Tumorigenicity of hamster and mouse cells transformed ... - Europe PMC

1 downloads 0 Views 1MB Size Report
tViral Pathogenesis Section, Laboratory of Immunopathology, and §Laboratory of Immunogenetics, ... in Syrian hamsters by Trentin and coworkers in 1962 (1).
Proc. Nati. Acad. Sci. USA

Vol. 83, pp. 9684-9688, December 1986 Immunology

Tumorigenicity of hamster and mouse cells transformed by adenovirus types 2 and 5 is not influenced by the level of class I major histocompatibility antigens expressed on the cells (neoplasia/celiular immunity/immune surveillance)

H. HADDADA*, A. M. LEWIS, JR.tt, J. A. SOGN§, J. E. COLIGAN§, J. L. COOK¶, T. A. WALKER¶, AND A. S. LEVINE* *Section on Viruses and Cellular Biology, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20892;

tViral Pathogenesis Section, Laboratory of Immunopathology, and §Laboratory of Immunogenetics, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892; and lRobert W. Lisle Research Laboratory in Immunology and Tumor Cell Biology, Department of Medicine, National Jewish Center for Immunology and Respiratory Medicine, Denver, CO 80206 Communicated by Igor B. Dawid, September 3, 1986

role in the tumorigenicity of Ad-transformed cells and, by analogy, in the oncogenicity of the viruses themselves. The cellular immune compartment is the division of the immune system of mammals that is involved in the recognition and elimination of tumor cells. Within the cellular immune compartment, there are two components (specific and nonspecific) that act against potentially neoplastic cells. The specific component is mediated by the capacity of immune system-stimulated cytotoxic T lymphocytes (CTL) to recognize and destroy virus-transformed cells that express both virus-specific cell-surface proteins and class I major histocompatibility complex (MHC) proteins. The nonspecific component is mediated by the capacity of unstimulated natural killer (NK) cells and macrophages to attack and eliminate neoplastic cells. Much of the current thinking about the mechanisms of carcinogenesis of human adenoviruses in rodents revolves around the ability of different Ad serotypes to induce in transformed cells the ability to succumb to or resist destruction by one or both of these components of the cellular immune system. Two recent reports (3, 4) have shown that the early region 1A (EIA) gene of nononcogenic Ad2 and Ad5 can induce high levels of susceptibility to lysis by NK cells, while the EIA gene of Adl2 either induces lower levels of susceptibility or induces resistance to NK lysis. Other investigators (5, 6) have found that the Adl2 EJA gene inhibits the expression of class I MHC proteins on the surfaces of transformed cells, thus removing a critical signal for recognition by CTL (5, 7). In an attempt to clarify the specific contributions toward tumorigenicity made by the expression of class I MHC proteins on the surfaces of Ad-transformed rodent cells, and by the expression of virus-induced susceptibility to NK cell lysis, we have begun a systematic evaluation of the correlation between tumor-inducing capacity and class I expression in a series of carefully characterized Ad2-, AdS-, and Adl2transformed hamster and mouse cells that are also being studied for their relative susceptibility to lysis in vitro by various types of nonspecific cellular immune effector cells. In this report, we show that both Ad2-transformed hamster cells and AdS-transformed mouse cells can express either high or low levels of cell-surface class I MHC proteins and

Inbred hamster and mouse cells transformed ABSTRACT by the nononcogenic adenovirus (Ad) serotypes, Ad2 and AdS, are nontumorigenic in syngeneic adult animals, while cells from these species transformed by the highly oncogenic Adl2 are tumorigenic in such rodents. By immunoprecipitation and flow cytometry, cells from four of six Ad2- and Ad5-transformed hamster and mouse lines expressed high levels of cell-surface class I major histocompatibility complex (MHC) antigens, while cells from two of these six lines expressed low levels of cell-surface class I MHC antigens. The levels of class I MHC proteins expressed by cells from these latter two lines were comparable to the levels of cell-surface class I MHC proteins expressed by cells from Adl2-transformed hamster and mouse lines. Moreover, an Ad2-transformed line that had become highly oncogenic after in vivo adaptation showed the same high level of MHC expression as the nononcogenic parent. The amounts of class I mRNA, analyzed by RNA blotting, were, in general, consistent with the levels of class I antigens expressed on the surfaces of these cells. These results indicate that there is no correlation between the tumorigenicity in immunocompetent syngeneic adult rodents of Ad2- and Ad5-transformed hamster and mouse cells and the level of class I MHC antigens expressed on the surfaces of these cells. Thus, the expression of different levels of class I MHC proteins does not seem to explain the differences in the oncogenicity between nononcogenic and highly oncogenic human Ad serotypes. Human adenoviruses (Ad) were first shown to be oncogenic in Syrian hamsters by Trentin and coworkers in 1962 (1). Following this discovery, several Ad serotypes were categorized into nononcogenic (Adl, Ad2, Ad5) and highly oncogenic (Adl2, Adl8, Ad3M) groups by their capacity to induce tumors in hamsters and other rodents (2). In spite of the differences in oncogenicity of different Ad serotypes in vivo, all of these viruses have the capacity to transform normal rodent cells into immortalized neoplastic cells during infections in tissue culture. In general, the tumor-inducing capacity of these transformed cells reflects the oncogenic phenotype of the transforming virus. Cells transformed by nononcogenic Ad2 and AdS are either nontumorigenic or tumorigenic only when injected into immunodeficient animals, while cells transformed by highly oncogenic Adl2 are tumorigenic in immunocompetent as well as immunodeficient animals. The dependence of tumorigenicity on the immune status of the recipient suggests that surveillance by the immune system in the injected rodent plays a determining

Abbreviations: Ad2, Ad5, and Adl2, human adenovirus serotypes 2, 5, and 12; CTL, cytotoxic T lymphocytes; MHC, major histocompatibility complex; El, E3, and ElA, early regions 1 and 3 of the adenovirus genome, and the A gene that constitutes one of the two transcription units of the El region; (32m, P2-microglobulin; FACS, fluorescence-activated cell sorter. *To whom reprint requests should be addressed at: National Institute of Allergy and Infectious Diseases, Bldg. 5, Room B1-32, National Institutes of Health, Bethesda, MD 20892.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

9684

Immunology: Haddada et al. that the differences in the expression of these class I transplantation antigens does not affect the tumorigenic phenotypes of these transformed cells in adult hosts that possess mature thymus-dependent cellular immune functions. Thus, there are barriers to tumorigenicity for cells transformed by nononcogenic adenoviruses that are not released by low-level expression on these transformed cells of class I MHC antigens.

MATERIALS AND METHODS Transformed Cell Lines. Transformation of hamster embryo cells by infectious or UV light-inactivated Ad has been described (8, 9). BALB/c mouse embryo cells were transformed by similar procedures. The Ad2HE3ATL-1 (adultadapted tumor line 1) cell line was derived by serial passage of tumors induced by Ad2HE3 cells first in newborns and then in adult LSH hamsters; Ad2HE3ATL-1 cells were obtained by growing cells from an adult-adapted tumor in tissue culture. The Ad5- and Adl2-transformed BALB/c mouse embryo cell lines were established from discrete foci of morphologically transformed cells that were growing in separate tissue culture flasks. Virus-free (by cocultivation on HeLa cells) cells present in each of these mouse cell lines expressed virus-specific early antigens or proteins that were detectable by immunofluorescence or immunoprecipitation using a variety of broadly reactive antisera from tumor-bearing hamsters or specific antisera prepared against early proteins themselves (4, 9). Cells from each of these lines also contained DNA sequences of the transforming virus that were detectable by reacting cellular DNA with Ad5 or Adl2 El region probes using the technique described by Southern (10). Animal Studies. The tumor-inducing capacities of rapidly dividing Ad-transformed cells were determined by injecting serial dilutions (1:10) of cells subcutaneously into 8- to 10-wk-old syngeneic (LSH) hamsters or (BALB/c) mice; the incidence of fatal tumors (number of animals with tumors >30 mm in diameter per number of animals surviving for 90 days) was interpolated by the method of Karber (11, 12) to determine the number of cells that would produce tumors in 50% of the injected animals. Preparation of Radiolabeled Cell Proteins. Cells in the subconfluent exponential phase of growth were cultured with [35S]methionine (100 ,Ci/ml; 1 Ci = 37 GBq), lysed, and their proteins were extracted by described techniques (13). The protein concentration of each lysate was measured with Coomassie brilliant blue (Bio-Rad) (14). Immunoprecipitation and NaDodSO4/PAGE. Hamster class I MHC proteins were immunoprecipitated by the IgG fraction of a rabbit anti-human 182-microglobulin antiserum (rabbit anti-,82m) (lot 114; Accurate Chemicals, Wesbury, NY) and mouse class I MHC proteins were immunoprecipitated by a monoclonal antibody (34-2-12) specific for H-2Dd (15) (kindly provided by Keiko Ozato). For these immunoprecipitations, equal amounts of proteins (80 ,g of protein in an equal vol of extraction buffer) from each cell lysate were extensively preabsorbed with heat-inactivated Staphylococcus aureus (Bethesda Research Laboratories) (three times with 1/4th of the lysate vol) to remove all nonspecific protein fixation to S. aureus. Then, the cleared lysates were incubated with optimal (determined by titration in preliminary assays) concentrations of antibody for 90 min at 4°C. The immune complexes were adsorbed to S. aureus (10 times the vol of antibody), washed four times, and eluted by heating for 5 min in boiling water with 50 ,ul of sample buffer [75 mM Tris HC1, pH 6.8/2% NaDodSO4/0.7 M 2-mercaptoethanol/0.01% bromophenol blue/50% (vol/vol) glycerol/0.2 unit of aprotinin in 1 ml]. Electrophoresis was

Proc. Natl. Acad. Sci. USA 83 (1986)

9685

carried out on polyacrylamide gels (12-13%) as described by Laemmli (16). In preliminary experiments, samples of the lysates were immunoprecipitated with anti-actin, anti-class I, or with rabbit anti-f32m. We found that actin and class I proteins migrated at different and easily distinguishable positions in our gels. Fluorescence-Activated Cell Sorter (FACS) Analysis. FACS analysis was used to determine the cell-surface density of class I MHC antigens. Briefly, 106 viable cells were washed with 1 ml of phosphate-buffered saline containing 1% bovine serum albumin and 0.01% sodium azide and reacted with optimal (as determined by initial titrations) concentrations of rabbit anti-p82m or with the monoclonal antibody 34-2-12 for 30 min on ice. The cells were washed twice and stained with the fluorescein-isothiocyanate-conjugated IgG fraction of sheep anti-mouse IgG or goat anti-rabbit IgG. After 30 min of incubation on ice, the cells were washed four times and immediately analyzed for surface fluorescence in a Becton Dickinson FACS II. The values for mean fluorescence were calculated by deducting the mean values of reactivity of normal mouse ascites fluids from that of 34-2-12 or of normal rabbit IgG from that of rabbit anti-f32m. Class I mRNA Analysis. The level of class I MHC mRNA was analyzed by RNA blotting (17). The probe used to analyze hamster RNA was a fragment of a hamster class I genomic clone, extending from the beginning of exon 4 to the beginning of the 3' untranslated region. The clone was kindly provided by Philip Tucker. The probe used for mouse RNA was a H-2Ld cDNA (pMHC-lLd) isolated by Evans et al. (18). RESULTS Tumorigenicity of Transformed Cells. To begin our assessment of the role of cell-surface class I MHC proteins on Ad-transformed rodent cells in tumorigenicity, we compared the tumor-inducing capacity of several Ad2- and Adl2transformed hamster cells in immunocompetent syngeneic adult hamsters with the tumor-inducing capacity of AdS- and Adl2-transformed BALB/c mouse cells in immunocompetent syngeneic adult mice (Table 1). From these data, it is apparent that Ad2-transformed LSH hamster embryo cells and AdS-transformed BALB/c mouse embryo cells either are not tumorigenic or are very weakly tumorigenic in immunocompetent syngeneic adult rodents, even when massive inocula of 108 cells were injected into each animal. In Table 1. Tumor induction in syngeneic adult hamsters or mice by Ad-transformed hamster or mouse embryo cells No. of cells injected per animal, log1o Cell line 6 8 7 5 4 3 TPD5o Ad2HE7 0/4 0/23 28.5 Ad2HE9 -8.5 0/4 0/15 Ad2HE1 .8.5 0/20 0/30 2/20 Ad2HE3 0/7 0/20 0/10 28.5 Ad2HE3ATL-1 NT 31/31 30/31 24/31 17/31 2/26 4.1 Ad12HE4 5/5 5/5 5/5 5/5 0/5 3.5 AdSME1 0/5 0/5 0/5 0/5 0/5 28.5 Ad5ME2 0/5 0/5 0/5 0/5 0/5 28.5 Ad12ME1 6.0 4/4 4/4 2/4 0/4 0/4 No. in numerator = no. of animals that developed fatal (diameter, >30 mm) tumors during 3-month observation period. No. in denominator = no. of animals with fatal tumors + no. that survived the assay without developing tumors or that developed small tumors (diameter, usually