growth factors which are strictly required by normal HME cells for growth under defined conditions. In our experiments, escape from the requirements of exogenous epidermal growth factor (EGF) has been observed in cells from four patient-derived samples. Interestingly, the cellular mechanisms by which cells become EGF-independent for growth is different in cell lines isolated from different patients. Two breast cancer cell lines isolated in our laboratory proliferate continuously in serum-free, EGF-free medium and do not express EGF receptors. Thus, these cells are completely independent of EGF-mediated signalling pathways for their growth. A third cell line isolated in our laboratory has an amplified EGF receptor gene and overexpresses EGF receptor protein. Western blot analysis indicates that the tyrosine residues of the EGF receptor proteins in these cells are highly phosphorylated. These cells do not secrete any EGF-like growth factors that could be activating the receptors in an autocrine manner. This suggests that amplification and overexpression of EGF receptors can yield constitutively activated receptors that provide a mitogenic signal in the absence of a stimulatory ligand. Finally, we have analyzed the EGF requirements of human breast cancer cells that overexpress the erbB-2 receptor as a result of gene amplification. The results of these experiments indicated that overexpression of erbB-2 is, by itself, insufficient to overcome the EGF requirements of human breast cancer cells. However, one cell line that has a 15- to 20-fold amplification of erbB-2 and which expresses very high levels of tyrosine phosphorylated erbB-2 protein, is EGF-independent for growth. Thus, cell culture systems that allow proliferation of normal HME cells and HBC cells under well-defined culture conditions can result in identification of altered growth phenotypes associated with the neoplastic progression of breast cancer cells. In addition, isolation of cells exhibiting altered growth phenotypes may lead to insights as to the genetic mechanisms resulting in altered growth regulation in breast cancer cells.

A monolayer culture system has recently been developed for the extended growth and serial passage of normal rat mammary epithelial (RME) cells. In this system the cells undergo greater than 20 population doublings when grown on type I collagen-coated tissue culture dishes in Ham's F12 medium supplemented with insulin, hydrocortisone, epidermal growth factor, prolactin, progesterone, cholera toxin, and 5% fetal bovine serum (FBS). The purpose of the present studies was to define additional growth factors that would allow equivalent RME cell proliferation in serum-free medium. Ethanolamine (EA) was effective at reducing the FBS requirements for RME cell proliferation and at its optimum concentration did so by greater than 20-fold. Even with optimum levels of EA there was essentially no cell proliferation in the absence of FBS. However, addition of bovine serum albumin (BSA) to the hormone, growth factor, and EA-supplemented medium resulted in substantial proliferation in the absence of serum, and the further addition of transferring (T) potentiated this effect. Thus, in this culture system, replacement of FBS with EA, BSA, and T resulted in RME cell proliferation in primary culture which was equivalent to that obtained in the 5% FBS-containing medium.

We have reported that HT29 colon cancer cells, which are radiosensitized by fluorodeoxyuridine (FdUrd), exhibit a greater increase in cyclin E-dependent kinase activity and progress further into S phase in the presence of FdUrd than do SW620 colon cancer cells, which are only minimally sensitized by this drug (Cancer Res 56: 3203, 1996). Although these findings suggested that the ability to progress into S phase in the presence of FdUrd permits cells to be radiosensitized, we wished to test this hypothesis by attempting to drive SW620 human colon cells into S phase by transducing them with the HPV16-E7 gene. Two-parameter flow cytometry showed that E7-transduced cells progressed through S phase after radiation and FdUrd treatment more rapidly than SW620 parental cells. We found that E7-transduced SW620 cells were significantly radiosensitized by FdUrd (100 nmol/L, 14 hours) with an enhancement ratio for 2 clones of 1.47 +/- 0.03 and 1.51 +/- 0.14, compared with 1.24 +/- 0.04 in SW620 parental cells. These data strongly support the hypothesis that dysregulation of S-phase progression is an important factor in FdUrd-mediated radiosensitization.

SummaryA handful of tumor-derived cell lines form the mainstay of cancer therapeutic development, yielding drugs with an impact typically measured as months to disease progression. To develop more effective breast cancer therapeutics and more readily understand their clinical impact, we constructed a functional metabolic portrait of 46 independently derived breast cell lines. Our analysis of glutamine uptake and dependence identified a subset of triple-negative samples that are glutamine auxotrophs. Ambient glutamine indirectly supports environmental cystine acquisition via the xCT antiporter, which is expressed on one-third of triple-negative tumors in vivo. xCT inhibition with the clinically approved anti-inflammatory sulfasalazine decreases tumor growth, revealing a therapeutic target in breast tumors of poorest prognosis and a lead compound for rapid, effective drug development.

A newly developed culture system was used to examine the proliferative potential of rat mammary epithelial (RME) cells in vitro. RME cells were obtained by enzymatic dissociation of mammary tissues of 45- to 50-year-old virgin female LEW rats. The tissues were dissociated to small aggregates (10-50 cells per aggregate) separated from stromal cells and plated at a density of 10(5) cells per 60-mm tissue culture dish. The cells were grown in Ham's medium F12 supplemented with 5% fetal bovine serum, insulin, hydrocortisone, epidermal growth factor, prolactin, progesterone, and cholera toxin. Plating of 10(5) cells as small aggregates resulted in the attachment of 1,000-1,500 aggregates per plate. When grown on tissue culture plastic, approximately 1-2% of these aggregates gave rise to rapidly proliferating epithelial colonies. Individual colonies expanded with a population-doubling time of 24-34 hours and grew for about 3 weeks. Although these cells grew well in primary culture, they were not subculturable. When RME cells were plated onto dishes coated with type I collagen, the number of rapidly proliferating epithelial colonies per dish increased fivefold to tenfold. Cells grown on type I collagen-coated dishes expanded with a population-doubling time of approximately 27 hours and after 2 weeks in primary culture were nearly confluent. Unlike cells grown on plastic, RME cells grown on type I collagen were readily subculturable and serial subculture resulted in the cells undergoing 15-20 population doublings (5-6 passages) before exhibiting any loss of growth potential. Continued feeding of senescent cultures resulted in the emergence of discrete RME cell foci that retained proliferative potential and that eventually developed into rapidly growing cell strains. Exposure of primary cultures to the carcinogen N-methyl-N'-nitro-N'-nitrosoguanidine (CAS: 70-25-7) enhanced the proliferative potential of RME cells in early passages and in later passages either delayed or eliminated the "senescent" phase of cell growth. Carcinogen treatment of RME cells also facilitated the establishment of rapidly growing cell strains with long-term growth potential (greater than 20 passages).

Agents that elevate intracellular cAMP levels are required for growth of many cell types in culture including normal rat mammary epithelial (RME) cells. To determine if the intracellular levels of cAMP that result from stimulation by agents such as cholera toxin (CT) or prostaglandin E-1 (PGE-1) are within the physiological range, cAMP levels were determined in RME cells growing in primary culture and compared to levels measured in freshly isolated mammary epithelium. The results indicate that the cAMP levels of mammary epithelial organoids obtained from 45-day-old virgin rats are 4 to 6 pmol/106 cells. Growth of RME cells in primary culture in the presence of CT results in cAMP levels of approximately 15 to 20 pmol/106 cells early in culture when cells are proliferating rapidly. As cells approach confluence, cAMP concentrations decrease to levels observed in fresh organoids. CT-stimulated cAMP levels appear to be within the range of those found in pregnant mammary epithelium in vivo. Growth of RME cells in medium supplemented with PGE-1 instead of CT results in cAMP levels equivalent to those found in fresh mammary epithelial organoids and under these conditions the growth rate is approximately half that found in CT-stimulated cells. These results indicate cAMP to be a positive regulator of cell growth in vivo at levels that are within the physiological range.

These experiments were aimed at using a recently developed serum-free culture system for growth of normal rat mammary epithelial (RME) cells in vitro to examine the interactions of specific hormones and growth factors on the proliferative potential of these cells. RME cells were obtained by enzymatic dissociation of mammary tissues of Lewis rats. Primary cultures were started by plating 2 X 10(5) RME cells per 60-mm type I collagen-coated tissue culture dish. Cultures were maintained in a basal medium that consisted of Ham's F-12 medium supplemented with bovine serum albumin (BSA), ethanolamine (EA), and transferrin (Tf), which, by itself, did not support RME cell proliferation. Insulin (I), hydrocortisone (HC), and epidermal growth factor (EGF), when added to the basal medium interacted synergistically to stimulate RME cell proliferation, but this effect was dependent on the additional presence of cholera toxin (CT). Under these conditions a greater-than-tenfold increase in cell number over a 10-day culture period was obtained. Insulin could be replaced by physiological levels of insulin-like growth factor-I (IGF-I). CT could be replaced by other agents that elevate intracellular levels of cyclic adenosine 3':5' monophosphate (cAMP) such as dibutyryl-cAMP (db-cAMP), prostaglandin E1 (PGE-1), and/or isobutylmethylxanthine (IBMX). Prolactin (M) or progesterone (P) potentiated the effect of I, HC, EGF, and CT, resulting in an additional twofold increase in cell number over that found in their absence. However, addition of both hormones was no more effective than either one alone. Furthermore, addition of M or P in the absence of EGF had no effect on RME cell proliferation. Addition of 17-B-estradiol (E2) to the I-, HC-, EGF-, and CT-containing medium also resulted in enhanced RME cell proliferation. These results point to a number of hormone and growth factor interactions that influence the proliferation of normal RME cells in vitro.

Introduction Toll-like receptors (TLRs) are a family of pattern recognition receptors that are expressed on cells of the innate immune system. The ligands can be pathogen derived (pathogen associated molecular patterns; PAMPs) or endogenous (damage associated molecular patters; DAMPs) that when bound induces activation of nuclear factor kappa B (NF-κB) and transcription of pro-inflammatory genes. TLRs have also been discovered in various malignant cell types, but with unknown function. Methods In this study we performed a detailed analysis of TLR and co-receptor expression pattern and function in breast cancer. Expression patterns were examined using real-time quantitative polymerase chain reaction (RT-qPCR) and immunohistochemistry (IHC) on three estrogen receptor-positive (ER+) and four estrogen receptor/progesterone receptor-negative (ER−/PR−; ER/PR-negative) breast cancer cell lines, and a breast cancer cohort consisting of 144 primary breast cancer samples. The function was investigated using in vitro assays comprising PAMP/DAMP-stimulation, downstream signaling and TLR-silencing experiments. Results We found that TLR4 was expressed in a biologically active form and responded to both PAMPs and DAMPs primarily in ER/PR-negative breast cancers. Stimulation of TLR2/4 in vitro induced expression of pro-inflammatory genes and a gene expression analysis of primary breast cancers showed a strong correlation between TLR4 expression and expression of pro-inflammatory mediators. In line with this, TLR4 protein expression correlated with a decreased survival. Conclusions These findings suggest that TLR4 is expressed in a functional form in ER/PR-negative breast cancers. Studies regarding TLR4-antagonist therapies should be focusing on ER/PR-negative breast cancer particularly. Electronic supplementary material The online version of this article (doi:10.1186/s13058-015-0640-x) contains supplementary material, which is available to authorized users.

The fibroblast growth factor receptor (FGFR)2 gene has been shown to be amplified in 5-10% of breast cancer patients. A breast cancer cell line developed in our laboratory, SUM-52PE, was shown to have a 12-fold amplification of the FGFR2 gene, and FGFR2 message was found to be overexpressed 40-fold in SUM-52PE cells as compared with normal human mammary epithelial (HME) cells. Both human breast cancer (HBC) cell lines and HME cells expressed two FGFR2 isoforms, whereas SUM-52PE cells overexpressed those two isoforms, as well as several unique FGFR2 polypeptides. SUM-52PE cells expressed exclusively FGFR2-IIIb isoforms, which are high-affinity receptors for fibroblast growth factor (FGF)-1 and FGF-7. Differences were identified in the expression of the extracellular Ig-like domains, acid box and carboxyl termini, and several variants not previously reported were isolated from these cells., Introduction: The FGFR family of receptor tyrosine kinases includes four members, all of which are highly alternatively spliced and glycosylated. For FGFR2, alternative splicing of the second half of the third Ig-like domain, involving exons IIIb and IIIc, is a mutually exclusive choice that affects ligand binding specificity and affinity [1,2,3]. It appears that the second half of the third Ig-like domain can dictate high affinity for FGF-2 or keratinocyte growth factor (KGF), whereas affinity for FGF-1 appears to remain the same [3]. Alternative splicing of the carboxyl terminus has been shown to involve at least two different exons that can produce at least three different variants. The C1-type and C2-type carboxyl termini are encoded by the same exon, and have two different splice acceptor sites, whereas the C3-type carboxyl terminus is encoded by a separate exon [4]. The biologic significance of the C1 carboxyl terminus, as compared with the shorter C3 variant found primarily in tumorigenic samples, has been studied in NIH3T3 transfection assays, in which C3 variants were able to produce three times more transformed foci in soft agar than C1 variants (both IIIb), whereas full length FGFR2 and FGFR1 (both IIIc variants) showed no transforming activity [4]. Previous studies [5,6] have found amplification and overexpression of FGFR2 in 5-10% of primary breast cancer specimens. A recent study [7] done using a tissue array consisting of 372 primary breast cancer specimens found a 5% incidence of FGFR2 amplification. To our knowledge, none of the HBC cell lines studied thus far have an FGFR2 gene amplification, although overexpression of FGFR2 message and protein has been documented for some breast cancer cell lines [6,8,9]. SUM-52PE is a breast cancer cell line previously isolated in our laboratory that grows under serum-free and epidermal growth factor-free conditions, has high levels of tyrosine-phosphorylated membrane proteins, and has the capacity to invade and grow under anchorage-independent conditions [10,11,12]. This cell line exhibits all of the important hallmarks of transformed, highly malignant cells. Therefore, SUM-52PE was used as a model to study the diversity of FGFR2 expression in a breast cancer cell line that has true amplification and overexpression of FGFR2. Objectives: This study was conducted to examine the degree of FGFR2 amplification and overexpression in the breast cancer cell line SUM-52PE. Subsequent sequencing and characterization of individual FGFR2 variants cloned from the SUM-52PE cell line was completed to determine the complexity of FGFR2 alternative splicing in the context of a highly metastatic breast cancer cell line. Methods: Southern, Northern and Western blot analyses were done in order to determine the degree of FGFR2 amplification and overexpression in the breast cancer cell line SUM-52PE. Individual FGFR2 variants were cloned out of SUM-52PE using FGFR2-specific primers in a reverse transcription (RT) polymerase chain reaction (PCR). FGFR2 cDNAs were characterized by restriction fragment analysis, sequencing and transient transfection into 293 cells to examine the protein expression of each FGFR2 clone. Results: The results of the Southern blot showed that there was a 12-fold amplification of FGFR2 in the SUM-52PE cell line. Northern blot analysis of SUM-52PE showed FGFR2 transcripts to be highly overexpressed compared with other breast cancer cell lines and normal HME cells. Several overexpressed bands of approximately 6.3, 5.0, 4.0, and 2.8kb were observed in SUM-52PE cells. The most prominent band, at 2.8kb, was so abundant that it was difficult to discern other individual bands clearly. Western blot analysis showed that both normal HME and HBC cells expressed two FGFR2 variants of 95 and 135kDa. The SUM-52PE cell line greatly overexpressed not only these two polypeptides, as compared with HME and HBC cells, but also overexpressed two unique variants of FGFR2 - 85 and 109kDa polypeptides - as well as several smaller polypeptides in the 46-53kDa range. The antibody used in Western blot analysis only recognizes FGFR2 isoforms that express the C1 carboxyl termini, therefore greatly underestimating the actual number of different FGFR2 variants that are overexpressed in this cell line. PCR was performed to determine the proportion of C1/C2 variants as compared with C3 variants in the SUM-52PE cell line. Results of this analysis indicated the presence of all three types of variants in this cell line, although the C1/C2 variants were predominant as compared with the C3 variants in SUM-52PE. Four different FGFR2-C1 clones were isolated and sequenced from SUM-52PE cells, which differed in their signal sequence, first Ig-like loop, and acid box. Two FGFR2-C2 clones were isolated from the SUM-52PE cell line, which were identical to each other except for the variable expression of the number of Ig-like domains (two or three). Three C3 clones were isolated and sequenced, two of which have not previously been described in the literature. Clone C3-#3 contained two Ig-like domains, but no acid box. C3-#5 was missing the first two Ig-like domains and the acid box, but did contain the third Ig-like domain. Discussion: There is an extensive amount of evidence implicating erbB-2, a gene that is overexpressed in approximately 30% of breast cancer cases, as a breast cancer gene [13]. The identification of other breast oncogenes that function in the remaining 70% of cases is an ongoing challenge, as is establishing a causal role for such oncogenes in HME cell transformation. FGFR1 and FGFR2, previously established oncogenes, were found to be amplified within large amplicons on 8p11 and 10q26, respectively, in the breast cancer cell line SUM-52PE [14]. Previous studies have shown that the FGFR2 gene is amplified in about 5-10% of breast cancer cases. Our results showed that SUM-52PE cells overexpressed many alternatively spliced isoforms of FGFR2 at both the transcript and protein level as compared with normal HME cells. The variability in FGFR2 isoform expression is complex and involves exon IIIb/c, which encodes the second half of the third Ig-like loop; variations in the carboxyl terminal end of the receptor, involving the C1/C2 or C3 domains; and variable expression of the Ig-like loops and acid box in the extracellular portion of the receptor. The characterization of three unique FGFR2 isoforms that were cloned from SUM-52PE may build on the findings of others concerning the transforming potential of FGFR2 variants [4]. In particular, because it has been demonstrated that expression of C3-IIIb variants may have more transforming activity than C1-IIIb variants, differences between the three C3 clones we have isolated may provide information regarding the influence of particular structural domains on transforming potential. Ongoing studies are aimed at characterizing the transforming ability of FGFR2 isoforms obtained from SUM-52PE cells by transducing these genes into normal HME cells. By overexpressing FGFR2 isoforms in a physiologically relevant system, we hope to determine the isoform(s) that acts in a dominant way in the process of cell transformation, and to determine whether different regions present in individual clones drive specific phenotypes associated with transformation.

We describe a survey of genetic changes by comparative genomic hybridization (CGH) in 11 human breast cancer cell lines recently established in our laboratory. The most common gains took place at 8q (73%), 1q (64%), 7q (64%), 3q (45%) and 7p (45%), whereas losses were most frequent at Xp (54%), 8p (45%), 18q (45%) and Xq (45%). Many of the cell lines displayed prominent, localized DNA amplifications by CGH. One-third of these loci affected breast cancer oncogenes, whose amplifications were validated with specific probes: 17q12 (two cell lines with ERBB2 amplifications), 11q13 (two with cyclin-D1), 8p11–p12 (two with FGFR1) and 10q25 (one with FGFR2). Gains and amplifications affecting 8q were the most common genetic alterations in these cell lines with the minimal, common region of involvement at 8q22–q23. No high-level MYC (at 8q24) amplifications were found in any of the cell lines. Two-thirds of the amplification sites took place at loci not associated with established oncogenes, such as 1q41–q43, 7q21–q22, 7q31, 8q23, 9p21–p23, 11p12–p14, 15q12–q14, 16q13–q21, 17q23, 20p11–p12 and 20q13. Several of these locations have not been previously reported and may harbour important genes whose amplification is selected for during cancer development. In summary, this set of breast cancer cell lines displaying prominent DNA amplifications should facilitate discovery and functional analysis of genes and signal transduction pathways contributing to breast cancer development. © 1999 Cancer Research Campaign