Generation of Immunoprotection Against Squamous Cell Carcinomas by In Vitro Cultivation and a Possible Mechanism of Action

The immunogenicity of individual diethylnitrosamine (DEN)-induced forestomach carcinomas in female BALB/c mice was investigated following in vitro and in vivo cultivation. Of the five transplantable tumor lines studied, (DEN1? DEN3, DEN6, DEN8, and DEN^ only two (DEN6 and DEN8) showed some degree of immunogenicity. DENX, E»EN3, and DEN9 were highly tumorigenic with very little immunogenic potency as judged by tumor transplantation-excision assay, Winn neutralization, and antibody binding tests. These three tumors grew rapidly and showed a high degree of malignancy. DENX and DEN3 also metastasized readily. Cell lines from DEN6 and DEN9 lost their tumorigenicity at the 5th and 50th passage of culture, respectively. Although DENa and DEN3 did not lose their tumorigenicity, the number of tumor cells required to produce tumors increased substantially and their ability to metastasize was lost. Tumor transplantation studies, with these cultured cell lines in normal and x-irradiated recipients, suggested that the decrease in tumorigenicity may be immunologically mediated. Mice immunized with the in vitro lines demonstrated transplantation resistance against the respective in vitro and in vivo lines. The treatment of in vivo or in vitro propagated cells with periodic acid or neuraminidase enhanced antigen-antibody binding significantly. The effect of these chemicals became less pronounced as in vitro culture continued. It appears that during in vivo cultivation the antigenic determinants are masked or modulated by some glycoprotein or glycolipid molecules which render them non-, or very weakly, immunogenic. OHIO J. SCI. 94 (1): 14-23, 1994 INTRODUCTION Multiple factors appear to be involved in the control and development of oncogenic diseases. Among these factors, elements of immune response undoubtedly play an important role in some tumor systems. However, for the immune mechanisms to be effective in controlling oncogenic diseases, tumors must express appropriate tumor-associated antigen(s). The presence of tumorassociated antigen(s) is common on cell surfaces of tumors induced by chemical carcinogens or oncogenic viruses (North 1984, Schreiber et al. 1988). Such antigen(s), however, has not been demonstrated as a common feature of many spontaneously arising tumors (Hewitt et. al. 1976, Klein and Klein 1977). For this reason, many attempts have been made using different chemicals (Bonmassar et al. 1970, Contessa et. al. 1981, Frost et. al. 1984, Ishikawa et. al. 1987), viruses (Kobayashi et. al. 1969, Boon 1983, Altevogt 1986, Fearon et. al. 1988), mutagens (Boon 1983), and factors such as cholesterol derivatives (Ludes et al. 1990) and interleukin 6 (Mullen et al. 1992), to enhance immunogenicity and to reduce tumorigenicity of some malignant tumor cells. In some of the previous investigations, we reported that with increased in vitro passage, the immunogenicity of several respiratory squamous cell carcinomas (that is, the ability of tumor cells to induce cellular and humoral immunity in syngeneic hosts) was increased while their tumorigenicity decreased (Jamasbi and Nettesheim 1977, 'Manuscript received 17 December 1992 and in revised form 14 October 1993 (#92-32). Correspondence and reprint requests should be addressed to: Dr. R. J. Jamasbi, Department of Medical Technology, Bowling Green State University, Bowling Green, OH 43403, U.S.A. 1979). Similar observations have also been reported by other investigators using different tumor models (Ossowski and Reich 1980, Correll et. al. 1983, Yamashina et. al. 1986, Chiba et. al. 1987). Although some speculations concerning the mechanism(s) of increased immunogenicity of cultured cell lines have been made, the exact nature of this phenomenon remains unknown. The main objectives of the present investigation were to study the biological and immunological characteristics of five different mouse forestomach carcinoma lines and to determine whether the tumorigenicity and immunogenicity of some of the highly malignant tumor lines could be altered by in vitro cultivation. If so, efforts would be made to determine whether the underlying mechanism(s) can be elucidated. MATERIALS AND METHODS Animals Syngeneic BALB/c female mice, 8-12 weeks of age were used throughout these experiments. They were bred and maintained in a conventional animal facility and had free access to food and water. Tumors Five forestomach carcinoma lines were used. These tumors developed in the forestomach of female BALB/c mice which had received DEN in their drinking water for eleven weeks (40 fig/1, cumulative dose 350 |ig/kg of body weight) as reported previously Qamasbi and Perkins 1990). These tumors were designated DENi; DEN3, DEN6, DEN8, and DENy. Histologic examination of transplanted tumors showed that all of the tumors were invasive squamous cell carcinomas. The syngeneic sarcoma cell line, designated MSC (Kennel et al. 1985), was used as a control. OHIO JOURNAL OF SCIENCE R. J. JAMASBI 15 In all experiments tumor cells were inoculated intramuscularly (i.m.) in the thigh of syngeneic mice. The transplantability of each tumor line was determined by serial passage in vivo. At the fifth in vivo passage, tumors were removed, single cell suspensions were prepared by trypsinization, and a large cell pool was established and stored in liquid nitrogen. Cells from the fifth in vivo passage were used for the majority of in vivo experiments and for the establishment of tissue culture cell lines. The relative tumorigenicity of each carcinoma line was determined by injecting graded doses of tumor cells. Tumor development, growth, rates, and time and incidence of mortality were determined. Metastatic characteristics of these tumors were studied as described previously (Jamasbi and Perkins 1990). Briefly, mice were inoculated with 10 tumor cells and the tumor was removed when it reached 1-1.5 cm in diameter. Thirty days later the animals were killed and examined for the incidence of metastasis. In Vitro Culture Culture media, sera, and antibiotics were obtained from Grand Island Biological Co. (Grand Island, NY). Tissue culture procedures and the establishment of epithelial cell lines from these tumors were as previously described (Jamasbi and Perkins 1990). Briefly, cell lines were derived from the fifth in vivo tumor passage by explant culture of tumor fragments in McCoy's medium supplemented with 10% fetal bovine serum (FBS). All cultures were incubated at 37° C in a humidified atmosphere of 5% CO2 in air. When sufficient epithelial outgrowth was observed, cells were subcultured. All of the cell lines were screened periodically for mycoplasma contamination (Hayflick 1965). Tissue culture cell lines were developed from all tumors except DENg. In Vitro Growth Properties For determination of seeding efficiency, growth rate, and cell doubling times, cell lines were plated at a density of 10 cells per 100 mm plastic dish. Cells from two dishes were independently harvested by trypsinization each day and counted using a hemacytometer. Results were plotted on semilogarithmic paper, and doubling times calculated. Transplantation Tests To test the tumorigenicity of in vivo or in vitro passaged tumor cells, different doses of cells were inoculated into syngeneic hosts. Cultured cells were tested for their tumorigenicity after every fifth passage for 25-50 passages. In vivo propagated cells were serially transplanted into syngeneic hosts for an additional five times. Thighs of inoculated mice were checked at weekly intervals for tumor development. Rates of tumor growth and incidence of mortality were recorded. Fluorescent Antibody Assay In order to analyze cells for viral antigens, tumor cells were grown on coverslips. When the cells reached near confluency, the coverslips were washed in PBS and adherent cells were fixed in cold acetone. The cells were then incubated with broadly reactive antibodies from a rat immunized against mouse Moloney leukemia and Moloney sarcoma viruses. The cells were washed in PBS, incubated with goat anti-rat serum (fluorescein conjugated), and counter-stained with rhodamine. Positive cells were observed using an ultraviolet microscope. Immunization and Challenge (In Vivo Tumor Lines) To determine tumor immunogenicity, previously described immunization procedures were used (Jamasbi et al. 1978). For this purpose, mice were injected i.m. with 1O-1O live tumor cells. Tumors were removed 3-4 weeks later when tumor size reached 1-1.5 cm in diameter. One week later the animals were challenged with graded doses of in vivo parental tumor cell lines. Because of the apparent absence of immunogenicity of some of these tumor lines, transplantation-excision procedures were repeated 2-3 times to determine whether immune responses (resistance) could be generated. Immunization by x-irradiated cells (Jamasbi and Nettesheim 1979) was also attempted. Mice immunized with MSC or receiving sham surgery served as controls. Immunization and Challenge (In Vitro Tumor Lines) For immunization with in vitro grown cell lines, 10 cells in 0.1 ml PBS from different in vitro passages were inoculated into syngeneic recipients. Tumors which developed in some of the hosts were surgically removed when they grew to about 1-1.5 cm in diameter (3-5 weeks later), and the animals were challenged with appropriate in vivo or in vitro cell lines 7-10 days later. In all of the preceding experiments normal mice which had undergone sham surgery or mice immunized with MSC served as controls. Winn Neutralization Test To determine the cytotoxic effect of the immune spleen cells on the tumor development, the Winn neutralization test was conducted. For this purpose, spleens from either normal or immunized mice were aseptically removed and pooled (3-5 spleens/group). Single-cell suspensions were prepared and mixed at a ratio of 200:1 with tumor cells using the smallest tumor cell dose that produced 100% tumor takes for each respective tumor line. The mixture was incubated at 37° C for 1 hr, and inoculated i.m. into the thighs of normal or x-irradiated mice (450r x-rays). Control groups received tumor cells only or tumor cells mixed with spleen cells from either normal mice or mice which had been immunized with MSC tumor cells. Antibody-binding Test To demonstrate an antibody in the sera of mice immunized against each tumor, an enzyme-linked immunosorbent assay (ELISA) was used. To perform ELISA, in vitro propagated cells were cultured into 96-well plates (Corning) and fixed with 0.1% glutaraldehyde (Sigma) for 10 min at room temperature. The fixed cells were treated with 0.1 M glycine for 5 min and were then 16 GENERATION OF IMMUNOPROTECTION VOL. 94 exposed to 1% bovine serum albumin in phosphatebuffered saline, pH 7.2 (BSA-PBS) for 1 h. Properly diluted (1:500) immune or non-immune sera were added at 100 |il/well and allowed to react for 1 h at 37° C. The plates were washed three times with PBS and incubated at 37° C for 1 h with 100 |il/well of a 1:250 dilution (in 2% BSA-PBS) of 13-galactosidase-conjugated goat antimouse immunoglobulin (Southern Biotechnology Associates, Inc.). After the incubation, the plates were washed three times with PBS and incubated with 100 fil/ well of p-nitrophenyl-fi-D-galactopyranoside (Sigma) solution (1 mg/ml in phosphate buffer, pH 8.5) at 37° C for 1 h. The optical density (at 410 nm) was measured with an MR-600 microwell plate reader (Dynatech Laboratories, Inc.). In some ELISA experiments, when in vivo propagated cells were to be tested, a 10 cells/well of either in vivo or in vitro (for comparison) were attached to poly-L-lysine treated wells, fixed and examined as above. Cultured MSC cells served as controls. Periodic Acid Treatment To determine whether the binding of immune sera to the tumor cells can be altered by periodic acid treatment, glutaraldehyde-fixed cells in 96-well plates were treated with 100 mM periodic acid solution (Sigma) at 37° C for 30 min. The wells were washed with PBS three times and the reactivity of the antibody with the treated and untreated cells was determined by ELISA. Enzyme Treatment The effect of neuraminidase (Sigma) treatment on the binding of immune sera to in vivo and in vitro propagated cells was also assessed. Glutaraldehyde-fixed cells in 96well plates were exposed to a 0.5 mU/well of neuraminidase (in 50 mM sodium acetate buffer, pH 5.5), for 1 h at 37° C. Control wells were exposed to 100 uj of sodium acetate buffer. Following enzyme treatment, the wells were washed 3 times with PBS; then antigen-antibody binding was assessed by ELISA. RESULTS Growth Characteristics of In Vivo Propagated Tumor Lines The degree of tumorigenicity varied considerably among the forestomach carcinomas of BALB/c mice (Table 1). Whereas the 10 viable cells of DEN:, DEN3, or DEN9 produced tumors in 100% of the inoculated hosts, a much higher (approximately 100-fold) tumor cell dose (10) of DEN6 or DENg was needed to obtain similar results. This experiment was repeated several times using a different range of tumor cell concentrations. Only those which gave approximately 10-100% tumor takes are presented (Table 1). Growth rates of in vivo propagated DEN17 DEN3, and DEN tumor lines (data not shown) were also much greater than those observed for DEN6 and DENg. The rate of mortality (Table 2) also differed considerably among mice that received cells from different tumor lines, i.e., 10 cells of DENj killed all mice between 6-8 weeks whereas the same dose of DEN3, DEN6, DENg, or DEN? tumor cells killed the injected mice between 9-11, 20-24, 18-23, and 7-11 weeks, respectively. Highly malignant, fast-growing DENX and DEN metastasized readily (Table 2). DENX metastasized faster and produced micrometastases and secondary tumors in many organs, including the lungs, kidneys, and renal lymph nodes, while DEN3 primarily metastasized to the lungs and occasionally to the renal lymph nodes. Surprisingly, aggressively growing, highly malignant DENg did not metastasize. Relatively slow growing DEN6 and DENg did not metastasize, either. Growth Characteristics of In Vitro Propagated Tumor Cell Lines Epithelial outgrowths from the tumor explants in the culture grew slowly and subcultures were generally not possible until 5-6 weeks after the primary culture (Jamasbi and Perkins 1990). Once fibroblast-free epithelial cultures were established, the cells grew more rapidly and subcultures were routinely made at weekly intervals.