Search

The observation that depletion or inhibition of regulatory T cells (Tregs) unleashes efficient antitumor effector immune responses that can lead to tumor eradication in mice has opened new perspectives for the development of cancer immunotherapy. The quality and overall efficiency of the effector immune responses induced in the absence of Tregs seem to depend on multiple factors that determine the result of a battle involving effector T cells (Teffs), Tregs and tumor cells. In this study, we investigated the quality of tumor-associated antigens (TAAs) as one such factor. We show that the presence of a strong dominant antigen is required for the induction of effector responses capable of tumor eradication in the absence of Tregs. The sole addition of a dominant antigen on tumor cells does not change tumor growth in unmanipulated mice, but improves tumor eradication rate from a few to almost 100% in the absence of Tregs. This eradication can be shown to result from the recruitment and activation of specific Teffs recognizing this antigen. We also show that the presence of such dominant antigens has the side effect of restricting the breadth of the immune response to other TAAs, which could favor the generation of escape mutant by tumor editing. Taken together, our results highlight the potential, and some requirements for cancer immunotherapy based on Treg depletion. They also show that, ultimately, tumor fate depends on multiple factors that should all be taken into consideration for the design of more efficient immunotherapy.

In humans, no efficient treatment exists not only against multifocal liver metastases (LM) but also against recurrent microscopic liver metastases within the liver remnant following curative liver resection. Furthermore, in nonmultifocal LM, partial liver resection could be performed, but in more than 50% of the patients, tumor recurrence within liver remnant is observed, partly due to the growth of dormant cancer cells in the setting of postoperative host immune dysfunction. We investigated the therapeutic potential of interleukin-12 (IL-12) immuno-gene therapies in these experimental models under total vascular exclusion (TVE) of the liver. In rats with multiple LM of DHDK12 colon cancer cells, we observed a significant reduction in tumor volume after retroviral-mediated gene transfer of either herpes simplex virus thymidine kinase (HSV1-TK) and ganciclovir (GCV) administration, or IL-12. Combined treatment with HSV1-TK/GCV and IL-12 resulted in improved tumor volume reduction and even survival. In rats with recurrent microscopic DHDK12 LM established after partial liver resection, we observed significantly decreased recurrent tumor volumes and increased survival after retroviral-mediated IL-12 gene transfer. In both settings, immunohistological analysis revealed that IL-12 immuno-gene therapy was accompanied by an infiltration of CD8+ T lymphocytes within the tumors. Altogether, our results suggest that IL-12 adjuvant gene therapy could improve the management of patients with either resectable or unresectable LM.

Poor efficiency of gene transfer into cancer cells constitutes the major bottleneck of current cancer gene therapy. We reasoned that because tumors are masses of rapidly dividing cells, they would be most efficiently transduced with vector systems allowing transgene propagation. We thus designed two replicative retrovirus-derived vector systems: one inherently replicative vector, and one defective vector propagated by a helper retrovirus. In vitro, both systems achieved very efficient transgene propagation. In immunocompetent mice, replicative vectors transduced >85% tumor cells, whereas defective vectors transduced <1% under similar conditions. It is noteworthy that viral propagation could be efficiently blocked by azido-thymidine, in vitro and in vivo. In a model of established brain tumors treated with suicide genes, replicative retroviral vectors (RRVs) were approximately 1000 times more efficient than defective adenoviral vectors. These results demonstrate the advantage and potential of RRVs and strongly support their development for cancer gene therapy.

Adenovirus DNA is rapidly lost in actively dividing cells. In addition, first-generation (E1-defective) vectors trigger a strong cytotoxicity that impairs the duration of transgene expression. To solve these issues, we have developed a chimeric vector system that uses E1/E4 doubly defective adenoviruses for efficient production of infectious retroviral vectors. The retroviral vector sequences and packaging functions were split into two E1/E3/E4-deleted adenoviral vectors: the Moloney murine leukemia virus gag-pol cistron was expressed from the human EF1 alpha (elongation factor) promoter (AdGAG/POL), whereas the thymidine kinase transgene, embedded in a retroviral vector context, and an amphotropic retroviral envelope cassette were included within a second adenovirus (AdTK/ENV). This chimeric vector system was evaluated with a special emphasis on recombinant retrovirus production in vitro, as well as transgene amplification and persistence in vivo. Retrovirus titers of >10(5) infectious units/mL were routinely obtained in W162 cells coinfected with both recombinant adenoviruses. Long-term transgene persistence (up to 3 months) was demonstrated in vitro in two different cell lines coinfected with AdGAG/POL and AdTK/ENV, and correlated with the detection of specific provirus sequences. A 10- to 50-fold transgene amplification also was demonstrated in an in vivo tumor model infected with the Ad/Rt chimeric vector system. The chimeric vector system described herein combines the efficiency of gene delivery by recombinant adenoviruses with the integrative properties of infectious retroviral vectors. This versatile vector system may open up new avenues for efficient production of oncogenic, but also non-oncogenic, retroviruses from cells of non-murine origin.

Thymidine kinase (TK) genes from three alpha-herpesviruses (i.e., human herpes simplex type 1, varicella-zoster virus, equid herpesvirus 4) and two y-herpesviruses (i.e., Epstein-Barr virus and Saimiri herpesvirus 2) were cloned in expression vectors based on zeocin resistance by complementation of a TK-defective Escherichia coli strain. In vivo complementation of an appropriate yeast strain and in vitro enzymatic measurements demonstrated that all viral TKs possess a second phosphorylating activity corresponding to the thymidylate kinase function in contrast to the E coli TK, which is deprived of this activity. When expressed in an engineered E coli strain rendered resistant to purine and pyrimidine nucleoside analogs, the viral TKs sensitize host bacteria to 3'-azido-3'-deoxythymidine (AZT), 3'-deoxy-2',3'-didehydrothymidine (D4T), dideoxyinosine, or fluorodeoxyuridine (5-FUdR). The extent of activation of all these analogs, in this bacterial assay, was found to be greatly superior for the two gamma-virus TKs, compared to the alpha-virus TKs, including the reference suicide gene, HSV1-TK. TK from the two gamma-Epstein-Barr and Saimiri 2 viruses were also found to be more efficient in sensitizing murine melanoma B16 tumor cells to pyrimide nucleoside analogs.

We recently described a novel nonviral/viral vector for gene transfer, the plasmovirus (Noguiez-Hellin P, Robert-le Meur M, Laune S, et al. C R Acad Sci Paris, Sciences de la Vie. 1996;319:45-50; Noguiez-Hellin P, Robert-le Meur M, Salzmann J-L, et al. Proc Natl Acad Sci USA. 1996;93:4175-4180). Plasmoviruses are plasmids capable of expressing all the viral genes required for generating infectious particles and packaging a defective genome containing a transgene. Transfected as plasmids, plasmoviruses transform the transduced cells into packaging cells that release infectious replication-defective retrovirus vectors (RV) containing a transgene, which are capable of infecting nearby cells. We previously showed that such a vector can efficiently "propagate" the transgene after transfection. Here we examine in greater detail the different steps of plasmovirus replication in vitro in human (143 B TK-) and murine (NIH 3T3 TK-) cells. Molecular-biological analysis revealed plasmovirus-coded protein expression starting from 24 hours post-transfection, followed by the detection of infectious RV 48 hours post-transfection. The gag proteins were correctly processed in the released particles. Electron microscopic analysis revealed typical type C particles. Nonintegrated plasmovirus DNA was not toxic for the cells and could be detected for at least 14 days post-transfection. While the transfected gag gene and the transgene could also be detected throughout this period, we observed that env-coded proteins decreased after 72 hours post-transfection. Nevertheless, the production of RV resulted in the propagation of the transgene in the culture, with stable integration of plasmovirus proviral DNA into the host genome of infected cells. We show that this propagation results in a major improvement in therapeutic efficacy using an HSV1-TK transgene and ganciclovir treatment, when compared to that of plasmovirus constructs that cannot propagate. Altogether, these results demonstrate the functionality of this gene transfer method and suggest that improvements in the vector design enhance its efficacy.

The integration and expression of the herpes simplex virus type 1 thymidine kinase (HSV1-TK) gene in localized tumors results in tumor regression after the administration of the specific nucleoside analogue ganciclovir (GCV). Although only 10% to 20% of the tumor cells take up the HSV1-TK gene, the neighboring cells also die, a phenomenon termed "bystander effect.". In the present study, coinjection of the MC26 mouse colon carcinoma cell line and the HSV1-TK expressing retroviral packaging cell line followed after 7 days by the intraperitoneal administration of GCV resulted in almost total tumor regression in the immunocompetent BALB/c mice but not in immunocompromised athymic BALB/c mice. This suggested a strong cell-mediated immune component to the bystander effect.