Search

Although CD8(+) T cells play a central role as immune effectors, CD4(+) T cells act to control the activation and persistence of the CD8(+) T cell response in autoimmune disease, antiviral immunity, and experimental systems with immunogenic model tumor Ag. However, little information is available on the effects of CD4(+) T cells on the function of endogenous CD8(+) T lymphocytes recognizing authentic tumor rejection Ag with limited immunogenicity. We report here that the prophylactic or postchallenge administration of T helper Th1-type and Th2-type CD4(+) clones specific for an unmutated rejection Ag (murine P815AB, resembling tumor-specific shared Ag in humans) leads to the induction of P815AB-specific reactivity in vivo and concomitant tumor destruction, with quantitative rather than qualitative differences characterizing the antitumor activity of Th1 vs Th2 cells. Because the transferred CD4(+) cells lacked direct antitumor activity in vitro and required the de novo generation of P815AB-specific CD8(+) T cells in vivo, these findings suggest that CD4(+) lymphocytes can enhance the ability of host APC to initiate an endogenous CD8(+) T cell response to authentic, poorly immunogenic tumor rejection Ag.

Using an in vivo model of tumor/self peptide presentation for induction of class I-restricted skin test reactivity, we have previously shown that a minority population of CD8+ dendritic cells (DC) negatively regulates the induction of T cell reactivity by peptide-loaded CD8- DC in DBA/2 mice. However, the CD8- fraction can be primed by IL-12 to overcome inhibition by the CD8+ subset when the two types of DC are cotransferred into recipient hosts. We report here that exposure of CD8+ DC to IFN-gamma greatly enhances their inhibitory activity on Ag presentation by the other subset, blocking the ability of IL-12-treated CD8- DC to overcome suppression. In contrast, IFN-gamma has no direct effects on the APC function of the latter cells and does not interfere with IL-12 signaling. The negative regulatory effect triggered by IFN-gamma in CD8+ DC appears to involve interference with tryptophan metabolism in vivo. Through tryptophan depletion affecting T cell responses, IFN-gamma acting on CD8+ DC may thus contribute to regulation of immunity to tumor/self peptides presented by the CD8- subset.

Previous work has shown that a significant proportion of murine splenic dendritic cells (DC) express a high affinity receptor for IL-12, thus accounting for the adjuvanticity of the cytokine when DBA/2 mice are transferred with syngeneic DC exposed in vitro to rIL-12 and an otherwise poorly immunogenic tumor peptide. In DBA/2 mice, splenic DC consist of 90-95% CD8- and 5-10% CD8+ cells. To detect any difference in IL-12 responsiveness among phenotypically distinct DC subtypes, enriched CD8- (>99% pure) and CD8+ ( approximately 80% pure) populations of DC from DBA/2 spleens were assayed for APC function in vivo following exposure to rIL-12 and tumor peptide in vitro. Unlike unfractionated DC, the CD8- fraction was capable of effective presentation of the peptide even when the cells had not been pretreated with IL-12 before peptide pulsing. The addition of as few as 3% CD8+ cells during pulsing blocked in vivo priming by the CD8- fraction. However, pretreatment of CD8- DC with IL-12 before cell mixing and peptide pulsing ablated the inhibitory effect of the CD8+ fraction. CD8-, but not CD8+, DC showed significant message expression for the beta 1 and beta 2 subunits of the IL-12 receptor. These data suggest that a minority population of CD8+ DC, which appeared to secrete IL-10 in vitro, negatively regulates the induction of T cell reactivity by peptide-loaded CD8- DC in DBA/2 mice. However, the CD8- fraction can be primed by IL-12 to overcome the inhibitory effect of the CD8+ subtype.

Cell-mediated immunity involving CD8+ lymphocytes is effective in mediating rejection of murine mastocytoma cells bearing P815AB, a tumor-associated and self antigen showing similarity to tumor-specific shared antigens in humans. Although this antigen may act as an efficient target for class I-restricted responses in immunized mice, neither P815AB expressed on tumor cells nor a related synthetic nonapeptide will activate unprimed CD8+ cells for in vivo reactivity, measured by skin test assay. We review evidence showing that the failure of P815AB to initiate CD8+ cell reactivity may be due to defective recruitment of accessory and Th1-like cells to the afferent phase of the response initiated by transfer of mice with dendritic cells pulsed in vitro with the P815AB peptide. Although the copresence of a T helper peptide in dendritic cell priming in vitro with P815AB may compensate for the poor generation of accessory and Th1 cells in the adoptively transferred mice, recombinant IL-12 can replace the helper peptide in both effects. Effective priming to P815AB in vivo is achieved by either exposing dendritic cells to IL-12 prior to P815AB priming or administering the recombinant cytokine in vivo. Different approaches suggest that IL-12 may act both on accessory cells to improve presentation of previously undescribed class II-restricted epitopes of P815AB and on CD4+ cells to improve recognition of such epitopes. In particular, at the CD4+ cell level, IL-12 apparently acts as an adjuvant and an inhibitor of anergy induction. These data offer useful information for developing vaccination strategies using dendritic cells and class I-restricted tumor peptides in humans.

Ag-specific CD8+ cell responses, including delayed-type hypersensitivity in vivo and IFN-gamma production in vitro, are initiated by host immunization with P815AB, a self peptide bearing CTL epitopes and expressed by murine mastocytoma cells. Using P815AB-pulsed dendritic cells (DC) and monitoring class I-restricted skin test reactivity in DC-primed mice, we have previously shown that the development of a Th1-like response to P815AB requires T helper effects, such as those mediated by coimmunization with class II-restricted (helper) peptides or by the use of rIL-12. The adjuvanticity of IL-12 was suggested to involve improved recognition of class II-restricted epitopes of P815AB. In the present study, we provide evidence for the occurrence of I-A(d)-restricted epitopes in the tumor peptide. We also show that in the absence of helper peptide or rIL-12, P81 5AB not only failed to initiate CD8+ cell responses in vivo and in vitro, but resulted in a transient state of functional unresponsiveness, characterized by a profound inability of CD4+ cells to produce IL-2 in vitro. Ag-specific T cell anergy was also observed after neutralization of endogenous IL-12 at the time of priming with P815AB plus helper peptide. All of these effects were reversed by rIL-12, which was added to DC cultures and administered to the DC-recipient mice. Anergy induction may thus contribute to P81 5AB unresponsiveness in vivo. IL-12 may act to prevent or revert anergy to this tumor-associated and self peptide.

Delayed-type hypersensitivity responses, mediated by CD8+ cells and detected by skin test assay, occur in sensitized mice in response to challenge with a class I-restricted synthetic peptide related to a poorly immunogenic tumor rejection Ag, P815AB, of murine mastocytoma cells. Efficient priming for this response, which requires functional CD4+ cells and production of IFN-gamma in the host, is achieved by transfer of dendritic cells (DC) pulsed in vitro with a physical mixture of P815AB and T helper peptides, such as a class II-restricted synthetic peptide of tetanus toxin. We now show that the adjuvant effect of the T helper peptide was associated with the appearance of early and late IL-12 transcripts in the spleens of DC recipient mice, correlated with a late IFN-gamma response, and was negated by serologic ablation of endogenous IL-12 at the time of cell transfer. rIL-12, administered in vivo to the DC recipient mice, could substitute for the T helper peptide in initiating skin test reactivity following transfer of DC pulsed with P815AB alone, leading to Ag-specific production of IFN-gamma by CD4+ and CD8+ T cells. In vitro and in vivo cell depletion experiments suggested the following: 1) the exogenous IL-12 required both CD4+ and CD8+ cells for activity; 2) the immune response initiated by IL-12 relied on later production of IL-12 by the host; and 3) the early adjuvanticity of the exogenous IL-12 involved improved recognition of class II-restricted epitopes of this otherwise poorly immunogenic tumor peptide.

Delayed-type hypersensitivity (DTH) responses, mediated by CD8+ cells and detected by skin test assay, occur in sensitized mice in response to challenge with class I-restricted antigenic peptides of mutagenized (tum-) P815 mastocytoma cells. In contrast, a nonapeptide related to a tumor rejection antigen, P815AB, failed in this study to elicit DTH after sensitization of mice with irradiated tumor cells or adoptive transfer of P815AB-pulsed dendritic cells. Unresponsiveness, however, could be overcome by immunization with tumor cells co-expressing P815AB and tum- antigens. When used for cell pulsing in vitro, a mixture of P815AB and tum- peptides was also highly effective in inducing anti-P815AB reactivity, as was the combined use of P815AB and class II-restricted peptides of tetanus toxin or Plasmodium berghei circumsporozoite protein. While the effector phase of the CD8+ cell-mediated DTH to P815AB was unaffected by the ablation of CD4+ cells, the same treatment, or neutralization of IFN-gamma, negated the induction of reactivity if it occurred at the time of sensitization. Thus, defective activation of CD4+ cells may contribute to the poor immunogenicity of P815AB. Besides providing an insight into the mechanisms of anti-tumor protection induced by tum- cells, these data offer useful information for the design of vaccination strategies against identified tumor antigens.

We have observed delayed-type hypersensitivity (DTH) reactions in immunized mice challenged subcutaneously with class I-binding peptides related to rejection antigens recognized by cytotoxic T lymphocytes on mutagenized (tum-) variants of mastocytoma P815. As observed by skin test in virally infected mice challenged with viral peptides, the intrafootpad injection of tum- peptides resulted in a dose-dependent DTH that peaked at approximately 24 h. The response was mediated by CD8+ cells and could be induced by previous vaccination of mice with live tumor cells, intrasplenic deposition of the eliciting peptide, or adoptive transfer with peptide-pulsed syngeneic dendritic cells. These sensitization procedures resulted in an immunologically specific footpad reaction detectable for up to 2-6 months after priming. The evaluation by DTH in cancer patients of long-lived CD8+ anti-tumor T cell responses following local challenge with tumor-specific peptides may be of great interest in human immunotherapy trials involving immunization against identified tumor antigens.

Highly immunogenic tumor variants are generated by in vitro or in vivo treatment of the murine L5178Y lymphoma line with triazene derivatives. Most of these variants express new transplantation- and antibody-defined antigens that previous studies have shown to be closely related. One such 80-kDa protein on the surface of clone-D cells was found to be related to xenotropic MuLV gp70 molecules. To investigate the possible relevance of clone-D data to general properties of immunogenic variants in this tumor model system, polyclonal syngeneic antisera raised to a panel of immunogenic clones (including clone D) of the drug-treated L5178Y lymphoma line were employed in the immunoprecipitation of cell-surface and intrinsically labeled variant cells. In all clones, 1- and 2-dimensional electrophoretic analysis of the immunoprecipitates detected an antigen of approximately 80 kDa, and 35S-labeled 80-kDa molecules could be cross-precipitated from all clones by the panel of clone-specific antisera. In addition, 45- and 30-kDa components were also found in metabolically labeled variant cells. While the surface 80-kDa component was reactive with anti-xenotropic gp70 antibodies, the 30-kDa molecule was removed by anti-gag p30 antibody in sequential immunoprecipitation experiments. These data suggest that expression of aberrant, retrovirus-related proteins is a common finding in immunogenic cells of the drug-treated L5178Y lymphoma line.

Highly immunogenic tumor variants are generated by in vitro or in vivo treatment of L5178Y murine lymphoma cells with triazene derivatives. Most of these variants express new transplantation antigens which are not present on the original L5178Y tumor cells. In this study, a polyclonal syngeneic antiserum raised to one such variant (L5178Y/DTIC) was employed in immunoprecipitation studies of cell surface and metabolically labeled L5178Y/DTIC cells. One- and two-dimensional electrophoretic analyses of the immunoprecipitates detected a surface antigen of approximately 80 kDa. Additionally, a 45-kDa component was detected in the lysate of [35S]methionine-labeled cells. Anti-xenotropic MuLV gp70 serum precipitated material whose electrophoretic pattern was similar to that of the 80-kDa surface antigen. Sequential immunoprecipitation analysis revealed that the molecules reactive with the variant-specific antiserum were removed by the anti-xenotropic gp70 antibodies, whereas immunodepletion was only partial when the cell extract was first treated with the variant-specific antibodies. After Western blotting, the 80- and 45-kDa antigens precipitated by the variant-specific antibodies were injected intrasplenically into recipient mice. Only the animals sensitized with the 80-kDa antigen developed specific immunity to L5178Y/DTIC cells in that they displayed an increased frequency in CTL precursors (CTLp) to the variant cells. Sera from mice sensitized to the 80-kDa protein specifically inhibited the development of a primary CTL response to L5178Y/DTIC cells.

Previous work has shown that murine lymphoma cells antigenically altered ("xenogenized") by drug treatment elicit strong in vivo resistance to the original cells. Moreover, splenocytes immune to a drug-treated variant (L5178Y/DTIC) of a murine lymphoma exert anti-parental tumor activity in an adoptive transfer system, an effect mediated by L3T4+ lymphocytes and associated with the detection of an anti-L5178Y delayed-type hypersensitivity (DTH) response. We now report that the in vivo activity of the tumor-immune L3T4+ lymphocytes is a radio-sensitive (2,500 rad in vitro) phenomenon that requires collaboration with radio-resistant, silica-sensitive syngeneic cells in the host, and is inhibited by treatment of recipient mice with monoclonal antibodies (MAbs) to the L3T4 antigen or murine interferon-gamma (IFN-gamma). In vitro, the tumor-immune L3T4+ lymphocytes produce interleukin 2 (IL-2), lymphotoxin (LT) and IFN-gamma activities on incubation with L5178Y cells and spleen-adherent cells. These results suggest that the mechanisms of anti-parental tumor protection by xenogenized cells involve specific induction of a DTH response mediated by the "inflammatory" (THI) subset of L3T4+ T lymphocytes and IFN-gamma activated macrophages.

T cell-mediated proliferative and cytotoxic responses occur in vitro to syngeneic tumor cells antigenically altered by mutagen treatment. One such xenogenized variant of the murine L5178Y lymphoma elicits IgG antibodies reactive with determinants on variant cells that are not expressed at detectable levels on parental or normal cells of the same H-2d haplotype and are also unrelated to public specificities of H-2b or H-2k histocompatibility antigens. In the present study we investigated the effect of those antibodies on development of cell-mediated responses in vitro to the xenogenized cells used for induction of the humoral response. The proliferative reaction, generation of cytolytic activity and target cell lysis were all inhibited by the anti-xenogenized tumor immune serum, whereas the corresponding reactions to the parental cells by syngeneic or allogeneic effector lymphocytes were not. In order to investigate the possible H-2 association of T cell-mediated responses to xenogenized cells, we also examined the effect on those reactions of antibodies specific for Class I or Class II products of the H-2d complex. The results obtained suggested a role for I-Ad molecules in the T cell proliferative response to the xenogenized cells, and also indicated a preferential association of the cytotoxic response with H-2Kd determinants.

Because the antigenic changes occurring after in vivo treatment of murine lymphoma cells with 5-(3,3-dimethyl-1-triazenyl)-1H-imidazole-4-carboxamide (DTIC) were suggested to result from DTIC-induced somatic mutation(s), quinacrine dihydrochloride, an antimutagenic compound, was tested for possible antagonistic activity related to this phenomenon. The increased immunogenicity of L1210Ha leukemia occurring at the first transplant generation of DTIC-treated histocompatible (BALB/cCr x DBA/2Cr)F1 male mice or the appearance of strong lymphoma-associated transplantation antigens at transplant generation 6 was prevented by simultaneous administration of quinacrine. However, the compound did not modify the antitumor or immunodepressive activity of DTIC in the mouse. We concluded that the selective antagonistic effect of quinacrine on DTIC-mediated immunogenic changes (DMIC) supported the hypothesis that the molecular mechanism of DMIC could be related to somatic mutation(s).

Murine leukemia cells transformed by in vivo treatment with 5-(3,3-dimethyl-1-triazeno)imidazole-4-carboxamide (DTIC) are rejected by histocompatible recipients following inocula of 10(7) cells i.p. Progressive tumor growth or tumor growth and regression was monitored measuring the extent of DNA synthesis in the peritoneal cavity of mice using the [125I]5-iodo-2'-deoxyuridine uptake method. In addition, the results were confirmed by cell count and mortality data. Comparable growth rate was found initially in both DTIC and parental lines in histocompatible hosts. Later, mice challenged with parental lines died, whereas hosts inoculated with DTIC-treated sublines rejected the tumor. On the other hand, lethal growth occurred in mice inoculated with DTIC-treated sublines when immunodepressed by cyclophosphamide given before tumor challenge, or by methotrexate given after challenge of a methotrexate-resistant DTIC-treated subline. The similarity between the growth rate of the parental and DTIC-treated lines in histocompatible hosts does not support the hypothesis of impaired "oncogenic potential" of such DTIC-treated lines. Furthermore, the growth and rejection pattern of a parental line in H-2-incompatible hosts was similar to that observed for DTIC-treated lines in histocompatible hosts, suggesting that comparable immune mechanisms were involved in both cases.

A chemically induced lymphosarcoma line (LSBM-1) of C57BL/10 (H-2b) origin lacks detectable TATA and is scarcely immunogenic for H-2-incompatible congenic recipients. New antigenic specificities, defined as drug-mediated tumor antigen(s) (DMTA) were found in a subline (LBD-1) obtained by in vivo treatment of LSBM-1 with the anti-neoplastic agent dimethyl-1-triazeno-imidazole-carboxamide (DTIC) over several transplant generations. It was concluded that the presence of detectable TATA is not a prerequisite for the induction of new antigenic specificities (DMTA).

To determine whether antigen-presenting ability might be involved in the superior immunogenicity of chemically xenogenized tumors over that of parental cells, we tested a murine lymphoma line xenogenized by a triazene derivative for expression of Ia antigens, ability to present soluble antigen in vitro, and production of factor(s) active in a mouse thymocyte assay. Results showed that Ia antigens, absent on nonimmunogenic parental L5178Y cells, were expressed on a xenogenized, highly immunogenic tumor variant (clone D), as detected by immunofluorescence. While the ability of parental cells to stimulate lymphocyte proliferation in vitro was lost on removal of Ia+ cells from the responder population, considerable augmentation of reactivity was observed upon depletion of Ia+ cells from the population of splenocytes responding to the xenogenized cells. Under these conditions, stimulation was blocked by anti-Ia antibodies, or an anti-L3T4 reagent or antibodies to the novel antigenic determinants induced by xenogenization. In addition, no stimulating activity was observed following exposure of clone D cells to glutaraldehyde or lysosomotropic agents such as chloroquine and ammonia. When the ability of clone D cells to present ovalbumin in vitro was assayed, it was found that the xenogenized cells could present the soluble antigen to specifically primed lymphocytes. Moreover, clone D cells could substitute for splenic adherent cells in the proliferative reaction of splenocytes to concanavalin A. Finally, when the supernate from clone D-cell culture pulsed with phorbol myristic acetate was tested in a mouse thymocyte assay, considerable IL-1-like activity was disclosed.

To develop monoclonal antibodies (MAbs) recognizing drug-mediated tumor antigens on a chemically xenogenized murine lymphoma, hybridomas were constructed with splenocytes from histocompatible mice hyperimmunized with L5178Y cells antigenically altered by triazene treatment in vivo (clone D, derived from a polyclonal L5178Y/DTIC subline). Screening of supernatants with parental and xenogenized cells showed that nine MAbs displayed exclusive or preferential reactivity with clone D cells as detected by immunofluorescence, and failed, as a rule, to bind normal or unrelated malignant cells of the same or different haplotype. Moreover, no reactivity was displayed to the triazene-xenogenized variants of antigenically unrelated tumors. All nine MAbs, however, were capable of binding a panel of L5178Y/DTIC clones in addition to clone D. When the ability of these antibodies to interfere with the development of cell-mediated immunity to clone D cells in vitro was tested, it was found that the proliferative reaction and generation of cytolytic activity by syngeneic lymphocytes were inhibited by addition of several MAbs to the tumor--lymphocyte co-cultures.

The relationship between induction of novel immunogenicity by xenogenizing chemicals and DNA-methylating activity in murine tumors was investigated at the clonal level in L1210Ha cells treated with 5-azacytidine, N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) or 1-(p-chlorophenyl)-3,3-dimethyltriazene (DM-Cl). Cells were exposed to the drugs in vitro, cloned by limiting dilution, and assayed for transplantation immunogenicity and 5-methylcytosine content. The results showed that 0% (0/29, 5-azacytidine), 6.8% (2/29, MNNG) and 87.5% (28/32, DM-Cl) of the resulting clones were highly immunogenic, as judged by their tumorigenicity in intact compared to immunodepressed hosts. Frequency distribution analysis of the 5-methylcytosine content of drug-treated and parental clones showed that the methylation pattern was not significantly modified by tumor exposure to either 5-azacytidine or MNNG, and the two immunogenic clones induced by MNNG had methylcytosine levels very close to the 50th percentile value. In contrast, the extent of DNA methylation was increased in the cells treated with DM-Cl, but no obvious association was found between methylation status and immunogenicity of the drug-treated clones. In four 5-azacytidine-treated clones that displayed little or no immunogenicity, additional rounds of drug exposure led to progressive DNA demethylation, but failed, as a rule, to enhance tumor cell immunogenicity. Taken together, the present data indicate that, at least for the examined tumor, immunogenic variants are generated by mutagen treatment at high (MNNG) or very high (DM-Cl) frequencies under conditions in which hypomethylation-induced antigen amplification is unlikely.

Novel drug-mediated tumor antigens (DMTA) have been detected in chemically induced L5178Y lymphoma of DBA/2 origin, following treatment of tumor-bearing hosts with 5-(3,3-dimethyl-1-triazenyl)-1H-imidazole-4-carboxamide (DTIC). Studies were conducted on the susceptibility of the DTIC-treated subline, maintained in tissue culture, to NK-mediated cytolysis in vitro. The results of the experiments showed that DTIC-treated lymphoma cells are less susceptible to NK cell lysis than the parental line (i.e. L5178Y); the DTIC-treated lymphoma, maintained in vivo as ascitic form, behaved as the corresponding in vitro line, thus suggesting that NK-resistant phenotype was not dependent upon propagation conditions. The intrinsic susceptibility to cell-mediated lysis of the DTIC-treated tumor was unchanged, as demonstrated by lysis produced by alloimmune cytotoxic cells or by natural cytotoxic mesenteric lymph node effectors. "Cold" competition experiments and target binding assay performed with L5178Y/tc and its DTIC-treated subline showed that DTIC-altered cells are less efficient than the parental line as "cold" competitor cells for NK-mediated lysis and bind less efficiently than L5178Y tumor to NK lymphocytes. These data suggested that the NK resistance of the DTIC-treated lymphoma may result from a failure to bind to effector cells, as a consequence of a profound rearrangement of the cell surface produced by DTIC treatment of cancer cells.

A series of dimethyl aryl-triazene derivatives and related monomethyl compounds were studied for their efficacy in mediating a strong increase in immunogenicity (i.e., chemical xenogenization, CX) of murine leukemic cells following in vitro treatment. It was found that all compounds under investigation were able to induce CX. The dimethyl derivatives were able to induce CX only after metabolic activation, whereas related monomethyl compounds were active per se. The antigenicity acquired by triazene-treated leukemic cells was very marked; intact hosts histocompatible with the parental line were able to reject up to 10(7) cells. Antigenic tumor cells retained their immunogenic properties even after a large number of transplant generations in the absence of the drug. This means that marked immunogenicity of triazene-treated cells is a stable and heritable characteristic.