Combinational Immunotherapy with Allo-DRibble Vaccines and Anti-OX40 Co-Stimulation Leads to Generation of Cross-Reactive Effector T Cells and Tumor Regression

Abstract

It is well-known that vaccines comprising of irradiated whole tumor cells or tumor-derived heat shock proteins can generate tumor-specific immune responses. In contrast, we showed recently that vaccines composed of autophagosomes (DRibbles) derived from syngeneic sarcomas could induce cross-reactive T-cell responses and cross-protection against the tumor. This unusual property of DRibbles was related to the selective recruitment of defective ribosomal products (DRiPs) and other short-lived proteins (SLiPs) into autophagosomes via sequestosome (SQSTM1, p62) mediated association of ubiquitinated SLiPs to the autophagy gene product LC3. Here, we extend our observations to mammary carcinomas from mice of different genetic background. We demonstrated that combined of intranodal administration of autologous or allogeneic DRibbles together with anti-OX40 antibody led to robust proliferation, expansion, and differentiation of memory and effector T cells. We also showed that SLiPs is an excellent source of antigen for cross-priming of CD8⁺ T-cells that recognize shared tumor antigens in the context of host MHC class I molecules. Thus, our results provide a strong basis for novel clinical trials that combine allogeneic "off-the-shelf" DRibble vaccines together with antibodies against co-stimulatory molecules.

Figure 1. The effect of vaccine administration route and costimulation with anti-OX40 antibodies on the T-cell expansion and differentiation.

To determine the most efficient route for DRibbles in the cross-priming of CD8⁺ T cells, DRibbles derived from 3LL cells expressing the OVA antigen were injected at the indicated doses into C57BL/6 mice via s.c. or i.n. injection after adoptive transfer of 1 × 10⁶ CFSE-labeled TCR transgenic Thy1.1⁺ OT-I T cells. The degree of OT-I proliferation was evaluated by staining OT-I T cells with antibodies against CD8, Thy1.1 and Vα2, flow cytometry analysis of CFSE dilution at day 5 after vaccination. (a) CFSE division (b) Frequency of OT-I T cells after s.c. injection of 10~300 μg of DRibbles. (c) CFSE division and (d) Frequency of OT-I T cells after i.n. injection of 1~10 μg DRibbles. To investigate the effect of anti-OX40 costimulation on the T-cell expansion and differentiation induced by DRibbles, naïve C57BL/6 mice were adoptively transferred with 10,000 splenocytes from Thy1.1 OT-I transgenic mice. OVA-DRibbles 10 μg were administrated into two inguinal lymph nodes and 100 μg anti-OX40 antibody was given i.p. twice 3 days apart, on the day of DRibble injection and day 3. Seven days later, spleens were harvested from control mice that received adoptive transfer of OT-I T cells alone, anti-OX40 alone, DRibble vaccination only, or a combination of DRibble vaccination and anti-OX40. (e) The percentage of OT-I T cells was determined by flow cytometry analysis (anti-CD8, anti-Thy1.1, Vα2). Data represent the mean and standard error of the mean from three mice per group. (f) Additional antibodies against CD127 and KLRG1 were used to determine the phenotypes of OT-I T cells from different mice. The plot represents a typical result from one of three mice. The experiment was repeated once and similar results were obtained. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 2. Functional analysis of tumor cross-reactive T cells elicited by DRibble vac DRibble vaccination and in combination with anti-OX40 costimulation.

(a) Naïve BALB/c mice were vaccinated with 4T1, C57MG, and MMC DRibbles intranodally and splenocytes were isolated 7 days after vaccination and restimulated with either 4T1 DRibbles pulsed onto naïve spleen APCs (left panel) or directly with whole 4T1 tumor cells (right panel). The production of intracellular IFN-γ after restimulation was determined as described in the method section. Untreated naïve mice were used as negative controls. (b) Same as (a) with additional i.p. injection of anti-OX40 antibody at 3 and 5 after DRibble injection. Mice treated with anti-OX40 alone were included as an additional control. (c) Spleen cells from primed mice with C57MG DRibbles and anti-OX40 antibody were kept in culture with complete media without adding exogenous cytokine for 5 days before they were used as effector T cells for the measurement of effector functions by intracellular staining and flow cytometry analysis. (d,e) Beside the production of IFN-γ, both freshly isolated and cultured spleen CD8⁺ T cells were capable of producing granzyme A and B upon restimulation with 4T1 DRibbles or tumor cells. Data represent the mean and standard error of the mean from three or four mice per group. Independent experiments were performed two to three times. *P < 0.05; **P < 0.01.

Figure 3. A combination of intranodal injection of 4T1 DRibbles and anti-OX40 co-stimulation induced potent anti-tumor immune responses and mediated significant tumor regression and cure of mice bearing established 4T1 tumor cells.

(a) A schema of the immunotherapy strategy that combines intranodal vaccination with DRibbles and co-stimulation with the anti-OX40 antibody. 4T1 cancer cells were s.c. injected (20,000) into the right mammary fat pad of BALB/c mice. On day 13, mice received DRibble vaccine (20 μg, i.n.), followed by two boosts with DRibble-loaded DCs (20 μg/3 million, s.c.) on day 15 and 18. Anti-OX40 was given i.p. immediately after each DRibble vaccination at 100 μg per mouse. (b) DRibble vaccination combined with anti-OX40 co-stimulation led to remarkable tumor regression and cure of 60% of the treated mice. (c) DRibble vaccination in conjunction with anti-OX40 significantly increased the frequency of both CD4⁺ and CD8⁺ T cells in the peripheral blood mononuclear cells (PBMC) of tumor-bearing mice. Blood was collected at day 32 and PBMC were analyzed by flow cytometry to determine the percentage of CD4 and CD8 subset in total PBMC (CD45⁺) after purification through a density gradient. Data represent the mean and standard error from three mice per group. (d,e) DRibble/anti-OX40 combination vaccine cured mice and control naïve mice were challenged with 100,000 Delta-actA-Lm-AH1-A5 i.v. Splenocytes were collected and re-stimulated with AH1 peptide a week later. ICS of IFN-γ producing CD8⁺ T cells was performed. Independent experiments were performed three times. *P < 0.05; **P < 0.01.

Figure 4. The allogeneic DRibble vaccination in combination with anti-OX40 antibody mediated a potent antitumor response and required both CD4⁺ and CD8⁺ T-cell subsets for its efficacy.

(a) To evaluate whether allo-DRibbles can confer anti-tumor immunity against 4T1, we administrated allo-DRibbles from C57MG (b,c) or MMC (d) mammary carcinoma cells. BALB/c mice bearing 13-day 4T1 tumors were established as in Fig 2 and treated with DRibbles from C57MG tumors, 4T1 tumors, followed by two boosts with DC s.c. at 2 day interval. Anti-OX40 (100 μg) was injected i.p along with each immunization. C57MG allo-DRibbles resulted in 80% cure rate of BALB/c mice bearing 4T1 tumors. To determine the role of CD4⁺ or CD8⁺ T cells for the anti-tumor efficacy, 4T1 tumor cells (30,000) were injected into the right mammary pads of BALB/c mice. At day 5, mice with palpable tumors were divided into four experimental groups: untreated, DRibble vaccination and anti-OX40, DRibble vaccination and anti-OX40 with CD8 depletion, DRibble and anti-OX40 with CD4 depletion. Anti-CD8 (100 μg) and anti-CD4 (200 μg) was given twice i.p. to deplete CD8⁺ and CD4⁺ T cells respectively on day 5, 9. Mice received DRibbles (i.n., 20 μg) from MMC tumors on day 6. Anti-OX40 (100 μg) was given i.p. at day 6 and 9. Tumor growth was measured every other day on weekdays starting from day 10 (c), and mouse survival was monitored (d). Independent experiments were performed two times. *P < 0.05; **P < 0.01.

Figure 5. Cross-priming of tumor cross-reactive T cells using SLiPs isolated from allogeneic or MHC class Ia deficient tumor cells.

To evaluate whether isolated ubiquitinated SLiPs from allogeneic mammary cancer cells (C57MG) or syngeneic colon cancer cells (CT26), we administered isolated SLiPs from tumor cells via the i.n. route into BALB/c mice. Seven days later, mice were sacrificed and splenocytes were restimulated with 4T1 cells or CT26 cells. Intracellular IFN-γ staining was performed to determine the frequency of tumor-reactive CD8⁺ T cells induced by vaccination with SLiPs. (a) CD8⁺ T cell responses to 4T1 cells. (b) CD8⁺ T cell responses to CT26 cells. To evaluate whether SLiPs isolated from tumor cells could induce tumor-reactive T cells, we vaccinated C57BL/6 mice with SLiPs from B78H1 melanoma cells together with alumina nanoparticles (HS) as the adjuvant. Mice vaccinated with HS adjuvant alone or SLiPs from normal liver tissue plus HS were included as controls. Seven days later, mice were sacrificed and splenocytes were restimulated with irradiated B78H1 (d), B78H1-D^b (e), B78H1-K^b (f) and B78H1-D^bK^b (g) tumor cells, CM as a negative control (c). Intracellular IFN-γ staining was performed to determine the frequency of tumor-reactive CD8⁺ T cells induced by vaccination with SLiPs. Data represent the mean and standard error of the mean from five mice per group. Independent experiments were performed three times. *P < 0.05; **P < 0.01; ***P < 0.001.