Luteinizing Hormone/Human Chorionic Gonadotropin Receptors in Breast Cancer'

Home , Luteinizing hormone

ICANCER RESEARCH57, 857-864. March I, 19971 Luteinizing Hormone/Human Chorionic Gonadotropin Receptors in Breast Cancer'

Geri Meduri, Nathalie Charnaux, Hugues Loosfelt, AndrÃ©Jolivet,FrÃ©dÃ©riqueSpyratos,Sylvie Brailly, and Edwin Milgrom2

institut National de la Sante et de Ia Recherche MÃ©dicaleU135, Hormones et Reproduction, HÃ´pital BicÃªtre, 78 rue du GÃ©nÃ©ralLeclerc,94275 Le Kremlin BicÃªtre Cedex 1G. M., N. C., H. L, A. J., S. B., E. MI, and Centre RenÃ©Huguenin, 92210 Saint Cloud (F. 5.1. France

ABSTRACT related cancer markers that also includes carcinoembryonic antigen and a-fetoprotein. Recent studies have suggested that human choriogonadotropin (hCG), A stimulation by hCG of the growth of several tumor cell lines has in addition to its function in regulating steroidogenesis, may also play a been reported (I 1, 12, 16, 22). Kumar et a!. (12) have described the role as a growth factor. Immunocytochemistry using two different mono secretion of hCG by human lung cancer cells. Incubation with anti clonal antibodies (LHR29 and LHR1OSS) raised against the human lu teinizing hormone/human chorionic gonadotropin (LH/hCG) receptor al hCG antibodies or with specific antisense RNAs caused these cells to lowed us to detect this receptor in breast cancer cell lines (T47D, MCF7, lose their transformed phenotype. Furthermore, when transplanted and ZR75) in individual cancer biopsies and in benign breast lesions. The into nude mice, the tumors underwent necrosis following injection of receptor was also present in epithelial cells of normal human and sow anti-hCG antibodies. These observations have led to Phase I clinical breast. In the latter, its concentration increased after ovulation. trials of an anti-hCG vaccine in cancer patients (23). Very recently, The presence of LH/hCG receptor mRNA was confirmed by reverse Lunardi-Iskandar et a!. (24) showed that AIDS Kaposi syndrome transcription-PCR using primers extending over exons 2â€”4,5â€”11,and derived cells could be killed by hCGI3. The same authors also reported 9â€”11. cases of remission of Kaposi sarcoma in pregnant women. The proportion of LHIhCG-receptor positive cells and the intensity of Pregnancy and, especially, early pregnancy have been shown to be the immunolabeling varied in individual biopsies, but there was no obvi protective for breast cancer (25, 26). This observation led Russo et a!. ous correlation with the histological type of the cancer. (13, 14) to establish corresponding animal models. They have shown These results are compatible with previous studies suggesting that during pregnancy, hCG is involved In the differentiation of breast glan that in the rat, the susceptibility to breast neoplastic transformation by dular epithelium and that this hormone may play an inhibitory role in chemical carcinogenesis or radiation is decreased by the differentia mammary carcinogenesis and in the growth of breast tumors. tion of the gland in parous animals. The same result could be obtained by treatment with hCG (15). Because all of these observations were made in noncastrated animals, the mechanism of action of hCG INTRODUCTION remained unclear. However, Alvarado et a!. (27) showed that hCG could exert a direct antiproliferative effect on human breast epithelial LH3 is a pituitary hormone that regulates ovarian and testicular cells and on breast cancer cells in culture. This effect was possibly steroidogenesis. hCG is secreted by the placenta and also stimulates mediated through secretion of inhibin (28, 29). Because hCG pre ovarian steroid secretion. Both hormones act through the same spe vented both initiation and progression of mammary cancers, Russo cific receptor, which is coupled to proteins G and thus displays the suggested that this hormone might be used clinically to prevent the characteristic pattern of 7 transmembrane spans (reviews in Refs. 1 occurrence of this cancer in high-risk patients (30). and 2). The LH/hCG, follicle-stimulating hormone, and thyroid-stim A direct action of hCG and LH on normal breast and breast tumors ulating hormone receptors are characterized by a large NH2-terminal implies the existence of the corresponding receptors in these tissues. extracellular domain, which is the site of hormone binding (3â€”6). Furthermore, because most hormonal markers, including estrogen, However, recent evidence shows that in addition to their classical progesterone, EGF receptors, and so forth, are present in only a subset endocrine effects, hCG and LH may also be ectopically synthesized of the cancers (31â€”33),it appeared important to devise a method and exert paracrine effects regulating the growth of various cell types. allowing the detection of tumors containing LHIhCG receptors. We The role of these hormones in the occurrence and development of describe here the characterization of such receptors in normal breast various cancers has been extensively discussed (7â€”16). and in breast tumors and also report an immunocytochemical method Crystallization and X-ray diffraction studies have established the allowing the analysis of individual biopsies. three-dimensional structure of hCG (17). This structure shows this hormone to be a member of the structural superfamily of cystine knot MATERIALS AND METHODS growth factors, which also includes nerve growth factor, transforming growth factor-a, and platelet-derived growth factor-@3. Antibodies. The preparation of mouse monoclonal antihuman LH/hCG Synthesis of hCG@ by a variety of cancer cell lines of various origin receptor antibodies will be described in detail elsewhere.4 In summary, mice has been demonstrated (18â€”20).Concomitant expression of LHf3 has were given (100 @.tg/injection)arecombinant protein (human LHIhCGreceptor often been observed suggesting an activation of the whole hCGf3-LH@ amino acids 75â€”406 fused to ubiquitin) produced in Escherichia co/i. After gene cluster (21). hCG thus belongs to the group of embryonically three s.c. injections at 15-day intervals and an i.p. injection 3 days before the sacrifice, the splenocytes of the mouse were fused to myeloma Sp2 cells. Hybridoma clones were selected, expanded in ascites fluid and antibodies Received 8/5/96; accepted 1/4/97. The costs of publication of this article were defrayed in part by the payment of page purified on protein A-Sepharose. The specificity of the antibodies was verified charges. This article must therefore be hereby marked advertisement in accordance with by their ability to react with receptor-@-glucuronidaserecombinant fusion @ 18 U.S.C. Section 1734 solely to indicate this fact. protein to immunoprecipitate and immunopurify 251-hCG-receptorcomplexes t This work was partially supported by the Institut National de Ia Sante et de la prepared from L-cells transfected with an LH!hCG receptor expression vector. Recherche MÃ©dicale,the Association pour Ia Recherche sur Ic Cancer, the Ligue contre Ic Cancer, the Fondation pour Ia Recherche MÃ©dicale,andthe FacultÃ©deMÃ©decineParis-Sud. The expression vector for LH/hCG receptor was prepared and used for trans 2 To whom requests for reprints should be addressed. Phone: (33) 1 45-21-33-29; Fax: fection as described for similar studies with the follicle-stimulating hormone (33) 1 45-21-27-51. receptor vector (34). The antibodies were shown to immunostain cells trans 3 The abbreviations used are: LH, luteinizing hormone; hCG, human chorionic gona dotropin; ATCC, American Type Culture Collection; ER, estrogen receptor; PR. proges

terone receptor. 4 H. Loosfelt, A. Jolivet, and E. Milgrom, manuscript in preparation. 857

cancerInvadedPRPatientAgeMenopausal Table 1immunocytochemical study ofL1-I/hCG receptors in patients withbreast

scoredI32Pre351lnf. statusâ€•PT (mm)nodes (no.)Histological typeGradeER (fmol/mg protein)(fmol/mgprotein)Stained cells (%)5Stainingintensity'ICC duct.III29403224235Pre350Inf. duct.III225424338Pre400Inf. lob.II781003024445Pre150Inf. lob.114832136136547Pre303Inf. duct.II179129000648Pre100DIS0027I2750Pre600ID

papill.25281612859Post350lnf. duct.III24000961Post220Inf. duct.II16821464121067Post500tnf. duct.IIII21611I 39I01270Post600Inf.I69Post504Colloid101 lob.I115559361373Post180Inf. duct.II3521920001476Post707Inf. duct.III2291335241577Post4010Inf. duct.III2311320111678Post6021Inf. duct.II4230241783Post200lnf. duct.III58400001887Post373Inf. duct.III2132310199 1Post306Inf. duct.II140424024 a Pre, premenopausal; post, postmenopausal; PT, primary tumor size; inf. duct., infiltrating ductal carcinoma; inf. lob., infiltrating lobular carcinoma; DIS, ductal carcinoma in situ; ID @,apiIl.,intraductalpapillary carcinoma; colloid, colloid carcinoma. The percentage of cells stained by anti-LH/hCG receptor antibody was graded as follows: 0, <14%; 1, 15â€”24%;2, 25â€”64%;3. >65%. â€˜The staining intensity was divided into four categories: 0, negative; I, low; 2, medium; 3, high.

(I ICC, immunocytochemical. The final score is the product of the percentage grade multiplied by the intensity grade.

lesionsAssociationMenopausalwithTable 2immunocvtochemical study of LH/ICG receptors in normal breast and inbenign breast

breastStainedStainingPatientAgestatusâ€•Histologycarcinomacells scoreâ€•I23PreNormal (%)â€œintensityâ€•ICC breastNo7739225PreNormal breastNo7926328PreNormal breastNo7226436PreMild atypiaNo6813537PreMild hyperplasia without atypiaNo7313637PreMild hyperplasia without atypiaYes5836738PreMild hyperplasia without changeYes5324847PreBlunt hyperplasia; apocrine adenosisNo9226948PreMild duct changeYes5924105 hyperplasia; cystic changeNo6424I 1PostMild hyperplasia; cystic I65PostAdenosisYes35361265PostModerate epitheliosisYes6626a hyperplasia; ductal

See TableI , footnote a,for definition.

k â€˜

;,@

Fig. I. Use of anti-LH/hCG receptor monoclonal antibod ies for immunolabeling of breast tumors. A, invasive ducts) carcinoma (patient 16) immunostained with antibody LHR29. The intraductal malignant cells are labeled. X 100. I B, no labeling is visible on a serial section of the same tumor incubated with an unrelated control antibody (IDA 10) of the same subclass and at the same concentration. X 100. C, B immunolabeling of a serial section of the same ductal car C cinoma with antibody LHRIO55, which recognizes a differ ent epitope of the LH/hCG receptor. X 100. D, invasive lobular carcinoma (patient I2) immunostained with antibody LHR29; the malignant intralobular cells adjacent to the basement membrane are stained. X200. E, no immuno staining is found in the same region of the tumor using the same antibody previously incubated with recombinant LH/ hCG receptor ectodomain. X200. I-.

S @.1@ E

858

â€˜â€˜a

â€˜, @ Fig. 2. Immunocytochemical study of LH/hCG receptors in @5.&r @. @ human breast cancer. A. intraductal carcinoma (patient 6). Only :@ .!@4 @ the basal cells and the cells facing the luminal cavity (L) are S intensely stained. X450. B, highly cellular invasive ducts) carci noma (patient 17). Only strands of cells adjacent to the fibrovas cular septa are labeled. X 100. Bar, 100 ,sm. C. invasive lobular carcinoma (patient 3). All cells are labeled with variable staining intensity. x200. Bar, 50 pm. D, papillary carcinoma (patient 7). ,- The expression of LH/hCG receptors is more intense on the peripheral cells. X400. Bar, 10 pin. E, intraductal carcinoma (patient 6). The staining is localized to the plasma membrane. x 1420. Bar, 10 zrn. F, colloid carcinoma (patient I 1). The immunolabding is intense and mainly intracytoplasmic. and sometimes has a band-like perinuclear disposition. X2000. Bar, 10 pin. LHR29 antibody was used. â€˜@ __ @c

. â€˜S

I@1 lOttm . â€˜,@[email protected]

fected with an expression vector encoding LHIhCG receptor, whereas no Three samples of normal breast tissue were obtained during reductive reactionwasseenwithmock-transfectedcontrolcells.AntibodiesLHR29and surgery. The blocks of tissue were snap-frozen in liquid nitrogen and LHR1O55 (both of the IgGi subclass) were used in this study to identify preserved at â€”196Â°Cuntiluse. LHJhCG receptor-bearing cells. Porcine mammary tissue was obtained from four Large White sows of Antibodies against cytokeratin 19 (CK 19 clone BA 17; Dako, Carpinteria, reproductive age (Institut National de la Recherche Agricole, Nouzilly, CA), epithelialmembraneantigen(clone E 29; Dako), and a-smooth muscle France). The animals were treated with a hormonal cycle regulator (Regu actin (clone 1 A4; Sigma Chemical Co., St. Louis, MO) also were used to mate 228, Roussel Uclaf, Romanville, France). Two sows were in the late differentiate normal and neoplastic epithelial cells from myoepithelial cells in preovulatory phase (day 1 of the estrus cycle) and two were in the early the sections studied. IDA1Oantibody (35) was used as a non-receptor related luteal phase (day 5 of the estrus cycle). The length of the estrus cycle in the antibody control. sow is about 21 days. The tissue was excised immediately after sacrifice, Tissues. Breast cancer specimens were obtained from 19 patients un trimmed of excess fat, cut into blocks of 5 mm width, and snap-frozen in dergoing mastectomy or lumpectomy (Table 1). Samples of benign breast liquid nitrogen after embedding with tissue-Tek OTC embedding medium lesions were obtained from four patients with benign breast disease and (Miles Inc., Elckart, IL). The blocks were preserved at â€”196Â°Cuntil from five patients undergoing mastectomy for breast cancer (Table 2). furtheruse. 859

LHJ8CGRECEPTORSIN BREASTCANCER

subclass, used at the same concentrations (anti-common leukocyte antigen, Dako LC, clones PD 26 and 2B 1 1). .. â€¢. . a For the immunocytochemical scoring, at least 300 epithelial cells were p counted at high magnification in three different areas of each biopsy. These areas were assigned to four grades according to the percentage of cells stained (Table 1). The staining intensity also was assessed and subdivided into the following four grades: negative (grade 0), low (grade I), medium (grade 2), and high (grade 3). The final score was the product of the percentage grade I. (1*@. multiplied by the intensity grade (36). ERs and PRs were assayed on cytosolic extracts of the surgical specimens, B using Abbott enzyme immunoassay kits (ER-EIA and PR-EIA, Abbott Labo p @A' ratories, Abbott Park, IL). Briefly, the tumor material was homogenized and centrifuged at 150,000 X g for I h. ERs and PRs were assayed in the supernatant (cytosol). Beads coated with monoclonal anti-ER (H222) or pro gesterone (JZB39) antibodies were incubated overnight with aliquots of tumor cytosol or standard solutions of ER and PR. After washing, the beads were S. @% â€˜, â€¢ , incubated I h with horseradish peroxidase-conjugated monoclonal anti-ER @S (D547) or anti-PR (KD68) antibodies. The beads were washed, and then incubated for 30 mm with hydrogen peroxide and o-phenylenediamine. Ab sorbance was measured at 492 nm and steroid receptor concentration was derived from a standard curve. Proteins were assayed by Pierce protein assay (Pierce, Rockford, IL). Results were expressed in fmol/mg proteins. Cutoff Dwas RT-PCRofLH/hCGReceptormRNA.TotalRNAwasextractedwith set at 15 fmol/mg protein for both receptors. the RNA Flash T (Bioprobe) from human tissues collected at surgery, from Fig. 3. Immunocytochemical study of LH/hCG receptor in human breast cancer cell lines. A, permanently transfected L-cells uniformly expressing the LH/hCG receptor. T47D, ZR75, and MCF-7 cell lines and from an L-cell line transfected X200. B, T47D breast cancer cells. The receptor is found in only a fraction of the cells, permanently with an expression vector encoding the human LH/hCG receptor. and the staining is strong. X200. C, MCF7 breast cancer cells. The expression of the The cDNAs were synthesized using the first-strand cDNA kit from Clontech (4 receptor is more ubiquitous but the labeling is fainter than in the previous cell line. X200. 1.Lgof total RNA in a 20-pJ standard reaction). RNA of LHR recombinant D, ZR75 breast cancer cells. Only a small minority of the cells are stained. X200. LHR29 antibody was used. L-cells was diluted 100-fold with RNA extracted from mock-transfected L-cells. PCR was carried out (volume, 100 pA) with 5 p1 of the cDNA mixture and the following synthetic oligonucleotides (upstream/downstream primers): Breast Cancer Cell Lines. MCF7 (ATCC, Rockville, MD; HTB 22), nucleotides 170â€”190(exon2)1336â€”360(exon4), 836â€”858(exon9)11089â€” T47D (HTB 33), and ZR75â€”l(CRL1500)were used. The cells were cultured III1(exonII),and500â€”524(exon5)/1009â€”1031(exonI1;nucleotidesare in DMEM with 10% FCS, 2 mM glutamine, 2 mM sodium pyruvate and 400 numbered from the initiation ATG). After 35 cycles of amplification, DNA ;.d/lofinsulin(UmulineRapide,Lilly,France)inanincubatorat37Â°Cwith5% fragments (one-thirtieth of the PCR reaction) were electrophoresed in I% carbon dioxyde. Before immunohistochemistry,the cells were plated in Labtek agarose and analyzed by Southern blot using an internal 32P-labeled oligonu chamber slide culture chambers (Nunc Inc., Naperville, IL) at a concentration cleotide probe (l0@ cpm/ml; the probes corresponded to nucleotides 283â€”310, of l0@cells per chamber. Just before confluence (18 -24 h growth) the cells 1009â€”1031, and 836â€”858, respectively). Control experiments were performed were removed from the incubator and washed twice with PBS. The upper to assess (3-actin mRNA content in human tissue samples (not shown). structure of the chamber slides was removed, and the slides were fixed with cold (â€”20Â°C) acetone for 5 mm and then rehydrated in PBS prior to imniunostaining. RESULTS Immunohistochemistry. Six-g.srn-thickserialfrozensectionswerecut with an 0TM56 Bright cryomicrotome, air dried for 30 mm at room temperature, Specificity of LHIhCG Receptor Immunostaining in Human fixed in cold (â€”20Â°C)acetonefor 5 mm, and dried for 20 mm. The sections BreastTissue. Weusedforthesestudiesbothneoplasticandnormal were individually wrapped in cellulose and tinfoil and stored at â€”20Â°Cuntil breast tissue. Representative experiments are shown in Fig. I. Immu further use; the storage time never exceeded 3 months. Before immunohisto nolabeling of an invasive ductal carcinoma (Fig. 1A) and of an chemical procedures, the slides were left unwrapped to thaw for 10 mm at invasive lobular carcinoma (Fig. 1D) was observed using LHR29 room temperature and were then rehydrated in two successive baths of 10 mM monoclonal antibody. If we used a control unrelated monoclonal PBS, pH 7.2. The sections were then pretreated for 20 mm with preimmune antibody of the same IgGI subclass, there was no staining (Fig. 1B). sheep serum diluted 20-fold in PBS containing 5% BSA before incubation with A similar result was obtained if the primary antibody was omitted (not the anti-LHR primary antibodies. shown) or if it was previously incubated with recombinant LH/hCG Primary antibody incubation was performed overnight in a humid chamber at 4Â°C.The monoclonal antibodies LHR29 and LHRIO55 were used at a receptor ectodomain (Fig. lE). concentration of 10 @g/ml.Thebound immunoglobulins were subsequently Furthermore, we compared the staining obtained with antibody revealed with biotinylated antimouse secondary antibodies (Amersham Inter LHRIO55 with that observed with antibody LHR29. These two anti national, Bucks, England) that had been diluted 1:200 and incubated for 1 h in bodies bind additively to the antigen, showing that they recognize a humid chamber at room temperature, and then with streptavidin-biotin different epitopes. As shown in Fig. 1, A and C, the labeling observed peroxidase complexes (Amersham International) that had been diluted I:100 on consecutive sections was similar for the two antibodies. for 30 mm at room temperature. Endogenous peroxidases were inhibited by Immunocytochemical Study of LH/hCG Receptor in Breast incubation with 3% H,O, in PBS for 5 mm before addition of streptavidin Cancers. Intraductal carcinomas were frequently strongly immuno biotin complexes. Aminoethylcarbazole (Sigma Chemical Co.) was used as a labeled either on the majority of the cells or on both the basal layers chromogen. The sections were lightly counterstained with Meyer's hematox and the layers close to the lumen. In the latter case, the intermediary ylin and mounted with an aqueous medium (Glycergel, Dako). Replacement of the specific primary monoclonal antibody with preimmune mouse immuno layers of cells appeared devoid of LH/hCG receptor (Fig. 14). globulins (Sigma Chemical Co.) or IDA1Oantibody (35) at the same concen In ductal invasive cancers, the number of stained cells varied tration resulted in absence of epithelial staining, as did the replacement of the widely between carcinomas of the same histological type. A stronger primary antibodies with other unrelated monoclonal antibodies of the same staining intensity was often observed in the cells situated at the 860

LH/hCG RECEFFORS IN BREAST CANCER

. @: 5

$ S

100

.â€˜.

â€˜1@ @p',â€˜

@ Fig. 4. Immunocytochemical study of LHIhCG receptor in be S. .@ â€˜ nign breast lesions without atypia. A. adenosis (patient I 1). About ,., EL.' 30% of the epithelial cells of the hyperplastic lobule are stained â€˜I., (arrow). X 120. Bar, 100 ,[email protected], epitheliosis (patient 12). All of the @? benign proliferating epithelial cells lining the duct are stained. x 100.Bar,100jsm.C,apocrinechange(patient7).Themetaplas tic epithelial cells are lightly labeled; the staining is more evident on the basal cells. X400. D, blunt duct adenosis (patient 8). The epithelial cells lining the blunt ducts are uniformly stained. X 100. E, small cyst (patient 10). All of the epithelial cells are immuno stained. X800. Bar, 10 @sm.F,larger cyst showing a discontinuous epithelial layer (patient 9). Some of the epithelial cells are stained (arrows). X200. Bar, 50 sm. LHR29 antibody was used.

periphery of the tumoral formations. The most cellular carcinomas The great majority of tumoral cells were stained on the plasma frequently showed a particular staining pattern: positive cells were membrane (Fig. 2E). In only one case of colloid carcinoma, the arranged in strands adjacent to the fibrovascular septa (Fig. 2B). staining observed in some cells appeared very intense and mainly The majority of the cells of invasive lobular carcinomas were intracytoplasmic, sometimes forming a band parallel to the nucleus stained with variable intensity (Fig. 2C). Papillary carcinomas ap (Fig. 2F). Such a localization has been described for this family of peared heterogenous: some papillas were stained whereas other were receptors in cases in which immature precursors of the receptor not. The proportion of labeled cells was usually maximal at the base accumulate (5). of the papilla (Fig. 2D). Nineteen different biopsies were examined (7 premenopausal and 861

LH/hCG RECEPTORS IN BREAST CANCER

Fig. 5. Immunocytochemical study of LH/hCG re ( ceptor in normal human breast tissue. A, immuno r. staining of the epithelial cells of some acini and small terminal ducts (patient I). x240. Bar. 50 @sm.B.label ..\ ing of theepithelialcells lining a duct(patient3). X 300. Bar. 50 @tm.LHR29 antibody was used. I

5O@tm B

12 postmenopausal). There was no obvious relation between the Immunocytochemical Studies of LH/hCG Receptors in Benign LH/hCG receptor immunostaining score and the histological type, the Breast Lesions. Twelve different biopsies were analyzed. Seven size of the tumor, the presence and number of invaded lymph nodes, were taken from patients with benign breast disease, whereas 5 the content of ERs and PRs, or the histological grading (Table 1). consisted of nonmalignant dysplastic breast tissue obtained during Immunocytochemical Study of LH/hCG Receptors in Breast surgery for breast cancer. The lesions that were observed included foci Cancer Cell Lines. T47D, MCF7, and ZR75 breast cancer cell lines of adenosis, hyperplasia of ductal epithelium, foci of apocrine meta were studied. As a control, L-cells permanently transfected with an plasia of epithelial cells, and small cysts. expression vector encoding LH/hCG receptor also were examined. In foci of adenosis, about 30% of the cells were labeled by anti The latter were uniformly labeled (Fig. 3A) because all of the cells LH/hCG receptor antibodies (Fig. 4A). The hyperplastic ductal epi expressed the receptor. Only one-third of the T47D cells were in thelium was more intensively stained (Fig. 4B). In most cases, all of tensely labeled by the anti-LH/hCG receptor antibody (Fig. 3B). the hyperplastic layers were labeled, whereas in some cases, only the About two-thirds of MCF7 cells were stained less intensely (Fig. 3C). luminal and the basal layers were immunostained. Less than 1% of the ZR75 cells were immunolabeled: the staining was The foci of apocrine metaplasia were only focally and faintly stained intense (Fig. 3D). The immunolabeling was identical using antibodies (Fig. 4C). All of the epithelial cells oflesions of blunt duct adenosis were LHR29 and LHRIO55, whereas a nonrelated antibody (of the same IgG1 subclass) did not stain these cells (not shown). labeled (Fig. 4D). In cysts, there was a contrast between the intensely

@.. ,#

,%\ â€˜@ S @@ .â€˜@::I'. @.

Fig. 6. Immunocytochemical study of LH receptor in preovu latory and postovulatory sow breast. A. preovulatory breast sam pIe. A minority of the epithelial cells of the acini and small ducts ,@..: are labeled. X 150. Bar, 50 ,sm. B, postovulatory breast sample. All epithelial cells of acini and small ducts express the receptor. The labeling is uniformly distributed on the cell membranes. x 100. Bar, 100 @m.LHR38 antibody (37) was used. No labeling was observed when the antibody was replaced by a control unre Iated antibody of the same subclass (IgGI).

862

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1997 American Association for Cancer Research. LH/hCG RECEPTORS IN BREAST CANCER labeled cuboidal epithelium lining the smaller cysts (Fig. 4E) and the faintly stained (or negative) cells lining the larger cysts (Fig. 4F). Ovary Immunocytochemical Study of LH/hCG Receptors in Normal Human Breast Tissue. Observation of LH/hCG receptors in breast Testis cancers led us to ask whether their presence was due to the disease or whether they also existed in normal breast tissue. In normal human Breast breast, labeling was observed that increased from the acini to the small ducts (Fig. 5A) and to the large ducts (Fig. 5B). The proportion of Liver labeled cells varied from one lobule to another: some lobules were @1 more intensely stained than others. Only epithelial cells were immu nostained; no labeling was found on myoepithelial cells or fibroblasts. 2 In individual cells, there was no polarization of the receptor toward the apical or the basal pole. 3 Immunocytochemical Studies of LH/hCG Receptor in Pre- and Postovulatory Sow Breast. If LHJhCG receptors play an important 4 physiological role in breast function, they should be present in various Human 5 species. Furthermore, the use of an animal model should allow an breast easier analysis of a possible hormonal regulation of these receptors. 6 Because high-affinity antibodies for pig LH receptors were avail tumors able (37, 38), we studied the breast tissue of sows biopsied during 7 either the follicular or the luteal phase. A clear-cut difference was observed: during both phases, only the epithelial cells, not the stromal 8 or myoepithelial cells, were immunostained. During the follicular 9 stage, the ducts (Fig. 6A) were labeled, whereas a heterogeneity of labeling was observed between lobules and in the same lobule be 10 tween acini. Human After ovulation, during the luteal phase (Fig. 6B), a marked increase MCF7 in labeling was observed: all of the acini in all of the lobules were now breast immunostained. tumor I ZR75 These results indicate that the LH receptor in sow breast undergoes a hormonal regulation, suggesting either up-regulation by the LH peak cell lines LT47D or induction by progesterone. No labeling was observed whenever antibody LHR38 was replaced Transfected TLHR - with a control unrelated antibody of the same subclass at the same mouse LLHR+ concentration. LH/hCG Receptor mRNA in Human Breast. We used RT-PCR Lcells to study the presence of LHIhCG receptor mRNA in breast tissue. Fig. 7. LH/hCG receptor mRNA in various human tissues and in breast cancer. Total RNA was extracted from normal human tissues (ovaries, testes. breast, and liver), breast Three pairs of primers were used, spanning regions of the receptor tumor biopsies (from 10 patients), and breast tumor cell lines (T47D, MCF7, and ZR75). encoded by exons 2â€”4,5â€”11,and 9â€”11(the LH/hCG receptor gene Control experiments were performed using an L-cell line permanently expressing the contains I 1 exons; Ref. 39). Positive controls included mRNA pre recombinant LH/hCG receptor (LHR+) and with mock-transfected L-cells (LHRâ€”). cDNA synthesis and PCR were carried out as described in â€˜Materialsand Methods.â€•DNA pared from human ovaries, testes, and an L-cell line permanently fragments were analyzed by Southem blot using internal oligonucleotide probes (32P expressing human LH/hCG receptor (Fig. 7). Negative controls in labeled). Several PCR primers were used to detect different regions of the receptor mRNA (see â€œMaterialsand Methodsâ€•).The experiment shown here was performed using PCR cluded human liver and L cells not transfected with human LH/hCG primers specific for exons 2 and 4, which amplify a l90-bp DNA fragment. receptor. Hybridization was performed with internal 32P-labeled oh gonucleotide probes. Using this methodology, LHIhCG receptor mRNA was detected in vation of the presence of LH/hCG receptors in normal breast tissue. normal breast tissue, in 10 different breast cancers, and in MCF7, The receptors seem to undergo a hormonal regulation; their concen ZR75, and T47D breast cancer cell lines (Fig. 7). In all cases, the size tration is increased after ovulation. of the amplified DNA corresponded to that observed in ovaries and In breast cancer, the number of cells immunostained with anti-LW testes. hCG receptor antibodies and the intensity of the staining varied among individual cancers. There was no obvious relation to the DISCUSSION histological type. The receptor was also detected at various concen trations in different breast cancer cell lines. Further studies in a greater Resolution of the three-dimensional structure of hCG has shown number of patients will be necessary to determine whether the pres this hormone to be part of the structural superfamily of cystine knot ence and the concentration of LH/hCG receptors are correlated with growth factors (17). Indeed, hCG has been shown to exert either the aggressiveness of the tumor and the prognosis of the patients. The stimulatory ( 11, 12, 16, 22, 40) or inhibitory (24, 27, 41 , 42) effects preparation of high-affinity antibodies against the human LH/hCG on the growth of various cancer cell lines. receptor, able to immunostain the receptor even in tissues and cells In the breast, it has been proposed that hCG provokes differentia where its concentration is relatively low, will allow such studies. tion, which in turn renders the cells less susceptible to neoplastic The receptor present in breast tumors interacts with two different transformation (13, 14, 30). This mechanism would explain the de antibodies recognizing different epitopes. The RT-PCR method al creased occurrence of breast cancer in women who had an early lowed detection of the receptor messenger corresponding to exons pregnancy (25, 26). Such a hypothesis is compatible with the obser 2â€”4,5â€”11,9â€”11. This suggests that the receptor present in the breast 863

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1997 American Association for Cancer Research. LH/bCG RECEPTORS IN BREAST CANCER is similar to that expressed in the gonads; however, total sequencing with gene expression in cultured human cancer and fetal cells. Endocrinology, 136: would be necessary to definitively solve this question. 1034â€”1039, 1995. 22. Rivera. R. T., Pasion. S. G., and Wong. D. T. Loss of tumorigenic potential by human It is also unknown if the LHIhCG receptor is activated only by lung tumor cells in the presence of antisense RNA specific to the ectopically circulating hCG (and possibly LH) or if there exists a local secretion synthesized alpha subunit of human chorionic gonadotropin. J. Cell Biol., 108: of this hormone and thus paracrine regulations occurring through the 2423â€”2434,1989. 23. Synder, L. L., Woo, D. W., Triozzi, P. L., and Stevens, V. C. Synthetic hormone/ LH/hCG receptor. growth factor subunit vaccine with application to antifertility and cancer. In: M. F. Powell and M. J. Newman (eds.). Vaccine Design: The Subunit Approach, pp. 907â€”930.New York: Plenum Publishing Corp., 1995. REFERENCES 24. Lunardi-Iskandar, Y., Bryant, J. L., Zeman, R. A., Lam, V. H., Samaniego, F.. Besnier, J. M., Hermans, P., Thierry. A. R., Gill, P., and Gallo, R. C. Tumori I. Misrahi, M., Vu Hai, M. T.. Ghinea, N., Loosfelt, H., Meduri. G., Atger, M., Jolivet, genesis and metastasis of neoplastic Kaposi's sarcoma cell line in immunodefi A., Gross, B., Savouret, J. F., Deren, P., and Milgrom, E. Molecular and cellular cient mice blocked by a human pregnancy hormone. Nature (Lond.), 375: 64â€”68, biology of gonadotropin receptors. In: E. Y. Adashi and P. C. K. Leung (eds.), The 1995. Ovary, pp. 57â€”92.New York: Raven Press, 1993. 25. Mac-Mahon. B., Cole, P., Liu, M., Lowe, C. R., Mirra, A. P., Ravinihar, B., Salber, 2. Segaloff, D. L., and Ascoli, M. The lutropin-gonadotropin receptor 4 years later. E. G., Valaoras, V. G., and Yuasa, S. Age at first birth and breast cancer risk. Bull. Endocr. Rev., 14: 324â€”347, 1993. 3. Xie, J. B., Wang, H., and Segaloff, D. L. Extracellular domain of lutropin/choriogo WHO, 43: 209â€”221, I970. nadotropin receptor expressed in transfected cells binds choriogonadotropin with high 26. Kelsey, J. L., Gammon, M. D., and John, E. M. Reproductive factors and breast affinity. J. Biol. Chem., 265: 2141 1â€”21414,1990. cancer. Epidemiol. Rev.. 15: 36â€”47, 1993. 4. Nagayama, Y., Wadsworth, H. L., Chazenbalk, 0. D., Russo, D., Seto, P., and 27. Alvarado, M. V., Russo, J., Alvarado, N. E.. and Russo, I. H. Human chorionic Rapoport, B. Thyrotropin-luteinizing hormone/chorionic gonadotropin receptor cx gonadotropin inhibits proliferation and induces expression of inhibin in human breast tracellular domain chimeras as probes for thyrotropin receptor function. Proc. Natl. epithelial cells in vitro. In Vitro Cell & Dcv. Biol., 30A: 4â€”8.1994. Acad. Sci. USA, 88: 902â€”905,1991. 28. Alvarado, M. V., Russo, J., and Russo, I. H. Immunolocalization of inhibin in the 5. Vu Hal Luu Thi, M. T., Misrahi, M., Houllier. A., Jolivet, A., and Milgrom. E. mammary gland of rats treated with hCG. J. Histochem. Cytochem., 41: 29â€”34, Variant forms of the pig lutropin/choriogonadotropin receptor. Biochemistry, 31: 1993. 8377â€”8383,1992. 29. Alvarado, M. V., Ho, T. Y., Russo, J., and Russo, I. H. Human chorionic gonado 6. Tsai-Moms, C. H., Buczko, E., Wang, W., Xie, X. Z., and Dufau, M. L. Intronic tropin regulates the synthesis of inhibin in the ovary and mammary gland of rat. nature of the rat luteinizing hormone receptor subspecies with hormone binding Endocrine. 2: 1107â€”1114, 1994. activity. J. Biol. Chem., 265: 19385â€”19388, 1990. 30. Russo, J., and Russo, I. H. Hormonally induced differentiation: a novel approach to 7. Acevedo, H. F., Tong, J. Y., and Hartsock. R. J. Human chorionic gonadotropin @3 breast cancer prevention. J. Cell. Biochem.. 22 (Suppl.): 58â€”64. 1995. subunit gene expression in cultured human fetal and cancer cells of different types and 31. Wiuliff, J. L. Hormone and growth factor receptors. In: W. L. Donegan and J. S. origins. Cancer (Phila.), 76: 1467â€”1475,1995. Spratt (eds.), Cancer of the breast, Ed. 4, pp. 346â€”369.Philadelphia: W. B. Saunders 8. Moutzouris, G., Yannopoulos,D., Barbatis,C.. Zaharof, A., and Theodorou,C. Is Co.. 1995. f3-human chorionic gonadotropin production by transitional cell carcinoma of the 32. Stierer, M., Rosen, H.. Weber, R., Hanak, H., Spona. J., and TÃ¼chler,H. Immuno bladdera markerforaggressivediseaseandresistancetoradiotherapy?Br.J. Urol., histochemical and biochemical measurement of estrogen and progesterone receptors 72: 907â€”909,1993. in primary breast cancer. Correlations of histopathology and prognostic factors. Ann. 9. lIes, R. @3hCGexpression by bladder cancers. Br. J. Cancer, 65: 305, 1992. Surg., 218: 13â€”21,1993. 10. Nishimura, R.. Koizumi, T., Morisue, K., Yamanaka, N., Lalwani, R., Yoshimura, M., 33. Fekete, M., Wittliff, J.. and Schally. A. V. Characteristics and distribution of receptors Nakagawa. T., Shii, K., Hasegawa. K., and Baba, S. Expression and secretion of the for (DTRP6) luteinizing-hormone releasing hormone. somatostatin, epidermal growth p subunitofhumanchorionicgonadotropinbybladdercarcinomainvivoandinvitro. factors and sex steroids in 500 biopsy samples of human breast cancer. J. Clin. Lab. Cancer Rca., 55: 1479â€”1484, 1995. Anal., 3: 137â€”147,1989. 11. Gillou, D. J., lies, R. K., and Chard, T. The effects of @3-humanchotionicgonadotropin 34. Vannier, B., Loosfelt, H., Meduri, G., Pichon, C., and Milgrom, E. Anti-human FSH on the in vitro growth of bladder cancer cell lines. Br. J. Cancer, 73: 323â€”326, 1996. receptor monoclonal antibodies: immunochemical and immunocytochemical charac 12. Kumar, S., Talwar, G. P., and Biswas, D. K. Necrosis and inhibition of growth of terization of the receptor. Biochemistry, 35: 1358 â€”1366,1996. human lung tumor by anti-a human chorionic gonadotropin antibody. J. NatI. Cancer 35. Legrain. P., Juy, D., and Buttin, P. Rosette-forming cell assay for detection of Inst., 84: 42â€”47,1992. antibody synthesizing hybridomas. In: J. J. Langone and H. V. Vunakis (eds.), 13. Russo, I. H., Koszalka, M., and Russo, J. Protective effect of chorionic gonadotropin Methods in Enzymology: Immunochemical Techniques. Monoclonal Antibodies and on DMBA-induced mammary carcinogenesis. Br. J. Cancer, 62: 243â€”247,1990. General Immunoassay Methods, Vol. 92, pp. 175-182. New York: Academic Press 14. Russo, I. H., Koszalka, M., and Russo, J. Human chorionic gonadotropin and rat Inc., 1983. mammary cancer prevention. J. NatI. Cancer Inst., 82: 1286â€”1289, 1990. 36. Tesch, M., Shawa, A., and Henderson, R. Immunohistochemical determination of 15. Russo, I. H., Koszalka, M., and Russo, J. Comparative study of the influence of estrogen and progesterone receptor status in breast cancer. Am. J. Clin. Pathol., 99: pregnancy and hormonal treatment on mammary carcinogenesis. Br. J. Cancer, 64: 8â€”12,1992. 481â€”484,1991. 37. Vu Hal, M. T., Jolivet, A., Jallal, B., Salesse, R., Bidart. J. M., Houllier, A., 16. Muckherjee, K.. and Das. S. A placental glycoprotein, human chorionic gonadotropin, Guiochon-Mantel, A., Gamier, J., and Milgrom, E. Monoclonal antibodies against LH as a growth stimulant of murine tumor cells. IRCS Med. Sci., 12: 1101â€”1102,1984. (luteinizing hormone) receptor. Immunochemical characterization of the receptor. 17. Lapthorn. A. J., Harris, D. C.. Littlejohn, A., Lustbader, J. W., Canfield, R. E., Machin, K. J., Morgan, F. J., and lsaacs, N. W. Crystal structure of human chorionic Endocrinology,117:2090â€”2098,1990. gonadotropin. Nature (Land.), 369: 455-461, 1994. 38. Meduri, G., Vu Hai, M. T., Jolivet, A., and Milgrom, E. New functional zonation in 18. Acevedo, H. F., Krichevsky, A., Campbell-Acevedo, E. A., Galyon, J. C., Buffo, the ovary as shown by immunohistochemistry of luteinizing hormone receptor. M. J., and Hartsock, R. J. Flow cytometry method for the analysis of membrane Endocrinology, 131: 366â€”373, 1992. associated human chorionic gonadotropin, its subunits and fragments on human 39. Atger. M.. Misrahi, M., Sar, S., Le Hem, L., Dessen, P., and Milgrom, E. Structure cancer cells. Cancer (Phila.), 69: 1818â€”1828, 1992. of the human LH/CG receptor gene: unusual promoter and 5' noncoding regions. Mol. 19. Acevedo, H. F., Krichevsky, A.. Campbell-Acevedo, E. A., Galyon, J. C., Buffo, Cell. Endocrinol., 111: 113â€”123,1995. M. J., and Hartsock, R. J. Expression of membrane-associated human chorionic 40. Melmed, S., and Braunstein, G. D. Human chorionic gonadotropin stimulated pro gonadotropin, its subunits, and fragments by cultured human cancer cells. Cancer liferation of Nb2 rat lymphoma cells. J. Clin. Endocrinol. Metab., 56: 1068â€”1070, (Phila.), 69: 1829â€”1842. 1992. 1983. 20. Lazar, V., Diez, S. G., Laurent, A., Giovangrandi. Y., Radvanyi, F., Chopin, D.. 41 . Moviglia, G. A., and Russo, J. Antimitogenic effect of human chorionic gonadotropin Bidart, J. M., Bellet, D., and Vidaud, M. Expression of human chorionic gonadotropin (hCG) on rat mammary epithelial cells in culture. J. Cell Biol., 99: 1439a, 1984. 13subunitgenesinsuperficialandinvasivebladdercarcinomas.CancerRes.,55: 42. Alvarado, N., Alvarado, H. T., Russo, M. E., Russo, I. H., and Russo, J. Human 3735â€”3738, 1995. chorionic gonadotropin (hCG) inhibits proliferation and induces the expression of 21 . Krichevsky. A., Campbell-Acevedo. E. A.. Tong, J. Y., and Acevedo, H. F. Immu inhibin in human breast epithelial cells in vitro. Proc. Am. Assoc. Cancer Res., 33: nological detection of membrane-associated human luteinizing hormone correlates 280, 1992.

864

Geri Meduri, Nathalie Charnaux, Hugues Loosfelt, et al.

Cancer Res 1997;57:857-864.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/57/5/857

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/57/5/857. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.