Hk2) in the Breast Cancer Cell Line T47.D1

[CANCERRESEARCH57,2651â€”2656.July1, 19971 Expression of Human Prostate-specific Glandular Kallikrein Protein (hK2) in the Breast Cancer Cell Line T47.D1

Ming-Li Hsieh, M. Cristine Charlesworth, Marcia Goodmanson, Shaobo Zhang, Thomas Seay, George G. Klee, Donald J. Tindall, and Charles Y. F. Young@

Departments of Urology (M-L H., S. Z, T. S., D. J. T., C. Y. F. Y.], Biochemistry and Molecular Biology [M. C. C., D. J. T., C. Y. F. Y.j, and Laboratory Medicine IM. G.. G. G. K.], Mayo Graduate Schools, Mayo Clinic/Foundation, Rochester, Minnesota 55905

ABSTRACT whether PSA plays a physiological role in these tissues in women remains to be elucidated. Human glandular kallikrein (hK2) protein, like prostate-specific anti The hK2 is another member of the human kallikrein gene family gen (PSA), is produced mainly In prostatic epithellum. It may be useful as (16). The hK2 gene was first discovered in 1987 (17). However, its a new diagnostic indicator for prostate cancer. Recently, a number of hK2-speciflc monoclonal antibodies have been developed that enable us to complete cDNA was not isolated until 1992 (18). Studies of hK2 detect hK2 protein In human prostate tissue, seminal fluid, and scm. mRNA indicate that hK2 is specific for prostatic tissue and is regu Whether hK2 can be expressed, like PSA, in nonprostatic cells is not lated by androgens via the androgen receptor (18â€”20).To define the known. In this study, we have characterized the presence of hK2 in an biological role and potential clinical utility of hK2 protein for prostate androgen-responsive breast cancer cell line T47-D at both the protein and cancer, recombinant hK2 proteins and hK2 peptides have been syn mRNA levels with an immunoassay, Western blot analysis, Northern blot thesized and used to generate monospecific antibodies (mAbs; Ref. analysis, and the reverse tmanscription-PCR. 21). A mAb HK1AS23, specific for hK2, has been used to detect hK2 Using a sensitive lmmunoassay with monoclonal antibodies to hK2, we protein in spent media from the androgen-dependent prostate cancer found that T47-D cells could be induced with androgens, minemalocorti cell line, LNCaP, and in human prostate tissues by Western blot coids, glucoeerticoids, and progestins to produce significantly more hK2 analysis and immunohistochemistry (22, 23). More recently, other than PSA. Estrogens failed to mimic the effect of the other steroids, blocking instead the stimulatory effect of androgens. Androgen induction high affinity mAbs have been developed for detecting hK2 protein in of hK2 in T47-D cells was dose dependent. More interestingly, we found serum (24â€”26). that the hK2 in androgen-induced T47-D cell spent media appears to be The similarities of PSA and hK2 suggest that hK2 might also be the pro-form of hK2 rather than mature hK2. produced by cells other than those of prostatic origin. We demonstrate Our study demonstrates that hK2, a serine protease thought to be in the current work that hK2 can be produced at a higher level than found only in prostate-related tissues and fluids, is also produced in a PSA by a breast cancer cell line T47-D under androgen stimulation. breast cancer cell line T47-D after steroid stimulation. This finding sug This is the first study to our knowledge that demonstrates the produc gests that hK2 may have a potential role in breast cancer as well as tion of hK2 protein by nonprostate cells. This finding further under prostatic cancer and will be the Impetus for further studies of hK2 scores the similarities between breast and prostate epithelial tissue and distribution and function. may extend the potential applications of hK2 as a biomarker for breast tumor in addition to that of prostate cancer. INTRODUCTION

PSA3 is a kallikrein-like renne protease that was thought to be MATERIALS AND METHODS exclusively produced by epithelial cells lining the acini and ducts of the prostate gland (1â€”3).Because of its tissue specificity, PSA has Materials. mAbs were prepared as described by Grauer et a!. (22) and been widely used as a marker for diagnosis and monitoring of prostate Kumar et a!. (25). HK1AS23, HK1H449, and HK1H599 are specific for hK2, and HK1H464 is specific for the pro-region of phK2. The pro-region of phK2 cancer because the introduction of PSA testing in 1986 (4, 5). How contains a seven-amino acid extension at the NH2 terminus of mature hK2 (or ever, the tissue specificity has been questioned in several recent hK2). Recombinant hK2 and phK2 were expressed and purified using the reports. A number of studies have demonstrated the presence of PSA adenovirus-induced AV12 hamster tumor cell line (25). PSA was purified from in the periurethral and perianal glands (6â€”8).It has been demon seminal fluid according to the procedure of Sensabaugh and Blake (27) and stated that the steroid hormone receptor-positive breast carcinoma provided by Dr. Roger Sokoloff (Hybritech, Inc., San Diego, CA). cell line T47-D can be stimulated by androgens, progestins, miner Cell Cultures and Steroid Treatments. Several human cell lines (ob alocorticoids, and glucocorticoids to produce PSA (9). More recently, tamed from American Type Culture Collection, Rockville, MD) including using a newly developed ultrasensitive PSA assay, it has been re T47D (breast), ZR75â€”l(breast), Hs5787 (breast), BT-20 (breast), MCF-7 ported that 30% of female breast tumor cytosolic extracts contained (breast), OVCAR-3 (ovary), Hep-G2 (liver), HT-27 (colon), and LNCaP PSA immunoreactivity, accounting for greater than 0.03 ng/mg of (prostate) were propagated in Coming 24-well culture dishes or Tl75 culture total protein (10, 11). Also, a study has shown that PSA may possibly flasks with RPM! 1640 supplemented with 5% FCS at 37Â°Cand 5% CO2. After reaching approximately 80% confluency, cells were incubated in phenol be used as a prognostic marker for breast cancer (12). Moreover, low red-free, serum-free RPMI 1640 for 24 h to deplete undesired steroids prior to levels of PSA were found in normal breast, normal endometrial tissue, experiments. milk of lactating women, and amniotic fluid (13â€”15). However, The initial stimulation experiment was performed by adding 11 different steroids to androgen receptor-positive T47-D cells. Ethanol was used as a Received 1/1(997;accepted 4/29/97. solvent to dissolve the steroids. Equivalent amounts of solvent were added to Thecostsof publicationofthisarticleweredefrayedinpartby thepaymentofpage control wells. Steroids were replenished every 24 h. After a 4-day incubation, charges. This article must therefore be hereby marked advertisement in accordance with spent medium was harvested for PSA and hK2 quantitation by an immuno 18 U.S.C. Section 1734 solely to indicate this fact. I Supported in part by the NIH Grants CA70892 and DK41995. metric assay as described below. Cell density was assayed by incubating cells 2 To whom requests for reprints should be addressed, at Mayo Clinic, 200 First Street, with 150 @.dof3-[4,5-dimethylthiazol-2-ylj-2,5-diphenyl-tetrazolium bromide SW, Guggenheim Building, l7-1742B, Rochester, MN 55905. (Sigma Chemical Co., St. Louis, MO; 5 mg/mI in PBS) in I ml of RPMI 1640 3 The abbreviations used are: PSA, prostate-specific antigen; mAb, monoclonal anti per well at 37Â°Cfor4 h. The insoluble product of 3-[4,5-dimethylthiazol-2- body; hK2, human glandular kallikrein; phK2, prohK2; DHT, dihydrotestosterone; RT PCR, reverse transcription-PCR; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; yll-2,5-diphenyl-tetrazolium bromide retained in the cells was then dissolved ER, estrogen receptor; AR. androgen receptor. in DMSO (Sigma) for absorbance measurement at 570 nm. The concentrations 2651

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1997 American Association for Cancer Research. EXPRESSIONOFhK2 IN BREASTCANCERCELLS of PSA and hK2 were normalized by the above cell density measurements and conditions for each cycle were as follows: denaturation at 94Â°Cfor 60 s, expressed as nanograms/milliliter/A570. For dose effect, the same procedure annealing at 58Â°C (for PSA) or 60Â°C (for hK2) for 90 s, and DNA polymer was also performed on T47D cells after treatment with various DHT concen ization at 72Â°Cfor90 s. After PCR cycling, the reactions underwent DNA trations from 0.2 to 1000 nM. Similarly, the time course effects of 200 nM DHT extension at 72Â°Cfor 10 mm. An aliquot of PCR reaction product was used for stimulation on expression of PSA and hK2 were also examined. 1% agarose gel electrophoresis. PCR DNA products in agarose gel were To test whether different human cell lines as listed above can produce PSA visualized by ethidium bromide staining. DNA bands corresponding to PSA and hK2 proteins, these cells were incubated in the presence or absence of a and hK2 were eluted from the gel, directly subcloned into pCR-II plasmid nonmetabolizable steroid, mibolerone, for 7 days. Finally, experiments were (Invitrogen, San Diego, CA), and transformed into Escherichia coli to obtain conducted to test whether hK2 induction by androgens in T47D cells can be enough DNA for sequencing. blocked by estrogens. The experiments were performed by stimulating the cells Northern Blot Analysis. Steroid-depleted T47-D cells were cultured in with 50 nM DHT in the presence of different concentrations of estradiol. The serum-free RPM! 1640 with the addition of DHT (100 nM) for 28 h. Total RNA spent media were harvested 4 days later. was extracted by the acidic phenol-chloroform-guanidinium thiocyanate Immunoassays for PSA and hK2 Quantitation. PSA levels in culture method (18). Equal amounts of RNA (30 j.&gflane)werefractionated in the supernatants were determined by an immunoenzymatic assay using the Tan presence of ethidium bromide by denaturing gel electrophoresis and trans dem-E PSA kit (Hybritech, Inc., San Diego, CA). Cross-reactivity ofthis assay ferred to a Zeta Probe membrane (Bio-Rad, Richmond, CA). The amount of with hK2 is negligible. The levels of hK2 were measured by an immunometric RNA used was quantified by spectrophotometric assay at wavelength 260 @tm sequential (sandwich) assay with two monoclonal hK2 antibodies, HK1H449 in H2O. The Zeta Probe membranes were hybridized with the 32P-labeled hK2 and HKI H599. The cross-reactivity of this assay with PSA is also negligible. or PSA probe prepared as described previously (18) and washed sequentially HK1H449 (the capture antibody) was coated on quarter-inch polystyrene beads at 50Â°Cfor 5 mm with 50 mM NaH2PO4 (pH 7.2), 5% SDS, and 1 mM EDTA; (Clifton Plastics) at I @gofantibody per bead, and HK1H599 (the detection and with 50 mM NaH2PO4 (pH 7.2), 1% SDS, and 1 mM EDTA. The GAPDH antibody) was prelabeled with acridinium ester. Recombinant hK2 (2500 cDNA labeled with 32Pby random primer labeling was used for normalization ng/ml, kindly provided by Steve Mikolajczyk, Hybritech, Inc.) was serially of hK2 and PSA mRNAs. Autoradiographs were obtained by exposing the diluted into assay diluent and used as an antigen in the standard curve. The membrane for 20 h at â€”70Â°C. capture antibody on the beads was incubated with the analyte in the sample or standard hK2 solution for 2 h at 37Â°C.Any unbound antigen was then washed RESULTS away. The detecting antibody labeled with acridinium ester was then added to the bead and incubated for 2 h at 37Â°C.Washing removed any unreacted excess The effect of various steroids on hK2 and PSA production by the antibody, leaving the antigen sandwiched between the two mAbs. The chemi breast cancer cell line T47-D is shown in Fig. 1A. After 4 days of luminescent signal (expressed as relative light units) was then detected in a treatment, androgens, progestins, glucocorticoids, and mineralocorti Magic lite Analyzer II. The unknown sample hK2 concentrations were calcu coids were able to induce significant PSA and hK2 expression, lated from the standard curve derived from linear regression of the relative light unit data. whereas estrogens and an antiestrogen, tamoxifen, were not stimula Western Blot Analysis. For direct immunoblottingof hK2 and PSA in tory. The expression of hK2 paralleled that of PSA. Thus, both PSA spent media from androgen-stimulated T47-D cells, mAbs HKIA523 and a and hK2 proteins are inducible by a number of different steroids in polyclonal PSA antibody (provided by Dr. Judith Finlay, Hybritech, Inc.) for this cell line. hK2 and PSA, respectively, were used as primary antibodies. Serum-free, Fig. 2 shows that DHT promoted the secretion of hK2 in a dose phenol red-free RPMI 1640 media were harvested from the same number of dependent manner in T47-D cells. The hK2 was stimulated by as little T47-D cells incubated with or without mibolerone for 7 days. Media were as 0.2 nM DHT, which suggests that the response is mediated via the concentrated 20-fold by ultrafiltration using Centriplus-3 concentrators (Ami androgen receptor. PSA induction was also stimulated by 0.2 flM con, Beverly, MA). Aliquots of concentrated media were subjected to SDS DHT, but the total amount of PSA protein induced was lower than that PAGE on 12% Tris-glycine gels. Protein markers (molecular weight range, 2,530â€”46,000;AmershamCorp., Arlington Heights, IL) were included. Sep of hK2 protein, which is consistent with results displayed in Fig. 1. arated proteins were electro-transferred to nitrocellulose membranes (Bio-Rad The time course of DHT stimulation of hK2 and PSA proteins was Laboratories, Hercules, CA). After blocking with 5% nonfat dry milk in TBST next examined. The levels of PSA and hK2 secreted into the culture buffer (20 mM Tris-HC1, 137 mr@iNaC1,and 0.1 % Tween 20, pH 7.6), the blots media were significantly increased within 24 h after the initial stim were incubated with mAb HK1A523 at 1:5000 dilution or PSA polyclonal ulation and continued to increase with time (data not shown). Again, antibody at 1:2000 dilution for I h at room temperature. The blots were washed more hK2 than PSA accumulated in the media with time. with TBST buffer and incubated with horseradish peroxidase-conjugated anti Several human cancer cell lines such as breast cells (MCF-7, mouse or anti-rabbit immunoglobulin (Amersham) at 1: 10,000 dilution for 1 h Hs5758, ZR75â€”l, and BT2O), liver cells (Hep-G2), ovarian cells at room temperature. Following three washes, the blots were incubated for I (OVCAR-3), and colon cells (HT27) were tested for expression of mm with enhanced chemiluminescence reagents (ECL; Amersham) and cx hK2 and PSA in the presence or absence of androgen stimulation. posed to X-ray film for detection of immunoreactive protein bands. The None of these cells produced significant amounts of PSA or hK2, membrane probed with HK1A523 antibody was then stripped and reprobed with phK2-specific HK1H464 antibody. regardless of the presence or absence of androgens. Interestingly, it RT-PCR. Total RNA was extracted from mibolerone-stimulatedT47D was found that in LNCaP cells (an androgen-sensitive prostate cell cells. The cultured cells were harvested and centrifuged at 1000 X g at 4Â°Cfor line) the level of secreted PSA was higher than that of hK2 (Fig. 1B). 15 mm. Total RNA was isolated by using acidic phenol-chloroform-guani Thus, although the absolute amount of secreted hK2 is similar in dinium thiocyanate method (1 8). Isolated RNA was then treated with RNase LNCaP and T47-D cells, PSA secretion differs widely between the free DNase I to remove contaminating genomic DNA. two cell types. For RT-PCR, an aliquot (I @g)ofRNA was used for the first-strand cDNA Estrogens did not induce hK2 or PSA secretion in T47-D cells (Fig. synthesis (25 p.1 of final total reaction volume) using either PSA-specific 1). To determine whether estrogens could block the effect of andro (5'-TCATCTCTGTATCC-3') or hK2-specific (5'-CATACACCTGTGTC-3') gens, T47-D cells were treated with both estradiol and DHT simulta primer ( 100 pmol) and Moloney leukemia virus reverse transcriptase (Life neously. It was found that estradiol could decrease the stimulatory Technologies, Inc., Gaithersburg, MD). The first-strand cDNA was then am plified with a pair of 5' and 3' primers (50 pmol each) for PSA [PSA-l effect of DHT in a dose-dependent manner (Fig. 3). (5'-GATGACTCCAGCCACGACCT-3') and PSA-2 (5'-CACAG ACAC To further characterize the molecular size of hK2 in T47-D cells, a CCCATCCTATC-3')] or hK2 [hK2-l (5'-GGTGGCTGTGTACAGTCATG total of 40 ml of spent media from T47-D cells incubated with or GAT-3') and hK2-2 (5'-CAGAAAGCACAGGTCAGTAGGCA-3')] using without 10 flMmibolerone was collected and concentrated to a volume Taq polymerase. Thirty-five to 40 cycles of PCR were performed. PCR of 2 ml for each sample. An equal volume of each sample was then 2652

200@ DhK2 DhK2 U PSA R PSA Fig. I. A, production of PSA and hK2 by T47-D cells after stimulation by various steroids. T47-D cells cultured in serum-free RPMI 1640 were treated with various steroids for 4 days. The spent media were harvested 150@ for hK2 and PSA quantitative analysis by immunoassay. Mib, mibo

0 lerone, 3.2 nM;DYT, 200 nM;1', testosterone, 200 nM;Ald, aldosterone, 1'- â€˜A I psi; Estrone, I @avi;Dex,dexamethasone, I gssi; HC, hydrocortisone. I p.ti; HP, I l-@-hydroxy-progesterone, I psi; Tam, tamoxifen, I ,.u,i; TAA, 100 triamcinolone acetonide, I psi. B, production of PSA and hK2 in breast cancer cell line (T47-D) and prostate cancer cell line (LNCaP). Cell lines @ II were incubated in serum-free media in the presence (+) or absence (â€”) of a nonmetabolizable steroid, mibolerone, for 7 days. The concentration 50â€¢ of mibolerone was 3.2 nMfor T47-D cells and I nsi for LNCaP cells. Bars, SE.

0@ . + zero Mib DHT T AId E2 Estrone Dcx HC HP Tam TAA T47-D LNCaP

subjected to SDS-PAGE under reducing conditions and blotted onto a 100- @hK2 nitrocellulose membrane. Using a highly specific hK2 antibody, U PSA HK1A523, a major immunoreactive band of Mr 34,00035,000 ac companied by a minor band was detected (Fig. 4A). These immuno 80 reactive protein bands were androgen inducible. The hK2 protein from T47-D cells seems, unexpectedly, to be larger than the mature form of recombinant hK2, which was used as a control (Fig. 4A). This could be due to a precursor or zymogen form of hK2 protein, an unproc E en essed, secretory form of hK2 (phK2). To test the hypothesis, the mAb a HK1H464, specific to the pro-segment of hK2 protein, was used for reprobing the membrane. Indeed, the major protein band was recog nized by the phK2-specific mAb and comigrated with recombinant phK2 protein (Fig. 4A). The minor band was also detected by the phK2 antibody. PSA was also detected in androgen-induced spent media of T47-D cells by a PSA-specific polyclonal antibody by 0.1 5 10 Western blot analysis (Fig. 4B). Its size was slightly larger than that of purified PSA from seminal fluid. Whether this is due to the Estradiol (@.tM) pro-form of PSA remains to be determined. Fig. 3. Inhibition of the stimulatory effect of androgens by estradiol in T47-D cells. To demonstrate that T47-D cells produce hK2 mRNA, total RNA T47-D cells in serum-free media were incubated with 50 n@,iDHT and different concen U'ationsof estradiol. The spent media were harvested 4 days after stimulation. Horiwntal was subjected to RT-PCR using primers specific for either the hK2 or axis. estradiol concentrations. Bars, SE. the PSA transcript. As seen in Fig. 5A, androgen-induced T47-D cells did express transcripts for not only hK2 but also PSA. The identities of the PCR products shown in the figure were verified by DNA sequencing. Moreover, a Northern blot (Fig. SB) shows that the steady-state levels of both hK2 and PSA mRNAs in T47-D cells were stimulated by androgens but not by estrogens. The molecular size of 60@ .1. 0 hK2 both hK2 and PSA mRNAs in T47-D cells is approximately 1.6 kb in . PSA length, which is about the same size as both mRNAs detected in prostate cells (18). The above results demonstrate that androgens up-regulate the expression of hK2 and PSA genes at both protein and 40. 1 I mRNA levels in T47-D cells. C -I DISCUSSION

Early efforts to identify the hK2 protein proved difficult due to its I:20 similarity to PSA. The protein sequence for hK2 deduced from the cDNA shares 80% sequence homology with PSA (17). However, specific hK2 mAbs have been developed that have minimal cross reactivity to PSA (21â€”26).The hK2 protein has been identified in both illi LNCaP cells and sera of patients with prostate cancer and, therefore, 0P .2 2 20 200 1000 may be a useful marker for prostate cancer (22â€”24).This report DHT (nM) demonstrates that hK2 is present in breast cancer cells and regulated by steroid hormones. To our knowledge, this is the first report of hK2 Fig. 2. PSA and hK2 production by T47-D cells in response to DHT. T47-D cells cultured in serum-free media were stimulated with DHT (0.2-1000 nM) The spent media protein expression in cells of nonprostate origin. The Mr 33@000 were collected 4 days poststimulation. Bars, SE. protein was identified as hK2 in two ways: (a) by its strong reactivity 2653

EXPRESSION OF hK2 IN BREAST CANCER CELLS A mAb HK1A523 mAb HK1H464 A 1234 1234

46K@ 46K

30K -@ 30K

21.5K@ 21.5K 1018

14.3K@ 14.3Krn

123 .@ B46K- -@ 506

30K- M PSA hK2 (7lObp) (819bp) 21.5K@

@ 14.3K -@ B

Fig. 4. Westem blot analysis of T47-D cell spent media for hK2 and PSA immuno 0 DHTE2 reactivity. Proteins in concentrated spent media were separated by SDS-PAGE (12% polyacrylamide) and blotted onto nitrocellulose. The blots were incubated with anti-hK2 mAb HK1A523 (A) or anti-PSA polyclonal antibody (B). Secondary antibody was horseradish peroxidase-labeled anti-mouse or anti-rabbit IgG, and detection was done by enhanced chemiluminescence. The membrane blotted with HKIA 523 antibody was then hK2@ stripped and reprobed with anti-phK2 mAb HK1H464 antibody. A: Lane I. mature form of recombinant hK2 (50 ng); Lane 2. pro-form of recombinant hK2 (50 ng); Lane 3, concentrated androgen-induced T47-D cell spent media; Lane 4, concentrated T47-D cell spent media without androgen stimulation. B: Lane I. purified PSA (8 ng); Lane 2, concentrated androgen-induced T47-D cell spent media; Lane 3, concentrated T47-D cell spent media without androgen stimulation. PSA with the anti-hK2 mAb HK1A523 in Western blotting; and (b) by its detection in a specific sandwich immunoassay using two other hK2 mAbs, HK1H449 and HK1H599. In addition, with the RT-PCR and Northern blot analysis, we also demonstrate the presence of hK2 mRNA in cytosols of T47-D cells, which is also up-regulated by androgens. GAPDH We chose to examine the breast cancer cell line T47-D for produc tion of hK2, because it had been reported that this cell line could produce significant amounts of PSA (9). We have shown that secre tion of both hK2 and PSA proteins are stimulated by androgens, Fig. 5. A, RT-PCR detection of PSA and hK2 mRNAs in mibolerone-stimulated T47-D cells. Total RNAs were extracted from mibolerone-stimulated T47-D cells. An aliquot (I progestins, glucocorticoids, and mineralocorticoids, whereas estro @Lg)ofRNA was used for the first-strand cDNA synthesis using a primer for either PSA gens are not active. From dose-response and time course experiments, or hK2 and Moloney leukemia virus transcriptase. The first-strand cDNA was then we have found that both proteins have similar secretion patterns in the amplified with a pair of 5' and 3' primers for PSA and hK2. The amplified PCR DNA fragments for PSA and hK2 were excised from electrophoresed agarose gel and subcloned breast cancer cell line T47-D. These observations suggest that the into PCR II vector for DNA sequencing. The results of partial sequencing proved that the mechanism of hK2 gene expression in T47-D cells may be similar to PCRDNAswerederivedfrommRNAsofPSAandhK2genes.Al.I kbDNAladder(Life Technologies, Inc., Gaithersburg, MD). B, effects of androgens and estrogens on steady PSA gene expression. state levels of hK2 and PSA in T47-D cells. After 24 h of steroid depletion, cells were It is known that the induction of gene regulation by glucocorticoids, treated as follows: Lane 0, no treatment; Lane DHT, 100ns@DHT;and E2, 1 @u,iestradiol. androgens, progestins, and mineralocorticoids is mediated by a com Total RNA was extracted at 28 h, and Northern blot analysis was carried out as described in â€œMaterialsandMethods.â€•mRNAsfor hK2, PSA, and the housekeeping gene, GAPDH, mon response element (GRE/PRE/ARE) in the target DNA (28). The in T47-D cells were detected with 32P-labeled hK2, PSA, or GAPDH cDNA probe on the glucocorticoid, mineralocorticoid, progesterone, or androgen recep same membrane. 2654

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1997 American Association for Cancer Research. EXPRESSIONOFhK2 IN BREASTCANCERCELLS tor-hormone complex could interact with the GRE/PREIARE, which dykinin production. Therefore, these studies, together with our finding regulates the hK2 or PSA expression. On the other hand, the ER of hK2 in this steroid hormone receptor-positive, nonprostatic origin complex cannot bind to this element and, if this hypothesis is true, cell line, suggest that, like PSA, hK2 may be involved in growth would have no effect on either PSA or hK2 gene regulation. However, control and/or other cellular processes in not only prostate but other the effects of estradiol in inhibiting the androgen stimulation of hK2 AR-positive tissues, although final elucidation of the role of hK2 might be mediated by modulating the transactivation activity of the awaits further in vitro and in vivo studies. AR via either competing androgen binding to AR or by the interaction In summary, our study demonstrates that hK2, a serine protease of AR and ER (29). At the present time, our data cannot distinguish thought to be found only in prostate-related tissues and fluids, could the mechanism by which androgen-induced hK2 or PSA is repressed also be produced by at least one breast cancer cell line after steroid by estrogens. Moreover, if the quantitative difference between hK2 hormone stimulation. These findings extend our understanding of hK2 and PSA in T47-D cells is considered, the levels of hK2 are higher expression and raise the possibility that hK2 may have some clinical than PSA under the same stimulatory conditions. This situation is the significance in breast cancers, although, by analogy to PSA, hK2 reverse of what is seen in the LNCaP cells. Thus, the regulation of might be expressed in quite a low level in female tissues compared to secretion of PSA and hK2 in the two cell lines is obviously different. the prostate gland. Studies have shown that estrogen and progesterone receptors are favorable prognostic indicators as well as predictive factors for endo crine manipulation in the clinical management of breast cancer (30, REFERENCES 3 1). Androgen receptors are present in most breast cancer cells, and I. Papsidero. L. D.. Kuriyma. M., Wang, M. C., Horosewicz, J., Leong, S. S., their presence is closely related to the presense of ER and PR in breast Valenzuela, L., Murphy. G. P., and Chu, T. M. Prostate antigen: a marker for human prostate epithelial cells. J. NatI. Cancer Inst., 66: 37â€”41,1981. cancer cells (32). In addition, it has been shown that androgens can 2. Lilja. H. A. Kallikrein-like serine protease in prostate fluid cleaves the predominant inhibit the proliferation of breast cancer cells and counteract the seminal vesicle protein. J. Clin. Invest., 76: 1899â€”1903, 1985. 3. Watt, K. W. K., Lee, P-J., M'Timkulu, T., Chan, W-P., and Loor, R. Human effects of estrogen (33, 34). The possible significance of our findings prostate-specific antigen: structural and functional similarity with senne proteases. that hK2 is expressed only in an androgen-sensitive breast cell line Proc. NatI. Acal. Sci. USA, 83: 3166â€”3170,1986. and inhibited by estrogens is the following: (a) antiestrogen agents 4. Stamey, T. A,. Yang, N., Hay, A. R., McNeal, J. E., Freiha, F. S., and Redwine, E. Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. have been shown to be effective in the treatment of some breast N. EngI. J. Med., 317: 909â€”916, 1987. cancers. It is plausible to speculate that the level of hK2 may be 5. Catalona, W. J., Smith, D. S., Ratliff, T. L., Dodds, K. M., Caplen, D. E., Yuan, J. J.. potentially of relevance to the efficiency of antiestrogen therapy; and Petros, J. A., and Andriole, G. L. Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. N. EngI. J. Med., 324: 1156â€”1161,1991. (b) the expression of hK2 protein, like PSA (12), may be a favorable 6. Fraser, H. A., Humphrey, P. A., Burchette, J. L., and Paulson, D. F. Immunoreactive prognostic indicator for breast cancer. prostatic antigen in male periurethral glands. J. Urol., 147: 246â€”248,1992. 7. Iwakiri, J., Grandbois, K., Wehner, N., Graves, H. C. B., and Stamey, T. An analysis Another important observation in our study is the finding of two of urinary prostatic specific antigen before and after radical prostatectomy: evidence hK2 immunoreactive bands in the Western blot analysis of T47-D cell for secretion of prostatic specific antigen by the periurethral glands. J. Urol.. 149: media. The hK2 band was divided into one major component with a 783â€”786,1993. 8. Elgamal, A. A., Voorde, W. V., Poppel, H. V., Lauweryns, J., and Baert, L. slightly higher molecular weight than the mature form of recombinant Immunohistochemical localization of prostate-specific markers within the accessory hK2 and one minor lower molecular weight component. This major male sex glands of Cowper, Litre, and Morgagni. Urology, 44: 84â€”90,1994. 9. Yu, H., Diamandis, E. P., Zarghami, N., and Grass, L. Induction of prostate-specific protein is likely the pro-form of hK2. After stripping and reprobing antigen production by steroids and tamoxifen in breast cancer cell lines. Breast the same membrane with a mAb specific to the pro-segment of hK2, Cancer Res. Treat., 32: 291â€”300,1994. both major and minor bands were detected only in the lane containing 10. Yu, H., Diamandis, E. P., and Sutherland, D. J. A. lmmunoreactive prostate-specific antigen levels in female and male breast tumors and its association with steroid T47-D cell media. The minor band may be a degraded or clipped hormone receptors and patient age. Clin. Biochem., 27: 75â€”79,1994. fragment of hK2 retaining the pro-segment epitope. Similarly, an 11. Diamandis, E. P., Yu, H., and Sutherland, D. J. A. Detection of prostate-specific immunoreactive PSA band in androgen-induced T47-D cells was antigen immunoreactivity in breast tumors. Breast Cancer Res. Treat., 32: 301â€”310, 1994. detected with a higher apparent molecular weight than that of purified 12. Yu, H., Giai, M., Diamandis, E. P., Katsaros, D., Sutherland, D. J. A., Levesque, PSA from seminal fluid. In the future, it will be interesting to deter M. A., Roagna, R., Ponzone, R., and Sismondi, P. Prostate-specific antigen is a new favorable prognostic indicator for women with breast cancer. Cancer Ret., 55: mine whether T47-D cells secrete pro-PSA. It has been reported 2104â€”2110,1995. recently that recombinant hk2 is secreted as the pro-form initially by 13. Clements, A., and Mukhtar, A. Glandular kallikrein and prostatic specific antigen are the stably transfected AV12 hamster tumor cell line and the human expressed in the human endometrium. J. Clin. Endocrinol. Metab., 78: 1536â€”1539, 1994. prostate carcinoma cell lines, PC-3 and DU145 (25). Nevertheless, 14. Yu, H., and Diamandis, E. P. Prostate specific immunoreactivity in amniotic fluid. these cells seemed to be able to process the pro-hK2 into the mature Clin. Chem., 41: 204â€”210, 1995. hK2 extracellularly. In addition, the LNCaP cell line produces mainly 15. Yu, H., and Diamandis, E. P. Prostate specific antigen in milk of lactating women. Clin. Chem., 41: 54-60, 1994. mature hK2 protein. Thus, our results suggest that T47-D cells prob 16. Clements, J. The human kallikrein gene family: a diversity of expression and ably produce very little of an hK2-processing protease. However, function. Mol. Cell Endocrinol., 99: Clâ€”C6, 1994. 17. Schedlich, L. J., Benneus, B. H., and Morris, B. J. Primary structure of a human further confirmation is needed by verifying the identity of these bands glandular kallikrein gene. DNA, 6: 429â€”437, 1987. as the pro-hK2 by amino acid sequence analysis. 18. Young, C. Y. F.. Andrews, P. E., Montgomery, B. T., and Tindall, D. J. Tissue The physiological role of hK2 is presently unknown. Due to its high specific and hormonal regulation of human prostate-specific glandular kallikrein. Biochemistry, 31: 818â€”824,1992. sequence homology with PSA, it may have similar functions. The 19. Murtha, P., Tindall, D. J., and Young, C. Y. Androgen induction of a human primary biological role of PSA is to cleave the seminal coagulum (or prostate-specific kallikrein, hKLK2: characterization of an androgen response dc seminogelin) in seminal fluid into peptides, resulting in increased ment in the 5' promoter region of the gene. Biochemistry, 32: 6459â€”6464,1993. 20. Young, C. Y. F., Andrews, P. E., and Tindall, D. J., Expression and androgenic sperm motility (35). Recently, it was shown that PSA also can cleave regulation of human prostate-specific kallikrein. J. Androl., 16: 97â€”99,1995. insulin-like growth factor/binding protein 3 to increase the availability 21. Saedi, M. S., Cass, M. M. J., God, A. S., Grauer, L., Hogen, K. L.. Okaneya, T., Griffin, B. Y., Klee, G. G., Young, C. Y. F., and Tindall, D. 3. Overexpression of a of insulin-growth factor I, which is thought to be involved in normal human prostate-specific glandular kallikrein, hK2, in E. coli and generation of and malignant cell proliferation (36). It was suggested that hK2 may antibodies. Mol. Cell. Endocrinol., 109: 237â€”241,1995. be the protease responsible for the clipping and inactivation of PSA 22. Grauer, L. S.. Charlesworth, M. C., Saedi, M. S., Finlay, J. A., Liu, R-S., Kuus Reichel, K., Young, C. Y. F., and Tindall, D. J. Identification of human glandular and thereby serve as a regulator of PSA activity (37). Additionally, kallikrein hK2 from LNCaP cells. J. Androl., 17: 353â€”359.1996. hK2 might potentially be a kininogenase, an enzyme catalyzing bra 23. Young, C. Y. F., Seay. T., Hogen. K., Charlesworth, M. C., Roche, P. C., Klee, G. G., 2655

and Tindall, D. J. Prostate-specific human kallikrein (hK2) as a novel marker for Project (NSABP), prognostic discriminants for 8-year survival for node-negative prostate cancer. Prostate, 7: 17â€”24,1996. invasive breast cancer patients. Cancer (Phila.), 65: 2121â€”2128,1990. 24. Charlesworth, C. M., Young, C. Y. F., Klee, G. G., Saedi, M. S., Mikolajczyk, S. D., 32. Kuenen-Boumeester, V., Van der Kwast, T. H., van Putten, W. L. J., Claassen, C., van Finlay, J. A., and Tindall, D. J. Detection of a prostate-specific protein, human Ooijen, B., and Henzen-Logmans, S. C. Immunohistochemical determination of glandular, kallikrein (hK2), in sera of patients with elevated prostate-specific antigen androgen receptors in relation to oestrogen and progesterone receptors in female levels. Urology, 49: 487â€”493, 1997. breast cancer. Int. J. Cancer, 52: 581â€”584,1992. 25. Kumar, A., God, A. S., Hill, T. M ., Mikolajczyk, S. D., Millar, L. S., Kuus-Reichel, 33. Dauvois, S., Geng, S. C., Levesque, C., Merand, Y., and Labrie, F. Additive K., and Saedi, M. S. Expression of human glandular kallikrein, hK2, in mammalian inhibitory effects of an androgen and antiestrogen EM-hO on estrogen-stimulated cells. Cancer Ret., 56: 5397â€”5402,1996. growth of human ZR-75â€”lbreasttumors in athymic mice. Cancer Res., 51: 3131â€” 26. Rittenhouse, H., Tindall, D. J., Klee, G., Young, C. Y. F., Bostwick, D., Saedi, 3135, 1991. M. S., Grauer, L. S., Mikolajczyk, S. D., Finlay, J. A., Kuus-Reichel, K., Kumar, 34. de Launoit, Y., Dauvois, S., Dufour, M., Simard, J., and Labrie, F. Inhibition of cell A., and Wolfert, R. Characterization and evaluation of hK2, a potential prostate cycle kinetics and proliferation by the androgen 5a-dihydrotestosterone and antics cancer marker, closely related to PSA. Proc. 1st mt. Consult. Prostate Cancer, in trogen N.n-butyl-N-methyl-1l-[l6'a-chloro-3',l7@-dihydroxy-estra-l'.3',5'- press, 1997. (l0')triene-7'a-yl] undecanamide in human breast cancer ZR-75â€”lcells. Cancer 27. Sensabaugh, G. F., and Blake, E. T. Seminal plasma protein p3O: simplified purifi Res., 51: 2797â€”2802,1991. cation and evidence for identity with prostate specific antigen. J. Urol., 144: 1523-. 35. Ulja, H., Oldbring, J., Rannevik, G., and Laurell, C-B. Seminal vesicle-secretion 1526, 1990. proteins and their reactions during gelation liquification of human semen. I. Clin. 28. Beato, M. Gene regulation by steroid hormones. Cell, 56: 335â€”344,1989. Invest., 80: 281â€”285,1993. 29. Kumar, M. V., Leo, M. E., and Tindall, D. J. Modulation of androgen receptor 36. Cohen, P., Graves, H. C. B., Peehi, D. M.. Kamarci, M., Giudice, L C., and transcriptional activity by the estrogen receptor. J. Androl., 15: 534â€”542,1994. Rosenfeld, R. G. Prostate-specific antigen (PSA) is an insulin-like growth factor 30. Thorpe, S. M. Estrogen and progesterone receptor determinations in breast cancer. binding protein-3 protease found in seminal plasma. J. Clin. Endocrinol. Metab. 75: ActaOncol., 27: 1â€”18,1988. 1046â€”1053, 1992. 31. Fisher, E. R., Redmond, C., Fisher, B., Bass, G., and contributing NSABP investi 37, Dube, J. Y. Prostatic kallikreins: biochemistry and physiology. Comp. Biochem. gators. Pathologic findings from the National Surgical Adjuvant Breast and Bowel Physiol., 107C: 13â€”20.1994.

2656

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1997 American Association for Cancer Research. Expression of Human Prostate-specific Glandular Kallikrein Protein (hK2) in the Breast Cancer Cell Line T47-D

Ming-Li Hsieh, M. Cristine Charlesworth, Marcia Goodmanson, et al.

Cancer Res 1997;57:2651-2656.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/57/13/2651

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/57/13/2651. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.