Growth differentiation factor-15 suppresses maturation and function of dendritic cells and inhibits tumor-specific immune response

Abstract

Dendritic cells (DCs) play a key role in the initiation stage of an antigen-specific immune response. A variety of tumor-derived factors (TDFs) can suppress DC maturation and function, resulting in defects in the tumor-specific immune response. To identify unknown TDFs that may suppress DCs maturation and function, we established a high-throughput screening technology based on a human liver tumor T7 phage cDNA library and screened all of the proteins derived from hepatoma cells that potentially interact with immature DCs. Growth/differentiation factor-15 (GDF-15) was detected and chosen for further study. By incubation of DCs cultures with GDF-15, we demonstrate that GDF-15 can inhibit surface protrusion formation during DC maturation; suppress the membrane expression of CD83, CD86 and HLA-DR on DCs; enhance phagocytosis by DCs; reduce IL-12 and elevate TGF-β1 secretion by DCs; inhibit T cell stimulation and cytotoxic T lymphocyte (CTL) activation by DCs. By building tumor-bearing mouse models, we demonstrate that GDF-15 can inhibit the ability of DCs to stimulate a tumor-specific immune response in vivo. These results indicate that GDF-15 may be one of the critical molecules that inhibit DC maturation and function and are involved in tumor immune escape. Thus, GDF-15 may be a novel target in tumor immunotherapy.

Figure 1. Preparation and treatment time lines for human and murine DCs.

(A) Human iDCs were obtained from CD14⁺ cells cultured in GM-CSF and IL-4 for 8 days. Human mDCs were obtained from CD14⁺ cells stimulated with GM-CSF and IL-4 for 8 days and stimulated by TNF-α in the last 2 days. Human GDF-15 treated DCs were obtained from CD14⁺ cells cultured in GM-CSF, IL-4 and GDF-15 for 8 days and stimulated by TNF-α in the last 2 days. (B) Murine iDCs were obtained from bone marrow cells cultured in GM-CSF and IL-4 for 8 days. Murine mDCs were obtained from bone marrow cells cultured in GM-CSF and IL-4 for 8 days and stimulated by CT26 cells lysate in the last 2 days. Murine GDF-15 treated DCs were obtained from bone marrow cells cultured in GM-CSF, IL-4 and GDF-15 for 8 days and stimulated by CT26 cells lysate in the last 2 days.

Figure 2. Confirmation of in vitro binding by cell-based ELISA.

(A) Identification of the binding selectivity of the 38 clones by cell-based ELISA. The iDCs (CD14⁺ cells and mDCs were used as negative controls) were incubated with the amplified T7 phage in 96-well plates at 37°C for 1 h. Phage clones binding to cells were detected by the T7 Tail Fiber monoclonal antibody and following sheep anti-mouse IgG conjugated with HRP. (B) Specificity of the GDF-15-expressing phage to iDCs. Control phage: uncombined phage in the first round of panning. All assays were carried out in triplicate and the error bars indicate standard deviation.

Figure 3. GDF-15 affects the morphologies of DCs.

Scanning electron micrographs of an iDC, an mDC and a GDF-15-treated (50 ng/mL) DC. CD14+ cells were grown on glass coverslips and induced to form DCs. Samples for scanning electron microscopy were then prepared and investigated using an S-3400N microscope. Original magnification ×3000 or ×2500.

Figure 4. GDF-15 affects the phenotypes of DCs.

(A) Quantitative PCR analysis of DC phenotypes based on the expression of CD83, CD86 and HLA-DR. The quantitative values for the genes of interest were normalized using the housekeeping gene β-actin as an endogenous reference. The fold-increase over the control was calculated using the relative quantification method of 2^−ΔΔCt. Mean ± SD, n = 3. (B, C) Flow cytometry analysis of DC phenotypes based on the expression of CD83, CD86 and HLA-DR. Mean ± SD, n = 7. * P<0.05, ** P<0.01 and *** P<0.001 compared with the mDCs. The experiments were conducted in triplicate.

Figure 5. GDF-15 affects phagocytosis and cytokine secretion by DCs.

(A) Confocal microscopy. CD14+ cells were cultured in a 15 mm confocal dish and induced to form DCs under different culture conditions. After incubation with FITC-dextran, fixation and staining with DAPI, the cells were observed by confocal microscopy. Original magnification ×400. The scale bars are equal to 20 μm. (B) Flow cytometry. The results are shown as the ΔMFI. For each sample, the background (MFI of the fluorescence of the cells pulsed at 4°C) was subtracted from the MFI of the cells incubated at 37°C. (C, D) Levels of cytokine secretion in DCs. The culture medium was collected when the cells were harvested for analysis. After centrifugation, IL-12 (C) and TGF-β1 (D) levels in the supernatants were detected using an ELISA kit according to the manufacturer's instructions. The OD_450nm was recorded using a spectrophotometer. Mean ± SD, n = 3. * P<0.05, ** P<0.01 and *** P<0.001 compared with the mDCs. The experiments were conducted in triplicate.

Figure 6. GDF-15 inhibited T cell stimulation and CTL activation induced by DCs.

(A) MLR. DC-mediated T cell stimulation was analyzed in an MLR by incubating DCs obtained under different culture conditions with T lymphocytes at the indicated ratios. (B) At a DC: T cell ratio of 1∶5, the T cell-stimulatory capacity of DCs was significantly reduced by GDF-15. Mean ± SD, n = 4. (C) CTL assay. DCs obtained under different culture conditions were co-cultured with T cells at a ratio of 1∶5 for 3–4 days to generate effector cells. SW480 cells were then co-cultured with the activated T cells at a ratio of 1∶10, 1∶20, 1∶40 or 1∶80 for 36 h. The LDH in the supernatants was detected using a kit. Mean ± SD, n = 8. *** P<0.001 compared with the mDCs. The experiments were conducted in triplicate.

Figure 7. GDF-15 inhibited the ability of DCs to stimulate a tumor-specific immune response in vivo.

(A) Flow cytometry. rhGDF-15 suppressed the expression of the murine BMDC phenotype-associated molecules H-2D^b, I-A^b and CD80. (B) Tumor growth curves for different groups. CT26 tumor cells (2×10⁶) were dorsally subcutaneously injected or co-injected with murine iDCs, GDF-15-treated DCs or mature DCs (1×10⁶) into BALB/c mice (10- to 12-week-old, male). Two weeks after the tumor challenge, the tumor size was calculated. (C) Tumor weights for different groups. Each tumor was excised four weeks after inoculation, and the tumor weight was measured. Mean ± SD. * P<0.05 and ** P<0.01 compared with the mDCs.