Search

// Jana Simova 1 , Olena Sapega 1 , Terezie Imrichova 2 , Ivan Stepanek 1 , Lenka Kyjacova 2 , Romana Mikyskova 1 , Marie Indrova 1 , Jana Bieblova 1 , Jan Bubenik 3 , Jiri Bartek 2, 4, 5 , Zdenek Hodny 2 , Milan Reinis 1 1 Immunology Unit, Czech Centre for Phenogenomics, BIOCEV and Department of Transgenic Models of Diseases, Institute of Molecular Genetics of the ASCR, v.v.i., Prague 14220, Czech Republic 2 Department of Genome Integrity, Institute of Molecular Genetics, v.v.i., Academy of Sciences of the Czech Republic, Prague 14220, Czech Republic 3 First Faculty of Medicine, Charles University in Prague, Prague 12000, Czech Republic 4 Danish Cancer Society Research Center, Copenhagen DK-2100, Denmark 5 Department of Medical Biochemistry and Biophysics, Science For Life Laboratory, Division of Translational Medicine and Chemical Biology, Karolinska Institute, 17121 Solna, Sweden Correspondence to: Zdenek Hodny, email: hodny@img.cas.cz Milan Reinis, email: reinis@img.cas.cz Keywords: cellular senescence, cancer chemotherapy, docetaxel, IL-12, cell therapy Received: April 18, 2016 Accepted: May 29, 2016 Published: July 19, 2016 ABSTRACT Standard-of-care chemo- or radio-therapy can induce, besides tumor cell death, also tumor cell senescence. While senescence is considered to be a principal barrier against tumorigenesis, senescent cells can survive in the organism for protracted periods of time and they can promote tumor development. Based on this emerging concept, we hypothesized that elimination of such potentially cancer-promoting senescent cells could offer a therapeutic benefit. To assess this possibility, here we first show that tumor growth of proliferating mouse TC-1 HPV-16-associated cancer cells in syngeneic mice becomes accelerated by co-administration of TC-1 or TRAMP-C2 prostate cancer cells made senescent by pre-treatment with the anti-cancer drug docetaxel, or lethally irradiated. Phenotypic analyses of tumor-explanted cells indicated that the observed acceleration of tumor growth was attributable to a protumorigenic environment created by the co-injected senescent and proliferating cancer cells rather than to escape of the docetaxel-treated cells from senescence. Notably, accelerated tumor growth was effectively inhibited by cell immunotherapy using irradiated TC-1 cells engineered to produce interleukin IL-12. Collectively, our data document that immunotherapy, such as the IL-12 treatment, can provide an effective strategy for elimination of the detrimental effects caused by bystander senescent tumor cells in vivo .

The present study aimed to elucidate the role of cluster of differentiation (CD)8+, CD4+, natural killer (NK), and myeloid (CD11b+) cells in the course of the growth and rejection of experimental major histocompatibility complex (MHC) class I-deficient, HPV16 E6/E7-associated TC-1/A9 tumors in mice. Stable mouse lines (F30) generated by inbreeding of Balb/c and C57BL/6 strains, which were characterized by H-2Db+d-NK1.1neg (B6-neg) and H-2Db-d+NK1.1high (Balb-high) phenotypes, were used for the present study. The novel strains spontaneously regressed tumors in 70–90% of cases. Ex vivo histological analysis of the tumor microenvironment in cryosections showed an indirect correlation between the growth of the transplanted tumor (progressor vs. regressor mice) and the proportion of immunocompetent cell infiltration in the tumors. The regressor mice exhibited a higher infiltration of tumors with CD4+ and CD8+ cells, and in Balb-high with NK cells as well, compared with the progressors. All tumor transplants also indicated a huge infiltration of CD11b+ cells, but this infiltration was not dependent on the stage of the TC-1/A9 tumor development. Depletion of individual cell subpopulations in vivo exhibited different effects on the tumor development in the two strains. Elimination of CD8-positive cells enhanced growth of TC-1/A9 tumor transplants in both hybrid stains, whereas CD4+ cell depletion affected rejection of TC-1/A9 tumors in the B6-neg mice only. Depletion of NK cells with anti-asialo GM1 antibody in the Balb-high strain led to enhancement of tumor growth, which was more pronounced after depletion of the NK1.1+ subpopulation. On the other hand, depletion of NK cells with anti-asialo GM1 in B6-neg mice did not affect the regression of TC-1/A9 tumor transplants, but increased the CD11b+ cell infiltration. In summary, these results indicate that co-operation of particular subsets of immunocompetent cells is essential for the rejection of TC-1/A9 tumor transplants. In B6-neg mice, the co-operative action of CD8+ and CD4+ cells is required, whereas in Balb-high mice, the synergy of CD8+ and NK1.1+ cells is of major importance.

Dendritic cell (DC)-based vaccines pulsed with high hydrostatic pressure (HHP)-inactivated tumor cells have recently been shown to be a promising tool for prostate cancer chemoimmunotherapy. In this study, DC-based vaccines, both pulsed and unpulsed, were as effective as docetaxel (DTX) in reducing prostate tumors in the orthotopic transgenic adenocarcinoma of the mouse prostate (TRAMP) model. However, we did not observe any additive or synergic effects of chemoimmunotherapy on the tumor growth, while only the combination of DTX and pulsed dendritic cells resulted in significantly lower proliferation detected by Ki67 staining in histological samples. The DC-based vaccine pulsed with HHP-treated tumor cells was also combined with another type of cytostatic, cyclophosphamide, with similar results. In another clinically relevant setting, minimal residual tumor disease after surgery, administration of DC-based vaccines after the surgery of poorly immunogenic transplanted TRAMP-C2, as well as in immunogenic TC-1 tumors, reduced the growth of tumor recurrences. To identify the effector cell populations after DC vaccine application, mice were twice immunized with both pulsed and unpulsed DC vaccine, and the cytotoxicity of the spleen cells populations was tested. The effector cell subpopulations were defined as CD4+ and NK1.1+, which suggests rather unspecific therapeutic effects of the DC-based vaccines in our settings. Taken together, our data demonstrate that DC-based vaccines represent a rational tool for the treatment of human prostate cancer.

Background: Epigenetic mechanisms have important roles in the tumour escape from immune responses, such as in MHC class I downregulation or altered expression of other components involved in antigen presentation. Chemotherapy with DNA methyltransferase inhibitors (DNMTi) can thus influence the tumour cell interactions with the immune system and their sensitivity to immunotherapy. Methods: We evaluated the therapeutic effects of the DNMTi 5-azacytidine (5AC) against experimental MHC class I-deficient and -positive tumours. The 5AC therapy was combined with immunotherapy, using a murine model for HPV16-associated tumours. Results: We have demonstrated 5AC additive effects against MHC class I-positive and -deficient tumours when combined with unmethylated CpG oligodeoxynucleotides or with IL-12-producing cellular vaccine. The efficacy of the combined chemoimmunotherapy against originally MHC class I-deficient tumours was partially dependent on the CD8+-mediated immune responses. Increased cell surface expression of MHC class I cell molecules, associated with upregulation of the antigen-presenting machinery-related genes, as well as of genes encoding selected components of the IFNγ-signalling pathway in tumours explanted from 5AC-treated animals, were observed. Conclusion: Our data suggest that chemotherapy of MHC class I-deficient tumours with 5AC combined with immunotherapy is an attractive setting in the treatment of MHC class I-deficient tumours.

Dendritic cells (DCs) are believed to be the most potent antigen-presenting cells able to link the innate and adaptive immune systems. Many studies have focused on different immunotherapeutic approaches to applying DCs as tools to improve anticancer therapy. Although a number of investigations suggesting the benefit of DC-based vaccination during anticancer therapy have been reported, the general knowledge regarding the ultimate methods of DC-vaccine preparation is still unsatisfactory. In this article, the perspectives of DC-based anti-tumor immunotherapy and optimizing strategies of DC vaccination in humans in light of results obtained in mouse models are discussed.

High hydrostatic pressure (HHP) has been shown to induce immunogenic cell death of cancer cells, facilitating their uptake by dendritic cells (DC) and subsequent presentation of tumor antigens. In the present study, we demonstrated immunogenicity of the HHP-treated tumor cells in mice. HHP was able to induce immunogenic cell death of both TC-1 and TRAMP-C2 tumor cells, representing murine models for human papilloma virus-associated tumors and prostate cancer, respectively. HHP-treated cells induced stronger immune responses in mice immunized with these tumor cells, documented by higher spleen cell cytotoxicity and increased IFNγ production as compared to irradiated tumor cells, accompanied by suppression of tumor growth in vivo in the case of TC-1 tumors, but not TRAMP-C2 tumors. Furthermore, HHP-treated cells were used for DC-based vaccine antigen pulsing. DC co-cultured with HHP-treated tumor cells and matured by a TLR 9 agonist exhibited higher cell surface expression of maturation markers and production of IL-12 and other cytokines, as compared to the DC pulsed with irradiated tumor cells. Immunization with DC cell-based vaccines pulsed with HHP-treated tumor cells induced high immune responses, detected by increased spleen cell cytotoxicity and elevated IFNγ production. The DC-based vaccine pulsed with HHP-treated tumor cells combined with docetaxel chemotherapy significantly inhibited growth of both TC-1 and TRAMP-C2 tumors. Our results indicate that DC-based vaccines pulsed with HHP-inactivated tumor cells can be a suitable tool for chemoimmunotherapy, particularly with regard to the findings that poorly immunogenic TRAMP-C2 tumors were susceptible to this treatment modality.

Our study was designed to examine the effects of IL-2 gene therapy in a surgical minimal residual tumour disease (SMRTD). Mice were inoculated s.c. with methylcholanthrene (MC)-induced MC12 sarcoma cells. When the tumours reached 8 to 12 mm in diameter, they were excised, either completely (“microscopic SMRTD”) or incompletely (“macroscopic SMRTD”). On day 90 after surgery, the tumour recurrence rate in untreated mice with microscopic SMRTD was approximately 30%, whereas in those with macroscopic SMRTD it was 75%. After surgery, experimental mice were treated with 2 types of irradiated, IL-2 gene-modified, IL-2-producing tumour cell vaccine. One type of vaccine was derived from the MC12 sarcoma cells (MC12-IL2/IV-3); the other type was derived from an unrelated X63-Ag8.653 plasmacytoma (X63-m-IL-2). Both types of vaccine failed to cure the macroscopic SMRTD. Whereas the X63-m-IL-2 vaccine was also ineffective in the microscopic SMRTD, the MC12-IL2/IV-3 vaccine was capable of preventing growth in all but one mouse (1/64) with microscopic SMRTD when administered 2 to 5 days after surgery. If the vaccination took place 2 days before surgery or later than 5 days after surgery, the therapeutic activity was lost. Vaccination with irradiated parental MC12 cells did not produce any significant benefit compared to the operated-only mice. The protective effect of the MC12-IL2/IV-3 vaccine was specific and comparatively long-lasting. Vaccinated mice, which had rejected the MC12 tumour residuum, were capable of rejecting a second inoculum of the MC12 sarcoma cells injected on days 35 to 110 after surgery but succumbed to the growth of 2 other unrelated murine sarcomas carrying different tumour-rejection antigens. Int. J. Cancer 76:115–119, 1998.© 1998 Wiley-Liss, Inc.

Experiments were designed to investigate a possible therapeutic role of interleukin-2 (IL-2) gene transfer in the model of murine (EL-4) leukaemia pretreated with cyclophosphamide. It has been found that i. p. pretreatment of the leukaemic mice with cyclophosphamide, followed by i. v. administration of irradiated cells, genetically engineered to produce IL-2 and used as a source of the cytokine (IR-IL-2 cells), cured a substantial percentage of the leukaemic mice. Neither treatment with cyclophosphamide nor administration of the IR-IL-2 cells alone had any significant therapeutic effect. Labelling of the EL-4 and IR-IL-2 cells with different fluorescent cell linkers followed by i. v. injection and detection of the labelled cells in cryostat sections of various organs has shown that both cell populations can be detected almost exclusively in the red pulp of the spleen, close to the white pulp nodules, thus providing the possibility of short-range local interactions among the IL-2-producing cells, IL-2-responsive defence effector cells and EL-4 leukaemia targets.

The present prospective study was designed to assess whether the renal cell carcinoma (RCC) patients treated with recombinant interferon alpha (IFN alpha), whose tumours respond (responders) and do not respond (non-responders) to IFN alpha therapy, differ with regard to in vitro sensitivity of peripheral blood lymphocytes (PBL) to interleukin 2 (IL-2), IFN alpha, and IFN gamma signals prior to therapy. Twenty-one patients with advanced RCC after nephrectomy, 15 responders and 6 non-responders, were entered into a protocol. The protocol involved isolation and freezing of PBL samples followed by IFN alpha treatment of patients, assessment of proliferative and activating PBL responses, and evaluation of the therapeutic results. Freezing of PBL samples allowed us to compare the in vitro reactivity of PBL from individual RCC patients, repeatedly and under standard conditions. Substantial differences in proliferative responses to the mitogenic IL-2 signal of PBL derived from IFN alpha responders and nonresponders were found. Whereas the IL-2-induced proliferative responses of PBL from normal blood donors and IFN alpha responders were comparable, the proliferative responses of PBL from IFN alpha non-responders were significantly decreased, suggesting an immune dysfunction in non-responders. Cultivation of PBL from RCC patients in medium supplemented with IFN alpha increased the lytic activity of PBL from IFN alpha responders directed against RCC targets; no such increase could be observed with non-RCC targets, with PBL from IFN alpha non-responders, or with PBL from normal blood donors. Detection of phytohaemagglutinin (PHA)-stimulated IFN gamma secretion by PBL at the single cell level using enzyme-linked immunospot (ELISpot) assay revealed that the ability to produce IFN gamma was substantially decreased in IFN gamma non-reponders, as compared to IFN alpha responders and to normal blood donors.

Genetically modified tumour cells producing cytokines such as interleukin 12 (IL-12) are potent activators of the antitumour immune responses and represent a promising therapeutic modality when combined with chemotherapy. The objective of this study was to examine whether IL-12-producing cellular vaccines can augment chemotherapy of human papilloma virus (HPV) 16-associated murine tumours with the cytostatic agent gemcitabine (GEM). We found that peritumoral administration of IL-12-producing tumour vaccines enhanced the effect of cytoreductive therapy with GEM both in non-metastasizing murine carcinoma TC-1 and in metastasizing murine carcinoma MK16. The percentage of mice with MK16 metastases and the number of lung metastatic nodules was substantially decreased. In another clinically relevant setting, surgical minimal residual tumour disease, the administration of IL-12-producing tumour vaccine and GEM after the MK16 tumour surgery reduced the percentage of mice with tumour recurrences; similarly, the percentage of metastasis-bearing mice and the number of metastatic nodules was decreased. Tumour inhibitory effects exerted by GEM plus IL-12 were associated with high production of interferon-γ (IFNγ) by splenocytes. Our results suggest that the IL-12-producing vaccine can enhance the effect of GEM chemotherapy in some HPV16-associated murine tumour models.

Loss or downregulation of MHC class I molecules on tumour cells is a common mechanism by which tumours can escape T-cell mediated immune responses. In this study, we examined the role of different immune cell lineages in the development of immunity against tumours of the same aetiology but with different MHC class I expression. In vivo depletion of CD8+ cells, but not of CD4+ or NK1.1+ cells in the immunization period resulted in complete elimination of the protective effects of immunization with irradiated TC-1 cells (MHC class I-positive cell line) against the TC-1 tumour challenge. After immunization with irradiated TC-1/A9 or with MK16 tumour cells (MHC class I-deficient sublines) a remarkable dependence on the presence of NK1.1+ cells was observed, while the tumour growth inhibition after CD4+ or CD8+ depletion was not efficient. Cytotoxic activity induced by TC-1 cell immunization was significantly abrogated in the CD8+ and CD4+ but not NK1.1+ cell-depleted mice, as compared to the immunized only controls. After MK16 or TC-1/A9 cell immunization, NK1.1+ but not CD8+ and CD4+ cell-depleted mice displayed significant reduction of specific cytotoxicity. Mice immunized with TC-1 cells showed similar percentage of IFNγ producing cells in CD8+, CD4+ and NK1.1+ cell populations. On the other hand, the highest proportion of IFNγ producing cells after immunization with TC-1/A9 or MK16 cells was concentrated into the NK1.1-positive spleen cell population. Our data demonstrate that the development of immunity against MHC class I-deficient tumours is highly dependent on the activity NK1.1+ cell population.

Natural killer T (NKT) cells are potent modulators of antitumor immunity. Their protective effects can be achieved upon their activation by glycolipid ligands presented in the context of the CD1d molecule. These CD1d-binding glycolipid antigens have been described as potent therapeutic agents against tumors, infections, as well as autoimmune diseases. Immunoregulatory and therapeutic effects of glycolipid ligands depend on their structure and modes of administration. Therefore, more studies are needed for optimization of the particular therapeutic settings. This study was focused on the tumor-inhibitory effects of 12 carbon acyl chain beta-galactosyl ceramide (C12 beta-D-Galactosyl Ceramide; beta-GalCer(C12)) on the growth of human papillomavirus type 16 (HPV16)-associated neoplasms transplanted in syngeneic mice. Treatment of tumor-bearing mice with beta-GalCer(C12) 3-14 days after tumor cell transplantation significantly inhibited the growth of the major histocompatibility complex (MHC) Class I-positive (TC-1), as well as MHC Class I-deficient (TC-1/A9) HPV16-associated tumors. Moreover, administration of beta-GalCer(C12) after surgical removal of TC-1 tumors inhibited the growth of tumor recurrences. Similar results were obtained in the treatment of tumors after chemotherapy. beta-GalCer(C12) treatment turned out to be also synergistic with immunotherapy based on administration of IL-12-producing cellular vaccines. These results suggest that beta-GalCer(C12), whose antitumor effects have so far not been studied in detail, can be effective for the treatment of minimal residual tumor disease as well as an adjuvant for cancer immunotherapy.

We have established animal models of HPV16-associated tumours with distinct levels of MHC class I expression. This model was used for examination of immune responses, production of cytokines and kinetics of immune cell subsets after IL-12 therapy of minimal residual tumour disease induced by CBA-4A (cyclophosphamide derivative) treatment. Upregulation of cytokine production was detected, compared to control animals without tumours. No differences in Th1/Th2 polarization of the immune responses after immunotherapy in animals bearing tumours with different surface expression of MHC class I molecules were observed. In the spleens of TC-1 (MHC class I+) but not of TC-1/A9 (MHC class I-) treated tumour-bearing animals, the cytotoxic CD8+ cells detectable in 51Cr microcytotoxicity assay, were found. In the spleens of TC-1/A9 but not of TC-1 tumour-treated animals, the NK activity measured as the lysis of NK-sensitive YAC-1 targets was detected. Down-regulation of the CD4+ and CD8+ subpopulations in spleens of tumour-bearing animals were not restored after therapy. The percentage of CD25+/CD4+ T regulatory (Treg) cells in lymph nodes remained unchanged. The cytoreductive chemotherapy led to strong upregulation and accumulation of immunosuppressive immature myeloid Gr-1+/CD11b+ cells (IMC) in the spleens of treated animals. The accumulation of Gr-1+/CD11b+ cells was significantly decreased after subsequent IL-12 immunotherapy. These data suggest that elimination of IMC after IL-12 immunotherapy may be responsible for the improvement of antitumour responses after adjuvant IL-12 vaccination for the treatment of CMRTD.

Oligodeoxynucleotides containing guanine-cytidine dimers (CpG ODN) are potent inducers of anti-tumour immune responses. In this study, we analyzed the capacity of CpG ODN to inhibit the growth of both MHC class I-positive and -deficient tumours after debulking the tumour mass by chemotherapy or surgery. We employed an animal model resembling human papillomavirus (HPV) 16-associated tumours. Tumour cell lines with distinct cell surface expression of the MHC class I molecules were injected into syngeneic C57BL/6 mice, and the growing tumours were either subjected to cytoreductive chemotherapy with ifosfamide derivative, CBM-4A, or surgically removed. Subsequent treatment with synthetic CpG ODN significantly blocked the growth of the recurrent tumours. Our results indicate that the therapy with CpG ODN can be effective for the treatment of minimal residual tumour disease of the tumours that have escaped from the immune surveillance by downmodulating the MHC class I expression.

Loss or downregulation of MHC class I molecules on tumour cells is a common mechanism by which tumours can escape from T-cell mediated immune responses. In this study we have investigated the immunologic crossreactivity between murine tumour cell lines expressing human papilloma virus (HPV) 16-derived E6/E7 oncoproteins with distinct surface expression of MHC class I molecules. The aims of this study were to demonstrate whether immune responses capable of coping with MHC class I-positive tumours can also be effective against their MHC class I-deficient derivatives and whether it is possible to induce immunity against MHC class I-deficient tumours by cellular vaccines based on MHC class I-deficient tumour cell lines. Our data showed that immunization with MHC class I-deficient but not with MHC class I positive tumour cells inhibited the growth of MHC class I-deficient tumours. In vivo depletion studies revealed that the mechanisms underlying effective immune responses against MHC class I-negative tumours in animals immunized with MHC class I-deficient tumour cells involved natural killer cells. The presented findings are of particular clinical relevance in the sense of construction of vaccines directed against a broad spectrum of HPV-associated tumours.

Moderately immunogenic HPV 16-associated murine tumour cell line mimicking human HPV 16-associated neoplasms TC-1 (MHC class I + ) and its variants, TC-1/P3C10 and TC-1/A9, with a marked down-regulation of MHC I molecules, were used to examine the effect of local interleukin 12 (IL-12) gene therapy for the treatment of early tumour transplants and minimal residual tumour disease obtained after cytoreductive chemotherapy (CMRTD). Experiments were designed to examine whether down-regulation of MHC class I molecules plays a role during chemotherapy and gene therapy of early tumour transplants. It was found that peritumoral administration of IL-12-producing tumour cell vaccines (single dose, day 8 after tumour cell administration) inhibited the growth of both TC-1 (MHC class I positive) tumours and their MHC class I-deficient variants. To investigate the antitumour effects in a clinically relevant setting, IL-12 gene therapy was utilised for the treatment of minimal residual tumour disease after cytoreductive chemotherapy. Intra-peritoneal treatment of tumour-bearing mice with ifosfamide derivative, CBM-4A, produced a significant tumour-inhibitory effect. This treatment was followed by peritumoral s.c. administration of genetically modified TC-1 (MHC class I positive) or MK16/I/IIIABC (MHC class I negative) vaccines producing IL-12 (single dose, day 7 after chemotherapy) or with recombinant interleukin 12 (rIL-12) in two cycles of 5 daily doses (days 8-19) after chemotherapy. This combined therapy significantly inhibited the growth of TC-1 and TC-1/A9 (MHC class I-) tumours. When the combined therapy of TC-1 (MHC class I positive) tumours was followed by peritumoral administration of bone marrow dendritic cell (BMDC) vaccines, the IL-12-mediated inhibitory effect was significantly boosted. In the next set of experiments, the impacts of chemotherapy and IL-12 adjuvant therapy on MHC class I surface expression were assessed. Chemotherapy and gene therapy of tumours led to the up-regulation of MHC I expression on MHC class I-deficient tumours (TC-1/A9 and TC-1/P3C10) and to down-regulation on MHC I-proficient tumours (TC-1). These findings indicate that the MHC I phenotype is not stable during tumour progression and treatment. Collectively, these results illustrate the efficacy of IL-12 gene therapy in combination with chemotherapy on HPV-associated tumours regardless of the level of MHC class I expression on the tumour cells.

Moderately immunogenic HPV16-associated tumours TC-1 (MHC class I+, HPV16 E6/E7+, G12V Ha-ras+) and MK16/1/IIIABC (MK16, MHC class I-, HPV16 E6/E7+, G12V Ha-ras+), both of the H-2b haplotype and transplanted in syngeneic mice, were used to examine the effects of local IL-2 and GM-CSF cytokine or gene therapy in the treatment of minimal residual tumour disease. The mice carrying MHC class I+ TC-1 tumour residua after surgery were injected into the site of the surgery either with irradiated, IL-2 gene-modified MK16 tumour cells, or with recombinant IL-2. It has been found that both, the recombinant IL-2 and the IL-2 gene-modified tumour vaccine substantially reduced the percentage of tumour recurrences in the operated mice. Similarly, when the mice carrying TC-1 tumour residua after surgery were injected with recombinant GM-CSF, the recombinant GM-CSF inhibited growth of the tumour residua in the operated mice. Gene therapy with irradiated, GM-CSF secreting MK16 cells did not produce any tumour-inhibitory effect. In further experiments, mice bearing s.c. TC-1 tumours were injected i.p. with ifosfamide derivative CBM-4A and 8 days later, peritumourally, either with IL-2 gene-modified and IL-2-producing MK16 cells, or with recombinant IL-2. It has been found that both, the recombinant IL-2 and the IL-2 gene therapy substantially reduced the percentage of tumour-bearing mice. When the mice bearing s.c. TC-1 tumours were injected i.p. with ifosfamide derivative CBM-4A and then, peritumourally, either with irradiated, GM-CSF gene-modified and GM-CSF-producing MK16 cells, or with recombinant GM-CSF, it was found that both, the recombinant GM-CSF and GM-CSF gene therapy inhibited growth of tumour residua. Comparative experiments were performed with the MHC class I-, metastasizing tumour MK16. It has been found that both, recombinant IL-2 and GM-CSF, can inhibit growth of the tumour residua after surgery or chemotherapy. The lung metastases in mice with surgical minimal residual tumour disease or in mice with tumour residua after chemotherapy were inhibited by IL-2 but not by GM-CSF. The MK16 tumour vaccine producing IL-2 inhibited growth of tumour residua after chemotherapy, but not the tumour residua after surgery. The GM-CSF-producing vaccine was without significant effect in both, surgically- and chemotherapeutically-induced minimal residual MK16 tumour disease. In conclusion, the MHC class I+ and MHC class I-, HPV16-associated tumours were found to be sensitive to IL-2 and GM-CSF therapy after surgery or after cytoreductive chemotherapy. It is yet to be addressed if this is more general case with HPV16-associated experimental tumours. If so, it would be of interest to further investigate whether such adjuvant therapy can also help to eradicate the residua after surgery and chemotherapy in patients carrying HPV16-associated neoplasms.

Oncogenic, moderately immunogenic, MHC class I- and class II-, B7-, MK16/1/III ABC (MK16) cells were previously established by co-transfection of HPV16 E6/E7 and activated H-ras oncogene DNA into C57BL/6 kidney cells. Subcutaneous transplantation of these cells produced progressively growing local neoplasms which metastasized spontaneously to lungs and lymph nodes. The MK16 cells were implanted into syngeneic mice and used to examine whether the tumour lacking the signal molecules required for the induction of and sensitivity to T cell immunity is susceptible to local IL-2 treatment and IL-2 gene therapy. Peritumoural administration of human rIL-2 or murine IL-2 gene-modified MK16 tumour vaccine inhibited growth of subcutaneous MK16 tumour transplants and reduced the number of their lung metastases. Spleen cells from MK16 tumour-immunized mice were not cytolytic when allowed to react with the MK16 target cells, although they efficiently lysed the MHC class I+ malignant TC1 cells, obtained from C57BL/6 lung cell cultures after transfection with the same plasmids as those used for the derivation of the MK16 cells. However, when the MK16 cells were cultivated in vitro in the presence of IFNgamma, they acquired, together with the expression of MHC class I molecules, the sensitivity to the cytolytic effect of spleen cells from the MK16 tumour-immunized mice. These results indicate that experimental tumours which are MHC class I- and mimick in this respect a high proportion of human HPV16-associated carcinomas are suitable for IL-2 treatment.

Interleukin-2 and CD80 transfectants of a methyl-cholanthrene-induced murine sarcoma Mc12 (Mc12-IL-2 and Mc12-CD80 cells) with similar tumorigenicity in euthymic mice were utilized for experiments designed to investigate a co-stimulatory role of the CD80 molecule in allogeneic, congenitally athymic (nu/nu) mice. The CD80-transfected cells were as tumorigenic in nu/nu mice as the parental Mc12 sarcoma. The IL-2-transfected cells grew only transiently and regressed in all nu/nu recipients during four weeks after challenge with doses up to 5x10 7 cells. The 1: 1 mixture of parental Mc12 with Mc12-CD80 cells grew progressively in all inoculated nu/nu mice; in a 1: 1 mixture with parental Mc12 cells, Mc12-IL-2 cells were able to cause regressions in approximately 50% of nu/nu mice; the 1: 1 mixture of Mc12-IL-2 and Mc12-CD80 transfectants showed only transient growth and regressed during four weeks in all inoculated nu/nu mice. Adoptive transfer of cell-mediated immunity revealed that spleen cells from tumor regressors were capable of transferring the resistance to Mc12 tumor in nu/nu mice. The spleen cells from tumor regressors were not cytolytic when allowed to react in vitro with Mc12, Mc12-IL-2, or Mc12-CD80 target cells. However, when grown in IL-2-containing medium, splenocytes from tumor regressors, but not the splenocytes from tumor progressors, could develop cytolytic activity directed against Mc12 target cells that was comparable to that of the splenocytes from tumor-free controls. These results suggest that the rejection of tumors in nu/nu mice was mediated by IL-2-dependent mechanisms in which the CD80 molecule played a co-stimulatory role; the results also indicate that the ability to be activated by IL-2 and to give rise to cytolytic activity of nu/nu splenocytes fro tumor progressors is decreased.

The effect of irradiation on the therapeutic efficacy of IL-2 gene-modified plasmacytoma cells used as a vaccine in the immunotherapy of parental murine plasmacytoma X63-Ag8.653 was examined. Local administration of the IL-2-secreting plasmacytoma irradiated with a dose of 50 Gy inhibited i.p. plasmacytoma growth more effectively than the administration of non-irradiated, live cell vaccines. Whereas the vaccination with the live cell vaccine could substantially prolong the survival of the tumour-bearing mice but did not significantly induce tumour regressions, the irradiated vaccines could substantially increase the number of tumour-free animals. The irradiated vaccines produce higher amounts of IL-2 than the live cell vaccines both in vitro and in vivo. Depletion of CD4+ and CD8+ effector cells with monoclonal antibodies has significantly decreased the effect of the vaccination. It can be concluded that both, CD4+ and CD8+ T lymphocytes are required for effective IL-2 gene therapy of the X63-Ag8.653 plasmacytoma and that the higher effect of the irradiated vaccines is probably due to their higher IL-2 production.

Two genes, the gene coding for IL-2 and the gene encoding the CD80 molecule, were inserted into murine sarcoma MC12 cells. Tumorigenicity of a variety of cell clones with different expression of the inserted genes was assessed. Most of the genetically manipulated MC12 cell clones were less tumorigenic than the parental MC12 cell population. Tumorigenicity of the clones declined with increasing production of IL-2 as well as with the increasing expression of the CD80 molecule. When the tumorigenicity of the clones carrying an inserted IL-2 gene was compared with that of the clones carrying an inserted CD80 gene, it was found that the insertion of the IL-2 gene suppresses tumorigenicity more efficiently than insertion of the CD80 gene. Admixture of the IL-2-producing MC12 clones to the tumorigenic CD80(+) MC12 cell doses could completely inhibit the tumorigenicity of the CD80(+) cells. Insertion of the CD80 gene into sarcoma cells substantially enhanced the adhesive interaction between the MC12 sarcoma and syngeneic T lymphocytes.

The present study was designed to examine the kinetics and function of peritoneal exudate cells (PEC) during local interleukin 2 (IL-2) gene therapy of the X63-Ag8.653 plasmacytoma growing in the peritoneal cavity. BALB/c mice were inoculated i.p. on day 0 with a tumorigenic dose of the syngeneic plasmacytoma and on day 1 with non-tumorigenic plasmacytoma cells carrying an inserted IL-2 gene and producing constitutively IL-2. At regular time intervals the experimental mice were sacrificed and their peritoneal exudate cells were used for phenotypic analysis and Cr-51 microcytotoxicity assay. On the first day after i.p. inoculation of the genetically modified plasmacytoma cells the percentage of Thy 1.2+, CD3+, TCR alphabeta+ T lymphocytes and NK+ cells in the peritoneal fluid dramatically increased. The levels of the positive cells continually decreased until day 11, when the values of normal, healthy mice were obtained. The percentage of Thy 1.2+ and CD3+ cells remained at these, or slightly lower values, until the end of the observation period. A similar, though more slowly descending kinetics was seen in the CD5+ cell population, whereas the CD8+ cells, compared to the controls, exhibited only a short-term peak between days 3 and 5, and the values of TCR alphabeta+ and NK+ cells exhibited a second peak between days 25 and 48. The percentage of TCR gammadelta+ cells showed a permanent, moderately elevated plateau from day 1 till the end of the observation period. In control, untreated mice, inoculated i.p. with the X63-Ag8.653 plasmacytoma, the kinetics of peritoneal exudate cells was different. A moderate, permanent elevation of all of the T and NK cell subsets examined occured during the observation period. In addition, the percentage of TCR alphabeta+, TCR gammadelta+ and NK+ cells further increased continually from day 11 till the end of the observation period. The cytolytic activity of the peritoneal exudate cells was examined in vitro concurrently with FACS phenotyping. Free tumour-specific killer cells generated in the peritoneal fluid due to the local IL-2 gene therapy were found only on day 6, and these cells were cytolytic for both, the parental X63-Ag8.653 and the genetically modified X63-m-IL-2 plasmacytoma cells.

Insertion of functional interleukin-2 (IL-2) gene into a plasmacytoma cell line X63-Ag8.653 substantially reduced tumorigenicity of the resulting cloned cells, designated as X63-m-IL-2. Peritumoral administration of the X63-m-IL-2 cells, producing constitutively large quantities of IL-2, resulted in regressions of established X63-Ag8.653 plasmacytomas growing in the peritoneal cavity of syngeneic mice. In vitro activation of BALB/c spleen cells by co-culture with X63-m-IL-2 cells or their supernatants gave rise to cytotoxic lymphocytes with lymphokine-activated killer (LAK) activity against syngeneic X63-Ag8.653 plasmacytoma and other tumor targets. In contrast, peritumoral administration of X63-Ag8.653 cells carrying an inserted interleukin-4 (IL-4) gene (designated X63-m-IL-4 cells) and producing constitutively large quantities of IL-4 did not result in a therapeutic effect. Moreover, the admixture of the X63-m-IL-4 and X63-m-IL-2 cells substantially diminished the X63-m-IL-2 cell-mediated therapeutic effect. Similarly, IL-4-containing supernatants generated from X63-m-IL-4 cell cultures substantially diminished LAK activation by X63-m-IL-2 cell produced supernatants.

We have examined the effect of IL-12-producing cellular vaccines on the cytotoxicity and proliferative potential of CD45+ tumour-infiltrating cells (TIL) in mice carrying syngeneic TC-1 and TC-1/A9 HPV 16-associated tumours after chemotherapy with CBM-4A ifosfamide derivative. The chemotherapy resulted in the decrease of the CD4+ and CD8+ TIL, increase of the Gr-1+/CD11b+ TIL, no changes in the infiltration with CD4+/CD25+ Treg TIL, and decrease of the cytolytic and proliferative potential of the CD45+ TIL. Subsequent immunotherapy with the IL-12-producing, genetically modified TC-1 (TC-1-IL-12) cells increased tumour infiltration with CD8+ and CD4+ cells, decreased the Gr-1+/CD11b+ cells, and increased the cytolytic and proliferative potential of the CD45+ TIL. Taken together, these findings suggest that peritumoral administration of the IL-12-producing cellular vaccine can restore the cytolytic potential and inhibit immunosuppressive TIL-dependent mechanisms in the individuals bearing HPV 16-associated tumours, and explain our previously described tumour-inhibitory effects of the vaccine in mice with minimal residual disease after the tumour chemotherapy.

The effectiveness of combined chemoimmunotherapy with ifosfamide derivative CBM-4A and granulocyte-macrophage colony-stimulating factor (GM-CSF) was investigated in two experimental tumor models, 3MC-induced MHC class I + sarcoma Mc12 and HPV16 E6/E7 oncogene-induced MHC class I - carcinoma MK16, transplanted in syngeneic mice. Treatment of Mc12 and MK16 tumor-bearing mice with GM-CSF or CBM-4A alone produced moderate anti-tumor effects. However, when the tumor-bearing mice were first treated i.p. with a single dose of CBM-4A (150 mg/kg) and three days later peritumorally with five daily doses of GM-CSF (100 ng/day), substantially stronger tumor-inhibitory effects were observed. The results indicate that in both, MHC class I + and MHC class I- tumors, the combined chemoimmunotherapy can inhibit tumor progression more effectively than GM-CSF therapy or chemotherapy alone, and they suggest that GM-CSF should be considered as adjuvant to chemotherapy in clinical trials with HPV 16-associated neoplasms.

Cell surface adhesiveness, immunogenicity and immunosensitivity of tumour vaccines modified by the CD80 gene transfection was examined and compared to that of the parental MC12 murine sarcoma. Insertion of the CD80 gene substantially enhanced the adhesiveness of the genetically modified tumour cells to nylon wool non-adherent (T) but not to nylon wool adherent (B) lymphocytes. The increased adhesive interaction could be inhibited by anti-CD80 monoclonal antibody. CD80+ transfectants were more sensitive to the cytotolytic effect of MC12-immune splenocytes and IL-2-activated spleen cells than the parental MC12 sarcoma. Similarly, spleen cells from syngeneic mice immunized with CD80+ transfectants displayed a higher cytolytic activity when allowed to react with MC12 cells than splenocytes from mice immunized with the parental MC12 cells. These results suggest that a positive correlation exists among the expression of the CD80 molecules, T cell adhesion to the genetically modified cells, immunosensitivity of the CD80+ transfectants and the capacity of the transfectants to activate cytolytic, tumour-reactive effector cells in vivo. This correlation provides a rationale for gene therapy based on the construction of CD80- modified tumour vaccines.

Malignant transformation of somatic cells followed by selection of the transformed cell populations can give rise to tumours that display an immune escape phenotype, MHC class I deficient neoplasms. Experiments were designed to examine whether the immune escape phenotype of HPV16-associated tumours is stable or whether the MHC class I expression can change during tumour progression and therapy. It has been found that temporary growth of MHC class I- tumour MK16/1/IIIABC in syngeneic mice can lead to up-regulation of the low MHC class I expression, both in the subcutaneous tumour inocula and in their lung metastases. Mimicking this process in vitro by co-cultivation of tumour and spleen cell populations revealed that the spleen cells produce IFNgamma, which upregulates MHC class I expression on the MK16/1/IIIABC cells as well as their sensitivity to T cell-mediated cytolysis (CTLs). The up-regulation could be prevented by admixture of anti-IFNgamma antibody to the tumour/spleen cell mixtures. Similar up-regulation of the MHC class I expression was observed in HPV16-associated tumour cell lines, MK16/1/IIIABC, MK16/MET/M1, TC-1, TC-1/A9 and TC-1/P3C10 grown in vitro in the presence of IFNgamma. The up-regulation was found to be IFNgamma dose-dependent and the level of the MHC class I expression required for in vitro cytolysis of the tumour cells by CTLs could be characterized in cytofluorometry with anti-H-2 antibody. After removal of the IFNgamma from the cultivation medium or after injection of the IFNgamma-treated cells into syngeneic mice the MHC class I expression gradually dropped back to the original level or to the level observed on the tumours growing in vivo. These findings indicate that the immune escape phenotype of at least some HPV16-associated tumours is not stable and that up-regulation of the MHC class I expression can occur in vivo during progression of the MHC I- tumours, apparently due to production of IFNgamma by the immune cells in the tumour microenviroment and its vicinity. In vitro irradiation of HPV16-associated MHC class I-deficient tumour cell lines MK16/MET/M1 and TC-1/P3C10 with a dose of 150 Gy up-regulated their MHC class I expression. Similarly, substantial up-regulation of the MHC class I expression was observed in TC-1/A9 tumour recurrences after surgery. The up-regulation observed in the recurrences after surgery or after irradiation has reached the level required for in vitro cytolysis of the tumour cells by CTLs. If confirmed also with other tumour types and in human tumour systems, the up-regulation of MHC class I molecule expression during radiotherapy and in tumour recurrences after surgery may have important implications in the development of immunotherapeutic strategies.

Downregulation of MHC class I expression on tumour cells, a common mechanism by which tumour cells can escape from specific immune responses, can be associated with coordinated silencing of antigen-presenting machinery genes. The expression of these genes can be restored by IFNγ. In this study we documented association of DNA demethylation of selected antigen-presenting machinery genes located in the MHC genomic locus (TAP-1, TAP-2, LMP-2, LMP-7) upon IFNγ treatment with MHC class I upregulation on tumour cells in several MHC class I-deficient murine tumour cell lines (TC-1/A9, TRAMP-C2, MK16 and MC15). Our data also documented higher methylation levels in these genes in TC-1/A9 cells, as compared to their parental MHC class I-positive TC-1 cells. IFNγ-mediated DNA demethylation was relatively fast in comparison with demethylation induced by DNA methyltransferase inhibitor 5-azacytidine, and associated with increased histone H3 acetylation in the promoter regions of APM genes. Comparative transcriptome analysis in distinct MHC class I-deficient cell lines upon their treatment with either IFNγ or epigenetic agents revealed that a set of genes, significantly enriched for the antigen presentation pathway, was regulated in the same manner. Our data demonstrate that IFNγ acts as an epigenetic modifier when upregulating the expression of antigen-presenting machinery genes.

Experiments were designed to investigate immunogenicity and therapeutic efficacy of tumour vaccines constructed by transfection of poorly immunogenic murine sarcoma Mc12 with synergistic CD80 and IL-2 genes. Immunization/challenge experiments demonstrated that both, IL-2(+) and IL-2(+) plus CD80(+) live cell vaccines can exert an immunizing stimulus, the IL-2(+) plus CD80(+) vaccine being superior to the IL-2(+) vaccine. Live CD80(+) Mc12 cell vaccine could not be tested, since the vaccine was highly tumorigenic in the doses required for immunization. Preimmunization with IL-2(+) and IL-2(+) plus CD80(+) vaccines induced regressions of a proportion of the parental Mc12 challenge inocula after their temporary growth. Areas of necrosis and extensive infiltration with Mac1(+) and CD4(+) leukocytes have been observed in the regressing sarcomas. When the therapeutic efficacy of the irradiated CD80(+), IL-2(+), and mixed CD80(+) plus IL-2(+) vaccines was compared, it was found that the insertion of the IL-2, but not CD80 gene alone was efficient. The mixed IL-2(+) plus CD80(+) tumour vaccine was able to protect and prolong survival of a higher proportion of mice than the IL-2(+) tumour vaccine.