Breast Cancer Research and Treatment 56: 91–97, 1999. © 1999 Kluwer Academic Publishers. Printed in the Netherlands.
Report
Association of in vitro invasiveness and gene expression of estrogen receptor, progesterone receptor, pS2 and plasminogen activator inhibitor-1 in human breast cancer cell lines Dan Tong1 , Klaus Czerwenka2 , Jan Sedlak3 , Christian Schneeberger4 , Ingrid Schiebel1 , Nicole Concin1 , Sepp Leodolter1,5 , and Robert Zeillinger1 1 Division
of Gynecology, Molecular Oncology Group; 2 Department of Clinical Pathology, University of Vienna, Vienna, Austria; 3 Cancer Research Institute, Bratislava, Slowakia; 4 Division of Gynecological Endocrinology and Reproductive Medicine, Department of Obstetrics and Gynecology, Vienna, Austria; 5 Ludwig Boltzmann Institute for Gynecological Oncology and Reproductive Medicine, A-1090 Vienna, Austria
Key words: breast cancer cell lines, gene expression pattern, invasiveness
Summary The invasive potential of tumor cells is usually tested either by in vitro invasion assays which evaluate cell spreading ability in basement membrane-like matrices or by in vivo invasion assays in nude mice. Both methods are laborious and time-consuming. Tumor invasiveness is accompanied by the changes in expression of various genes. The invasive behavior of cells is therefore represented by certain gene expression patterns. The purpose of this study was to investigate whether expression patterns of several genes are characteristic for the invasiveness of cultured cells. We examined the mRNA levels of estrogen receptor (ER), progesterone receptor (PR), estrogen inducible pS2 and plasminogen activator inhibitor-1 (PAI-1) in 23 cell lines derived from benign and malignant breast tissues using a competitive reverse transcription-polymerase chain reaction (cRT-PCR) system. We also evaluated the invasiveness of these cell lines by their ability to penetrate into a collagen-fibroblast matrix. We demonstrate that the gene expression pattern of breast cell lines is clearly associated with their in vitro invasiveness. In general, cells with ER, PR, pS2 but no PAI-1 expression showed a non-invasive phenotype, while cells expressing PAI-1 mRNA but not ER mRNA are invasive. Our study indicates that the invasiveness of breast cancer cell lines is characterized by PAI-1 gene expression and the lack of ER mRNA. This suggests that PAI-1 may participate in the invasive process. Introduction Metastasis causes death in approximately 50% of breast cancer patients [1]. The process of metastasis involves a complex series of cellular alterations and is influenced by different steroid hormones, their receptors, growth factors, oncogenes and tumor suppressor genes. Various enzymes also play important roles in this process by degrading basement membranes and extracellular matrix components. Estrogen receptor (ER) and progesterone receptor (PR) inversely correlate with tumor size, histological grade and lymph node involvement [2–4]. They are indicators of longer disease-free intervals and better overall survival for
patients with primary breast cancer [5]. It has been documented that breast cancer cells without ER are generally less differentiated and more aggressive than those containing functional ER [6]. Moreover, activation of ER by transfecting it into a receptor-negative breast cancer cell line decreases the metastatic and invasive potential of the cells [7]. The protein pS2 was first identified during a search for estrogen receptorassociated markers [8]. pS2 has been found to correlate with ER and PR status and to correlate inversely with tumor size and histological grade [2, 9, 10]. Like ER and PR, pS2 is also a marker for good prognosis [11, 12]. The urokinase-type plasminogen activator (uPA) is a serine protease that catalyzes the conver-
92
D Tong et al.
sion of inactive pro-enzyme plasminogen to the active enzyme plasmin [13]. Plasmin can activate type IV collagenase, which degrades collagen and other proteins of the basement membranes and thus facilitates the capacity of tumor cells to invade the surrounding tissues and finally to metastasize to distant organs. One of the regulators of the activity of uPA is plasminogen activator inhibitor-1 (PAI-1). PAI-1 blocks the activity of uPA by forming an enzyme-inhibitor complex [14]. A high level of PAI-1 has been found to be associated with low level of ER, PR and pS2 [15–17]. In lymph node positive breast cancers, PAI-1 expression is higher than in lymph node negative ones or benign lesions [15, 18, 19]. Clinical studies have also shown that a high level of PAI-1 in primary breast carcinomas is a significant prognostic indicator of shorter relapse free survival and shorter overall survival [1, 16, 17]. The invasive capacity of tumor cells is usually tested either by in vitro assays which evaluates the cell spreading ability in basement membrane-like matrices or by in vivo invasion assays in nude mice. Both methods are laborious and time-consuming. Tumor metastasis is a consequence of changes in expression of various genes. The invasive behavior of cells is therefore characterized by certain gene expression patterns. Using competitive reverse transcription-polymerase chain reaction (cRTPCR), the quantification of gene expression can be achieved rapidly. As ER, PR, pS2 and PAI-1 have been reported to be differentially expressed in breast carcinomas and in normal breast tissues, we were interested in the gene expression of these factors in cell lines derived from different breast tissues. Furthermore, we were also interested in the correlation of the gene expression pattern with the invasiveness status of these cell lines. A cRTPCR system which was developed and validated in our laboratory [20] enables the simultaneous mRNA determination of ER, PR, pS2 and PAI-1. Using this system, we evaluated mRNA expression levels of various genes in 21 breast cancer cell lines, one epithelial cell line isolated from the milk of a nursing mother and one cell line which was established from mammary tissue of a patient with fibrocystic breast disease. We compared the gene expression pattern of the cell lines with the invasiveness status obtained by a conventional in vitro invasion assay.
Methods Materials Breast cancer cell lines BT-20, BT-474, BT-483, BT-549, CAMA-1, DU4475, Hs 578T, SK-BR-3, MCF-7, MDA-MB-157, MDA-MB-175-VII, MDA-MB-231, MDA-MB-361, MDA-MB-435S, MDA-MB-453, MDA-MB-436, T-47D, UCAA812, UCAA893, ZR-75-1, ZR-75-30, epithelial cell line HBL-100 derived from the milk of a nursing mother and cell line MCF-12F which was established from the mammary tissue of a patient with fibrocystic breast disease, were purchased from American Type Culture Collection (ATCC; Rockville, MD). Cell culture and cellular RNA preparation All cell lines were cultivated according to ATCC’s instructions. Total RNA was extracted by isopycnic centrifugation as described previously [21]. cRT-PCR and quantification of mRNA expression of ER, PR, pS2 and PAI-1 For the quantification of several genes which play important roles in the pathogenesis of breast cancer, a competitive RT-PCR system has been established as previously described [20]. Briefly, a RNA multistandard and a glyceraldehyde-3- phosphate dehydrogenase (GAPDH) RNA standard containing sequences homologous to primer pairs for in vitro amplification of ER, PR, pS2, PAI-1 and GAPDH were constructed. The amplification of the standard sequences with each primer pair generated a standard product differing from its related endogenous wild-type product in size, allowing the discrimination between wild-type and standard PCR products by agarose gel electrophoresis. The amount of mRNA was calculated according to the intensities of the bands. To correct possible sample variations caused either by quantitative determination of total RNA or by RNA degradation, the quantitative RT-PCR results of the mRNAs were normalized to that of GAPDH. Invasion Assay The invasive capacity of the cell lines was analyzed by a method previously reported [22]. Briefly, 3 × 105 human foreskin fibroblasts were suspended in 6 ml minimal essential medium (MEM) containing 10% FCS and 0.05% rat-tail collagen and immediately
Invasiveness and gene expression pattern
93
Table 1. Invasiveness and gene expression∗ of ER, PR, pS2 and PAI-1 Cell line
ER
PR
pS2
PAI-1
Invasivness
BT-483 T-47D MDA-MB-361 CAMA-1 MCF-7 ZR-75-1 UCAA812 MDA-MB-175VII ZR-75-30 BT-474 UCAA893 DU4475 MDA-MB-157 BT-20 SK-BR-3 HBL-100 Hs 578T MCF-12F BT-549 MDA-MB-436 MDA-MB-231 MDA-MB-453 MDA-MB-435S
0.663 0.584 0.352 0.337 0.307 0.296 0.127 0.026 0.022 0.008 – – – – – – – – – – – – –
– 0.966 0.294 0.219 0.048 0.115 0.186 – – 0.104 – – – – – – – – – – – – –
2.943 – 0.420 0.481 2.301 1.629 6.086 0.972 0.047 0.603 0.177 – – – – – – – – – – – –
– – – – – – – – – – 0.009 0.018 0.026 0.039 0.059 0.437 0.941 0.991 1.233 1.340 3.825 – –
non-invasive non-invasive non-invasive non-invasive non-invasive non-invasive non-invasive non-invasive non-invasive invasive (cl) non-invasive non-invasive invasive (s) invasive (cl) non-invasive invasive (s) invasive (s) invasive (cl) invasive (s) invasive (s) invasive (s) non-invasive invasive (s+cl)
∗ mRNA levels were give as 103 of the relative expression of each gene
determined by cRT-PCR normalized to the corresponding expression of GAPDH. (s): invasion of single cells; (cl): invasion of cell clusters.
transferred into a well of the tissue culture cluster 12 plate (Costar, Cambridge, MA). A glass plate with 12 cylindrical inserts of 8 mm diameter was inserted into the mixture. After 2 h of incubation at 37◦ C in 5.0% CO2 , the glass plate was removed and residual liquid was soaked out of the chambers which had formed. Cell suspension (2 × 105 cells per 300 µl) was added to the chambers in the collagen matrix and incubated at 37◦C. After 12 h of incubation, the outer edges of the collagen matrix were gently loosened from the plastic. Cells were then maintained in the standard medium described by ATCC, the medium being changed three times per week. After 14 days, collagen matrix discs were fixed in 10% neutral buffered formalin and embedded in paraffin. Sections cut at 5 µm were stained with hematoxylin and eosin. The invasiveness of the cells was determined by microscopy.
Results The gene expression of ER, PR and pS2 were obviously correlated to each other (Table 1). In 13 of the 23 cell lines, there was no detectable expression of all these three genes, except for one cell line showing a very low level of pS2. In the rest of the cell lines, all had detectable ER expression, 7/10 and 9/10 had detectable PR and pS2, respectively. The expression of PAI-1 was inversely correlated with that of the hormone receptors and pS2. Eleven cell lines, which had detectable PAI-1 mRNA, showed no expression of ER, PR or pS2. Ten cell lines, which had no detectable PAI-1 mRNA levels expressed ER universally, and PR and pS2 mostly. Cell line MDA-MB-435S and MDA-MB-453 differed from the gene expression patterns described above. There was neither detectable expression of ER, PR, pS2 nor of PAI-1. Figure 1 shows examples of invasive and noninvasive cell growth in the collagen fibroblast mat-
94
D Tong et al.
rix under a microscope. Invasive cells penetrated through the matrix either in cell clusters or as single cells (Figure 1c and 1b). In contrast, non-invasive cells remained at the surface of the collagen matrix
(Figure 1a). We found a clear correlation between the gene expression patterns and the invasiveness status in 18 cell lines. Cells expressing ER, PR and pS2 mRNA, but not PAI-1 mRNA, were non-invasive, while the cells with detectable PAI-1 mRNA, but not an expression of ER, PR and pS2, were invasive (Table 1). Cell line BT-474, which expressed the lowest amount of ER mRNA, showed an invasive phenotype. Similarly, the cell lines DU4475 and UCAA893, which presented the lowest expression of PAI-1 mRNA were non-invasive. SK-BR-3 had the gene expression pattern characteristic for invasive cell lines, but presented a non-invasive phenotype. The two cell lines MDAMB-435S and MDA-MB-453 had a typical gene expression patterns, but differed in their penetrating abilities into the collagen matrix. While MDA-MB453 was non-invasive, the MDA-MB-435S was clearly invasive.
Discussion
Figure 1. Examples of cell growth in collagen-fibroblast matrix (magnification 400×). A: non-invasive cells (ZR-75-30) remain at the surface of the matrix; B: invasion of single cells (MDA-MB-231); C: cells (MCF-12F) invading in form of clusters.
The invasiveness part of the breast cancer cell lines used in our study and the levels of ER, vimentin and several markers of epithelial and fibroblastic differentiation were studied previously [23, 24]. In these reports, the Boyden chamber chemoinvasion assay and the Matrigel outgrowth assay were used to evaluate the invasiveness of the cells. Our results accorded completely with these reports regarding the invasiveness of the cell lines tested, confirming that the collagenfibroblast assay used in our study is reliable. The ER status of the cell lines is also in concordance with the results from these groups obtained from an enzyme-immunoassay, with the exception of MDAMB-175-VII which showed a low expression of ER in our study while no ER was detected by any one of the groups [24]. In one of our previous studies, PAI-1 protein levels in breast cancer cell lines were assessed by an enzyme-linked immunosorbent assay (ELISA, 25). Most of the PAI-1 mRNA data are in agreement with the protein measurement. But three cell lines which had detectable PAI-1 mRNA failed to have measurable PAI-1 protein. One of the possible reasons is that the sensitivity of our cRT-PCR system differs from that of the ELISA. On the other hand, the protein level is also dependent on the regulation of translation and protein processing and therefore might not proportionally reflect the measurable amount of mRNA. We showed that in most of the cell lines, PAI-1 expression was detectable when no ER was present in the
Invasiveness and gene expression pattern cells (exceptions were MDA-MB-453 and MDA-MB435S, which had neither ER nor PAI-1 expression). This suggests that the regulation of PAI-1 expression might be estrogen dependent. So far, there is little data on the regulation of PAI-1 gene expression by estrogens. However, estradiol-stimulated fibrinolytic activity increase is a well-known phenomenon. Plasma levels of PAI-1 usually increase in women after menopause. Hormone replacement therapy in postmenopausal women enhances the fibrinolytic capacity by lowering the PAI-1 protein level [26, 27]. Estrogens might repress PAI-1 expression through other regulatory factors, but not directly, as the promoter region of PAI-1 does not contain any estrogen response element. However, the PAI-1 gene contains transforming growth factor-beta (TGF-β) response elements and its expression is regulated by TGF-β1 [28–30]. Estrogens appear to restrain TGF-β1 because depletion of estrogen induces a transient increase in TGF-β protein [31]. In addition, antiestrogens like tamoxifen and toremifen induce TGF-β1 expression [32, 33]. We hypothesize that estrogens restrain TGF-β and thus indirectly repress the PAI-1 gene expression. BT-474 presented the lowest level of ER mRNA in all of the ER positive cell lines and was invasive, while UCAA893 and DU4475 presented the lowest level of PAI-1 mRNA in all of the PAI-1 positive cell lines and were non-invasive. This indicates that a very low level of ER might not be able to repress the invasive activities of the cells while a very low level of PAI-1 might not be able to start the invasive activities of the cells. This could indicate indirectly that the invasive potential of the cells might increase with the decrease of the ER mRNA levels and the increase of PAI-1 gene expression level. Increasing ER protein level or blocking PAI-1 activity in invasive cells might therefore reduce their invasive potential. The expression of vimentin, an intermediate filament in human breast cancer is associated with the lack of ER and high histological grade [23, 34]. Comparing our result concerning PAI-1 expression with published data on vimentin expression, it seems that there is a correlation between the expression of these two factors [24]. A positive correlation was observed in 13 out of 15 cell lines with SK-BR-3 and MDAMB-435S as exceptions. This interesting correlation suggests a possible common regulation of these two genes. Further studies will be needed for the understanding of this observation. SK-BR-3 showed a gene expression pattern characteristic for the invasive cell line. It expressed PAI-1
95
and had no detectable expression of ER, PR and pS2. However, it was non-invasive in our in vitro test as well as in the test of other groups [23, 24]. The lack of vimentin expression might be responsible for the non-invasive behavior of this cell line. MDA-MB-453 and MDA-MB-435S showed the same unusual gene expression pattern, but differed in their invasive behavior. This suggests that the regulation of tumor cell invasion might involve different mechanisms in these two cell lines. To study the details of the plasminogen activation system of the MDA-MB-453 and MDA-MB-435S cell lines might provide useful information for the understanding of tumor invasion. The importance of PAI-1 in tumor cell invasion is likely due to its ability to mediate the balance between cell adhesion and cell detachment other than its ability to function as a protease inhibitor [35]. A recent report indicates that PAI-1 produced by cancer cells is not sufficient for tumor cell invasion and tumor vascularization of transplanted malignant cells. Stroma-derived PAI-1 is also necessary and even more essential for progression of angiogenesis and tumor invasion [36]. In our study, PAI-1 expression was found to be positively correlated with cell invasiveness in general. However, there were three cell lines that showed PAI-1 gene expression and non- invasive phenotype and two cell lines that were PAI-1 negative and invasive. This also suggests that the invasiveness of breast cancer cells is not dependent solely on PAI-1 gene expression. In conclusion, our data show that invasiveness of breast cancer cells is commonly accompanied by PAI1 expression, suggesting that PAI-1 may participate in the invasive process. Moreover, our results suggest a possible regulation of PAI-1 by estrogens.
Acknowledgement This project was supported by the Austrian Science Fund (FWF; project No. P10032-Med).
References 1.
Foekens JA, Schmitt M, Van Putten WLJ, Peters HA, Kramer MD, Jänicke F, Klijn JGM: Plasminogen activator inhibitor1 and prognosis in primary breast cancer. J Clin Oncol 12: 1648–1658, 1994
96
D Tong et al.
2. Stonelake PS, Baker PG, Gillespie WM, Dunn JA, Spooner D, Morrison JM, Bundred NJ, Oates GD, Lee MJR, Neoptolemos JP, Chan SY, Baker PR: Steroid receptors, pS2 and cathepsin D in early clinically node-negative breast cancer. Eur J Cancer 30: 5–11, 1994 3. Stal O, Brisfors A, Carstensen J, Ferraud L, Hatschek T, Nordenskjöld B: The South-East Sweden Breast Cancer Group. Relationships of DNA ploidy, S-phase fraction and hormone receptor status to tumor stage in breast cancers detected by population screening. Int J Cancer 51: 28–33, 1992 4. Alghanem AA, Hussain S: The effect of tumor size and axillary lymph node metastasis on estrogen and progesterone receptors in primary breast cancer. J Surg Oncol 31: 218–221, 1986 5. Clark GM, McGuire WL: Prognostic factors in primary breast cancer. Breast Cancer Res Treat 3 (Suppl): S69–S72, 1983 6. McGuire WL: Hormone receptor in breast cancer. In: Bulbrook RD (ed) Cancer Surveys. Vol 5. Oxford University Press, Oxford, 1986, pp 527–536 7. Garcia M, Derocq D, Freiss G, Rochefort H: Activation of estrogen receptor transfected into a receptor-negative breast cancer cell line decreases the metastatic and invasive potential of the cells. Proc Natl Acad Sci USA 89: 11538–11542, 1992 8. Masiakowski P, Breathnach R, Bloch J, Gannon F, Krust A, Chambon P: Cloning of cDNA sequences of hormon-regulated genes from the MCF-7 human breast cancer cell line. Nucleic Acids Res 10: 7895–7903, 1982 9. Soubeyran I, Wafflart J, Bonichon F, DeMascarel I, Trojani M, Durand M, Avril A, Coindre JM: Immunohistochemical determination of pS2 in invasive breast carcinomas: a study on 942 cases. Breast Cancer Res Treat 34: 119–128, 1995 10. Racca S, Conti G, Pietribiasi F, Stramignoni D, Tampellini M, Valetto MR, Ghezzo F, Di Carlo F: Correlation between pS2 protein positivity, steroid receptor status and other prognostic factors in breast cancer. Int J Biol Markers 10: 87–93, 1995 11. Foekens JA, Rio MC, Seguin P, VanPutten WL, Fauque J, Nap M, Klijn JG, Chambon P: Prediction of relapse and survival in breast cancer patients by pS2 protein status. Cancer Res 50: 3832–3837, 1990 12. Horiguchi J, Lino Y, Takei H: Expression of pS2 estrogeninducible protein in primary breast cancer. Oncology 53: 12– 15, 1996 13. Liotta LA, Goldfarb RH, Brundage R, Siegal GP, Terranova V, Garbisa S: Effect of plasminogen activator (urokinase), plasmin, and thrombin on glycoprotein and collagenous components of basement membrane. Cancer Res 41: 4629–4636, 1981 14. VanMourik JA, Lawrence DA, Loskutoff DJ: Purification of an inhibitor of plasminogen activator (antiactivator) synthesized by endothelial cells. J Biol Chem 259: 14914–14921, 1984 15. Foucré D, Bouchet C, Hacène K, Pourreau-Schneider N, Gentile A, Martin PM, Desplaces A, Oglobine J: Relationship between cathepsin D, urokinase, and plasminogen activator inhibitors in malignant vs benign breast tumours. Br J Cancer 64: 926–932, 1991 16. Jänicke F, Schmitt M, Pache L, Ulm K, Harbeck N, Höfler H, Graeff H: Urokinase (uPA) and its inhibitor PAI-1 are strong and independent prognostic factors in node-negative breast cancer. Breast Cancer Res Treat 24: 195–208, 1993 17. Grøndahl-Hansen J, Christensen IJ, Rosenquist C, Brünner N, Mouridsen HT, Danø K, Blichert-Toft M: High levels of urokinase-type plasminogen activator and its inhibitor PAI-1 in cytosolic extracts of breast carcinomas are associated with poor prognosis. Cancer Res 53: 2513–2521, 1993
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
Sumiyoshi K, Baba S, Sakaguchi S, Urano T, Takada Y, Takada A: Increase in levels of plasminogen activator and type-1 plasminogen activator inhibitor in human breast cancer: possible roles in tumor progression and metastasis. Thromb Res 63: 59–71, 1991 Sumiyoshi K, Serizawa K, Urano T, Takada Y, Takada A, Baba S: Plasminogen activator system in human breast cancer. Int J Cancer 50: 345–348, 1992 Tong D, Schneeberger C, Leodolter S, Zeillinger R: Quantitative determination of gene expression by competitive reverse transcription-polymerase chain reaction in degraded RNA samples. Anal Biochem 251: 173–177, 1997 Kury FD, Schneeberger C, Sliutz G, Kubista E, Salzer H, Medl M, Leodolter S, Swoboda H, Zeillinger R, Spona J: Determination of HER-2/neu amplification and expression in tumor tissue and cultured cells using a simple, phenol free method for nucleic acid isolation. Oncogene 5: 1403–1408, 1990 Sedlák J, Sedláková O, Hlavcák P, Hunáková L, Bizik J, Grófová M, Chorváth B: Cell surface phenotype and increased penetration of human multidrug-resistant ovarian carcinoma cells into in vitro collagen-fibroblasts matrix. Neoplasma 43: 389–395, 1996 Thompson EW, Paik S, Brünner N, Sommers CL, Zugmaier G, Clarke R, Shima TB, Torri J, Donahue S, Lippman ME, Martin GR, Dickson RB: Association of increased basement membrane invasiveness with absence of estrogen receptor and expression of vimentin in human breast cancer cell lines. J Cell Physiol 150: 534–544, 1992 Sommers CL, Byers SW, Thompson EW, Torri JA, Gelmann EP: Differentiation state and invasiveness of human breast cancer cell lines. Breast Cancer Res Treat 31: 325–335, 1994 Sliutz G, Eder H, Koelbl H, Tempfer C, Auerbach L, Schneeberger C, Kainz C, Zeillinger R: Quantification of uPA receptor expression in human breast cancer cell lines by cRTPCR. Breast Cancer Res Treat 40: 257–263, 1996 Koh KK, Mincemoyer R, Bui MN, Csako G, Pucino F, Guetta V, Waclawiw M, Cannon RO III: Effects of hormonereplacement therapy on fibrinolysis in postmenopausal women. N Engl J Med 336: 683–690, 1997 Meilahn EN, Cauley JA, Tracy RP, Macy EO, Gutai JP, Kuller LH: Association of sex hormones and adiposity with plasma levels of fibrinogen and PAI-1 in postmenopausal women. Am J Epidemiol 143: 159–166, 1996 Dong G, Schulick AH, DeYoung MB, Dichek DA: Identification of a cis-acting sequence in the human plasminogen activator inhibitor type-1 gene that mediates transforming growth factor-β1 responsiveness in endothelium in vivo. J Biol Chem 271: 29969–29977, 1996 Keeton MR, Curriden SA, VanZonneveld AJ, Loskutoff DJ: Identification of regulatory sequences in the type 1 plasminogen activator inhibitor gene responsive to transforming growth factor-beta. J Biol Chem 266: 23048–23052, 1991 Farina AR, Coppa A, Tiberio A, Tacconelli A, Turco A, Colletta G, Gulino A, Mackay AR: Transforming growth factor-beta1 enhances the invasiveness of human MDA-MB231 breast cancer cells by up-regulating urokinase activity. Int J Cancer 75: 721–730, 1998 Herman ME, Katzenellenbogen BS: Alterations in transforming growth factor-alpha and -beta production and cell responsiveness during the progression of MCF-7 human breast cancer cells to estrogen-autonomous growth. Cancer Res 54: 5867–5874, 1994 Knabbe C, Zugmaier G, Schmahl M, Dietel M, Lippman ME, Dickson RB: Induction of transforming growth factor-beta by
Invasiveness and gene expression pattern the antiestrogens droloxifene, tamoxifen, and toremifene in MCF-7 cells. Am J Clin Oncol 14 (Suppl 2): S15–S20, 1991 33. Chen H, Tritton TR, Kenny N, Absher M, Chiu JF: Tamoxifen induces TGF-beta 1 activity and apoptosis of human MCF-7 breast cancer cells in vitro. J Cell Biochem 61: 9–17, 1996 34. Raymond WA, Leong AS: Co-expression of cytokeratin and vimentin intermediate filament proteins in benign and neoplastic breast epithelium. J Pathol 157: 299–306, 1989 35. Deng G, Curriden SA, Wang S, Rosenberg S, Loskutoff DJ: Is plasminogen activator inhibitor-1 the molecular switch that governs urokinase receptor-mediated cell adhesion and release? J Cell Biol 134: 1563–1571, 1996
36.
97
Bajou K, Noel A, Gerard RD, Masson V, Brunner N, HoistHansen C, Skobe M, Fusenig NE, Carmeliet P, Collen D, Foidart JM: Absence of host plasminogen activator inhibitor1 prevents cancer invasion and vacularization. Nat Med 4: 923–928, 1998
Address for offprints and correspondence: Robert Zeillinger, Department of Obstetrics and Gynecology, Division of Gynecology, Molecular Oncology Group, University of Vienna, Währinger Gürtel 18–20, EBO 050, A-1090 Vienna, Austria; Tel: +43-1-404007831; Fax: +43-1-40400-7832; E-mail:
[email protected]