Expression of Cell-Mediated Immunity and Blocking Factor Using a New Line of Ovarian Cancer Cells in Vitro1

Home , Cervical cancer

[CANCER RESEARCH 39, 1185-1 191, April 1979] 0008-5472/79/0039-0000$02.00 Expression of Cell-mediated Immunity and Blocking Factor Using a New Line of Ovarian Cancer Cells in Vitro1

Roland A. Pattillo,2 Michael T. Story, and Anna C. F. Ruckert

Department of Gynecology and Obstetrics, Medical College of Wisconsin, Milwaukee, Wisconsin 53226

ABSTRACT serous and mucinous cystadenocarcinoma exhibited inhi bition of leukocyte migration to autologous as well as A cell-mediated cytotoxicity test, quantitated by postla homologous tissue antigens derived from these tumors. beling with tritiated thymidine, was used to assess immune Their studies indicated that common antigens were present reactivity of cancer patients to the HeW cell line derived in serous and mucinous cystadenocarcinoma but not pres from serous cystadenocarcinoma of the ovary. Lympho ent in antigens derived from other gynecological neo cytes from 71.4% of serous and mucinous cystadenocarci plasms, including squamous cell carcinoma, leiomyosar noma patients demonstrated a cytotoxic response towards coma, and mixed mesodermal tumor of the uterus. Studies the HeW cells, whereas no reactivity was observed towards in which antisera were prepared against tumor extracts target cells derived from nonovarian cancer. These obser from patients with ovarian carcinoma (3-5, 11, 12, 22, 23) vations indicated that the HeW cells express tumor-associ or from ovarian cancer cells in long-term culture (24, 25) ated antigen. In some patients bearing similar tumors, leave little doubt that human ovarian cancer cells express cytotoxicity was blocked by ascitic fluid from other patients unique antigens. Although the specificity of xenogenic with cystadenocarcinoma. In addition, antigen obtained antisera for ovarian carcinoma has not been established in from the spent culture fluid of HeW cells exhibited blocking all cases, at least one common antigen appears to be activity in a typical dose-response fashion, suggesting that associated with serous and mucinous cystadenocarcinoma. blocking factor may be free tumor-associated antigen or an The antigen appears to be a high-molecular-weight glyco antigen-specific suppressor molecule. Thus, blocking of proteinunrelatedto blood group substances, carcinoem the lymphocytotoxic response of cystadenocarcinoma pa bryonic antigen, a-fetoprotein, and histocompatibility anti tients towards HeW cells may be utilized to monitor the gen (4, 5). After appropriate absorption, the antiserum isolation of ovarian carcinoma-associated antigen. failed to react with normal ovarian tissue, benign ovarian cystadenoma, ovarian granuloma, and unrelated types of INTRODUCTION carcinoma (12, 22). It remains to be determined if xenogenic antiserum to ovarian carcinoma-derived antigen will be of The high mortality rate among patients with epithelial value in early detection of neoplasia. In addition, experi carcinoma of the ovary has prompted a search for specific mental approaches utilizing antisera to tumor cells do not markers (TAA3) for ovarian carcinoma. It is anticipated that lend themselves to studies of blocking of lymphocytotoxic assays for TAA would allow early diagnosis of ovarian ity by serum factors (13, 15, 27), which has been of interest cancer and provide a means of monitoring patient therapy as a possible explanation for in vivo escape by tumor cells in a manner analogous to that of the HCG marker associ from immune surveillance. ated with trophoblastic disease. In this study, a recently established cell line (HeW) from The approaches used in identifying ovarian TAA include: serous cystadenocarcinoma of the ovary was used as a (a)invitrotestsforcell-mediatedimmunityincludingleu source of target cells in a LMC assay. The cytotoxicity kocyte migration (6, 26) and LMC (8, 9, 14); (b) production response of lymphocytes from some patients was blocked of xenogenic antisera to tumor extracts and demonstration by autologous serum and ascitic fluid from cystadenocar of immune reactivity by immunoprecipitation and immuno cinoma patients, as well as by antigens recovered from the fluorescence (3-5, 11, 12, 22-25); and (C) isolation of TAA culture fluid from HeW cells. These observations support and associated antibody from immune complexes re the hypothesis that serum blocking activity may be due to covered from ascitic fluid (10). free TAA or an antigen-specific suppressor molecule (29) Despite recent controversy regarding the immune speci and suggest that the blocking of cytotoxicity may be used ficity of LMC testing (30, 36, 39), immune reactivity to TAA to monitor the purification of TAA. in patients with ovarian carcinoma (8, 9, 14) and the observation that histologically similar tumors share com MATERIALS AND METHODS mon antigens (14) are supported by other experimental approaches. Chen et al. (6) reported that patients with Establishmentof the HeW Cell Line. HeW was derived from a 59-year-old Caucasian female who presented with a

1 Supported by The American Cancer Society (Milwaukee Division) and serous cystadenocarcinoma of the ovary widely metastatic the Cudahy Fund. to the peritoneal cavity in the presence of 12 liters of ascitic 2 To whom requests for reprints should be addressed.

3 The abbreviations used are: TAA, tumor-associated antigens; HCG, fluid. Within 72 hr after explantation of a solid tumor human chorionic gonadotropin ; LMC, lymphocyte-mediated cytotoxicity; fragment in culture, migration of tumor cells from the BSS, Gey's balanced salt solution; FCS, fetal bovine serum; G6PD, glucose primary explant was observed. Mechanical subculture was 6-phosphate dehydrogenase. Received June 12, 1978; accepted December 28, 1978. accomplished 16 days after explantation. Tumor cells were

APRIL 1979 1185

initiated in a medium consisting of 30% BSS, 50% Way 8 7 mouth's 752/1 , and 20% human umbilical cord serum 6 (3520). The cells have been adapted to 3520 medium sup U C 5 n a, plemented with 20% FCS and are now dispersed with lOx 4 trypsin:EDTA. Flasks (25 sq cm; Falcon, Oxnard, Calif.) are subcultured at weekly intervals on a 1:2 split ratio. At @3 â€˜flnH present, the line has been sustained through 75 passages. @flHH@THH@ All tissue culture media, sera, and enzymes, with the 6970727374757879808182838485868789929698 exception of human umbilical cord serum, were obtained Chromosome Counts from Grand Island Biological Co. , Grand Island, N. Y. Chart 1. HeW chromosomal number frequency. Chromosomes were pre Human umbilical cord serum was obtained after normal pared from cells at passage 19 by hypotonic swelling, fixed with methanol: delivery from the obstetrical service at this institution. acetic acid (3:1), and stained with 10% Giemsa in BSS. Seventy cells were analyzed. The omission of consecutive numbers on the abscissa is for the Phase-contrast microscopic visualization of HeW cells sake of brevity. No cells were found with a chromosomal count correspond (Fig. 1) reveals 2 major cell types. A population of large ing to the missing number. differentiatedcellscontaininginclusionvacuoles staining positively for mucoprotein was seen as the predominant pattern in these cultures. The second cell type is repre Table 1 sented by tightly packed islands and sheets of small angular PatientstestedHistological cells with uniform nuclear size and distribution. Cells inter attionGroupPatientstagetimeclassifica FIGOTreatment mediate in the 2 morphological patterns described above of testing suggest a progression from the smaller undifferentiated to Serous or mucinous IV None larger differentiated cell type. cystadenocarcino 2 iv None The cells are aneuploid with a modal number of 87 but mas 3 bIB Irradiation 4 IV None with a wide distribution in chromosomal number about the 5 Ill None mode (Chart 1). 6 IV None The cells have the â€œB-typeG6PD isoenzyme pattern4 and 7 ill Irradiation are free of Mycoplasma. 8 bIB None 9 None Radioimmunoassay and gel filtration analysis of the cul 10 None ture fluid from HeW cells for HCG and the cxand f3subunit, 11 IV None as previously described (2, 20), revealed low levels of 12 Ill None material indistinguishable in size and immune reactivity 13 None @ from highly purified HCG 14 0 None 15 lIB irradiation Other cells used in this study to assess patient's cell 16 IV Irradiation mediated immune response included the ElCo (38), CaSki 17 IV None (34), and DoT (33) lines. The ElCo line was derived from an 18 None American Indian female with infiltrative ductal adenocarci 19 None 20 IV None noma of the breast and has been reported to have the â€œAâ€•- 21 III None type G6PD isoenzyme pattern and share karyological markers with HeLa (28). The CaSki and DoT lines were Endometrioid adeno B III None derived from epidermoid carcinoma of the cervix. CaSki carcinomas 2 III Chemotherapy cells have been reported to express TAA (34). 3 III None 4 IC None Patients.The cancerpatientstestedfor LMC are shown 5 Ill None in Table 1 under histological classification, FIGO staging of ovarian carcinoma, and treatment at the time of testing. Poorly differentiated C 1 III None Normal donor lymphocytes were obtained from healthy adenocarcinomas 2 III None 3 III Chemotherapy laboratorypersonnel. 4 IV Chemotherapy Preparationof LymphocyteSuspensions.Fifteenml of 5 IV None venous blood, collected in preservative-free heparin, were 6 IV Chemotherapy dilutedwith an equal volume of sterile0.9%sodium chlo ride solution and layered on a Ficoll-sodium metrizoate Clear cell adenocarci D 1 III None noma density gradient (specific gravity, 1.08) and centrifuged at 1200 rpm and at 4Â°for 40 mm (Model PR-6; International Benign cystadenomas E 1 None Equipment Co. , Needham Heights, Mass.). The lympho 2 None cyte-rich layer was recovered and washed 3 times with Rosweil Park Memorial Institute Tissue Culture Medium Breast carcinomas F 1 None 2 None 1640. Lymphocytes were counted with the aid of a hemo cytometer, centrifuged, and brought to 2 x lO@cells/mi in Cervicalcarcinomas G 1 None 3520 medium with 20% heat-inactivated FCS containing 2 None 3 None 4 None 4 W. A. Nelson-Rees, personal communication. 5 Irradiation 5 R. 0. Hussa, unpublished observations.

1186 CANCER RESEARCH VOL. 39

penicillin (50 units/mI) and Streptomycin (50 pg/mi). of 6 days of maintaining cells on serum-free medium) was A second 7-mi sample of blood was collected without thawed, pooled, and centrifuged at 105,000 x g and at 4Â° anticoagulant. The serum was recovered after centrifuga for 90 mm (Model L265B, SW 27 rotor; Beckman Instru tion, heat inactivated, and added (20%) to 3500 medium for ments, Inc. , Fullerton, Calif.). The supernatant was concen use inthe blockingfactorassay. trated by molecular filtration (UM 20 membrane; Amicon Ascitic fluid was aseptically obtained from a patient with Corp., Lexington, Mass.) and sterilized by filtration. Protein cystadenocarcinoma and stored frozen at â€”70Â°.Afterthaw determinations, as described by Oyama and Eagle (31), ing, the fluid was clarified by centrifugation at 12,000 rpm were made on acid precipitates (32) dissolved in 1 N NaOH and at 4Â°for 40 mm (Model RC-2B; Ivan Sorvall Inc., and neutralized with 1 N HCI. Norwalk, Conn.). The Supernatant was filtered through Whatman No. 1 paper and sterilized by filtration (Millipore RESULTS Corp.,Bedford, Mass.).Asciticfluidwastestedforblocking activity by addition to FCS-containing medium with a pro Cell-mediated Reactivity. Table 2 presents data from portional decrease in the concentration of the FCS compo incubations of ovarian cancer-derived cells (HeW) with nent. The concentration of ascitic fluid neverexceeded 10% lymphocytes from patients with ovarian cancer, other types in the incubation medium. Both patient and normal donor of cancer, or healthy normal donors. In total, assays were lymphocytes tested in the same assay received identical performed with lymphocytes from 8 different normal do amounts of ascitic fluid-supplemented medium. nors, 21 patients diagnosed as having serous or mucinous LMC Test. The assayconsistedof incubationof target cystadenocarcinoma, and 14 patients with other histologi cells with donor lymphocytes. A pulse procedure with cal types of ovarian tumors, including endometrioid ade tritiated thymidine was used to enumerate remaining target nocarcinoma, poorly differentiated adenocarcinoma, clear cells after the incubation period. The procedure does not cell adenocarcinoma, and benign cystadenoma. In addi distinguish between cell lysis, altered incorporation of triti tion, lymphocytes from 5 patients with cervical cancer and ated thymidmne, or enhanced detachment of target cells as 2 patients with breast cancer were tested for cell-mediated a result of incubation with lymphocytes. reactivity. Target cells (10@cells/well) were incubated in 8-well The 8 normal donors were used as a source of control chambers (Lab-Tek Products, Naperville, Ill.) at 37Â°in an lymphocytes a total of 42 times. None showed significant atmosphere of 5% CO2in air. After 16 to 24 hr of incubation, cytotoxicity to the HeW cells. On the other hand, lympho the medium was removed and replaced with fresh 3520 cytes from 15 of 21 serous and mucinous cystadenocarci medium, 3520 containing 1O@lymphocytes, or 3520 with noma patients showed cytotoxicity to the HeW cells. Of the 20% patient's serum containing l0@ lymphocytes. After 15 patients whose lymphocytes showed positive cell-me incubation for 18 to 24 hr, the chambers were washed 4 diated activity, 7 showed blocking of cytotoxicity when their times with BSS to remove the lymphocytes. The remaining lymphocytes were incubated with target cells in autologous target cells were incubated for 2 hr in 3520 medium with serum-containing medium. As a testfor possiblenonspe [methyl-3H]thymidmne (10 @Ci/ml,5 to 30 Ci/mmol; Amer cific effects of autologous serum in the assay, the response sham/Searle Corp., Arlington Heights, Ill.). The pulse me of normal donor lymphocytes incubated in FCS-containing dium was removed, the cells were washed twice with BSS medium was compared with that of normal donor lympho containing excess unlabeled thymidine, and frozen. After cytes incubated in medium containing autologous serum. thawing, the amount of radioactivity incorporated into acid In 12 separate assays, no significant differences in normal precipitable material was determined as previously de donor lymphocyte response were observed (data not scribed (21). Determinations were made from triplicate or shown). quadruplicate wells plated with target cells. Values were In incubations of lymphocytes from patients with other expressed as percentage of radioactivity incorporated by histological types of ovarian cancer (Table 2, Groups B to target cells incubated in the absence of lymphocytes. Cyto E), 6 of 14 differed from the response seen with normal toxicity of patient lymphocytes was assessed by comparing donor lymphocytes. These included lymphocytes from 2 of values obtained from incubations of patient lymphocytes 5 patients with endometrioid adenocarcinoma, 4 of 6 pa versus normal donor lymphocytes incubated with target tients with poorly differentiated adenocarcinoma, 0 of 1 cells in the same assay. Blocking of cytotoxicity was as patient diagnosed as clear cell adenocarcinoma, and 0 of 2 sessed by comparing values obtained from incubation of patients with benign cystadenoma. None of these patients patient lymphocytes in FCS-containing medium versus au showed serum blocking activity. In addition, cytotoxicity tologous serum-containing medium, with target cells in the but no serum blocking activity was observed with lympho same assay. Significance was assessed by Student's t test, cytes from only 1 of 5 patients with cervical carcinoma and @ withp 0.05 considered significant. 0 of 2 patients with breast carcinoma. Preparation of Soluble Antigens from HeW Cells. HeW Lack of a Cell-mediatedReactivityof Serous or Mud cells were cultured in 75-sq cm flasks in 3520 medium for 6 nousCystadenocarcinomaPatientsto NonovarianCard days. The cells were washed 3 times with BSS and incu noma-derived Cells. Table 3 compares the response of bated in medium devoid of serum for 2 days. The spent serous or mucinous cystadenocarcinoma patient lympho culture fluid was removed and centrifuged (1500 rpm and 4Â° cytes to the HeW cells and to 3 nonovarian cancer-derived for 20 mm), and the supernatant was frozen. The cells were cell lines established in our laboratory. Whereas lympho returned to 3520 medium for 1 to 2 days, and collection of cytes from 15 of 21 serous or mucinous cystadenocarci the spent culture fluid was repeated. The fluid (from a total noma patients were cytotoxic to the HeW cells, lymphocytes

APRIL 1979 1187

lympho donor bym cytes with autobo Group'@'Patientâ€•NormalserumSignificanceA1122.6lymphocytesSignificancePatient phocytesSignificancePatient gous [email protected] Â± 5,7bND@ (I26.4 Â± 7.5+(â€˜51 .9 Â± 11.8ND3100.5Â±Â± 3.8ND36.2 Â± 5.5ND24.4 Â± 4.1+4100.5 6.2ND8.2Â± 1.5+68.0Â± 3.8ND5103.6Â±Â± 6.2ND7.2 Â± 1.5+17.0 Â± 12.5+692.6Â± 5.3ND10.1 Â± 1.2+66.0Â± 5.6+792.6Â± 7.9ND1.8Â± 0.5+89.4Â± 0.1ND8123.4Â±7.9ND1.5Â± 0.2+0.8Â± 9.0+987.0 7.7ND4.1 Â± 1.3+141.9Â± 4.6ND1096.1 Â±15.9ND67.3 Â± 0.8ND50.0 Â± 5.5ND1178.4Â± Â± 8.1ND27.1 Â±21.5ND79.5 Â± 2.8ND12107.3Â± 7.1ND6.8Â± 2.5+14.3Â± 1.1ND13104.8 11.8ND37.9Â± 6.4+18.3Â± 2.9ND1491 Â±16.1ND78.9 Â± 3.6ND60.4 Â± 5.5+g1591.3.3 Â± 4.1ND101 .5 Â± 5.5ND69.3 Â± 19.1ND16111.9Â±Â± 4.1ND15.7 Â± 1.7+69.3 Â± 3.6+111.9Â±21.8+1787.6Â±5.6ND17.5Â± 1.8+1866.8Â± 8.3ND15.8Â± 3.1+53.0Â± 2.3ND1980.5 7.2ND23.2Â± 2.9+16.0Â± .0ND2089.2Â± Â± 5.8ND23.0 Â± 6.5+12.6 Â± 1 5.0ND2179.2 5.0ND29.4Â± 1.1+22.5Â± Â± 2.2ND70.1 Â± 3.7ND54.5 Â± 2.9ND

B 1 98.7Â± 7.7 ND 51.1 Â± 1.7 + 36.2 Â± 4.8 ND 2 119.6Â± 6.6 ND 40.1 Â±24.1 + 76.5 Â± 16.6 ND 3 80.5Â± 7.0 ND 75.1 Â± 2.7 ND 55.6Â± 3.2 ND 4 111.9Â± 5.6 ND 100.7Â±32.2 ND Notdetermined 5 78.4 Â± 7.1 ND 92.4 Â±16.1 ND 91.1 Â± 7.5 ND

C 1 87.3Â± 1.1 ND 38.8 Â± 4.7 + 2 127.2Â± 8.7 ND 90.9Â± 1.7 + 96.9Â±11.6 ND 3 117.3Â± 2.4 ND 88.1 Â± 8.8 ND 62.1 Â± 4.2 ND 4 107.1Â± 7.6 ND 15.7Â± 3.2 + 16.9Â± 1.3 ND 5 96.1 Â± 8.1 ND 98.4 Â±26.3 ND 58.6Â± 3.4 ND 6 89.3Â± 1.1 ND 55.7 Â± 7.7 +30.0Â± NDND100.7Â±46.3Â±7.33.2ND D 1 68.3Â± 1.8 ND 52.6Â± 8.1 ND 64.8Â± 6.9

E 1 118.7Â± 6.9 ND 104.7Â± 4.5 ND 2 119.6Â± 6.6 ND 93.3 Â±15.0 ND 90.7ND89.2Â± Â±7.07.0ND F 1 110.8Â± 9.0 ND 106.4Â± 4.9 ND 2 91.2 Â± 4.0 ND 131.4Â± 4.8 ND 81.6Â±1.714.4ND ND

G 1 135.4Â± 8.7 ND 48.1 Â± 3.7 + Not determined 2 135.4Â± 8.7 ND 125.0Â± 4.1 ND Not determined 3 132.1Â±11.7 ND 125.8Â±35.4 ND 121.7Â± 5.4 ND 4 96.2Â± 8.1 ND 69.3 Â±13.5 ND 46.6Â± 2.5 ND 5 78.5 Â± 7.2 ND 74.0 Â± 4.1 ND 51.0Â± 6.4 ND a See Table 1. b Mean Â± SE. C Differs significantly (p 0.05) from target cells incubated without lymphocytes. d Not significantly different. e Differs significantly (p 0.05) from target cells incubated with normal donor lymphocytes. @ I Differs significantly (p 0.05) from target cells incubated with patient lymphocytes. 9 Significantly lower than target cells incubated with patient lymphocytes.

from 0 of 9 patients and 0 of 8 patients were cytotoxic to the significantly greater than was the influence of normal donor EICo and CaSki or DoT cells, respectively. Also shown in lymphocytes tested in the same assay. Similarly, lympho Table 3 is the response of normal donor lymphocytes to the cytes from 4 of 4 different cervical cancer patients were nonovarian cancer-derived target cells. It is evident that cytotoxic to the epidermoid cervical cancer-derived cells normal donor lymphocytes frequently influenced tritiated (CaSki or DoT). On the other hand, the response of serous thymide incorporation by the ElCo, CaSki, and DoT cells. and mucinous cystadenocarcinoma patient lymphocytes to Nevertheless, lymphocytes from 10 of 10 different breast the nonovarian cancer-derived cells was no different from cancer patients showed a cytotoxic response which was normal donor lymphocytes tested in the same assay. These

1188 CANCER RESEARCH VOL. 39

findings suggest that serous and mucinous cystadenocar compared to 125.9 Â±18.8% in FCS-containing medium. cinoma patient lymphocytes are selectively cytotoxic to the These values were not significantly different, demonstrating HeW cells. that addition of ascitic fluid did not influence tritiated Blocking Activity in Ascitic Fluid. Since some of the thymidine incorporation by HeW cells in the presence or cystadenocarcinoma patients demonstrated serum block absence of normal donor lymphocytes. ing activity (Table 2), it was of interest to determine if Blocking Activity In a Crude HeW-derived Antigen Prep blocking activity was also present in ascitic fluid from aration. Since soluble TAA may be responsible for blocking patients with the same histological type of ovarian cancer. activity in patient serum and ascitic fluid, it was of interest Lymphocytes from 3 different ovarian cancer patients were to test an antigen preparation from HeW cells for possible retested for cytotoxicity to the HeW cells and for blocking blocking activity. A crude soluble antigen preparation was by ascitic fluid. Ascitic fluid was obtained from a patient obtained from the culture fluid from HeW cells as indicated (see Table 2, Group A, Patient 1) with advanced disease. in â€œMaterialsandMethods.â€•The preparation was found to The findings (Table 4) indicated that the cell-mediated have blocking activity when tested in 3 separate assays cytotoxicity of all 3 patients was blocked by ascitic fluid (data not shown). The dose-response of blocking activity in (Chart 2). the crude HeW-derived preparation tested in an assay with Since ascitic fluid appeared to be a good source of lymphocytes from a cystadenocarcinoma patient (Table 2, blocking factor, it was of interest to determine the concen Group A, Patient 16) is shown in Chart 3. A 50% reduction tration of ascitic fluid required for blocking activity. The in blocking activity was seen at a concentration of approxi blocking activity of ascitic fluid from a patient with ad matelyo.12 @gproteinper assay well. vanced serous cystadenocarcinoma was determined in an The highest concentration of HeW antigen tested, 3.7 @g assay with lymphocytes from a patient with the same histo protein per assay well, appeared to be inhibitory to target logical type of ovarian cancer. The dose-response curve of cell growth. Target cells incubated in HeW antigen-supple this blocking activity was seen at a concentration of approx mented medium incorporated 83.7 Â± 12.1% of tritiated imately 0.7 @gofascitic fluid protein per assay well. HeW thymidine compared to control cells receiving FCS-contain cells incubated in ascitic fluid-supplemented medium ing medium. Target cells incubated with normal donor (tested at the highest concentration used) incorporated lymphocytes in HeW antigen-containing medium incorpo 113.8 Â± 17.8% (S.E.) of tritiated thymidine compared to rated 60.2 Â±15.3% of label compared to 95.9 Â±I 1.2% in control cells receiving FCS-containing medium. Target cells FCS-containing medium. Thus, the lower blocking index incubated with normal donor lymphocytes in ascitic fluid containing medium incorporated 121.3 Â±13.2% of label 120 3LMC Table responseof cancerpatients' and normal donors' I OC lymphocytesto ovarian cancer and nonovariancancer-derivedcell linesCell 80 line :Iso Cancer type or normal DoTSerousdonors HeW EICo CaSkior @4o or mucinous 15/2V' 0/9 0/8 cystadenocarcinomas 20 Breast carcinomas 0/2 10/10 Not determined Cervical carcinomas 1/5 1/7 4/4 0 Normaldonors 018b 15/21 5/17

@ a Number of patients whose responsediffered significantly from 50 5 005 0 that of normal donors per number of patients tested. Ascitic FluidProtein(sag/well) b Number of normal donors whose response differed signifi cantly from target cells incubatedwithout lymphocytesper number Chart 2. Dose response of ascitic fluid blocking activity. Values are the of normal donors tested. average of duplicate samples. Vertical bar, range.

Table 4 Blocking of cell-mediated reactivity by ascitic fluid mediumNo LymphocytesincubatedinFCS-containing Patients@' lymphocytesFCSlymphocytesPatient lymphocytes + GroupfluidA3106.1 No.% ascitic fluidControl Ascitic fluidFCSAscitic 17.7â€•A7Not Â±190b,c144.7 Â±12.2c Not determined20.2 Â± @127.6Â± 3.9â€•A11131.8 determined70.3 Â±6.2c 45.3 Â± 39!1 1.6 Â±45d67.8 Â± Â±26.3c112.8 Â±99C 76.7 Â±[email protected] Â±0.6â€•150.4 Â±16.0â€• a SeeTable2. b Mean Â± S.E. C Not significantly different from target cells incubated in the absence of lymphocytes in FCS-containing medium. l Differs (p 0.05) from control lymphocytes incubated in FCS-containing medium. @ (, Differs (p 0.05) from patient lymphocytes incubated in FCS-containing medium. I Not significantly different from control lymphocytes incubated in FCS-containing medium.

APRIL 1979 1189

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1979 American Association for Cancer Research. R. A. Pattillo et al. 100 possible significance of blocking factor in escape by tumor cells from immune surveillance (7, 17, 18) and the inherent 80 difficulty in interpreting LMC data from individual patients, it is essential that other approaches be found for assessing .@ 60 serum blocking activity. g40 Regardless of the nature of blocking factor (17, 18, 29), in g order to pursue this objective it is necessary to (a) deter @20 mine if blocking factor from one patient is immunologically similar to that found in another patient with the same type 0 of disease, (b) identify a good source of material with ______blocking activity, and (c) have a reliable biological endpoint @ 5 0'S 0.05 ., 6 for monitoringtheisolationofblockingfactor.The present HeWProtein(gg/well) study in part fulfills these requirements. Ascitic fluid ex Chart 3. Dose-response of HeW-derived antigen blocking activity. Values hibited blocking activity when tested against allogeneic are the mean of triplicate samples. Vertical bar, SE. lymphocytes (Table 4), suggesting that cystadenocarci noma patients have blocking factor which is immunologi seen when target cells are incubated with patient lympho cally similar or identical. The titration of ascitic fluid from a cytes in the presence of 3.7 @gHeW protein-containing patient with cystadenocarcinoma for blocking activity medium (Chart 3) appears to be a result of inhibition of (Chart 2) suggests that blocking factor in this fluid is target cell growth. present in high amounts or is active in very low concentra tions. Although ascitic fluid appears to be a good source of DISCUSSION blocking factor, it is perhaps more significant that concen trates of the culture fluid from HeW cells blocked the Cell-mediated immune testing of patients with cancer has cytotoxic response of lymphocytes from cystadenocarci provided information suggesting that their lymphocytes noma patients (Chart 3). The observation appears consist exert a cytotoxic response to autologous tumor cells or to ent with the hypothesis that blocking factor may be free tumor cells of the same histological type (13-15) and that TAA or an antigen-specific suppressor molecule (29) and factors are present in patient serum which modify lympho supports earlier reports that TAA is shed by tumor cells in cyte reactivity (13, 15, 27). However, much of what was culture (1, 34, 35). Finally, it is anticipated that the isolation thought to be reactivity to TAA may really be due to a of blocking factor will be facilitated by using spent culture reaction described as natural cytotoxicity (30, 36, 39). We fluid as a source of material rather than serum or ascitic have also observed lymphocyte reactivity in normal donors fluid since cell culture material represents a standard and in some cancer patients which is best described as source of antigen in contrast to the heterogenicity of patient natural cytotoxicity (Table 3). However, the observation that material. Efforts in this laboratory are currently directed lymphocytes from 71.4% of serous or mucinous cystadeno towards identifying common antigens shared by the HeW carcinoma patients demonstrated cytotoxicity to the ovar cells and serum and ascitic fluid from cystadenocarcinoma ian cancer-derived cells (Table 2, HeW) is not explained by patients. the phenomenon of natural cytotoxicity since lymphocytes from normal donors and from 6 of 7 nonovarian cancer patients failed to show reactivity to the HeW cells. Further ACKNOWLEDGMENTS more, this explanation appears inconsistent with the obser The authors wish to thank Dr. Robert 0. Hussa for providing HCG analysis vation that lymphocytes from cystadenocarcinoma patients and for assistance in preparing the manuscript, and Dr. W. A. Nelson-Rees and Dr. A. A. Flandermeyer for G6PD and chromosomal analysis. showed no reactivity towards nonovarian carcinoma-de rived cells above the response of normal donor lymphocytes tested in the same assay (Table 3). On the other hand, REFERENCES lymphocytes from breast and cervical cancer patients were 1. Alexander, P. Escape from immune destruction by the host through reactive towards ElCo and CaSki or DoT cells, respectively. shedding of surface antigens: is this a characteristic shared by malignant These observations suggest that lymphocytes from cystad and embryonic cells? Cancer Res., 34: 2077-2082, 1974. 2. Baldwin, R. W. In vitro assays of cell-mediated immunity to human solid enocarcinoma patients show selective cytotoxicity towards tumors: problems of quantitation, specificity, and interpretation. J. NatI. ovarian carcinoma-derived cells and, therefore, that the Cancer Inst., 55: 745â€”748,1975. HeW cells express TAA. 3. Bhattacharya, M., and Barlow, J. J. An immunologic comparison be tween serous cystadenocarcinoma of the ovary and other human gyne Although in the present study no attempt was made to cologic tumors. Am. J. Obstet. Gynecol., 117: 849-853, 1973. correlate patient lymphocyte reactivity or serum blocking 4. Bhattacharya, M., and Barlow, J. J. Immunologic studies of human activity with such factors as stage of disease or treatment serous cystadenocarcinoma of ovary. Cancer, 31: 588â€”595,1973. 5. Bhattacharya, M., and Barlow, J. J. Tumor-associated antigen for modality, other investigators (16, 27, 37) have attempted to cystadenocarcinomas of the ovary. NatI. Cancer Inst. Monogr., 42: 25â€” correlate in vitro immune response to TAA during therapy 32, 1975. 6. Chen, 5-Y., Koffler, 0., and Cohen, C. J. Cell-mediated immunity in and clinical status. However, because of the technical patients with ovarian carcinoma. Am. J. Obstet. Gynecob., 115: 467-470, difficulties encountered in microcytotoxicity testing, it has 1973. been difficult to determine whether the differences in pa 7. Coggin, J. H., Ambrose, J. A., Dierlam, P. J., and Anderson, N. G. Proposed mechanisms by which autochthonous neoplasms escape tient responses were due to biological variations or to day immune rejection. Cancer Res., 34: 2092-2101 , 1974. to-day changes in assay conditions (2, 19). Because of the 8. DiSaia, P. J., Rutledge, F. N., Smith, J. P., and Sinkovics, J. G. Cell

1190 CANCER RESEARCH VOL. 39

mediated immune reaction to two gynecobogic malignant tumors. Can Obstet. Gynecol., 119: 966-970, 1974. cer,28: 1129-1137, 1971. 24. Levi, M. M. Antigenicity of ovarian and cervical malignancies with a view 9. DiSaia,P. J., Sinkovics,J. G., Rutledge,F. N., and Smith, J. P. Cell toward possible immunodiagnosis. Am. J. Obstet. Gynecol., 109: 687- mediated immunity to human malignant cells. Am. J. Obstet. Gynecol., 698, 1971. 114:979-989,1972. 25. Levi, M. M., Parshley, M. S., and Mandl, I. Antigenicity of papillary 10. Dorsett, B. H., loachim, H. L., Stolbach, L., Walker, J., and Barber, H. R. serous cystadenocarcinoma tissue culture cells. Am. J. Obstet. Gyne K. Isolation of tumor-specific antibodies from effusion of ovarian carci col., 102: 433-439, 1968. nomas. Int. J. Cancer, 16: 779-786, 1975. 26. Melnick, H., and Barber, H. R. K. Cellular immunologic responsiveness 11. Gall, 5. A., Walling, J. , and Pearl, J. Demonstration of tumor-associated to extracts of ovarian epithelial tumors. Gynecol. Oncol., 3: 77-86, 1975. antigens in human gynecologic malignancies. Am. J. Obstet. Gynecol., 27. Mitchell, M. S.. and Kohorn, E. I. Cell-mediated immunity and blocking 115:387-393,1973. factor in ovarian carcinoma. Obstet. Gynecol., 48: 590â€”597,1976. 12. Gerber, M. A., Faiferman, I., Cohen, C. J., and Koffler, D. Studies on 28. Nelson-Rees, W. A., and Flandermeyer, R. R. HeLa cultures defined. xenogenic antisera to tumor-associated antigen of ovarian carcinoma. Science, 191: 96â€”98,1976. Clin. Immunol. Immunopathol., 8: 171-180, 1977. 29. Nepom, J. T., Helbstrom, I., and Hellstrom, K. E. Purification and partial 13. Hellstrom, I., Helbstrom, K. E., Evans, C. A., Heppner, G., Pierce, G. E., characterization of a tumor-specific blocking factor from sera of mice andYang,J. P.S.Serummediatedprotectionofneoplasticcellsfrom with growing chemically induced sarcomas. J. Immunol., 117: 1846-' inhibition by lymphocytes immune to their tumor specific antigens. 1852, 1976. Proc. NatI. Acad. Sci. U. S. A., 62: 362-368, 1969. 30. Oldham, R. K., Djeu, J. Y., Cannon, G. B., Siwarski, D., and Herberman, 14. Hellstrom, I., Hellstrom, K. E., Sjogren, H. 0., and Warner, G. A. R. B. Cellular microcytotoxicity in human tumor systems: analysis of Demonstration of cell-mediated immunity to human neoplasms of var results. J. NatI. Cancer Inst., 55: 1305-1318, 1975. ious histological types. Int. J. Cancer, 7: 1-16, 1971. 31. Oyama, V. I., and Eagle, H. Measurement of cell growth in tissue culture 15. Hellstrom, I., Sjogren, H. 0., Warner, G., and Hellstrom, K. E. Blocking with a phenol reagent (Fobin-Ciocalteu). Proc. Soc. Exp. Biol. Med., 91: of cell mediated tumor immunity by serum from patients with growing 305-307, 1956. neoplasms. Int. J. Cancer, 7: 226-237, 1971. 32. Paff, L. M., and Grimes, W. U. Cell surface glycolipid and glycoprotein 16. Hebbstrom, I., Warner, G. A., Hellstrom, K. E., and Sjogren, H. 0. glycosyltransferases of normal and transformed cells. J. Biol. Chem., Sequential studies on cell mediated tumor immunity and blocking serum 249:4157-4165,1974. activity in ten patients with malignant melanoma. Int. J. Cancer, 11: 280- 33. Pattilbo, R. A., Hussa, R. 0., Ruckert, A. C. F., Story, M. T., and 292,1973. Mattingly, R. F. HCG-beta production by epidermoid carcinoma of the 17. Hellstrom, K. E., and Helbstrom, I. Immunological enhancement as cervix. Program and abstracts, 59th annual meeting, p. 179, Bethesda, studied by cell culture techniques. Annu. Rev. Microbiol., 24: 373-398, Md: The Endocrine Society, 1977. 1970. 34. Pattilbo, R. A., Hussa, R. 0., Story, M. T., Ruckert, A. C. F., 5halaby, M. 18. Hellstrom, K. E., and Helbstrom, I. Lymphocyte-mediated cytotoxicity R., and Mattingly, R. F. Tumor antigen and human chorionic gonadotro and blocking serum activity to tumor antigens. Adv. Immunol., 18: 209- pin in CaSki cells, a new epidermoid cervical cancer cell line. Science, 277, 1974. 196: 1456â€”1458,1977. 19. Heppner, G., Henry, E., Stolbach, L., Cummings, F., McDonough, E., 35. Pellis, N. R., and Kahan, B. D. Immunoprotection by fractions of tumor and Calabresi, P. Problems in the clinical use of the microcytotoxicity cell cultures: The immunogenicity of exhausted culture medium. Fed. assay for measuring cell-mediated immunity to tumor cells. Cancer Res., Proc., 34: 1042, 1975. 35: 1931-1937, 1975. 36. Pross, H. F., and Baines, M. G. Spontaneous human lymphocyte-me 20. Hussa, R. 0. Immunologic and physical characterization of human diated cytotoxicity against tumor target cells. 1. The effect of malignant chorionic gonadotropin and its subunits; cultures of human malignant disease. Int. J. Cancer, 18: 593â€”604,1976. trophoblast. J. Clin. Endocrinol. Metab., 44: 1154â€”1162,1977. 37. Sinkovics, J. G., Cabiness, J. R., and 5chullenberger, C. C. Disappear 21. Hussa, R. 0., Pattilbo, R. A., Garancis, J. C., Ruckert, A. C. F., and ance after chemotherapy of blocking serum factors as measured in vitro Mattingly, R. F. A malignant trophoblastic cell line: High rates of DNA with lymphocyte cytotoxic to tumor cells. Cancer, 30: 1428â€”1437,1972. synthesis measured by a rapid disc method. J. NatI. Cancer Inst., 45: 38. Story, M. T., Pattibbo,R. A., Ruckert, A. C. F., and Shalaby, M. R. On the 1119â€”1130,1970. isolation and characterization of soluble antigens from human breast 22. boachim, H. L., Dorsett, B. H., Sabbath, M., and Barber, H. R. K. Electron tumor cells in continuous culture. In Vitro, 10: 388-389, 1974. microscopy, tissue culture, and immunology of ovarian carcinoma. NatI. 39. Takasugi, M.. Mickey, M. A., and Terasaki, P. I. Reactivity of normal Cancer Inst. Monogr., 42: 45â€”62,1975. lymphocytes from normal persons on cultured tumor cells. Cancer Res., 23. Knauf, S., and Urbach, G. I. Ovarian tumor-specific antigens. Am. J. 33: 2898-2902,1972.

APRIL 1979 1191

., Pt, I

jâ€•. â€¢1d1,,f @. 4/ @â€œ

-@:@

p. @ \@. fr @-,,iv:.1: 4. -- i4

Fig. 1. Phase-contrast photomicrograph of HeW cells. x 100.

II92 CANCER RESEARCH VOL. 39

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1979 American Association for Cancer Research. Expression of Cell-mediated Immunity and Blocking Factor Using a New Line of Ovarian Cancer Cells in Vitro

Roland A. Pattillo, Michael T. Story and Anna C. F. Ruckert

Cancer Res 1979;39:1185-1192.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/39/4/1185

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/39/4/1185. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.