Evaluation of hGM-CSF/hTNFα surface-modified prostate cancer therapeutic vaccine in the huPBL-SCID chimeric mouse model

To validate its efficacy in the context of the human immune system, a novel therapeutic vaccine of hGM-CSF/hTNFα surface-modified PC-3 cells against human prostate cancer was evaluated in the human peripheral blood lymphocytes-severe combined immunodeficiency (huPBL-SCID) chimeric mouse model. The hGM-CSF or/and hTNFα modified vaccines inhibited prostate cancer growth effectively so as to prolong the mouse survival significantly. The splenocytes from the hGM-CSF/hTNFα vaccine-inoculated mice showed the strongest tumor-specific cytotoxicity against PC-3 cells and the highest production of IFNɤ. These features indicated that type 1 protective immune response was induced efficiently against human prostate cancer and further enhanced through synergetic adjuvant effects of hGM-CSF and hTNFα. Electronic supplementary material The online version of this article (doi:10.1186/s13045-015-0175-8) contains supplementary material, which is available to authorized users.

Prostate cancer is currently the second leading cause of cancer-related death in elderly men and likely develops into androgen-independent at advanced stages, which is refractory to conventional treatments. Cancer vaccine is a rational option for the treatment of androgenindependent prostate cancer [1]. Cancer immunotherapy is getting more and more important, remarkably evidenced by recent checkpoint blockade or chimeric antigen receptor-engineered T cell-based clinical trials with the impressive efficacy in different types of metastatic cancers [2,3].
Through simultaneous immobilization of streptavidintagged bioactive GM-CSF and TNFα on the biotinylated surface of cancer cells, we previously showed that the resultant cancer vaccines could induce a strong antitumor T cell immunity [4][5][6][7]. In the current study, human peripheral blood lymphocytes-severe combined immunodeficiency (huPBL-SCID) model was utilized to mimic the human immune system for evaluating the efficacy of these therapeutic vaccines. The materials and methods used in this study are detailed in Additional file 1.
We demonstrated that SA-hGM-CSF or/and SA-hTNFα could be efficiently immobilized on the biotinylated surface of ethanol-fixed PC-3 cells (Additional file 2) and SA-hGM-CSF or/and SA-hTNFα immobilized on the PC-3 cells still retained their bioactivity (Additional file 3).
The PC-3 cells inoculated subcutaneously in nonobese diabetic/severe combined immunodeficiency (NOD/ SCID) mice were found to maintain their original tumorigenicity so as to spread into the blood (Additional file 4). There was no phenomenon of immune leakage [8] in NOD/SCID mice used in this study. The activity of natural killer (NK) cells in the NOD/SCID mice was dramatically reduced by injection of anti-asialo-GM1 antibody (Additional file 5).
Flow cytometry detected human CD4 + , CD8 + , and CD45 + cells in peripheral blood and human CD45 + cells in spleen of huPBL-SCID mice. IHC staining revealed many human CD4 + and CD8 + lymphocytes present in spleen and fewer human CD4+ and CD8+ lymphocytes in liver tissue 8 weeks after transfer of huPBL. The results indicated that human T lymphocytes were successfully engrafted and homed in appropriate lymphoid organs of huPBL-SCID mice (Additional files 6 and 7).
We tested therapeutic effects of different PC-3 cell vaccines modified with hGM-CSF and/or hTNFα on human prostate cancer in the huPBL-SCID mouse model. Compared with other cancer vaccines, the hGM-CSF/hTNFα doubly modified cancer vaccine significantly inhibited prostate cancer growth in terms of tumor weight (Fig. 1a) and size (Fig. 1b) and effectively prolonged the mice survival (Fig. 1c).
There were more human leukocytes and CD4+ or CD8+ lymphocytes residing in the lymph nodes in the hGM-CSF/hTNFα doubly modified group than in other groups 8 weeks after vaccination (Additional file 8). Similarly, more human CD4 + or CD8 + T lymphocytes were found to infiltrate into the tumor tissues in the hGM-CSF/hTNFα doubly modified group than in other groups (Fig. 2a, b), indicating that the hGM-CSF/hTNFα doubly modified PC-3 cell vaccine could enhance its anti-tumor immunity by increasing the infiltration of human T lymphocytes into the tumors.
We finally analyzed the tumor-specific cytotoxicity of cytotoxic T lymphocytes (CTL) in the spleen and revealed that the GM-CSF/TNFα doubly modified vaccine did establish a stronger tumor-specific T cell immunity than other PC-3 cell vaccines (Fig. 2c). The supernatant from in vitro splenocyte culture in the GM-CSF/TNFα doubly modified group had the highest production of PC-3-specific IFNγ (Fig. 2d). Both results indicated that type 1 protective immunity was induced against the human prostate cancer.
Therefore, our current study provided a solid foundation for potential clinical application of this novel hGM-CSF/hTNFα surface-modified prostate cancer therapeutic vaccine. This unique approach can be easily adopted to generate a personalized whole cancer cell vaccine from individual autologous cancer cells, thereby potentially overcoming cancer antigen heterogeneity [9][10][11].   Immunohistochemical staining analysis of lymphocytes in tumor tissue and assessment of PC-3-specific cytotoxicity and IFNγ in spleen. Immunohistochemical staining analyzed CD4+ or CD8+ lymphocytes in the tumor tissues from huPBL-SCID mice 8 weeks after vaccination. The images of immunohistochemical staining were shown with ×200 magnification. Tumor tissues from different groups were stained with anti-hCD4 or anti-hCD8 antibody (a), and the quantitative analysis of the images was performed with integrated optic density (b). For PC-3specific cytotoxicity assay, spleen cells were isolated on day 21 after tumor injection from each experimental group. Effector cells were stimulated by recombinant human IL-2 and mitomycin-treated PC-3 cancer cells. The supernatants were collected for the non-radioactive cytotoxicity assay (c). Additional file 6: Flow cytometric analysis of human lymphocytes in the peripheral blood and spleen tissue of huPBL-SCID mice after huPBL transplantation. (A) The presence of CD45 + , CD8 + , or CD4 + cells was assessed respectively with phycoerythrin (PE)-labeled anti-hCD45, PE-Cy7-labeled anti-hCD8, and PE-Cy7-labeled anti-hCD4 monoclonal antibodies for flow cytometric analyses. (B) Flow cytometric analysis of human CD45 + cells in the spleen tissue of huPBL-SCID mice 8 weeks after huPBL transplantation. Spleen was isolated from huPBL-SCID 8 weeks after huPBL transplantation and single cell suspensions were prepared from the isolated spleen. The presence of CD45 + cells was assessed with PE-labeled anti-hCD45 monoclonal antibody through a flow cytometer. NOD/SCID mice without huPBL transplantation were used as the negative controls.
Additional file 7: Immunohistochemical staining of spleen and liver tissues from huPBL-SCID mice 8 weeks after huPBL transplantation. The images of immunohistochemical staining were shown with ×200 magnification. The spleen and liver tissues from huPBL-SCID mice were stained with anti-hCD4 or anti-hCD8 antibody (A), and the quantitative analysis of the images was performed with integrated optic density (B). The spleen and liver tissues from NOD/SCID mice without huPBL transplantation were used as negative controls.
Additional file 8: Immunohistochemical staining analysis of the lymph node tissues from huPBL-SCID mice 8 weeks after vaccination. The images of immunohistochemical staining were shown with ×200 magnification. Lymph node tissues from different groups were stained with rat anti-hCD45 (A), anti-hCD4 (B), or anti-hCD8 antibody (C). The quantitative analysis of those images was carried out with integrated optic density (D).