The Myc-evoked DNA damage response accounts for treatment resistance in primary lymphomas in vivo.

In addition to the ARF/p53 pathway, the DNA damage response (DDR) has been recognized as another oncogene-provoked anticancer barrier in early human tumorigenesis leading to apoptosis or cellular senescence. DDR mutations may promote tumor formation, but their impact on treatment outcome remains unclear. In this study, we generated ataxia telangiectasia mutated (Atm)-proficient and -deficient B-cell lymphomas in Emu-myc transgenic mice to examine the role of DDR defects in lymphomagenesis and treatment sensitivity. Atm inactivation accelerated development of lymphomas, and their DNA damage checkpoint defects were virtually indistinguishable from those observed in Atm+/+-derived lymphomas that spontaneously inactivated the proapoptotic Atm/p53 cascade in response to Myc-evoked reactive oxygen species (ROS). Importantly, acquisition of DDR defects, but not selection against the ARF pathway, could be prevented by lifelong exposure to the ROS scavenger N-acetylcysteine (NAC) in vivo. Following anticancer therapy, DDR-compromised lymphomas displayed apoptotic but, surprisingly, no senescence defects and achieved a much poorer long-term outcome when compared with DDR-competent lymphomas treated in vivo. Hence, Atm eliminates preneoplastic lesions by converting oncogenic signaling into apoptosis, and selection against an Atm-dependent response promotes formation of lymphomas with predetermined treatment insensitivity.

Gene transfer preferentially selects MHC class I positive tumour cells and enhances tumour immunogenicity.

The modulated expression of MHC class I on tumour tissue is well documented. Although the effect of MHC class I expression on the tumorigenicity and immunogenicity of MHC class I negative tumour cell lines has been rigorously studied, less is known about the validity of gene transfer and selection in cell lines with a mixed MHC class I phenotype. To address this issue we identified a C26 cell subline that consists of distinct populations of MHC class I (H-2D/K) positive and negative cells. Transient transfection experiments using liposome-based transfer showed a lower transgene expression in MHC class I negative cells. In addition, MHC class I negative cells were more sensitive to antibiotic selection. This led to the generation of fully MHC class I positive cell lines. In contrast to C26 cells, all transfectants were rejected in vivo and induced protection against the parental tumour cells in rechallenge experiments. Tumour cell specificity of the immune response was demonstrated in in vitro cytokine secretion and cytotoxicity assays. Transfectants expressing CD40 ligand and hygromycin phosphotransferase were not more immunogenic than cells expressing hygromycin resistance alone. We suggest that the MHC class I positive phenotype of the C26 transfectants had a bearing on their immunogenicity, because selected MHC class I positive cells were more immunogenic than parental C26 cells and could induce specific anti-tumour immune responses. These data demonstrate that the generation of tumour cell transfectants can lead to the selection of subpopulations that show an altered phenotype compared to the parental cell line and display altered immunogenicity independent of selection marker genes or other immune modulatory genes. Our results show the importance of monitoring gene transfer in the whole tumour cell population, especially for the evaluation of in vivo therapies targeted to heterogeneous tumour cell populations.