Motility Related Protein L (MRP-1/CD9) Expression: Inverse Correlation with MÃ©Tastasesin Breast Cancer1

ICANCERRESI-ARCH55.41:7-41.11.Scpicmhcr15.iw| Motility Related Protein l (MRP-1/CD9) Expression: Inverse Correlation with MÃ©tastasesin Breast Cancer1

Masayuki Miyake,2 Keiji Nakano, Yoshiaki leki, Masashi Adachi, Cheng-Long Huang, Shin-ichi Itoi, Takashi Koh, and Toshihiko Taki Department of Thoracic Surgery unti Department V of Oncology, Kilami Hospital. Ta-uke Kofukai Medical Research Institute, /.<-.' Kamiyamacho. Osaka 5JO Â¡M.M., K. N.. Y. I.. M. A., C-L. H., S-i. I.. T. T.I and Department of Pathology and Tumor Biology, r'aciilly of Medicine, Kyoto University, Kyoto 606 Â¡T.KJ, Japan

ABSTRACT MRP-1/CD9 in non-hematopoietic cells are still unknown. In an effort to clarify the functions of MRP-1/CD9 in such cells, we translected In our previous studies we showed that motility related protein 1 various types of cultured cells with plasmid constructs containing (MRP-1) is a glycoprotein recognized by niAh M31-15, and that the sequence of MRP-1 is identical to that of CI)'), a YVBC differentiation MRP-1/CD9 cDNA (10). These experiments revealed that cell motil antigen. Transfcction of MKP-1/CD9 cDNA into cultured nonhematopoi- ity and growth were suppressed in the MRP-l/CD9-expressing cells. etic cells suppresses cell motility. The extent of suppression is directly In addition, studies with the mouse melanoma BL6 cells BALB/c- related to the level of MRP-1/CD9 expression. In addition, the metastatic nu/nu mouse system (11) disclosed that the metastatic potential of all potential of MRP-l/CI)9-transfected melanoma BL6 cells is lower than transformants expressing MRP-1/CD9 was lower than that of the that of control HI.<>cells. To determine whether these experimental results parent BL6 cells. These findings would suggest that MRP-1/CD9 are of relevance with respect to actual human tumors, we investigated regulates cell motility and is a receptor for negative signal ligands. On MRP-1/CD9 expression in 143 invasive ductal carcinomas of the breast. the basis of these experimental results, we investigated whether there Of 97 patients with MRP-l/CD9-positive tumors, only 36 (37.1%) had is a relationship between primary human tumors and their mÃ©tastases lymph node involvement. In contrast, 21 of 39 (53.8%) patients whose tumors had reduced MRP-I/CD9 immunoreactivity and 5 of 7 patients with respect to MRP-1/CD9 expression. In this report we present the whose primary carcinomas were not stained by the anti-MRP-l/CD9 MAb results of such a study, carried out on surgically resected breast cancer had lymph node mÃ©tastases.The comparison of protein expression by 62 tissues and the respective metastatic lymph nodes. primary tumors and their respective metastatic lymph nodes revealed that in almost 50% of the cases, the latter had lower MRP-1/CD9 levels than the former. Moreover, reverse transcriptase-PCR-based analysis disclosed MATERIALS AND METHODS that MRP-I/CD9 gene expression in the metastatic lymph nodes of 17 of 32 patients was strikingly lower than in the primary invasive ductal carci Patients and Surgical Specimens. We studied specimens from 143 pa nomas. Gene overexpression was not observed in any of the samples tients with invasive ductal carcinoma of the breast who underwent surgery at studied. Our data suggest that low MRP-1/CD9 expression may be asso the Department of Thoracic Surgery of the Kitano Hospital, Medical Research ciated with the metastatic potential of certain human tumors. Institute of Osaka between May 1990 and December 1992. Ten patients with intraductal carcinoma and nine with double cancers were not included in the study. Four patients with distant mÃ©tastaseswere also excluded from this INTRODUCTION study. The salient clinical characteristics of the patients are presented in Table 1. One-half of each fresh tumor tissue and each lymph node obtained at surgery Cell motility is one of the rather essential complex functions were immediately embedded in optimum cutting temperature compound operative in inflammation, tissue repair, angiogenesis, and tumor (Miles, Elkhart, IN) and frozen at -80Â°C. Frozen sections were cut on the invasion ( I). We have shown previously that motility of MAC 10 cells, cryostat to a thickness of (i Â¡im.mounted on poly-D-lysine-coated slides, and derived from a human adenocarcinoma of the lung, is inhibited by either used immediately or stored at -8()Â°C until needed. In each case, one mAb M31-I5, and that this antibody recognizes MRP-1,3 a transmem additional section was stained with hematoxylin and eosin for verifying the brane glycoprotein (2). We also showed that the sequence of MRP-1 presence of cancer cells, especially in the metastatic lymph nodes. One-half of cDNA is identical to that of CD9 (2), reported in the same year by two a given lymph node specimen containing only cancer cells was used for other groups (3, 4). CD9 is a M, 24,000-27,000 glycoprotein that is Western blotting and RT-PCR. widely expressed on hematopoietic tissues, platelets, eosinophils, Immunohistochemical Assays. Because ine MRP-1/CD9 antigen is not basophils, stimulated T lymphocytes, and early B cells, but not on preserved in formalin-fixed, paraffin-embedded tissues, we used frozen sec resting T and B cells (3-5). Several studies have demonstrated that tions for these assays; they were performed as described previously (2, 12). mAbs to CD9 cause platelet activation and aggregation by a mecha Briefly, the sections were immersed for 30 min in 0.3% H2O: (in absolute nism similar to that of thrombin (3-6). Moreover, it has been sug methanol) to inhibit endogenous peroxidase activity and then treated with 5% BSA. Sections were incubated for 2 h with the anti-MRP-1/CD9 mAb M31-15, gested that this type of platelet activation is coupled to phospholipase A2 and to phospholipase C stimulation, which in turn induces an sequentially followed by incubation with biotinylated horse antimouse IgG and the avidin-biotin-peroxidase complex. Antibody binding was visualized with increase in cytoplasmic calcium and protein tyrosine phosphorylation 3.3'-di:iminohcn/idinc tetnihydrochloridc in the presence of 0.05% H2Oi. The (7-9), thus pointing to the direct participation of CD9 in signal sections were then counterstained with hematoxylin, dehydrated, and mounted. transduction. Although we found that MRP-1/CD9 is expressed on Biotinylated horse antimouse IgG, avidin. and biotin were purchased from most cell lines derived from solid human tumors (2), the functions of Vector Laboratories. Inc. (Burlingame, CA). Sections incubated with mouse myeloma SP, supernatant served as negative reaction controls. Received 3/23/95; accepted 7/18/95. Western Blotting. Cancer cell-containing tissue samples were selected and The costs of publication of this article were defrayed in part by the payment of page soluhilÃŒ7.edwith1% CHAPS. An aliquot of the soluble fraction was subjected charges. This article must therefore he hereby marked advertisement in accordance with IS U.S.C. Section 1734 solely to indicate this fact. to slab gel electrophoresis. followed by Western blotting and probing with 1This investigation was supported in part by Grants-in-Aid from the Ministry of M3I-15 mAb, rabbit antimouse IgG (-y-chain specific; Cappel, Malvern, PA) Education (to M. M.) and the Sagawa Foundation for Promotion of Cancer Research and I2SI protein A (DuPont New England Nuclear. Boston, MA; Refs. 13, 14). (lo M. M.). - To whom requests for reprints should be addressed. The total cellular protein content applied to each lane was adjusted to identical 'The abbreviations used are: MRP-1. motility-related protein I; RT-PC'R, reverse concentrations. Cells of the human breast cancer line ZR-75-30 were used as Iranscriplion-PCR. TM4SF. Iransmembrane 4 superfamily. positive controls. Band intensity was evaluated by densitometry. 4127

Fig. 1. Immunohisiochemical staining of human breast tissues with the MRP-1/CDQ-specific mAb M31-15. Avidin-biotin-peroxidase complex procedure. X 150. A, intense staining of normal ducts and lobules of the breast. R, strongly positive reaction in a benign fibroadenoma. C. positive immunostaining of an invasive ductal carcinoma of the breast. D, invasive ductal-carcinoma with reduced MRP-1/CDM expression. Â£,an MRP-l/CD9-negative invasive ductal carcinoma. F, metastatic lymph node corresponding to the invasive carcinoma shown in C. Note reduced staining intensity.

Specimen Classification Based on Immunohistochemical and Western RT-PCR Analysis. Total cellular RNA was isolated from fresh or frozen Blotting Results. When >50% of the carcinoma cells in a given specimen tumor tissues by the acid guanidinium thiocyanate procedure (15). Randomly were positively stained, the sample was classified as MRP-1/CD9 positive (+); primed cDNA was prepared from 5 /ig of total RNA by using Maloney murine when 5-50% were stained, as reduced (Â±);and when <5% were stained, as leukemia virus RT (Pharmacia, Piscataway. NJ) and following the manufac negative (-). To classify the specimens on the basis of Western blotting, the turer's protocol. One fil aliquot of the reaction was used for PCR ampli M,, 25.000 band obtained with the control cells ZR-75-30 was set at 100%. fication. MRP-l/CDu-specific oligonucleotides synthesized on the basis of the Patient samples with a band intensity of >30% were classified as MRP-1/CD9 nuclcotide sequence (2-4) served as primers; 5'-TGCATCTGTATC- positive (+), those with intensities between 3 and 30% as reduced (Â±).and <3% CAGCGCCA-3' was the sense primer, and 5'-CTCAGGGATGTAAGCT- as negative (-). GACT-3' was the antisense primer, ÃŸ-actinamplification was used as internal 4128

Fig. 2. Immunoblot analysis of MRP-1/CD9 ex pression in breast cancer. Lane I, breast cancer cell line ZR-75-30 (positive control); Lane 2, MRP-1/ CD9-positive primary invasive ductal carcinoma; Lane 3, primary invasive ductal carcinoma with reduced MRP-1/CD9 levels; Lane 4, invasive duc tal carcinoma with negative MRP-1/CD9; Lanes 5, 7, 9, and It, primary invasive carcinomas; Lanes 6, 8, 10, and 12, metastatic lymph nodes of these four carcinomas. Note that intensities of the lymph node 28 K bands are less than those of the respective primary breast tumors. K, molecular weight in thousands. 25 K

12345678 9101112

PCR control (16); the sense primer was 5'-CTGTCTGGCGGCACCACCAT- Immunoblot Analysis. The major M, 25,000 MRP-1/CD9 band 3', and the antisense primer was 5'-GCAACTAAGTCATAGTCCGC-3'. The was clearly evident in the Western blots of the soluble fraction of the MRP-1/CD9 primer pair amplifies an 800-bp fragment (nucleotides 19-818) primary tumors that were positively stained in the immunohistochem- that includes the full-length cDNA. Forty PCR amplification cycles were carried out in a thermal cycler, each consisting of 40 s at 94Â°C,40 s at 60Â°C, ical assays (Fig. 2). In contrast, the band intensity was weak or absent and 90 s at 72Â°C.The same conditions were used to amplify ÃŸ-actinDNA. in the primary invasive ductal carcinomas that had reduced or no immunohistochemically detectable MRP-1/CD9. Overall, the immu- Tubes containing all ingredients except templates were included in all runs as negative reaction controls. The amplified DNA samples were subjected to noblotting results agreed well with those of immunohistochemical agarose (1%) gel electrophoresis; bands were visualized with ethidium bro staining, and in case of discrepancy, the results of Western blotting mide and photographed with a Polaroid camera. Densitometric analysis of the were used in specimen classification. photographic negatives was used for band quantification (17). The values Relationship between Immunodetected MRP-1/CD9 and obtained for MRP-1/CD9 in a given sample were divided by that of ÃŸ-actinand Known Prognostic Factors. As shown in Table 1, we found no referred to as MRP-1/CD9 expression ratio. The expression ratio of a meta statistically significant relationship (x2) between protein expression static lymph node was divided by that of its corresponding primary tumor to and the age of patients at surgery, estrogen and progesterone receptor obtain the MRP-1/CD9 rate of reduction. Final values ranging from 0.5 to 2.0 status, or tumor size. In contrast, there was a significant relationship were considered as indicating no reduction in gene expression, and values of between MRP-1/CD9 expression and lymph node status (P = 0.045). <0.5 as denoting a decrease. Comparison of Immunohistochemical Staining Patterns of Pri mary Tumors and Their Respective Metastatic Lymph Nodes. RESULTS The analysis of the 62 cases from whom primary tumors and lymph Immunohistochemical Detection of MRP-1/CD9 in Normal, node mÃ©tastaseswere studied revealed that immunostaining intensity Fibroadenoma, and Cancer Tissues of the Breast. As shown in of the lymph nodes of all patients was equal or lower but never higher Fig. IA normal mammary ducts and glands were intensely stained by than that of the corresponding primary carcinoma (Table 2). Of the 36 the anti-MRP-1/CD9 antibody. Without exception, all epithelial cells patients whose primary tumors had high protein levels, expression in expressed the protein. On the other hand, there was a striking differ the involved lymph nodes was at the same level in 22 and lower in 14 ence between benign fibroadenomas and breast cancer tissues. The (Fig. IF). Moreover, the metastatic lymph nodes of 9 of 21 patients specimens from all 23 cases of fibroadenoma were strongly stained whose primary carcinoma had reduced protein levels were classified (Fig. IS); almost all benign tumor cells expressed MRP-1/CD9. The as MRP-1/CD9 negative (Table 2). reaction product deposits were located at the cell membrane. In RT-PCR Analysis of the MRP-1/CD9 Gene in Primary Tumors contrast, various distinct staining patterns were noted with the cancer and Their Metastatic Lymph Nodes. Lymph nodes with normal tissues. Of the 143 primary breast cancers studied, 97 cases (67.8%) lymphocytes were excluded from RT-PCR analysis. We found no were evaluated as MRP-1/CD9 positive. In these cases, expression evidence of PCR-mediated MRP-1/CD9 gene amplification in pri resembled that of normal glands and benign fibroadenomas because mary tumors and their respective lymph node mÃ©tastasesof patients immunostaining was intense and seen uniformly at the surface mem whose breast carcinoma cells were not positively stained by the brane (Fig. 1C). There were 39 cases (27.2%) with reduced (Â±) anti-MRP-l/CD9 antibody. These observations are in accord with our MRP-1/CD9 expression (Fig. ID). In most of these tumors, expres recent experimental findings that the levels of the MRP-1/CD9 gene sion was heterogeneous, and the immunostaining pattern along cell product depend on the quantity of transfected cDNA (10). On the junctions was not linear, but rather granular. No MRP-1/CD9 immu- other hand, a comparison between protein-positive tumors and their noreactivity was observed with the specimens of 7 (4.9%) of the 143 respective lymph nodes revealed that the band intensities of the breast cancer patients examined. An example of such a case is shown amplified MRP-1/CD9 DNAs were strikingly greater in the primary in Fig. IE. tumor specimens than in the metastatic lymph nodes (Fig. 3). We 4129

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1995 American Association for Cancer Research. MRP-I/CTW EXPRESSION IN BREAST CANCER found that the amount of MRP-1/CD9 cDNA was reduced (rate, <0.5) in the lymph nodes of 17 of 32 (53.1%) of the patients whose specimen pairs were subjected to RT-PCR analysis. This set of ob servations provided corroboration at the cDNA level of the results obtained in the immunohistochemical assays. Overexpression of the MRP-I/CD') gene was not seen with either type of specimen. Ampli fication of the ÃŸ-actingene, used as internal PCR control, was MRP-1 observed in all instances (Fig. 3).

DISCUSSION 12345678 Several cDNAs encoding for human MRP-1/CD9 have been iso lated (3, 4). From the characterization of the clones and homology searches it was determined that MRP-1/CD9 belongs to the TM4SF, which comprises 15 members that are variously expressed on leuko cytes, a variety of mammalian tissues, and even on the surface of two types of parasites (5, 19). The precise biochemical functions of the TM4SF are not known, but current data largely suggest that they are involved in the regulation of cell development, proliferation, activa tion, and motility (5, 19). The melanoma-associated antigen ME491/ actin CD63 is an interesting molecule that also belongs to the TM4SF. It is strongly expressed in dysplastic nevi and radial growth-phase primary melanomas, but expression declines significantly, and sometimes dis 12345678 appears entirely, as the melanoma cells progress to more malignant Fig. 3. A, agarose gel electrophoresis of RT-PCR-amplified 800-bp MRP-1/CD9 stages such as vertical growth-phase primary melanoma and meta- cDNA. Lane I, size marker; Ltmr 2, breast cancer cell line ZR-75-30 (positive control); Ltincs 3, 5, and 7, primary invasive breastcarcinomas;Lanes 4, 6, and (V,their respective static melanoma (20). On the other hand, an examination of the melastatic lymph nodes. Note that MRP-1/CD9 cDNA is reduced or almost nondelectable relationship between metastatic potential and other TM4SF mole in the lymph node samples, ÃŸ,agarosegel electrophoresis of amplified ÃŸ-actinDNA cules, including CD37, CD53, ME491/CD63, TAPA-1/CD81, as well (internal PCR control) of each specimen. us MRP-1/CD9 suggested that only the latter was expressed predom inantly by primary, rather than by metastatic tumors (21). These findings are very consistent with our data that MRP-1/CD9 expression

Table 1 Relationship of MRP- I/CD9 iniinunoslaininfi ami various prognostic factors diminishes as the clinical stage of a given breast cancer advances. In addition, recently it was reported that an exciting gene (KAI-I) was a status(+)"44 metastasis suppressor gene for prostate cancer. At least MRP-I/CD9, CharacteristicAge valueNSNSNS0.045NSME491/CD63, and KAl-I among TM4SF may play important roles in at Surgery tumor progression and metastasis (22). Furthermore, because the ^50 yr (68.3%) disruption of cell-to-cell contacts induces metastasis of tumor cells, it <50yrEstrogen 7962 53(67.1%)41 22(27.8%)19(30.6%)4(5.1%)2 is of interest in the present context that anti-CD9 mAbs elicit an receptor status enhancement of Fc-independent heterotypic adhesion of pre-B-cell ER+ (66.1%) (3.3%) ER-Progestcrone 81117 56(69.1%)79 20(24.7%)32 5(6.2%)6(5.1%) lines to bone marrow stromal fibroblasts but not to bone marrow statusPgR+receptor strema (6). Thus, on the basis of our previous experimental finding (67.5%) (27.4%) that the anti-MRP-1/CD9 mAb inhibits cell motility of various types PgR-Nodal 2681 18(69.2%)61 7(26.9%)18 1(3.9%)2 of cancer cells (1, 2), it is conceivable that MRP-1/CD9 affects statusMIN2Tumor motility and invasiveness by modulating the level of cellular adhesion. (75.3%) (22.2%) (2.5%) 43 27 (62.8%) 13(30.2%) 3 (7.0%) Various types of human carcinomas were recently examined for the 1964 9(47.4%)44 8(42.1%)18(28.1%)2(10.5%)2(3.1%) presence of E-cadherin, a member of the family of calcium-dependent statusTI cell adhesion molecules (23). It was shown that E-cadherin immuno- (68.8%) reactivity was inversely related to tumor differentiation and metastasis 61 40 (65.6%) 17(27.9%) 4 (6.5%) 10 8 (X().0%) 2 (20.0%) 0 (0.0%) (24, 25), and that mutations of E-cadherin gene may contribute to the TÃ•Total 8143MRP-1/CD95(62.5%)97(Â±)17(26.6%)2(25.0%)39(-)3(5.1%)1(12.5%)7P development of the diffuse type of gastric carcinomas (18). It is number of patientsTotal64 conceivable that this may also apply to MRP-1/CD9 because we " +, positive immunostaining; Â±,reducedexpression; -, negative immunostaining. found a significant relationship between MRP-1/CD9 immunoreactivity and lymph node status, as well as lower protein and MRP-1/ CD9 mRNA levels in the metastatic lymph nodes than in the respec Table 2 Relationship between MRP-1/CD9 expression in primary tumors and in tive primary carcinomas of around 50% of the patients. These metastatic lymph nodes iimmunohistochemiial assay) observations suggest that low expression of the MRP-I/CD9 gene MRP-1/CD9 expression in metastatic lymph node may be a feature of some breast tumors with high metastatic potential. However, it must be noted that there were certain patients with primarytumor(+) (-)14(38.9%) lymph node mÃ©tastaseswhose primary tumors were immunohisto- 0(0%) chemically MRP-1/CD9 positive, even at an early stage. These dif (Â±) 0 (0%) 12(57.1%) 9(42.9%) 21 ferences with respect to MRP-1/CD9 expression point to several (-) 0 (0%) 0(0%) 5(100%) 5 Total(+)"22(61.1%)22(Â±) 26 14Total36 62 possible explanations: (a) the anti-MRP-1/CD9 mAb M31-15 cannot ' +, positive immunostaining; Â±reducedexpression; -, negative immunostaining. discriminate between a normal and a mutant protein; (b) a gene 4130

mutation could give rise to an altered protein with diminished normal of TP82. a monoclonal antibody against CD9 antigen. Eur. J. Biochem., 199: function; or (c) because MRP-1/CD9 has one /V-linked and several 347-354, mqj. 10. Ikeyama. S.. Koyama. M.. Yamaoko, M.. Sasada. R., and Miyake. M. Suppression of O-linked glycosylation sites, an aberrant glycosylation could result in cell motility and metastasis by transfection with human motility-related protein loss of the protein's function. These and other questions about the (MRP-I/CD9) DNA. J. Exp. Med., 177: 1231-1237. 1993. role, if any, of MRP-1/CD9 in solid tumors remain to be answered in 11. Tullberg, K. F., and Burger, M. M. Selection of BI6 melanoma cells with increased metastatic potential and low intercellular cohesion using Nucleopore filters. Invasion the near future. Although it has not yet been determined that MRP- & Metastasis, 5: 1-15. 1985. 1/CD9 is an adhesive molecule, our results suggest that its diminished 12. Miyake, M., Ito, M.. Hitomi, S., Ikeda. S.. Taki. T., Kurata. M., HiÃ±o.A., Miyake, N.. and Kannagi, R. Generation of two murine monoclonal antibodies that can discrim expression appears to be associated with mÃ©tastases,and that the inate A'-glycolyl and yV-acetyl neuraminie acid residues of GM2 ganglioside. Cancer protein may represent a potentially useful prognostic marker. How Res., 48: 6154-6160, 1988. ever, large-scale prospective and retrospective studies on various 13. Laemmli, U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lond.), 227: 680-685. 1970. types of cancers are needed to establish whether MRP-1/CD9 is 14. Towbin, H., Stachelin, T., and Gordon, J. Klcclrophorctic transfer of proteins from indeed of practical utility as a prognostic predictor. polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Nati. Acad. Sci. USA, 76: 4350-4354. 1979. 15. Chomczynski, P., and Sacchi, N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., lf>2: ACKNOWLEDGMENTS 156-159, 1987. 16. Nakajima-Iijima. S., llamada. H.. Reddy. P.. and Kakunaga, T. Molecular structure of We thank Dr. Musato Vagita for editorial help with the manuscript. the human cytoplasmic ÃŸ-actingene: interspeeies homology of sequences in the introns. Prix. Nati. Acad. Sci. USA, Â«2:6133-6137. 1985. 17. Scheck, A. C., Mehia, B. M., Beikman, M. K., and Shapiro. J. R. BCNU-resistant REFERENCES human glioma cells with overepresentation of chromosomes 7 and 22 demonstrate increased copy number and expression of platelet-derived growth factor genes. Genes 1. Miyakc, M., and Hukomori, S. A specific ecll surface glycoconjugate controlling cell Chromosomes Cancer, 8: 137-148, 1993. motility: evidence by functional monoclonal antibodies that inhibit cell motility and 18. Becker. K. F., Atkinson. M. J., Reich. U.. Becker, L, Nekarda, H., Siewert, J. R.. and tumor cell metastasis. Biochemistry, JO: 3328-3334, 1991. Hofler, H. E-cadherin gene mutations provide clues to diffuse type gastric carcino 2. Miyake. M.. Kovama. M.. Seno. M., and Ikeyama. S. Identification of the motility- mas. Cancer Res.. 54: 3845-3852. 1994. related protein (MRP-1), recognized by monoclonal antibody M31-15. which inhibits 19. Horejsi. V.. and Vlcek, C. Novel structually distinct family of leukocyte surface cell motilily. J. Exp. Med., 174: 1347-1354, 1991. glycoproteins including CD9. CD37, CD53 and CD63. FEBS Lett., 288: 1-4. 1991. 3. Boucheix, C., Benoit, P.. Frachet, P.. Billard. M., Worlhington. R. E., Gagnon. J., 20. Holla, H.. Ross, A. H., Huebner, K., Isobe, M.. Wendeborn, S., Chao, M. V., and Uzan. G. Molecular cloning of the CD9 antigen. J. Biol. Chem., 266; Ricciardi, R. P., Tsujimoto, Y., Croce, C. M., and Koprowski, H. Molecular cloning 117-122, 1991. and characterization of an antigen associated with early stages of melanoma tumor 4. Lanza, F.. Wolf, D., Fox, C. F., Kicffer, N., Seyer, J. M., Fried, V. A.. Coughlin, S. R., progression. Cancer Res., 48: 2955-2962, 1988. Phillips. D. R., and Jennings, L. K. cDNA cloning and expression of platelet 21. Si, Z.. and Hersey, P. Expression of the neuroglandular antigen and analogues in p24/CD9. J. Biol. Chem., 266: 10638-10645, 1991. melanoma. CD9 expression appears inversely related to metastatic potential of 5. Wright, M. D.. and Tomlinson. M. G. The ins and outs of the transmembrane 4 melanoma. Int. J. Cancer, 54: 37-43, 1993. superfamily. Immunol. Today, 15: 588-594. 1994. 22. Dong, J-T.. Lamb. P. W., Rinker-Schaeffer. C. W.. Vukanovic. J.. Ichikawa, T., 6. Masellis-Smith. A., and Shaw. A. R. E. CD9-regulatcd adhesion: anti-CD9 mono Isaacs, J. T., and Barrett, J. C. KAl-1, a metastasis suppressor gene for pulsiate cancer clonal antibody induce pre-B cell adhesion to bone marrow fibroblasts through de on human chromosome llpll.2. Science (Washington DC). 26Â«:884-886. 1995. novo recognition of fibronectin. J. Immunol., 152: 2768-2777, 1994. 23. Takeichi. M. Cadherin cell adhesion receptors as a morphogenctic regulator. Science 7. Griffith. I... Slupsky, J., Seehafer. J., Boshkov, L.. and Shaw. A. R. E. Platelel (Washington DC). 251: 1451-1455, 1991. activation by immobilized monoclonal antibody: evidence for a CD9 proximal signal. 24. Oka. H.. Shiozaki. H., Kobayashi, K., Inoue, M., Tahara, H., Kobayashi, T., Bood, 78: 1753-1759, 1991. Takatsuka, Y., Matsuyoshi. N., Hirano. S., Takeichi. M.. and Mori, T. Expression 8. Yatomi, Y., Ozaki, Y., Satoh. K., and Kume. S. Anti-CD9 monoclonal antibody of E-cadherin cell adhesion molecules in human breast cancer tissues and its elicits staurosporine inhibilable phosphatidylinositol 4.5-bisphosphate hydrolysis, relationship of metastasis. Cancer Res.. 5.Ã•:1696-1701, 1993. phosphutidyiinositol 3.4-bisphosphate synthesis, and protein-tyrosine phosphoryla- 25. Umbas, R., Isaacs. W. B., Bringuicr, P. P.. Schaafsma. H. E.. Karthaus, H. F. M., tion in human platelets. FEES Lett., 322: 285-290, 1993. Oosterhof, G. O. N.. Debruyne, F. M. J., and Schalken, J. A. Decreased E-cadherin 9. Ozaki, Y.. Matsumoto. Y.. Yatomi. Y.. Higashihara, M., and Kume. S. Two-step expression is associated with poor prognosis in patients with prostate cancer. Cancer mobilization of arachidonic acid in platelet activation induced by low concentrations Res., 54: 3929-3933. 1994.

4131

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1995 American Association for Cancer Research. Motility Related Protein 1 (MRP-1/CD9) Expression: Inverse Correlation with Metastases in Breast Cancer

Masayuki Miyake, Keiji Nakano, Yoshiaki Ieki, et al.

Cancer Res 1995;55:4127-4131.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/55/18/4127

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/55/18/4127. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.