Summary of the differentially expressed protein between BC-M1 and MDA-MB-468

SILAC identification details for BC-M1 and MDA-MB-468

Quality control of the SILAC experiment

Immunofluorescence and western blot analyses; Grp78va in DTC

Supplementary methods and supplementary figure Legends

The transcription factor grainyhead-like 2 (GRHL2) plays a crucial role in various developmental processes. Although GRHL2 recently has attracted considerable interest in that it could be identified as a novel suppressor of the epithelial-to-mesenchymal transition, evidence is emerging that GRHL2 also exhibits tumour-promoting activities. Aim of the present study therefore was to help defining the relevance of GRHL2 for human cancers by performing a comprehensive immunohistochemical analysis of GRHL2 expression in normal (n = 608) and (n = 3,143) tumour tissues using tissue microarrays. Consistent with its accepted role in epithelial morphogenesis, GRHL2 expression preferentially but not exclusively was observed in epithelial cells. Regenerative and proliferating epithelial cells with stem cell features showed a strong GRHL2 expression. Highly complex GRHL2 expression patterns indicative of both reduced and elevated GRHL2 expression in tumours, possibly reflecting potential tumour-suppressing as well as oncogenic functions of GRHL2 in distinct human tumours, were observed. A dysregulation of GRHL2 expression for the first time was found in tumours of non-epithelial origin (e.g., astrocytomas, melanomas). We also report GRHL2 copy number gains which, however, did not necessarily translate into increased GRHL2 expression levels in cancer cells. Results obtained by meta-analysis of gene expression microarray data in conjunction with functional assays demonstrating a direct regulation of HER3 expression further point to a potential therapeutic relevance of GRHL2 in ovarian cancer. Hopefully, the results presented in this study may pave the way for a better understanding of the yet largely unknown function of GRHL2 in the initiation, progression and also therapy of cancers.

Abstract Disseminated tumor cells (DTC), which share mesenchymal and epithelial properties, are considered to be metastasis-initiating cells in breast cancer. However, the mechanisms supporting DTC survival are poorly understood. DTC extravasation into the bone marrow may be encouraged by low oxygen concentrations that trigger metabolic and molecular alterations contributing to DTC survival. Here, we investigated how the unfolded protein response (UPR), an important cytoprotective program induced by hypoxia, affects the behavior of stressed cancer cells. DTC cell lines established from the bone marrow of patients with breast cancer (BC-M1), lung cancer, (LC-M1), and prostate cancer (PC-E1) were subjected to hypoxic and hypoglycemic conditions. BC-M1 and LC-M1 exhibiting mesenchymal and epithelial properties adapted readily to hypoxia and glucose starvation. Upregulation of UPR proteins, such as the glucose-regulated protein Grp78, induced the formation of filamentous networks, resulting in proliferative advantages and sustained survival under total glucose deprivation. High Grp78 expression correlated with mesenchymal attributes of breast and lung cancer cells and with poor differentiation in clinical samples of primary breast and lung carcinomas. In DTCs isolated from bone marrow specimens from breast cancer patients, Grp78-positive stress granules were observed, consistent with the likelihood these cells were exposed to acute cell stress. Overall, our findings provide the first evidence that the UPR is activated in DTC in the bone marrow from cancer patients, warranting further study of this cell stress pathway as a predictive biomarker for recurrent metastatic disease. Cancer Res; 75(24); 5367–77. ©2015 AACR.

Using a retrovirus-mediated cDNA expression cloning approach, we identified the grainyhead-like 2 (GRHL2) transcription factor as novel protooncogene. Overexpression of GRHL2 in NIH3T3 cells induced striking morphological changes, an increase in cell proliferation, anchorage-independent growth, and tumor growth in vivo. By combining a microarray analysis and a phylogenetic footprinting analysis with various biochemical assays, we identified the epidermal growth factor receptor family member Erbb3 as a novel GRHL2 target gene. In breast cancer cell lines, shRNA-mediated knockdown of GRHL2 expression or functional inactivation of GRHL2 using dominant negative GRHL2 proteins induces down-regulation of ERBB3 gene expression, a striking reduction in cell proliferation, and morphological and phenotypical alterations characteristic of an epithelial-to-mesenchymal transition (EMT), thus implying contradictory roles of GRHL2 in breast carcinogenesis. Interestingly, we could further demonstrate that expression of GRHL2 is directly suppressed by the transcription factor zinc finger enhancer-binding protein 1 (ZEB1), which in turn is a direct target for repression by GRHL2, suggesting that the EMT transcription factors GRHL2 and ZEB1 form a double negative regulatory feedback loop in breast cancer cells. Finally, a comprehensive immunohistochemical analysis of GRHL2 expression in primary breast cancers showed loss of GRHL2 expression at the invasive front of primary tumors. A pathophysiological relevance of GRHL2 in breast cancer metastasis is further demonstrated by our finding of a statistically significant association between loss of GRHL2 expression in primary breast cancers and lymph node metastasis. We thus demonstrate a crucial role of GRHL2 in breast carcinogenesis.

The activated leukocyte cell adhesion molecule (ALCAM) is overexpressed in many mammary tumors, but controversial results about its role and prognostic impact in breast cancer have been reported. Therefore, we evaluated the biologic effects of ALCAM expression in two breast cancer cell lines and a larger cohort of mammary carcinomas. By stable transfections, MCF7 cells with ALCAM overexpression and MDA-MB231 cells with reduced ALCAM levels were generated and analyzed in functional assays and cDNA microarrays. In addition, an immunohistochemical study on 347 patients with breast cancer with long-term follow-up and analysis of disseminated tumor cells (DTCs) was performed. In both cell lines, high ALCAM expression was associated with reduced cell motility. In addition, ALCAM silencing in MDA-MB231 cells resulted in lower invasive potential, whereas high ALCAM expression was associated with increased apoptosis in both cell lines. Among genes which were differentially expressed in clones with altered ALCAM expression, there was an overlap of 15 genes between both cell lines, among them cathepsin D, keratin 7, gelsolin, and ets2 whose deregulation was validated by western blot analysis. In MDA-MB231 cells, we observed a correlation with VEGF expression which was validated by enzyme-linked immuno sorbent assay (ELISA). Our IHC results on primary breast carcinomas showed that ALCAM expression was associated with an estrogen receptor-positive phenotype. In addition, strong ALCAM immunostaining correlated with nodal involvement and the presence of tumor cells in bone marrow. By Kaplan-Meier analysis, strong ALCAM expression in ductal carcinomas correlated with shorter recurrence-free intervals (P=0.048) and overall survival (OAS, P=0.003). Our results indicate that the biologic role of ALCAM in breast cancer is complex, but overexpression might be relevant for outcome in ductal carcinomas.

Extracellular matrix metalloproteinase inducer expressed by tumor cells stimulates peritumoral fibroblasts to produce matrix metalloproteinases, thus contributing to tumor invasion and metastasis. To assess its suitability as potential therapeutic target, the overall incidence of EMMPRIN expression in normal and neoplastic tissues was analyzed. EMMPRIN expression was detected immunohistochemically using monoclonal antibodies MEM-M6/1 and HIM6 and tissue microarrays with 2,348 and 608 tissue samples from 129 distinct tumor types and 76 different normal tissues, respectively. Expression and glycosylation state of EMMPRIN in human breast cancer cells were analyzed by Western blot analysis with monoclonal antibodies recognizing distinct carbohydrate structures and biochemical methods. EMMPRIN expression was found in 112 of 129 tumor entities analyzed with malignant tumors being EMMPRIN positive more frequently than benign tumors. A remarkable heterogeneity in EMMPRIN expression between tumor entities was observed. Among others, squamous-cell carcinomas (60-100%), pancreatic (87%), chromophobic kidney (83%), hepatocellular (83%) or medullary breast (83%) adenocarcinomas as well as glioblastoma multiforme (79%) presented with a particular high incidence of EMMPRIN expression. There were a limited number of EMMPRIN-positive normal cell types including proliferatively active and differentiating epithelial cells, germ cells, myocardial cells in the left heart ventricle or vascular endothelial cells of the brain. We could further demonstrate that breast cancer cells expressed EMMPRIN isoforms differing in the presence or absence of Lewis(X) glycan structures. Our results may assist in defining the suitability of EMMPRIN as therapeutic target and predicting negative side effects. (c) 2006 Wiley-Liss, Inc.

PURPOSE:The bone marrow is a frequent and clinically important homing site for early disseminated breast cancer cells. Here, we aimed to profile the protein expression of these cells using unique cell line models and to evaluate the prognostic relevance of candidate gene expression for breast cancer patients.EXPERIMENTAL DESIGN:To identify expression patterns characteristic for micrometastatic cells, three different cell lines (BC-K1, BC-P1, and BC-S1) established by SV40 immortalization of cancer cells isolated from the bone marrow of patients with breast cancer were compared with MCF-7 breast cancer and SV40 immortalized normal breast ductal cells (MTSV-1.7) using two-dimensional gel electrophoresis followed by MALDI-ToF analysis. The prognostic significance and clinicopathologic associations of selected differentially expressed proteins were evaluated using high-density breast cancer tissue microarrays.RESULTS:In contrast to MCF-7 and MTSV1-7 reference cell lines, all micrometastatic cancer cell lines displayed loss of epithelial cytokeratins (CK8, CK18, and CK19) and ectopic expression of vimentin commonly present in mesenchymal cells. Immunohistochemical analysis of 2,517 samples of breast cancer further showed that loss of cytokeratin and ectopic vimentin expression were significantly associated with a higher tumor grade, high mitotic index, and negative estrogen/progesterone-receptor status. Although in univariate analyses significantly related to clinical outcome, none of the cytokeratins analyzed were independently associated with either overall or cancer-specific survival.CONCLUSIONS:Micrometastatic cancer cells exhibit marked changes in the expression pattern of cytoskeletal proteins indicative of an epithelial-mesenchymal transition. This phenotypical change could already be detected in primary tumors and is associated with the aggressive behavior of breast cancer cells in vivo.

Purpose: EMMPRIN (extracellular matrix metalloprotease inducer) is a glycosylated member of the immunoglobulin superfamily known to stimulate the production of matrix metalloproteases (MMPs) 1, 2, and 3 and MT1-MMP in peritumoral fibroblasts. We here evaluated whether EMMPRIN expression is related to tumor progression in human breast cancer. Experimental Design: An immunohistochemical study using high-density tissue microarrays (n = 2222 breast cancer samples) and EMMPRIN-specific antibodies HIM6 and MEM-M6/1 was performed, and staining results were statistically correlated with various clinicopathological parameters. To analyze the putative association between EMMPRIN expression and bone marrow (BM) micrometastasis, an additional set of 55 breast tumors from patients with or without micrometastatic cells as determined with anti-cytokeratin antibody A45-B/B3 were included in our study. Cytokeratin-positive cells in BM were costained with EMMPRIN-specific antibody 1G6.2. Results: Positive EMMPRIN staining correlated significantly with various histopathological risk factors (higher tumor grade, increased tumor size, negative estrogen receptor status and progesterone receptor status, and higher mitotic index) as well as decreased tumor-specific survival (log-rank, P = 0.0027). In particular, in patients > 50 years (i.e., postmenopausal women), EMMPRIN expression was an independent prognosticator as shown by Cox regression analysis (relative risk = 1.7, 95% confidence interval 1.4–4.3, P = 0.036). An involvement of EMMPRIN in tumor progression was also supported by the fact that it was expressed on ∼90% of micrometastatic cells in BM. Conclusions: EMMPRIN expression in primary tumor predicts an unfavorable prognosis in breast cancer, suggesting a crucial role of EMMPRIN in progression of human mammary carcinomas.

Hyaluronan or hyaluronic acid (HA) is a normal component of mammalian follicular, oviduct, and uterine fluids. Granulosa and expanding cumulus cells secrete large amounts of HA, and when HA is added in maturation and culture media, it improves the developmental potential of oocytes and embryos. HA regulates gene expression, signaling, proliferation, motility, adhesion, and morphogenesis. Many of these biological activities of HA are mediated through binding to the receptor for HA-mediated motility/intracellular HA-binding protein (RHAMM/IHABP). We evaluated the presence and dynamics of RHAMM/IHABP mRNA and protein expression in different stages of in vitro-produced bovine embryos using quantitative reverse transcriptase-real time-polymerase chain reaction and immunohistochemistry. We also analyzed the effects of different culture systems on the relative abundance of RHAMM/IHABP transcripts. RHAMM/IHABP mRNA levels decreased from the 2-cell to the 16-cell stage, increased again at the morula stage, and reached their highest level at the expanded blastocyst stage. RHAMM/IHABP mRNA abundance was significantly (P < 0.05) lower in embryos recovered in serum-containing medium than in embryos from serum-free media. Immunohistochemistry revealed the presence of RHAMM/IHABP first in 8-cell stages. Whereas RHAMM staining in 8-cell and morula stages was intense, it was weaker in blastocysts. Embryonic secretion of HA increased from the 2-cell stage until the 8-cell stage and then decreased in 16-cell embryos. After this, HA secretion increased in expanded and hatched blastocyst stages. These data suggest that the positive effects of HA on in vitro-produced bovine embryos may be mediated at least in part by RHAMM/IHABP.

In a quite spectacular report, a novel mechanism of cell transformation was proposed: a matrix receptor linked by a GPI anchor to the outer surface of the plasma membrane, called RHAMM (receptor for hyaluronic acid mediated motility), acted as a dominant oncogene and was required for transformation (4Hall C.L. Yang B. Yang X. Zhang S. Turley M. Samuel S. Lange L.A. Wang C. Curpen G.D. Savani R.C et al.Cell. 1995; 82: 19-26Abstract Full Text PDF PubMed Scopus (252) Google Scholar). This finding stimulated subsequent investigations that have revealed significant inconsistencies with the original study and with several other. The aim of this letter is to clarify these discrepancies and to stimulate discussion on this controversial issue. RHAMM was originally identified as a 56–58 kDa hyaluronate-binding protein present in the supernatant of murine fibroblasts and fibrosarcoma cell lines (9Turley E.A. Moore D. Hayden L.J Biochemistry. 1987; 26: 2997-3005Crossref PubMed Scopus (58) Google Scholar). A polyclonal antibody raised against this protein stained structures on the cell surface (10Turley E.A. Brassel P. Moore D Exp. Cell Res. 1990; 187: 243-249Crossref PubMed Scopus (26) Google Scholar, 6Hardwick C. Hoare K. Owens R. Hohn H.P. Hookm M. Moore D. Cripps V. Austen L. Nance D.M. Turley E.A J. Cell Biol. 1992; 117: 1343-1350Crossref PubMed Scopus (303) Google Scholar, 8Samuel S.K. Hurta R.A. Spearman M.A. Wright J.A. Turley E.A. Greenberg A.H J. Cell Biol. 1993; 123: 749-758Crossref PubMed Scopus (109) Google Scholar). A first cDNA sequence for murine RHAMM, which contained an open reading frame encoding a 52 kDa polypeptide, was isolated using this polyclonal antibody (6Hardwick C. Hoare K. Owens R. Hohn H.P. Hookm M. Moore D. Cripps V. Austen L. Nance D.M. Turley E.A J. Cell Biol. 1992; 117: 1343-1350Crossref PubMed Scopus (303) Google Scholar; see Figure 1). However, subsequent studies using antibodies based upon this cDNA clone sequence information have recognized numerous apparent RHAMM protein isoforms ranging from 52 to 125 kDa (e.g.6Hardwick C. Hoare K. Owens R. Hohn H.P. Hookm M. Moore D. Cripps V. Austen L. Nance D.M. Turley E.A J. Cell Biol. 1992; 117: 1343-1350Crossref PubMed Scopus (303) Google Scholar, 2Entwistle J. Zhang S. Yang B. Wong C. Li Q. Hall C.L. Jingbo A. Mowat M. Greenberg A.H. Turley E.A Gene. 1995; 163: 233-238Crossref PubMed Scopus (52) Google Scholar, 5Hall C.L. Lange L.A. Prober D.A. Zhang S. Turley E.A Oncogene. 1996; 13: 2213-2224PubMed Google Scholar). Molecular masses of 58, 60, 64, 70, and 84 kDa have been reported for the human RHAMM protein (see citations in 7Hofmann M. Fieber C. Assmann V. Göttlicher M. Sleeman J. Plug R. Howells N. von Stein O. Ponta H. Herrlich P J. Cell Sci. 1998; 111: 1673-1684Crossref PubMed Google Scholar). Later it became apparent that the cDNA described by 6Hardwick C. Hoare K. Owens R. Hohn H.P. Hookm M. Moore D. Cripps V. Austen L. Nance D.M. Turley E.A J. Cell Biol. 1992; 117: 1343-1350Crossref PubMed Scopus (303) Google Scholar was not full length. A subsequent RHAMM cDNA clone isolate (2Entwistle J. Zhang S. Yang B. Wong C. Li Q. Hall C.L. Jingbo A. Mowat M. Greenberg A.H. Turley E.A Gene. 1995; 163: 233-238Crossref PubMed Scopus (52) Google Scholar; see Figure 1) thought to be full-length murine RHAMM cDNA, contained an ORF encoding a polypeptide of 70 kDa. A putative splice variant of this cDNA clone contained an additional exon, named "v4," and encoded a 72 kDa "variant" isoform of RHAMM (Figure 1) that caused transformation and metastasis formation when overexpressed in immortalized murine fibroblasts (4Hall C.L. Yang B. Yang X. Zhang S. Turley M. Samuel S. Lange L.A. Wang C. Curpen G.D. Savani R.C et al.Cell. 1995; 82: 19-26Abstract Full Text PDF PubMed Scopus (252) Google Scholar). Based on the published cDNA sequence, we raised three polyclonal antibodies recognizing different epitopes in human and murine RHAMM proteins, each of which only recognizes either an 85 kDa protein in extracts of human cells (1Assmann V. Marshall Fieber C. Hofmann M. Hart I.R J. Cell Sci. 1998; 111: 1685-1694Crossref PubMed Google Scholar) or a single 95 kDa protein in extracts of immortalized murine cells and primary tissues (7Hofmann M. Fieber C. Assmann V. Göttlicher M. Sleeman J. Plug R. Howells N. von Stein O. Ponta H. Herrlich P J. Cell Sci. 1998; 111: 1673-1684Crossref PubMed Google Scholar, 3Fieber C. Plug R. Sleeman J. Dall P. Ponta H. Hofmann M Gene. 1998; in pressGoogle Scholar; see Figure 1). We also isolated additional sequences at the 5′ end of the murine RHAMM cDNA (7Hofmann M. Fieber C. Assmann V. Göttlicher M. Sleeman J. Plug R. Howells N. von Stein O. Ponta H. Herrlich P J. Cell Sci. 1998; 111: 1673-1684Crossref PubMed Google Scholar) that share high homology with the 5′ end of the newly isolated human RHAMM cDNA clones (11Wang C. Entwistle J. Hou G. Li Q. Turley E.A Gene. 1996; 174: 299-306Crossref PubMed Scopus (50) Google Scholar, 1Assmann V. Marshall Fieber C. Hofmann M. Hart I.R J. Cell Sci. 1998; 111: 1685-1694Crossref PubMed Google Scholar). The new 5′ sequences are encoded by additional exons that are contiguous with "RHAMM" genomic sequences (3Fieber C. Plug R. Sleeman J. Dall P. Ponta H. Hofmann M Gene. 1998; in pressGoogle Scholar; see Figure 1). Ectopic expression of our full-length murine cDNA in human fibroblasts leads to the expression of the 95 kDa protein only (7Hofmann M. Fieber C. Assmann V. Göttlicher M. Sleeman J. Plug R. Howells N. von Stein O. Ponta H. Herrlich P J. Cell Sci. 1998; 111: 1673-1684Crossref PubMed Google Scholar). In RHAMM null mice the 95 kDa protein exclusively disappears (unpublished data). Finally, extensive RT-PCR analyses of various tissues fail to detect alternatively spliced RNAs or variation in the expression of the v4 exon (exon 8). Taken as a whole, these data argue against major splicing-based variation in RHAMM gene products. Moreover, they suggest that the RHAMM cDNA clone used in the above transformation experiments (4Hall C.L. Yang B. Yang X. Zhang S. Turley M. Samuel S. Lange L.A. Wang C. Curpen G.D. Savani R.C et al.Cell. 1995; 82: 19-26Abstract Full Text PDF PubMed Scopus (252) Google Scholar) encodes an N-terminally truncated nonnative "RHAMM" protein. Structural predictions from the full-length cDNA sequences of both human and murine "RHAMM" revealed no sequences predisposing the protein to be transported to the plasma membrane. Consistent with this, we found only intracellular, mostly cytoplasmic distribution of RHAMM in all cell types examined, while neither surface staining nor protein shedding could be detected (1Assmann V. Marshall Fieber C. Hofmann M. Hart I.R J. Cell Sci. 1998; 111: 1685-1694Crossref PubMed Google Scholar, 7Hofmann M. Fieber C. Assmann V. Göttlicher M. Sleeman J. Plug R. Howells N. von Stein O. Ponta H. Herrlich P J. Cell Sci. 1998; 111: 1673-1684Crossref PubMed Google Scholar). In contrast4Hall C.L. Yang B. Yang X. Zhang S. Turley M. Samuel S. Lange L.A. Wang C. Curpen G.D. Savani R.C et al.Cell. 1995; 82: 19-26Abstract Full Text PDF PubMed Scopus (252) Google Scholar reported that overexpression of RHAMM1v4 in immortalized murine fibroblasts resulted in a 20-fold increase of RHAMM on the cell surface (see Table 1 of 4Hall C.L. Yang B. Yang X. Zhang S. Turley M. Samuel S. Lange L.A. Wang C. Curpen G.D. Savani R.C et al.Cell. 1995; 82: 19-26Abstract Full Text PDF PubMed Scopus (252) Google Scholar). Curiously, the same RHAMM1v4 isoform expressed in the same cells was later reported to be located exclusively in the cytoplasm (12Zhang S. Chang M.C.Y. Zylka D. Turley S. Harrison R. Turley E.A J. Biol. Chem. 1998; 273: 11342-11348Crossref PubMed Scopus (179) Google Scholar). Questions about the physiological significance of the RHAMM transformation data are raised by the fact that the link between surface hyaluronan binding and transformation has become doubtful and that the transformation studies used a truncated and incompletely processed RHAMM cDNA clone. The same is true of other work using these truncated murine RHAMM constructs (e.g.12Zhang S. Chang M.C.Y. Zylka D. Turley S. Harrison R. Turley E.A J. Biol. Chem. 1998; 273: 11342-11348Crossref PubMed Scopus (179) Google Scholar). Nonetheless the role of this protein in transformation still warrants further investigation and could yet represent a novel pathway of oncogenic activity.‡Present address: LION Bioscience AG, Im Neuenheimer Feld 517, D-69120 Heidelberg, Germany.

To explore the putative role of hyaluronan (HA) in tumor invasion in pancreatic cancer, we investigated the expression of the HA receptors CD44s and RHAMM in a panel of human pancreatic cancer cell lines. Expression of CD44s has been found in only 1 of 10 cell lines included in this study. This cell line exhibits a highly differentiated phenotype without any metastatic potential when injected into nude mice. Since it has previously been shown that normal pancreatic duct cells express a high level of CD44s, our results indicate that pancreatic cancer may be accompanied by an almost complete loss of CD44s expression. As demonstrated by PCR amplification, this loss of CD44s expression is due to alternative splicing of CD44 pre-RNA. Although most of the pancreatic cancer cell lines express a complex but identical pattern of variant CD44 gene transcripts, only one higher molecular weight CD44 isoform can be detected in a subset of pancreatic cancer cell lines in Western blot analysis. This variant CD44 molecule represents the epithelial CD44 isoform (CD44v8-v10). When cells are cultured on Matrigel, the expression of additional CD44 variants is induced, suggesting that the extracellular matrix can influence the expression of CD44 isoforms and thereby may facilitate tumor invasion. This induction could be due to a regulatory process in the translation of the CD44 variant mRNAs expressed in pancreatic tumor cells. Molecular cloning of a cDNA encoding human RHAMM reveals that both HA receptors are structurally unrelated. In addition, they share an inverse expression pattern. RHAMM mRNA is overexpressed in pancreatic cancer cell lines exhibiting a poorly differentiated phenotype and a high metastatic potential when injected into nude mice. These results indicate that CD44 and RHAMM differentially contribute to invasion of pancreatic adenocarcinoma; however, these functions still remain to be determined.