ORIGINAL 

Annals of Nuclear Medicine Vol. 7, No. 3, 133-139, 1993 

Changes in CA125 release and surface expression caused by drugs in uterine cervix adenocarcinoma cells 

Toshiharu NAKAI * Harumi SAKAHARA,* Keigo ENDO,** Makoto SHIRATO,* Hisataka KOBAYASHI,* Makoto HOSONO,* Tsuneo SAGA,* ' and Junji KONISHI* Masaru SAKAMOTO*** 

* Department of Nuclear Medicine, Kyoto University School of Medicine, Kyoto **Department of Nuclear Medicine, Gunma University School of Medicine, Maebashi *** Department of Gynecology and Obstetrics, Tokyo Medical College, Tokyo 

The effect of drugs on the release of CA125 antigen and the binding of anti-CA125 mono-clonal antibody (MOAb) to mali_gnant cells was evaluated in vitro. TMCC-1, uterine cervical adenocarcinoma cells, were exposed to dexamethasone (DEX), sodium n-butyrate (NaB), dibutyryl cyclic AMP (dbcAMP), retinoic acid (RA), calcitriol (VD3), and interferon-r (IFN-r)¥ NaB, RA and VD3 increased CA125 release per cell and 125I-labeled anti-CA125 MoAb binding to the cells. DEX also increased the 125I-labeled anti-CA125 MoAb binding to the cells, and CA125 antigen release per cell was also slight]y increased, IFN-r suppressed both CA125 release and 125I-labeled MoAb binding. A combination of DEX, VD3 and RA and increased the binding of MoAb to TMCC-1 cells, but the amount of bound MoAb was not significantly different from that obtained by single drug treatment. DbcAMP had no significant effect on enhancing MoAb binding. Drugs can increase the binding of anti-CA125 MoAb to malignant cells and they may be applied to increase the tumor uptake of radiolabeled MoAbs in vivo. 

Key words : CA125, TMCC-1, monoclonal antibody, ovarian cancer, VD3 

INTRODUCTION 

THOUGH A LOT OF TRIALS for the imaging and therapy of malignant tissues with radiolabeled monoclonal antibodies (MOAbs) have been reported, the applica-tion in vivo is still in the struggling stage.1,2 To in-crease the tumor uptake of radiolabeled MoAbs, the enhancement of cancer-associated antigens on the cell surface may be effective, and for that purpose a pharmacological approach has been investigated.3 In our previous report,4 the effect o-f dexamethasone (DEX), interferon-r (IFN-r) and sodium butyrate (NaB) on CA125 expression in 8 cell lines was ex-amined and it was observed that not all cells be-haved the same way. CA125 expression and 125I-labeled anti-CA125 MoAb binding to the cells were increased by NaB, which is known as a differentiator, in 3 cell lines and decreased by DEX in the same 3 cell lines. IFN-r increased them in one cell line but decreased them in two. These discrepancies among cells should be noted in the application of drug administration because they might modify the antigen expression in immunotherapy. To further as sess the effect of drugs on CA125 antigen expres-sion and MoAb binding, we used TMCC-1, uterine cervix adenocarcinoma cells which highly express CA125 antigen. The effect of calcitiol (VD3), an activated form of vitamin D3, and that of retinoic acid (RA) are of particular interest, as both drugs are known as differentiators . 

MATERIALS AND METHODS 

Cells 
TMCC-1 cells5 derived from uterine cervix adeno-carcinoma highly express CA125 antigen on the cell surface and release CA125 antigen into the culture medium supernatant. 

Drugs 
Dexamethasone (DEX; 9a-fluoro-16a-methylpredni-solone, Sigma, St. Louis, MO), sodium n-butyrate (NaB; C4H7NaO2 ; Kanto, Tokyo, Japan), dibutyryl cyclic adenosine 3',5'monophosphate (dbcAMP; Sigma), and retinoic acid (RA; all trans, Sigma) were commercially obtained. Calcitriol (VD3 ; 1.25-dihydroxyvitamin D3, Rosch, Basel, Switzerland), an active form of vitamin D3, was provided by Rosch Japan. Recombinant human interferon-r (IFN-r; produced by Genentech, South San Francisco, CA) was supplied by Toray. 

Cell culture 
All the cell cultures were performed in 25 cm2 plastic flasks with 10ml of culture medium (RPM11640, Nissui, Tokyo). containing 10% fetal calf serum at 37'C in 5% CO2. TMCC-1 cells (2 x 106) were seeded to each flask, and preincubated for 24 hr. The culture medium was changed every 24 hr to measure CA125 concentrations. During the first 24 hr after preincuba-tion, cells were cultured in medium without drugs. Then cells were cultured the next 5 days with DEX (10-9, 10-7, 10-6 M), NaB (10-7, 10-5, 10-4 M), dbcAMP (10-8, 10-6, 10-5 M), RA (10-8, 10-6. 10-5 M), VD3 (10-10, 10-8, 10-7 M), or IFN-r (10, 102, 103 U/ml). Each culture was done in duplicate. The first day after drug addition was designated as "day 1" and the final day as "day 5". DF.X, VD3 and RA were first dissolved in ethanol and diluted with culture medium to below 0.1 % of the concentration of ethanol. The medium containing each drug was freshly prepared on the first day of drug addition and stored at 4'C. The harvested medium was centrifuged at 500 g for 10 minutes and the supernatant was stored at -40'C. After ex-posure to each drug, the cells were harvested with 10 ml of phosphate buffer saline (Nissui Pharma-ceutical Co., Ltd., Tokyo) with 0.02% ethylendi-amine tetraacetic acid (Nakarai Co., Ltd.), and the number of cells in each culture was counted with a a hemacytometer. Cell viability after drug treatment was assessed by Trypan blue (Merck, Darmstadt, Germany) staining. The effect of each drug combina-tion was studied by mixing DEX, VD3 and RA. The least effective doses of each drug enhancing 125I-labeled MoAb binding to TMCC-1 cells were combined . 

MeasuremLmt of CA125 
To determine the CA 125 value for each culture super-natant, we used CA130 assay kits (Daiich, Tokyo). 
CA130 kits consist of beads coated with 145-9 MoAb and 125I-1abeled 130-22 MoAb, both of which re-cognize epitopes on CA125 antigen, but are different from OC125.6,7 Duplicate measurement was done with one step assay for each duplicate culture. Values obtained with the CA130 kit correlated well with CA125 values obtained with commercially available kits.8 Samples (100 /d) and 125I-labeled 130-22 MoAb (5 x 104 cpm/100 /d) were incubated with beads coated with 145-9 MoAb for 4 hr at room temperature. Each sample was measured in duplicate. The beads were washed three times with 3 ml of distilled water and the radioactivity of each bead was measured with a scintillation counter. The CA125 value was obtained from the standard curve plotted by standard sample value. 

Cell binding assay 
Cells treated with each concentration of drug were suspended in various numbers of cells ranging from 1 x 104 to 2 x 106 cells per 100 pl phosphate buffered saline. Then each sample was incubated for 2 hr with 125I-labeled 145-9 MoAb (5 x 104 cpm/100 ul) at 4'C in a 0.4 ml microtube. After incubation, the tubes were centrifuged at 10,000 g for 5 min, and the radioactivity of each pellet was measured in a well-type r-counter. The binding percentage of the radio-activity added was calculated. For Scatchard plot analysis, 2 x 105 cells were incubated with unlabeled 145-9 MoAb ranging from 12.5ng to 10peg and 125I-labeled 145-9 MoAb for 2 hr at 4'C. The numbers of MoAb binding sites and affinity constants were determined.9 

Statistics 
Data were expressed as the mean+-SD. The effect of drugs on the CA125 value of the culture super-natant, number of cells at harvest and binding of 125I-1abeled anti CA125 MoAb to cells were expressed as a percentage of the control. The statistical signi-ficance of the difference between drug treated and non treated cells was determined by the two sample t-test or Welch's test based on the results of tests for equal variance. 

RESULTS 

TMCC-1 cells were grown as an adherent line. In the repeated experiment, 2 x 106 TMCC-1 cells sus-pended in 10ml of RPMI1640 culture medium without any drugs in a 25 cm2 flask increased from 1.3+-0.2 x 107 to 1.6+-0.2 x 107 cells per flask after 7 days of culture. On day 7, cells were confluent in the flask. CA125 release of control cells on the last day varied from 8.2+-0.4x102 to 1.3+-0.1x103 U/ml and its viability was found to be 85.3+-7.0% by Trypan blue staining on the final day (day 5). 
The effect of 6 drugs on the number of cells, CA125 release on the final day and binding of 145-9 MoAb are summarized in Table 1. Among the 6 drugs ex-amined, the effect of DEX was of interest. CA125 release from TMCC-1 cells was suppressed by DEX at a concentration of 10-7 M, but CA125 release per cell was slightly increased since the cell growth was also suppressed. The binding of 125I-labeled 145-9 MoAb was significantly increased (Fig. 1). Cell viability was 89.3+-7,1% following 10-7 M DEX treatment with no significant difference from control cells, NaB (10-4 M) increased both CA125 release and 145-9 MoAb binding to TMCC-1 cells. VD3 did not increase the CA125 concentration of culture supernatant. However, because cell growth was sup-pressed by 10-7 M of VD3, CA125 release per cell seems to have increased. At the same time, VD3 (10-8 M) significantly increased 145-9 MoAb bind-ing to TMCC-1 cells (Fig. 2). The effect of RA on TMCC-1 cells was similar to that of VD3. RA at a concentration of 10-5 M increased the CA125 release per cell, and a lower concentration (10-6 M) of RA increased the binding of 125I-labeled 145-9 MoAb to the cells. IFN-r was strongly cytostatic or toxic to TMCC-1 cells. IFN-r (10U/ml) decreased both CA125 release and MoAb binding to TMCC-1 cells. The cell viability after treatment with 10 U/ml of IFN-r was 77.9+-13.8% and it was below 50% at higher concentrations of IFN-r. DbcAMP slightly increased the CA125 concentration in the culture supernatant, but did not significantly increase 145-9 MoAb binding. 
The number of binding sites of non treated TMCC-1 cells was 4.2x105/cell and the affinity constant was 1.4 x 10-9 M-1. The binding site was increased to 5.9 x 105/cell by 10-6 M of DEX, and to 6.4x105 by 10-7 M of VD3. Affinity constants were not significant]y changed by these drugs. The increase in the number of binding sites is compatible with increased binding of 145-9 MoAb to TMCC-1 cells by these drugs. 
The effects of combination treatment are sum-marized in Table 2. DEX (10-7 M) was included in all groups and VC3 (10-8 M) or RA (10-6 M) was added or both. Compared to non-treated cells, CA125 release was suppressed by the combination of DEX and VD3, and not significantly affected by the combination of DEX and RA, but increased by the combination of DEX, VD3 and RA. Both CA12S release per cell and the binding of 145-9 MoAb was increased by all of these three combinations (Fig. 3), but these effects do not seem to be significantly different from each other or from treatment with DEX. VD3 or RA alone. 

DISCUSSION 

Because early detection of ovarian cancer is not easy. and ovarian cancers are highly metastatic, the application of radioimmunodetection and therapy by means of CA125 antigen, which is highly expressed in ovarian cancer, is expected to play an important role. Various authors have already reported the administration of radiolabeled anti-tumor MoAbs in human studies.10-12 Although the concentration of MoAbs in xenografts of nude mice was as high as 10%. of the injected dose per gram of tumor,13-15 it was below 0.01 % in clinical human studiesro and it does not seem easy to reach sufficient radiation doses for treatment. Generally, tumor tissue has been pro-ven to have some heterogeneity of antigen expres-sion.16,17 That is, some cells of one tumor lesion highly express the antigen, but other do not express it at all or express it at low levels. Though MoAbs themselves are highly specific to their antigens, the heterogeneity of antigen expression within tumor tissues may offset the efficacy of MoAb accumula-tion. Increasing the antigen density in tumors will augment the tumor accumulation of radiolabeled MoAbs and result in successful tumor detection and targeting therapy by radiolabeled MoAbs. Several drugs such as DEX,18 NaB,19,20 dbcAMP,21 RA,20 and IFN-r22-25 have been reported to regulate the expression of tumor associated antigen. To improve the targetting to the CA125 antigen, we investigated the effects of these drugs on the CA125 antigen ex-pression and anti-CA125 MoAb binding to the cells. 
The binding of 125I-labeled MoAb to TMCC-1 cells increased after treatment with DEX, NaB. VD3 and RA. Of interest was the effect of DEX. The bind-ing of 125I-labeled MoAb increased and CA125 re-lease per cell was slightly increased. In other cell lines studied in the previous study,4 CA125 release and binding in the MoAb to cells were suppressed or not affected by DEX and changes in the surface expression of CA125 correlated well with changes in CA125 release. It has been reported that CA125 ex-pression of OVCA433 cells, a nonmucinous epithelial ovarian cancer, was suppressed by DEX.18 Therefore the effect of DEX on CA125 antigen expression and ,on the binding of radiolabeled MoAbs to cells is not always the same and it depends on the cell line. As Masuho et al. have suggested, CA125 antigen release may be associated with several different mecha-nisms,26 and the manner of regulation of CA125 expression by DEX may differ among cell lines. The affinity constant of the anti-CA125 MoAb in relation to TMCC-1 cells after DEX treatment was not affected, but the number of binding sites increased approximately 40%¥ This is compatible with the change in the binding of 1251-1abeled anti-CA125 antibody. 
Another noticeable finding of this study was the effect of VD3. VD3, is investiaged as one of the agents for differentiation therapy.27 As far as we know, this is the first report on the effect of VD3 in enhancing CA125 expression and MoAb binding to cells. The mechanism of CA125 modulation by VD3 is unknown, and awaits further investigation. 
Beside its effect of inducing cell differentiation. RA has other regulatory effects on cell proliferation, antigen expression, drug sensitivity and oncogene ,expression.28,29 As for CA125 expression. Langdon et al. have reported a reduction in the percentage of cells staining with OC125 after treating PE04, an ovarian carcinoma cell line, with RA.20 In this study, the binding of 125I-labeled 145-9 MoAb was in-creased by an average of 30% by treatment with as low a concentration of RA as 10-6 M. Therefore, the effect of RA on CA125 expression also seems to differ among cell lines. 
NaB, a 4 carbon fatty acid, is a classic differentiator which induces differentiation of malignant cells, and changes cell growth rate, enzyme activity, and glyco-protein synthesis.30 Some of the tumor associated antigen expressions of mammalian cells are also known to be modulated by NaB.19,20 In this study, CA125 expression and binding of 125I-labeled MoAb to TMCC-1 cells was increased by 10-4 M of NaB. In our previous study, 10-4 or 10-5 M of NaB had the same effect in 3 of the 8 cell lines examined. It is possible that this agent enhances MoAb binding to malignant cells presenting the CA125 antigen. Of interest was that VD3. RA and NaB, all of which have differentiating effects, affected CA125 antigen expression and MoAb binding to the cells. This phenomenon may indicate a relationship between CA125 antigen and the cell differentiation process. 
DbcAMP is also reported to induce cell differentiation31,32 and CA125 expression in several ovarian cancer cell lines also known to be increased by dbcAMP.21 DbcAMP mildly increased CA125 re-lease from TMCC-1 cells, but did not significantly increase 145-9 MoAb binding. 
IFNs are known to regulate major histocompa-tibility antigens,33 carcinoembryonic antigen34 and TAG72 antigen.35 Augmented localization of radiolabeled MoAb in tumors due to IFN-r was also reported.23,25 Marth et al. have reported that CA125 expression was increased by IFN-r in 2 of 3 ovarian cancer cell lines.36 In this study, IFN-r decreased antigen expression and antibody binding. But, because IFN-r was highly toxic to TMCC-1 cells, suppression of CA125 expression and MoAb binding by IFN-r may be associated with cytotoxic processes. 
The effects of combined treatment with DEX, VD3 and RA were not significantly different from single drug administration. But a combination of these drugs may be effective in increasing the total amount of antigens to be targetted in clinical radio-immunoimaging or radioimmunotherapy because malignant tissues have heterogeneity in their antigen expression and all cells do not always react to a particular drug in the same way. 
In conclusion, we have demonstrated that drugs changed the CA125 antigen expression of TMCC-1 cells and 125I-labeled MoAb binding to them in vitro. These drugs may be useful for application in vivo to increase the deposit of radiolabeled MoAbs in the tumor. 

REFERENCES 

1. Sands H : Experimental studies of radioimmuno-detection of cancer: An overview. Cancer Res 50 (Suppl) : 809-813, 1990 
2. Larson SM: Clinical radioimmunodetection, 1978-1988 : overview and suggestions for standardization of clinical trials. Cancer Res 50 (Suppl) : 892-898, 1990 
3. Schlom J, Hand PH, Greiner JW, et al : Innovations that influence the pharmacology of monoclonal antibody guided tumor targeting. Cancer Res 50 (Suppl) : 820-827, 1990 
4. Nakai T, Endo K, Hosono M, et al : Drug effects on CA125 antigen expression and antibody binding to cancer cells. Int J Cancer 48 : 463-467, 1991 
5. Sakamoto M, Okabe K, Negishi Y, et al: Cyto-biological and immunological characterization of a newly established cell line (TMCC-D derived from human uterine cervical adenocarcinoma. J Tokyo Med College 46: 925-936, 1988 
6. Matsuoka Y, Nakashima T, Endo K, et al : Recogni-tion of ovarian cancer antigen CA 125 by murine monoclonal antibody produced by immunization of lung cancer cells. Cancer Res 47: 6335-6340, 1987 
7. Saga T, Endo K, Nakashima T, et al : Construction of an immunoradiometric assay for ovarian cancer associated antigen CA125 recognizing different anti-genic determinant. Acta Obstet Gynecol Scand 69 : 175-181, 1990 
8. Endo K, Matsuoka Y, Nakashima T, et al : Develop-ment of a new sensitive immunoradiometric assay for CM125 : Mixed use of two monoclonal antibodies reactive with separate epitopes. J Tumor Marker Oncol 3: 65-71, 1988 
9. Scatchard G : The attraction of proteins for small molecules and ions. Ann NY, Acad Sci 51 : 660-672,1949 
10. Chatal JF, Saccavini J, Gestin J, et al : Biodistribution of indium-111-labeled OC125 monoclonal antibody intraperitoneally injected into patients operated on for ovarian carcinomas. Cancer Res 49 : 3087-3094, 1989 
11. Hnatowich DJ, Chinol M, Siebecker DA, et al: Patient biodistribution of intraperitoneally adminis-tered Yttrium-90-labeled antibody. J Nucl Med 29 : 1428-1434, 1988 
12. Haisma HJ, Moseley KR, Battaile AI, et al : Bio-distribution, pharmacokinetics and imaging of 131I-labelled OC125 in ovarian cancer. Int J Cancer (Suppl): 2: 109-113, 1988 
13. Colcher D, Keenan AM, Larson SM, et al: Pro-longed binding of a radiolabeled monoclonal antibody (B72.3) used for the in situ radioirnmunodetection of human colon carcinoma xenografts. Cancer Res 44 : 5744-5751, 1984 
14. Sakahara H, Endo K, Nakashima T, et al : Locali-zation of human osteogenic sarcoma xenografts in nude mice by a monoclonal antibody labeled with radioiodine and indium-111. J Nucl Med 28: 342-348, 1987 
15. Sakahara H, Endo K, Koizumi M, et al : Relationship between in vitro binding activity and in vivo tumor accumulation of radiolabeled monoclonal antibodies. J Nucl Med 29 : 235-240, 1988 
16. Hand PH, Nuti M, Colcher DL, et al : Definition of antigenic heterogeneity and modulation among human marnmary carcinoma cell populations using monoclonal antibodies to tumor-associated antigens. Cancer Res 43 : 728-735, 1983 
17. Fargion S, Carney D, Mulshine J, et al : Heterogeneity of cell surface antigen expressin of human small cell lung cancer detected by monoclonal antibodies. Cancer Res 46: 2633-2638, 1986 
18. Karlan BY, Amin W, Casper SE, et al : Hormonal regulation of CM125 tumor marker expression in human ovarian carcinoma cells: Inhibition by gluco-corticoids. Cancer Res 48: 3502-3506, 1988 
19. Cook JR, Schwartz CE, Fausel ED, et al : Effect of sodium butyrate on a-fetoprotein gene expression in rat hepatoma cells in vitro. Cancer Res 45 : 3215-3219, 1985 
20. Langdon SP, Hawkes MM, Hay FG, et al : Effect of sodium butyrate and other differentiation inducers on poorly differentiated human ovarian adenocarcinoma cell lines. Cancer Res 48: 6161-6165, 1988 
21. Ishiwata, I, Ishiwata C, Nozawa S, et al : Cal25 pro-duction by gynecologic tumors in vitro and its modulation induced by dibutyl cyclic adenosine monophosphate. Asia Oceania J Obstet Gynaecol 12: 285-290, 1986 
22. Rowlinson G, Balkwill F, Snook D, et al : Enhance-ment by r-interferon of in vivo tumor radiolocaliza-tion by a monoclonal antibody against HLA-DR antigen. Cancer Res 46: 6413-6417, 1986 
23. Greiner JW, Guadagni F, Noguchi P, et al : Re-combinant interferon enhances monoclonal antibody-targeting of carcinoma lesions in vivo. Science 235: 895-898, 1987 
24. Marth C, Fuith LC. Bock G, et al : Modulation of ovarian carcinoma tumor marker CM125 by r-interferon. Cancer Res 49 : 6538-6542, 1989 
25. Wosenblum MG, Lamki LM, Murray JL, et al : Interferon-induced changes in pharmacokinetics and tumor uptake of 111In-labeled antimelanoma anti-body 96.5 in melanoma patients. J Natl Cancer Inst 80: 160-165, 1988 
26. Masuho Y, Zalutsky M, Knapp RC, et al : Interaction of monoclonal antibodies with cell surface antigens of human ovarian carcinomas. Cancer Res 44: 2813-2819, 1984 
27. Lotan R, Francis GE, Freeman CS, et al : Differen-tiation therapy. Cancer Res 50 : 3435-3464, 1990 
28. Doyle LA, Giangiulo D, Hussain A, et al : Differentia-tion of human variant small cell lung cancer cell lines to a classic morphology by retinoic acid. Cancer Res 49: 6745-6751, 1989 
29. Yung, WKA, Lotan R, Lee P, et al : Modulation of growth and epidermal growth factor receptor activity by retinoic acid in hurnan glioma cells. Cancer Res 49: 1014-1019, 1989 
30. Kruh J : Effects of sodium butyrate, a new pharma-cological agent, on cells in culture. Mol Cell Biochim 42 : 65-82, 1982 
31. Nagamine S, Yanagawa T, Bando T, et al : Induction of cells with phenotypic features of neuronal cells by treatment with dibutyryl cyclic adenosine 3',5'-monophosphate in a human parotid gland adeno-carcinoma cell line in culture. Cancer Res 50 : 6396-6404, 1990 
32. Laskin DL. Beavis AJ. Sirak AA, et al : Differentiation of U-937 histiocytic lymphoma cells towards mature neutrophilic granulocytes by dibutyryl cyclic adeno-sine-3',5'-monophosphate. Cancer Res 50 : 20-25, 1990 
33. Gresser I : Interferons and the irnmune system. Interferon 2: 132-113, 1984
34. Attallah AM, Needy CF, Noguchi PD, et al: En-hancement of carcinoernbryonic antigen expression by interferon. Int J Cancer 24 : 49-52, 1979 
35. Greiner JW, Hand PH, Noguchi P, et al : Enhanced expression of surface tumor-associated antigens on human breast and colon tumor cells after recombinant human leukocyte oc-interferon treatment. Cancer Res 44: 3208-3214, 1984 
36. Marth C, Muller-Holzner E, Greiter E, et al : r interferon reduces expression of the protooncogene c-erbB-2 in human ovarian carcinoma cells. Cancer Res 50 : 7037-7041, 1990