Wilms’ tumor 1 (WT1) is a promising tumor-associated antigen for cancer immunotherapy. We developed an oral protein vaccine platform composed of WT1-anchored, genetically engineered Bifidobacterium longum (B. longum) and conducted an in vivo study in mice to examine its anticancer activity. Mice were orally treated with phosphate-buffered saline, wild-type B. longum105-A, B. longum 2012 displaying only galacto-N-biose/lacto-N-biose I-binding protein (GLBP), and WT1 protein- and GLBP-expressing B. longum 420. Tumor size reduced significantly in the B. longum 420 group than in the B. longum 105-A and 2012 groups (P < 0.00 l each), indicating B. longum 420’s antitumor activity via WT1-specific immune responses. CD8+ T cells played a major role in the antitumor activity of B. longum 420. The proportion of CD103+CD11b+CD11c+ dendritic cells (DCs) increased in the Peyer’s patches (PPs) from mice in the B. longum 420 group, indicating the definite activation of DCs. In the PPs, the number and proportion of CD8+ T cells capable of producing interferon-gamma were significantly greater in the B. longum 420 group than in the B. longum 2012 group (P < 0.05 or < 0.01). The production of WT1-specific IgG antibody was significantly higher in the B. longum 420 group than in the 2012 group (P < 0.05). The B. longum 420 group showed the most intense intratumoral infiltration of CD4+ and CD8+ T cells primed by activated DCs in the PPs of mice in the B. longum 420 group. Our findings provide insights into a novel, intestinal bacterium-based, cancer immunotherapy through intestinal immunity.

Backgrounds Wilms’ tumor gene 1 (WT1) peptide vaccine and anti-programmed cell death-1 (PD-1) antibody are expected as immunotherapies to improve the clinical outcome of glioblastoma. The aims of this study were to clarify how each immunotherapy affect tumor-infiltrating immune cells (TIIs) and to determine whether the combination of these two therapies could synergistically work. Methods Mice were transplanted with WT1 and programed cell death-ligand 1 (PD-L1) doubly expressing glioblastoma cells into brain followed by treatment with WT1 peptide vaccine, anti-PD-1 antibody, or the combination of the two, and survival of each therapy was compared. CD45 + cells were positively selected as TIIs from the brains with tumors, and TIIs were compared between WT1 peptide vaccine and anti-PD-1 antibody therapies. Results Most mice seemed to be cured by the combination therapy with WT1 peptide vaccine and anti-PD-1 antibody, which was much better survival than each monotherapy. A large number of CD4 + T cells, CD8 + T cells, NK cells including WT1-specific CD8 + and CD4 + T cells infiltrated into the glioblastoma in WT1 peptide vaccine-treated mice. On the other hand, the number of TIIs did not increase, but instead PD-1 molecule expression was decreased on the majority of the tumor-infiltrating CD8 + T cells in the anti-PD-1 antibody-treated mice. Conclusion Our results clearly demonstrated that WT1 peptide vaccine and anti-PD-1 antibody therapies worked in the different steps of cancer-immunity cycle and that the combination of the two therapies could work synergistically against glioblastoma.

No abstract

Background A Wilms’ tumor 1 (WT1) oral vaccine, Bifidobacterium longum (B. longum) 420, in which the bacterium is used as a vector for WT1 protein, triggers immune responses through cellular immunity consisting of cytotoxic T lymphocytes (CTLs) and other immunocompetent cells (e.g., helper T cells). We developed a novel, oral, helper epitope-containing WT1 protein vaccine (B. longum 2656) to examine whether or not B. longum 420/2656 combination further accelerates the CD4+ T cell help-enhanced antitumor activity in a model of murine leukemia. Methods C1498-murine WT1—a genetically-engineered, murine leukemia cell line to express murine WT1—was used as tumor cell. Female C57BL/6 J mice were allocated to the B. longum 420, 2656, and 420/2656 combination groups. The day of subcutaneous inoculation of tumor cells was considered as day 0, and successful engraftment was verified on day 7. The oral administration of the vaccine by gavage was initiated on day 8. Tumor volume, the frequency and phenotypes of WT1-specific CTLs in CD8+ T cells in peripheral blood (PB) and tumor-infiltrating lymphocytes (TILs), as well as the proportion of interferon-gamma (INF-γ)-producing CD3+CD4+ T cells pulsed with WT135–52 peptide in splenocytes and TILs were determined. Results Tumor volume was significantly smaller (p < 0.01) in the B. longum 420/2656 combination group than in the B. longum 420 group on day 24. WT1-specific CTL frequency in CD8+ T cells in PB was significantly greater in the B. longum 420/2656 combination group than in the B. longum 420 group at weeks 4 (p < 0.05) and 6 (p < 0.01). The proportion of WT1-specific, effector memory CTLs in PB increased significantly in the B. longum 420/2656 combination group than in the B. longum 420 group at weeks 4 and 6 (p < 0.05 each). WT1-specific CTL frequency in intratumoral CD8+ T cells and the proportion of IFN-γ-producing CD3+CD4+ T cells in intratumoral CD4+ T cells increased significantly (p < 0.05 each) in the B. longum 420/2656 combination group than in the 420 group. Conclusions B. longum 420/2656 combination further accelerated antitumor activity that relies on WT1-specific CTLs in the tumor compared with B. longum 420.

72 Background: The gut microbiota plays an important role in shaping systemic immune responses. We have developed a WT1 oral cancer vaccine using a recombinant Bifidobacterium Longum ( B. Longum) as a platform for displaying murine WT1 protein ( B. Longum-mWT1). The Wilms’ tumor 1 (WT1) gene, which encodes a zinc finger transcription factor, is reportedly overexpressing in various tumors and one of the most promising tumor-associated antigens for cancer immunotherapy. In order to examine anti-tumor effects of this oral vaccine, we administered it orally into mice inoculated with WT1+-expressing brain tumor which doesn’t respond to existing treatment. Methods: The synthesized murine-WT1 gene (117-439 amino acid residues) was fused to galacto-N-biose/lacto-N-biose I binding protein (GLBP) coding gene. GLBP is a membrane protein on the wild-type B. Longum cell wall, which is used as an anchor to display antigen. The resulting plasmid carrying GLBP-WT1 was introduced into B. Longum by electroporation. We inoculated mouse glioma cell lines (Gl261) subcutaneously into the C57BL/6J. C57BL/6J mice received oral administration of B. Longum-mWT1 every day or subcutaneous administration of WT1126 peptide with montanide adjuvant on days 1, 8, 15, 22, 29, and 36. Results: The tumor volume of the mice treated with B. Longum-mWT1 (n = 5) was significantly smaller than that of the mice given WT1 peptide vaccine (n = 5) ( P < 0.05). The frequency of CD8+/WT1-tetramer+ CTLs was higher in B. Longum-mWT1 and WT1 peptide vaccine groups than in PBS group, and the high frequency was maintained in B. Longum-mWT1 group. In the mouse with B. Longum-mWT1, the frequency CD8+/WT1-tetramer+ CTLs in mesenteric lymph node and spleen was higher than that in Peyer’s patch. The number of invasive CD8+ T cells in brain tumor was higher in the B. Longum-mWT1 group than in the PBS group. B. Longum-mWT1 induced significantly higher in vitro cytotoxicity against Gl261 cells than WT1 peptide. Conclusions: We confirmed that B. Longum-mWT1 can induce strong cellular immunity and the maintenance of this effect. These results suggest that it is a novel oral anti-cancer agent for treating glioblastoma, for which no effective treatment has been developed.