Immune surveillance against tumors usually depends on T cell recognition of tumor antigens presented by major histocompatibility complex (MHC) molecules, whereas MHC class I- tumors may be controlled by natural killer (NK) cells. Perforin-dependent cytotoxicity is a major effector function of CD8+ MHC class I-restricted T cells and of NK cells. Here, we used perforin-deficient C57BL/6 (PKO) mice to study involvement of perforin and Fas ligand in tumor surveillance in vivo. We induced tumors in PKO and normal C57BL/6 mice by (a) injection of different syngeneic tumor cell lines of different tissue origin in naive and primed mice; (b) administration of the chemical carcinogens methylcholanthrene (MCA) or 12-O-tetradecanoylphorbol-13-acetate (TPA) plus 7,12-dimethylbenzanthracene (DMBA), or (c) by injection of acutely oncogenic Moloney sarcoma virus. The first set of models analyzes the defense against a tumor load given at once, whereas the last two sets give information on immune defense against tumors at the very moment of their generation. Most of the tumor cell lines tested were eliminated 10-100-fold better by C57BL/6 mice in an unprimed situation; after priming, the differences were more pronounced. Lymphoma cells transfected with Fas were controlled 10-fold better by PKO and C57BL/6 mice when compared to untransfected control cells, indicating some role for FasL in tumor control. MCA-induced tumors arose more rapidly and with a higher incidence in PKO mice compared to C57BL/6 or CD8-deficient mice. DMBA+TPA-induced skin papillomas arose with similar high incidence and comparable kinetics in both mouse strains. C57BL/6 and PKO mice have a similar incidence of Moloney murine sarcoma and leukemia virus-induced sarcomas, but tumors are larger and regression is retarded in PKO mice. Thus, perforin-dependent cytotoxicity is not only a crucial mechanism of both cytotoxic T lymphocyte- and NK-dependent resistance to injected tumor cell lines, but also operates during viral and chemical carcinogenesis in vivo. Experiments addressing the role of Fas-dependent cytotoxicity by studying resistance to tumor cell lines that were stably transfected with Fas neither provided evidence for a major role of Fas nor excluded a minor contribution of Fas in tumor surveillance.