Tumor Microenvironment Induces Immature Dendritic Cells To Differentiate Into Regulatory Dendritic Cells

Dendritic cells (DCs) are the most potent antigen-presenting cells (APC) in the immune system. DCs play important roles in initiating innate and adaptive immune response. DCs have the potent capability to induce proliferation and response of Naive T cells, to promote the generation of cytotoxic T lymphocytes ( CTL) and Th1 cells. It has been well accepted that DC-initiated T cell immunity is critical for the anti-tumor immune response. Despite the presence of such tumor-infiltrating DC (TIDC), tumor cell outgrowth often occurs. The exact mechanism of tumor escape from immunological attack including escape from the recognition by DC remain to be fully understood.Tumor cells have been shown to be able to release many kinds of immunosuppressive cytokines, such as VEGF, TGF-β, IL-10, M-CSF, PGE2 , all of these cytokines could affect maturation, differentiation, antigen-presenting capability of DCs, Furthermore, after coculture with tumor cells, the DCs exhibit low responsive to the activating or maturation signals, indicating the function of DC was impaired by tumor cells. More recently, some kinds of DCs with negative regulatory functions have been reported, and such kind of DC have been designated as regulatory DC. We are wondering whether there were such kind of regulatory DC in the tumor-infiltrating DC, and tumor growth and metastasis will be promoted with help of those DCs. So, the primary aim of this study is to investigate whether tumor microenvironment could induce the generation of regulatory DC which inhibit T cell immunity, thus contributing to better understanding of the mechanisms for the tumor immune escape. In order to mimic the tumor microenvironment, we used the freshly isolated 3LL tumor cells to co-culture with immature DC, and then investigated the phenotype and function of DCs.At first, by using Transwell system, we found that the freshly isolated tumor cells could chemoattract immature DC more significantly, but almost had no chemoattracive effect on the mature DC. The data provided the explanation for whythere are immature DC inside the tumor tissue. Next, we selected the immature DC derived from bone marrow after 5 days' culture in IL-4 plus GM-CSF to coculturre with the freshly isolated tumor cells for 60 hours, then isolated the CD11c positive cells from the coculture system by CD1c beads. The CD11c positive DC were then analyzed for their phenotype and functions. We found that the DC, after cocultured with tumor cells, expressed low CD11c and Ia, but high CD11b. The CD40 and CD80 expresion remained unchanged. Even after LPS stimulation, the phenotype of the DC had no significant changes, indicating that the CD11c^lowIa^lowCD11b^highDC are in the stable state.CDllc^lowIalowCD11b^highDC, we isolated from the coculture system with tumor cells, display high phagocytic capacity, just like that of imDCs and even more potent than that of imDC. Interestingly, after the simulation of LPS, imDCs will be induced into mature DCs with low phagocytic capacity. However, the phagocytic capacity of the CD11c^lowIa^lowCD11b^highDC remained almost unchaged after LPS stimulation. So, this immature-like phenotype of the CD11c^lowIa^lowCD11b^highDC was stable, this suggesting that CD11c^lowIa^lowCD11b^highDC are different from imDC and mature DC.As to the cytokine profile, CD11c^lowIa^lowCD11b^highDC have been found to secrete high levels of IL-10,NO,PGE2, but low IL-12, even after the simulation by LPS. The data indicated that CD11c^lowIa^lowCD11b^highDC with stable profile of immunosuppressive cytokines may exert their negative regulation of immune response.To determine antigen-presenting ability of this type of DCs, CD11c^lowIa^lowCD11b^highDC were added to OVA323-339 peptide-specific CD4⁺T/mature DC/ OVA_323-339cocultute system . We found that antigen-presenting ability of CD11c^lowIa^lowCD11b^highDC was very weak. In contrast, CD11c^lowIa^lowCD11b^highDC could inhibt the antigen-specific CD4⁺T cell proliferation activated by mature DC.By dectecting the cytokines in the co-culture supernatants and in the CD4⁺ T cells, we unexpectedly found that the same levels of IL-2 and IFN-γ could be secreted by CD4 T cells induced by the CD11c^lowIa^lowCD11b^highDC. Taken together the phenotype and the inhibitory effect on the T cell proliferation, we designated such kind ofCD11c^lowIa^lowCD11b^highDC as the regulatory DC (DCreg).Next, we studied the mechanisms for the differentiation of immature DC into DCreg induced by tumor cells. We wanted to know whether the differentiation of DCreg was cell-cell contact-dependent or tumor soluble factor-dependent. As immature DCs exhibit strong endocytic ability, we could not use the fixed tumor cells to interact with the immature DC, so, we used transwell system to study the mechanisms. We added immature DC into the upper or lower chambers of the transwell, with tumor cells cultured in the lower chamber. We found that the proliferation of T cells in both systems was inhibited, and the inhibition was almost the same. So, the data suggest that the differentiation of immature DCs into DCregs is soluble factors-dependent. Then we detected the immunosuppressive cytokines secreted by tumor cells, the resluts revealed that freshly isolated 3LL tumor cells expressed high levels of M-CSF, TGF-β, VEGF etc. However, which kind of factors involved in differentiation of immature DCs to DCregs remains to be investigated in the future.CD11c^lowIa(low)CD11b^high DCreg we used were derived from the coculture system of tumor cells and immature DC. In order to exclude the possibility that the CD11c^lowIa^lowCD11b^high DCreg are artificially prepared, we must address the question that are there natural counterpart of CD11c^lowIa^lowCD11b^high DCreg in tumor tissue? So, we analyzed the CD11c⁺ cell population isolated from tumor cells. Condidering the CD11c^lowIa^lowCD11b^high phenotype , we labeled the mononuclear cell population with CD11b-APC, CD11c-FITC and Ia-PE. We found that there was a group of CD11b^highCD11c^low, and the expression of Ia was variable from high to low. In the future, we will sort the subset, and analyze the cytokine profile and confirm whether the subset also exhibit the inhibitory effect on T cell proliferation, so that the natural and functional counterpart of DCreg in tumor tissue can be identified.In conclusion, in this study , we demonstrate that the tumor microenvironment can chemoattract imDCs, and drive the differentiation of imDC into CD11c^lowIa^lowCD11b^high regulatory DC with high secretion of IL-10 but low secretion of IL-12, which inhibit the proliferation of antigen-specific CD4⁺ T cells. Thepreliminary data showed that there is a counterpart subset of CD11c^lowIa^lowCD11b^high regulatory DC in the tumor tissue.Our results provide new mechanistic explanation for the tumor immune escape.