[CANCER RESEARCH 48, 3502-3506, June 15, 1988]

Hormonal Regulation of CAI 25 Tumor Marker Expression in Human OvarianCarcinoma Cells: Inhibition by Glucocorticoids1

Beth Y. karlanr Waheeda Aniin, Scott E. Casper, and Bruce A. Littlefield3

Division of Gynecologic Oncology, Department of Obstetrics and Gynecology [B. Y. K., W. A., S. E. C., B. A. L.J, and Yale Comprehensive Cancer Center [B. A. L.J,Yale University School of Medicine, New Haven, Connecticut 06510

ABSTRACT

The CAI25 tumor marker is an antigenic determinant present on ahigh-molecular-weight glycoprotein expressed by more than 80% ofnewly diagnosed nonmucinous epithelial ovarian cancers. OVCA 433human ovarian carcinoma cells express the CA125 marker at the cellsurface and release large quantities of this antigen into culture medium.Here we show that release of CAI25 by OVCA 433 cells is 90 to 95%inhibited by treatment with 1 x 1(1" M dexamethasone, as determined

using a biotin-based enzyme-linked immunosorbent assay utilizing<>(! 25 monoclonal antibodies to CAI 25. The relative cell surface densityof CAI 25 is also decreased following dexamethasone treatment as determined by immunofluorescence techniques using (X 125 monoclonal antibodies. Inhibition of CAI25 expression is specific for glucocorticoids,such as cortisol and dexamethasone, and does not occur with estrogens,progestins, androgens, or mineralocorticoids. CA125 inhibition is alsodependent on the concentration of steroid used, with half-maximal andmaximal inhibition by dexamethasone occurring at about 3 x 10"' M and1 x IO"7M, respectively. Previous work has shown that OVCA 433 cells

are growth inhibited by glucocorticoids and contain 14,000 glucocorticoidreceptors per cell with an affinity for dexamethasone (Aj = 6.6 x 10~9

M) which corresponds well with the concentration required for half-maximal CAI25 inhibition. This correspondence, together with the specificity of CAI 25 inhibition for glucocorticoids, suggests that this effect ismediated by glucocorticoid receptors and is a specific biological effect ofglucocorticoids on this cell type. Our results demonstrate glucocorticoidinhibition of CAI25 expression by ovarian carcinoma cells and suggestthat endogenous or therapeutically administered glucocorticoids can influence CAI25 production by tumors in vivo.

INTRODUCTION

The CAI25 tumor marker is an antigenic determinant present on a high-molecular-weight nonmucinous glycoprotein (1).Circulating levels of this antigen are elevated in approximately80% of newly diagnosed nonmucinous epithelial ovarian carcinoma patients (2-7). In contrast, CAI25 levels in the vastmajority of healthy individuals or patients with benign diseaseare quite low (2-4, 6, 7). Rising levels of serum CAI25 duringfollowup of initial therapy are predictive of persistent cancer(2, 4-6, 8, 9). In one case where retrospective serum sampleswere fortuitously available, CAI25 levels were found to beelevated 10 to 12 mo prior to the initial clinical detection ofovarian carcinoma (10). Thus, measurements of serum CAI25levels may facilitate earlier detection and diagnosis of ovariancarcinoma as well as improved assessment of responses totherapy.

CAI25 was originally identified as an antigenic determinantrecognized by a murine monoclonal antibody termed OC 125(11). This antibody was generated from mice immunized with

Received 11/11/87; revised 3/2/88; accepted 3/16/88.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

' Supported by NIH Grant CA40010 to B. A. L. and fellowships from Ortho-ACOG and the James Hudson Brown-Alexander B. Coxe Fund to B. Y. K.

2 Present address: Department of Obstetrics and Gynecology, 24-127 CHS,

UCLA School of Medicine. Los Angeles, CA 90024.1To whom requests for reprints should be addressed, at Department of

Obstetrics and Gynecology, Yale University School of Medicine, 333 CedarStreet. New Haven. CT 06510.

OVCA 433 cells (11), a human ovarian carcinoma cell linederived from ascitic cells of a patient with serous cystadenocar-cinoma of the ovary (11, 12). More recently, the CA 125 antigenhas been identified as residing on a A/, 200,000 nonmucinousglycoprotein containing 24% carbohydrate (1). The circulatingform of CAI 25 in ovarian carcinoma patients, as well as CAI 25released by OVCA 433 cells, appears to be a higher molecularweight complex (M, > 10') of the M, 200,000 glycoprotein with

at least one other glycoprotein (1). Unlike many tumor markersdefined by monoclonal antibodies (13), the CAI25 determinantrecognized by OC 125 monoclonal antibodies appears to beproteinaceous rather than carbohydrate in nature (1). The function or biological significance of the glycoprotein on whichCAI25 is located is currently unknown.

Our laboratory has been using the OVCA 433 cell line toinvestigate the responsiveness of ovarian carcinoma cells tosteroid hormones in vitro. In a recent report (14), we demonstrated that OVCA 433 cells respond to glucocorticoid treatment with decreased cell growth, altered cellular morphology,and inhibition of both urokinase and tissue-type plasminogenactivator activities. Using the same criteria, no evidence wasfound for responsiveness of OVCA 433 cells to estrogens,progestins, androgens, or mineralocorticoids (14). OVCA 433cells were found to contain high affinity, steroid-specific glucocorticoid receptors (14), supporting the concept of this cellline as a model for glucocorticoid-sensitive ovarian carcinomas.In an attempt to further define the extent of glucocorticoidsensitivity of OVCA 433 cells, we have now examined theeffects of these hormones on expression of the CAI25 tumormarker. In this paper, we show that glucocorticoid treatmentof OVCA 433 cells markedly inhibits the release of CAI 25 intoculture medium and the density of cell surface expression ofthis antigen. These results clearly demonstrate that CAI25expression is subject to hormonal regulation, and they suggestthat CAI25 production by ovarian carcinoma cells in vivo maybe influenced by endogenous or therapeutically administeredglucocorticoids.

MATERIALS AND METHODS

Cell Culture Conditions, Preparation of Conditioned Medium forCA125 Measurement. OVCA 433 cells, originally derived from asciticcells of a patient with serous cystadenocarcinoma of the ovary (II. 12),were grown in culture medium containing 10% PCS4 as previously

described (14, 15). DEX and other steroids (Steraloids, Wilton, NH)were added to cell cultures as sterile lOx concentrates freshly preparedin culture medium (14). Preparation of conditioned medium for assessing hormonal effects on CAI25 release was performed as follows.

Since we previously demonstrated glucocorticoid inhibition ofOVCA 433 cell growth (14), we wished to avoid potential complicationsarising from differences in growth rates between control and steroid-treated cells. Thus, collection of conditioned medium for measurement

4The abbreviations used are: PCS, fetal calf serum; BS, borate-saline; BSA,bovine serum albumin; cAMP, 3',5'-cyclic adenosine monophosphate; DEX,dexamethasone (1.4-pregnadien-9-fluoro-16a-methyl-l liU7a.2l-triol-3.20-dione); ELISA, enzyme-linked immunosorbent assay; GR. glucocorticoid recep-tor(s).

3502

on May 25, 2018. © 1988 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

GLUCOCORTICOID INHIBITION OF CA125 TUMOR MARKER

of CAI25 release was initiated only after all cultures had reached thequiescent. Go-like phase occurring at confluence (Footnote 5; Ref. l),confluence being judged by visual inspection of the culture plates underphase-contrast microscopy. In addition, cells were seeded at a highenough density (5 x IO5 cells/25-cm2 area; 10 ml of growth medium)so that all control and steroid-treated cultures reached confluence after3 to 4 days. After reaching confluence, spent growth media werereplaced with fresh medium maintaining appropriate concentrations ofsteroids. This time represented "Day 0 postconfluence." It was not

necessary to use serum-free medium for preparation of conditionedmedium since OC125-recognizable material was not present in the PCSused in our experiments.5 Conditioned media were then harvested 1 to

7 days later, centrifugea at 4000 x g for 10 min, aliquotted in smallvolumes, and stored at -80°C until assayed for CAI25 content. To

normalize CAI25 data to cell number, cell monolayers were harvestedwith 0.25% (w/v) trypsin (Sigma Chemical Co., St. Louis, MO) in 130mM NaCl/3 mM KC1/2 mM EOT A/30 mM Ar-2-hydroxyethylpipera-zine-/V"-2-ethanesulfonic acid/1 mM sodium phosphate/10 mM glu

cose/3.3 nM phenol red (pH 7.6), washed once in medium containing10% PCS, and quantitated by hemacytometer counting.

Preparation of CAI25 Reference Standard from OVCA 433-condi-tioned Medium. A reference standard of CAI25 was prepared for usein our laboratory in the following manner. Two 75-cm2 culture flasks

of OVCA 433 cells at passage 69 were grown to confluence, at whichtime spent growth medium was replaced with fresh medium. Cultureswere then allowed to incubate for an additional 7 days. Under theseconditions, OVCA 433 cells are in Go-like phase and release largeamounts of CAI 25 into culture medium (1). The resultant conditionedmedium was pooled, clarified by centrifugation (300 x g, 20 min), andaliquotted into small volumes which were then stored at -•«()"('until

use. Using a commercially available CAI25 radioimmunoassay kit(Centocor, Inc., Malvern, PA), the CAI 25 content of this lot of conditioned medium was determined to be 2097 units/ml. Recently, Daviset al.(l) estimated that highly purified CAI25 has a specific activity of317 units/fig of protein; using this value, the 2097-unit/ml of CAI 25reference standard contains about 6.6 ¿igof CAI 25 per ml.

Biotinylation of (X 125 Monoclonal Antibodies. OC 125 monoclonalantibodies to CAI25 (11) were generously provided by Dr. Vincent R.Zurawski, Jr., of Centocor. Inc. (Malvern, PA). Biotinylation of antibodies was performed using the following modifications of the methodsof Heggeness and Ash (16). Concentrated OC 125 as supplied wasdiluted to 1 mg/ml in 0.1 M NaHCO3 (pH 8.4) and dialyzed overnightagainst 100 volumes of the same solution at 4°C.The solution was then

brought to room temperature, and 0.12 volumes of 1 mg/ml of biotin-A'-hydroxysuccinimide ester (Bethesda Research Laboratories, Gaith-

ersburg, MD) freshly prepared in dimethyl sulfoxide were added withrapid mixing. The solution was incubated at room temperature for 4 h,followed by dialysis overnight at 4°Cagainst 100 volumes of 10 mM

sodium phosphate/150 mM NaCl/6 mM sodium azide (pH 7.4). Theresultant biotinylated OC125 solution was then aliquotted into smallvolumes and stored at -80°Cuntil use.

Biotin-based Enzyme-linked Immunosorbent Assay for CA125.CAI25 released into conditioned medium was quantitated using anELISA based on biotinylated OC 125 monoclonal antibodies (see above)and streptavidin-alkaline phosphatase conjugate. The central 60 wellsof 96-well Immulon II ELISA plates (Dynatech Laboratories, Alexandria, VA) were precoated with 100 n\ of 10 /¿g/mlof OC125 in 25 mMsodium tetraborate (borax)/100 mM boric acid/75 mM NaCl (pH 8.4)(BS) for 4 h at room temperature. To prevent evaporation, this and allsubsequent incubations were performed in covered plastic freezer containers which were humidified with 2 to 3 layers of water-saturatedpaper towels in the bottoms. Following the incubation, wells werewashed once by filling with 0.85% (w/v) NaCl (saline), incubating for5 min, and emptying by a quick shake of the wrist in an invertedposition. Samples and standards were then added to the wells in 100-ii\ volumes. Samples consisted of empirically determined dilutions ofconditioned medium (typically 1:4 dilutions in culture medium containing 10% PCS) from experimentally treated cell cultures. Standards,prepared in culture medium containing 10% PCS, consisted of 0 to 210

units of CAI 25 from the CAI25 reference standard (see above). Incubations were continued overnight at 4"( '. Samples and standards were

then removed by inverted shaking as above, and wells were blocked forl h at room temperature by the addition of 200 iA of 5% (w/v) BSA inSolution BS (5% BSA/BS). Two 5-min washes with saline were performed after blocking, followed by addition of 100 »ilof 1 jiu/inl ofbiotinylated OC125 in Solution BS containing 1% BSA (1% BSA/BS).Incubations were continued for 4 h at room temperature, followed byblocking again with 5% BSA/BS and two saline washes. One hundredn\ of streptavidin:alkaline phosphatase conjugate (Bethesda ResearchLaboratories) diluted 1:1000 (manufacturer's recommendation) in 1%

BSA/BS were added to the wells, followed by incubation overnight at4°C.Wells were then washed 5 times with saline as above, followed by

the addition of 200 ^1 of 1 mg/ml of p-nitrophenylphosphate (Sigma)in l M diethanolamine/240 /¿MMgCl2 (pH 9.8). Reactions were thenallowed to occur at room temperature, and yellow color developmentfrom the generation of p-nitrophenol was monitored periodically at 405nm using a Titertek Multiskan Plus ELISA plate reader.

Immunohistochemical Evaluation of Cell Surface CAI 25 Levels. Theeffects of DEX on cell surface expression of CAI 25 were assessed usingimmunofluorescent staining techniques as follows. OVCA 433 cellswere grown for 3 days to confluence in Lab-Tek dual-chamber plastictissue culture slides (Miles Scientific, Naperville, IL) in the presence orabsence of 1 x 10~7M DEX. Cells then received fresh cultured medium,

with or without DEX, for an additional 24 h. Immunofluorescentstaining of cell surface CAI 25 with OC 125 monoclonal antibodies andfluorescein-conjugated goat anti-mouse IgG second antibody (Pel-Freez

Biologicals, Rogers, AR) was performed using methods described byDelbeke et al. (17). Immunofluorescence was assessed using a ZeissIM-35 fluorescence microscope with a 35-mm camera attachment.

RESULTS

OVCA 433 cells enter a quiescent, Go-like phase upon reaching confluence (Footnote 5; Ref. l). Such cells continuallyproduce and release CAI25, as demonstrated by the continuousaccumulation of CAI 25 in conditioned medium (Fig. 1). A lagperiod of about 1 day occurred before CAI25 levels in conditioned medium were immunologically detectable by ELISA.After 1 day, accumulated CAI25 levels in control culturesincreased rapidly, reaching 2990 units/IO6 cells by 7 days

postconfluence. In contrast, the amount of CAI 25 released by

' Personal observation.

246Days post-confluence

Fig. 1. Inhibition of CAI 25 release from OVCA 433 cells by dexamethasone.OVCA 433 cells were plated in 25-cm: culture flasks in the absence or presenceof 1 x 10~7 M DEX and grown for 3 days lo confluence. Spent growth mediawere then replaced with fresh media, with or without 1 x 10~7 M DEX asappropriate. This point represents "Day 0 postconfluence." Conditioned media

from individual culture flasks were then harvested at the indicated times andstored at -80'C for later analysis of CAI25 content by ELISA. Cells were

harvested and quantitated for normalization of CAI25 data to cell number.Details of all procedures are presented in "Materials and Methods." Points, mean

of triplicate determinations from single flasks; bars, SE.

3503

on May 25, 2018. © 1988 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

GLUCOCORTICOID INHIBITION OF CA125 TUMOR MARKER

O VGA 433 cells treated with 1 x IO"7 M concentrations of the

synthetic glucocorticoid DEX was dramatically reduced (Fig.1). In this experiment, CAI25 release by DEX-treated cells wasundetectable until at least 5 days postconfluence. At 7 days,CAI25 levels in conditioned medium from DEX-treated cellswere only 143 units/10* cells, or less than 5% of CAI 25 levels

in control cultures. Thus, DEX treatment of OVCA 433 cellsresulted in greater than 95% inhibition of CAI25 release intoculture medium.

The magnitude of glucocorticoid inhibition of CAI 25 releaseis dependent on the concentration of steroid used. In the experiment shown in Fig. 2, cells were grown to confluence (4 days)under control conditions or with 1 x 10~'°to 1 x IO"6 M

concentrations of DEX or the natural glucocorticoid cortisol.Cultures then received growth medium (maintaining steroids),and CAI25 levels in conditioned medium were analyzed 7 dayslater. Similar to the results of Fig. 1, treatment with 1 x 1(T7

M DEX led to virtually complete inhibition of CAI25 release.Half-maximal inhibition by DEX occurred at an estimatedconcentration of about 3 x IO"9 M. Inhibition of CAI 25 release

also occurred with the natural glucocorticoid cortisol, althoughthis steroid was less potent than DEX. Thus, maximal CAI25inhibition by cortisol was achieved at 1 x 10~6 M, and half-maximal inhibition occurred at 1 x 10~7 M. These data from

cortisol- and DEX-treated cells indicate that inhibition ofCA125 release by both natural and synthetic glucocorticoids isa concentration-dependent process.

To determine the steroid specificity of inhibition of CAI25release, OVCA 433 cells were grown to confluence (4 days)under control conditions or with 1 x 10~6 M concentrations of

170-estradiol, progesterone, dihydrotestosterone, aldosterone,

cortisol, and DEX. Cells then received fresh growth mediumcontaining the appropriate steroids, and CAI25 levels in conditioned medium were examined 7 days later. In agreementwith the results of Fig. 2, treatment with the glucocorticoidscortisol and DEX led to 71% and 90% inhibition of CAI25release, respectively (Fig. 3). In contrast, treatment with 170-estradiol, progesterone, dihydrotestosterone, or aldosterone didnot significantly affect CAI25 release. Thus, inhibition of

C 10 9 8 7 6-log [steroid]

Fig. 2. Inhibition of CAI25 release by cortisol and dexamethasone: concentration dependence. OVCA 433 cells were plated in 25-cm2 culture flasks in the

presence of the indicated concentrations of cortisol and DEX and grown for 4days to confluence. Spent growth media were then replaced with fresh mediacontaining the same concentrations of cortisol or DEX. Conditioned media werethen harvested 7 days later and analyzed for CAI25 content by ELISA. Cellswere harvested and quantitated for normalization of CAI25 data to cell number.Details of all procedures are presented in '•Materialsand Methods." Points, meanof triplicate determinations from single flasks, except for the steroid-free control(C), which represents the mean (±range) of two determinations from duplicateflasks, each determination being the mean of triplicate assay points. Bars, SE.

~ o3 ^

U-feL

Con E2 Pro DHT Aid Cort DexSteroid (IO"6M)

Fig. 3. Effects of steroid hormones on CAI 25 release from OVCA 433 cells.Cells were plated in 25-cm2 culture flasks in the presence of 1 x 10"* M

concentrations of the indicated steroids and grown for 4 days to confluence. Spentgrowth media were then replaced with fresh media containing the same concentration of the appropriate steroids. Conditioned media were then harvested 7 dayslater and analyzed for CAI25 content by ELISA. Cells were harvested andquantitated for normalization of CAI25 data to cell number. Details of allprocedures are presented in "Materials and Methods." Columns, mean of two

separate determinations from each of two flasks for each steroid, with eachdetermination being performed in triplicate; bars, range. Con, control; £j,17/3-estradiol; Pro, progesterone; DHT, dihydrotestosterone; Aid, aldosterone; Cort,cortisol; Dex, DEX.

CAI25 release from OVCA 433 cells is a glucocorticoid-specificeffect.

Finally, we used immunofluorescent staining techniques toexamine relationships between glucocorticoid inhibition ofCAI25 release into culture medium and cell surface expressionof this antigen. In the experiment of Fig. 4, cells were grownfor 3 days under control conditions or in the presence of 1 X10~7M DEX. Fresh culture medium with or without DEX was

then added, followed by analysis of cell surface CAI 25 expression 24 h later. In agreement with work by others (18), immu-nologically detectable CAI25 was present on the cell surfacesof control OVCA 433 cells, as indicated by a large degree ofimmunofluorescence (Fig. 4, left). The amount of fluorescenceon individual cells was variable, indicating that CAI25 expression in a given OVCA 433 cell population is heterogeneous.Although OVCA 433 cells treated with DEX showed someimmunofluorescence (Fig. 4, middle), the relative intensity ofstaining was markedly reduced in comparison to control cells,indicative of a reduced cell surface density of CA125 expression.The previously reported spreading and enlargement of DEX-treated cells (14) were also observable by immunofluorescence.Finally, nonspecific binding of the fluorescein-labeled secondantibody to DEX-treated cells (Fig. 4, right) and control cells(not shown) was quite low, indicating that the fluorescence seenwith OC 125 monoclonal antibodies accurately reflected thedensity of cell surface CAI 25 expression. Thus, the data of Fig.4 indicate that glucocorticoid inhibition of CAI25 release intoconditioned medium (Figs. 1 to 3) is accompanied by a decreased density of cell surface CAI 25 expression.

DISCUSSION

We previously demonstrated that OVCA 433 human ovariancarcinoma cells contain GR and respond to glucocorticoidtreatment with inhibited cell growth, morphological alterations,and inhibition of urokinase and tissue-type plasminogen activator activities (14). Such effects were not observed with exog-enously added estrogens, progestins, androgens, or mineralo-corticoids (14). The current findings now demonstrate glucocorticoid inhibition of CAI25 expression by OVCA 433 cells.

3504

on May 25, 2018. © 1988 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

GLUCOCORTICOID INHIBITION OF CAI 25 TUMOR MARKER

CON DEX DEXÕ-OCI25)Fig. 4. Effects of dexamethasone on cell surface density of CAI25. OVCA 433 cells were grown in dual chamber plastic tissue culture slides for 3 days toconfluence in the presence or absence of 1 x IO"7 M DEX. Growth media were then replaced with fresh media, with or without DEX, and incubations continued for

an additional 24 h. Cell surface CAI25 densities were then assessed by immunofluorescent staining procedures as described in "Materials and Methods." Left andmiddle, control and DEX-treated cells, respectively, stained with OC125 monoclonal antibodies followed by fluorescein-conjugated second antibodies; right, DEX-treated cells treated identically but without OC 125 monoclonal antibodies. Con, control.

Half-maximal CAI25 inhibition by DEX occurred at about 3X 10~9M, a value which corresponds well with the affinity (Ad)of OVCA 433 GR for [3H]DEX of 6.6 x IO"9 M (14). Similarly,

of the steroids tested, only cortisol and DEX inhibited CAI25expression, and only cortisol and DEX were capable of competing with [3H]DEX for OVCA 433 GR binding sites (14).Thus, the characteristics of CAI25 inhibition by glucocorti-coids, like those of inhibited cell growth and plasminogenactivator (14), are consistent with mediation by the GR presentin these cells.

An interesting aspect of our studies is that cortisol was about100-fold less potent than DEX. It has been known for yearsthat the biological potency of DEX in vivo can be greater than100 times that of cortisol (19), a larger discrepancy than mightbe expected from the smaller 10-fold difference in GR affinityfor these two steroids (20). Most of the discrepancy betweenGR binding affinity and biological potency in vivo can beaccounted for by different rates of metabolism of cortisol andDEX, since several of the substituents on the DEX moleculeare known to protect against metabolic inactivation relative tocortisol (21, 22). In our studies, we used relatively long 7-dayincubations with confluent cultures to maximize differences inCAI25 content of conditioned medium from control and steroid-treated cells. Such long incubations would allow ampletime for differences in metabolism of cortisol and DEX toinfluence the apparent degree of biological potency. Supportingthis interpretation is the correspondence of DEX concentrations required for half-maximal inhibition of CAI25, cellgrowth (14), and plasminogen activator (14) with the KA ofOVCA 433 GR for [3H]DEX (14). These correlations, as well

as the specificity of CAI25 inhibition for glucocorticoids,strongly support GR mediation of CAI25 inhibition despitethe 100-fold difference in potency between cortisol and DEX.

The immunofluorescence data indicate that decreased cellsurface density of immunoreactive CAI25 parallels decreased

CAI25 release into culture medium. However, it has yet to beestablished that this represents a true diminution of the numberof CAI 25 sites per cell, since dilution of a constant amount ofCAI25 over the larger surface area of a glucocorticoid-treatedcell would by itself lead to decreased cell surface density ofCAI25. However, we feel that the magnitude of the visualdecrease in immunofluorescence and the parallel inhibition ofCAI25 release both support the concept of glucocorticoid regulation of the number of immunoreactive cell surface CAI25sites per cell. We are currently performing quantitative cellsurface saturation binding studies with OC 125 monoclonalantibodies to directly address this issue.

The mechanisms of CAI25 inhibition by glucocorticoidsremain unknown. Our studies demonstrate that glucocorticoidtreatment of OVCA 433 cells results in decreased release andcell surface density of CAI 25 as assessed by OC 125 monoclonalantibodies. Although the determinant recognized by OC 125monoclonal antibodies resides on a M, 200,000 nonmucinousglycoprotein (1), our current results do not distinguish betweeninhibited production of this glycoprotein or altered proteinconformations or glycosylation states which would destroy orblock the CAI25 determinant. Alternatively, glucocorticoidsmight induce either a protein which would bind to the M,200,000 glycoprotein and block the CAI25 determinant, or anenzyme which would specifically destroy or alter the glycoprotein. Elucidation of the mechanisms of CAI25 inhibition byglucocorticoids will probably have to await the development ofeither polyclonal antibodies to the entire M, 200,000 glycoprotein or multiple monoclonal antibodies to different antigenicdeterminants.

Demonstration of CAI25 inhibition by glucocorticoids represents one of the first insights into potential regulatory mechanisms governing CAI25 expression and suggests that CAI25production is a dynamic, rather than constitutive, process.Recent work by two other groups also supports this concept.

3505

on May 25, 2018. © 1988 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

GLUCOCORTICOID INHIBITION OF CA125 TUMOR MARKER

Ishiwata et al. (23) found that treatment of four different humanovarian carcinoma cell lines with the dibutyryl derivative ofcAMP stimulated CAI25 release in a concentration-dependentmanner. Since cAMP transduces signals from a variety of cellsurface receptors, stimulation of CA125 by cAMP suggests thatunidentified hormones or growth factors might regulate CAI25expression. In endometrial stromal cells, Bischof et al. (24)demonstrated moderate inhibitory effects of medroxyprogester-one acetate on CAI25 production, an effect which was blockedby both 170-estradiol and the glucocorticoid/progestin antagonist RU38486. Taken together, our data and those of Ishiwataet al. (23) and Bischof étal.(24) indicate that CA125 expressionis a regulated cellular process. The significance of this regulation, however, will remain unknown until the role of CAI 25 inboth normal and cancer cell biology is uncovered.

ACKNOWLEDGMENTS

We are grateful to Dr. Vincent R. Zurawski, Jr., of Centocor, Inc.,for generously providing OC 125 monoclonal antibodies and to Dr.Edward Hawrot of the Department of Pharmacology, Yale University,for the use of his fluorescence microscope. We also thank Mary Amthorfor technical assistance and Dawn Bailey-Walton for her help in preparing the manuscript.

REFERENCES

1. Davis, H. M., Zurawski, V. R., Jr., Bast, R. C, Jr., and Klug, T. L.Characterization of the CAI 25 antigen associated with human epithelialovarian carcinomas. Cancer Res., 46: 6143-6148, 1986.

2. Bast, R. C, Jr., Klug, T. L.. St. John, E.. Jenison. E., Niloff, J. M., Lazarus,H., Berkowitz, R. S.. Leavitt, T., Griffiths, C. T., Parker, L., Zurawski, V.R., Jr., and Knapp, R. C. A radioimmunoassay using a monoclonal antibodyto monitor the course of epithelial ovarian cancer. N. Engl. J. Med., 309:883-887, 1983.

3. Klug, T. L., Bast. R. C.. Jr., Niloff, J. M., Knapp, R. C, and Zurawski, V.R., Jr. Monoclonal antibody immunoradiometric assay for an antigenicdeterminant (CA 125) associated with human epithelial ovarian carcinomas.Cancer Res., 44; 1048-1053, 1984.

4. Krebs, H.-B., Goplerud, D. R.. Kilpatrick, S. J., Myers, M. B., and Hunt, A.Role of Ca 125 as tumor marker in ovarian carcinoma. Obstet. Gynecol., 67:473-477, 1986.

5. Canney, P. A.. Moore, M., Wilkinson, P. M., and James, R. D. Ovariancancer antigen CAI25: a prospective clinical assessment of its role as atumour marker. Br. J. Cancer, 50: 765-769, 1984.

6. Schwartz, P. E., Chambers, S. K., Chambers, J. T., Gutmann. J., Katopodis,N., and Foemmel. R. Circulating tumor markers in the monitoring ofgynecologic malignancies. Cancer (Phila.), 60: 353-361, 1987.

7. Heinonen, P. K., Toniti, K., Koivula, T.. and Pystynen, P. Tumour-associated

antigen CA 125 in patients with ovarian cancer. Br. J. Obstet. Gynecol., 92:528-531, 1985.

8. Niloff, J. M., Bast, R. C., Jr., Schaetzl, E. M., and Knapp, R. C. Predictivevalue of CA 125 antigen levels in second-look procedures for ovarian cancer.Am. J. Obstet. Gynecol., ¡SI:981-986, 1985.

9. Atack, D. B., Nisker, J. A., Allen, H. H., Tustanoff, E. R., and Levin, L. CA125 surveillance and second-look laparotomy in ovarian carcinoma. Am. J.Obstet. Gynecol., 154: 287-289, 1986.

10. Bast, R. C, Jr., Siegal, F. P., Runowicz, C., Klug, T. L., Zurawski, V. R.,Jr., Schonholz, D., Cohen, C. J., and Knapp, R. C. Elevation of serum CA125 prior to diagnosis of an epithelial ovarian carcinoma. Gynecol. Oncol.,22:115-120, 1985.

11. Bast, R. C., Jr., Feeney, M., Lazarus, H., Nadler, L. M., Colvin, R. B., andKnapp, R. C. Reactivity of a monoclonal antibody with human ovariancarcinoma. J. Clin. Invest., 68: 1331-1337. 1981.

12. Introna, M., and Bast, R. C., Jr. Epithelial ovarian carcinoma: cell lines andantigenic tumor markers. In: C. J. Williams and J. M. A. Whitehead (eds.).Cancer of the Female Reproductive System, pp. 153-180. New York: JohnWiley and Sons, 1985.

13. Feizi, T. Demonstration by monoclonal antibodies that carbohydrate structures of glycoproteins and glycolipids are oncn developmental antigens.Nature (Lond.), 314: 53-57, 1985.

14. Amin, W., Karlan, B. Y . and Littlefield, B. A. Glucocorticoid sensitivity ofOVCA 433 human ovarian carcinoma cells: inhibition of plasminogen activators, cell growth, and morphological alterations. Cancer Res., 47: 6040-6045, 1987.

15. Karlan, B. V . Clark, A. S., and Littlefield, B. A. A highly sensitive chromo-géniemicrotiter plate assay for plasminogen activators which quantitativelydiscriminates between the urokinase and tissue-type activators. Biochem.Biophys. Res. Commun., 142: 147-154, 1987.

16. Heggeness, M. H., and Ash, J. F. Use of the avidin-biotin complex for thelocalization of actin and myosin with fluorescence microscopy. J. Cell Biol.,73: 783-788, 1977.

17. Delbeke, D., Scammell, J. G., Martinez-Campos, A., and Dannies, P. S.Dopamine inhibits prolactin release when cyclic adenosine 3',5'-monophos-phate levels are elevated. Endocrinology, //*: 1271-1277, 1986.

18. Masuho, Y., Zalutsky, M., Knapp, R. C., and Bast, R. C., Jr. Interaction ofmonoclonal antibodies with cell surface antigens of human ovarian carcinomas. Cancer Res., 44: 2813-2819, 1984.

19. Popper, T. L., and Watnick, A. S. Antiinflammatory steroids. In: R. A.Scherrer and M. W. Whitehouse (eds.), Antiinflammatory Agents, pp. 245-294. New York: Academic Press, 1974.

20. Wolff, M. E., Baxter, J. D., Kollman, P. A., Lee, D. L., Kuntz, I. D., Bloom,E., Matulich, D. T., and Morris, J. Nature of steroid-glucocorticoid receptorinteractions: thermodynamic analysis of the binding reaction. Biochemistry,17: 3201-3208, 1978.

21. Glenn, E. M., Stafford, R. O., Lyster, S. C., and Bowman, B. J. Relationbetween biological activity of hydrocortisone analogues and their rates ofinactivation by rat liver enzyme systems. Endocrinology, 61: 128-142, 1957.

22. Wolff, M. E. Structure-activity relationships in glucocorticoids. In: J. D.Baxter and G. G. Rousseau (eds.), Glucocorticoid Hormone Action, pp. 97-107. Berlin: Springer-Verlag, 1979.

23. Ishiwata, I., Ishiwata, C., Nozawa, S., and Ishikawa, H. CAI25 productionby gynecologic tumors in vitro and its modulation induced by dibutyl cyclicadenosine monophosphate. Asia-Oceania J. Obstet. Gynecol., 12: 285-290,1986.

24. Bischof, P., Tseng, L., Brioschi, P. A., and Herrmann, W. L. Cancer antigen125 is produced by human endometrial stromal cells. Human Reprod., I:423-426, 1986.

3506

on May 25, 2018. © 1988 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1988;48:3502-3506. Cancer Res   Beth Y. Karlan, Waheeda Amin, Scott E. Casper, et al.   Human Ovarian Carcinoma Cells: Inhibition by GlucocorticoidsHormonal Regulation of CA125 Tumor Marker Expression in

  Updated version

  http://cancerres.aacrjournals.org/content/48/12/3502

Access the most recent version of this article at:

   

   

   

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

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

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

.http://cancerres.aacrjournals.org/content/48/12/3502To request permission to re-use all or part of this article, use this link

on May 25, 2018. © 1988 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from