Modulation of dendritic cell differentiation by HLA-G and ILT4 requires the IL-6--STAT3 signaling pathway

Abstract

The expression of Ig-like transcript (ILT) inhibitory receptors is a characteristic of tolerogenic dendritic cells (DCs). However, the mechanisms of modulation of DCs via ILT receptors remain poorly defined. HLA-G is a preferential ligand for several ILTs. Recently, we demonstrated that triggering of ILT4 by HLA-G1 inhibits maturation of human monocyte-derived conventional DCs and murine DCs from ILT4 transgenic mice, resulting in diminished expression of MHC class II molecules, CD80 and CD86 costimulatory molecules, and prolongation of skin allograft survival. Different isoforms of HLA-G have diverse effects on the efficiency to induce ILT-mediated signaling. In this work, we show that HLA-G1 tetrameric complex and HLA-G5 dimer, but not HLA-G5 monomer, induce strong ILT-mediated signaling. We determined that the arrest of maturation of ILT4-positive DCs by HLA-G ligands involves the IL-6 signaling pathway and STAT3 activation. Ligation of ILT4 with HLA-G on DCs results in recruitment of SHP-1 and SHP-2 protein tyrosine phosphatases. We propose a model where SHP-2 and the IL-6-STAT3 signaling pathway play critical roles in the modulation of DC differentiation by ILT4 and HLA-G.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. 1.

Formation of HLA-G5 dimer in vitro and efficiency of different isoforms of HLA-G to induce ILT-mediated signaling. (A) HLA-G5 dimerizes in vitro. HLA-G5 monomer was incubated at 4°C for 7 days. A gel filtration chromatogram of HLA-G5 was analyzed by size exclusion chromatography using a Superdex 200 10/30 column. Arrows show molecular mass calibration standards. (B) Purified HLA-G5 proteins were electrophoresed under nonreducing and reducing conditions and blotted to nitrocellulose. HLA-G heavy chains were detected by using the HLA-G-specific mAb MEM-G/9. Approximate molecular masses of HLA-G5m and HLA-G5d were 37 and 74 kDa, respectively. (C) Efficiency of different isoforms of HLA-G to induce ILT2-mediated signaling. NFAT-GFP reporter cells expressing the ILT2-PILRβ chimera were stimulated with the indicated concentration of immobilized HLA-G5m, HLA-G5d, or HLA-G1t for 18 h. GFP expression on reporter cells was analyzed by flow cytometry. Numbers indicate the percentage of GFP-positive cells (Upper) and MFI of GFP (Lower). Data shown are from one of four independent experiments.

Fig. 2.

Elevated level of IL-6 transcription in BMDCs treated with HLA-G5 dimer and HLA-G1 tetramer. (A) Hybridization intensity of genes in HLA-G-treated and untreated ILT4-positive BMDCs on GEArray. BMDCs (2.5 × 105) were plated onto wells coated with 50 ng/ml HLA-G5m, HLA-G5d, or HLA-G1t. After a 3-h incubation, cells were stimulated with 100 ng/ml LPS for an additional 18 h. Total RNA was used for cDNA probe synthesis after hybridization to gene-specific cDNA fragments were spotted onto GEArray membranes. Arrows indicate the most differentially expressed gene, IL-6. Data were analyzed by ImageQuant 1.2 software (Amersham Bioscience) with STORM 840 gel and blot imaging system. The signal for each transcript was normalized by comparison with the housekeeping gene GAPDH. (B) The levels of IL-6 transcription in HLA-G-treated cells were compared with untreated cells in LPS-stimulated, ILT4-positive BMDCs. (C and D) RT-PCR analysis of ILT4-positive BMDCs treated with HLA-G1t and untreated cells after stimulation with 100 ng/ml LPS. Data represent three independent experiments.

Fig. 3.

Engagement of ILT4 receptors by HLA-G1t increased STAT3 activation, phosphorylation of ILT4, and recruitment of SHP-1 and SHP-2. (A and B) Levels of STAT3 activation in ILT4-positive BMDCs treated with HLA-G1t for 3 h with and without stimulation with 100 ng/ml LPS for the indicated time. ILT4-positive BMDCs (5 × 106) were treated for 3 h with 50 ng/ml HLA-G1t or were left untreated. DCs were stimulated for an additional 1 h with 100 ng/ml LPS or left unstimulated. Cells were lysed, and proteins were immunoprecipitated (IP) with anti-STAT3 or anti-STAT1 antibodies. Immunoprecipitates were analyzed on Western blot for STAT proteins (A, Upper) and reprobed with anti-phosphotyrosine antibody 4G10 (A, Lower). Proteins were visualized by enhanced chemiluminescence. (C) Levels of STAT1 activation in ILT4-positive BMDCs treated with HLA-G1t for 3 h with and without stimulation with 100 ng/ml LPS. (D) Phosphorylated ILT4 is associated with SHP-1 and SHP-2. BMDCs were generated from ILT4 transgenic and B6 (control) mice, and the receptors were immunoprecipitated from cell extracts by using an anti-ILT4 mAb (42D1). Immunoblotting was performed by using anti-ILT4 mAb or 4G10, an anti-phosphotyrosine antibody as indicated. Immunoblots were reprobed sequentially with anti-SHP-1 and anti-SHP-2 antibodies as indicated. Proteins were visualized by enhanced chemiluminescence. Approximate molecular masses of the ILT4, SHP-1, and SHP-2 proteins were 95, 68, and 72 kDa, respectively. Data are representative of four independent experiments.

Fig. 4.

SHP-2 is a central mediator of HLA-G–ILT4 inhibitory effect in DCs. (A and B) ILT4-positive BMDCs were infected with lentiviral particles expressing SHP-1 shRNA or SHP-2 shRNA. Cellular lysates were analyzed by Western blotting with antibodies against SHP-1, SHP-2, and actin as indicated. Actin was detected as a loading control. Bars represent a percentage of the maximum signal per nontarget control band measured by semiquantitative densitometry. Densities were calculated as an average of five measurements per band. (C) RNA was isolated 48 h after infection with the indicated lentiviral particles, and RT-PCR was performed with primers to detect β-actin (internal control) and IL-6. (D) Flow cytometry analysis of the expression of MHC class II molecules on ILT4-positive BMDCs. Cells were stimulated with 100 ng/ml LPS for 18 h or left unstimulated (left two panels). After transduction with the indicated lentiviral particles, cells were treated with HLA-G1t for 3 h and then stimulated with 100 ng/ml LPS for 18 h (right three panels). Cells were stained with APC-conjugated anti-CD11c and FITC-conjugated anti-MHC class II (I-Ab) mAbs. Histograms shown were gated on a CD11c+ population. Blue lines represent isotype control. Numbers indicate percentage of positive cells of total gated cells. The results are from one representative experiment of four performed.

Fig. 5.

Proposed model of arrest of maturation/activation of DCs via ILT4 receptor and HLA-G ligand. HLA-G induces phosphorylation of ILT4 receptor and recruitment of SHP-1 and SHP-2 phosphatases. SHP-2 enhances activation of NF-κB and downstream IL-6 production. IL-6 induces STAT3 activation, which decreases cystatin C level, the endogenous inhibitor of cathepsins, and enhanced cathepsin S activities. Cathepsin S decreased intracellular MHC class II αβ dimer levels, invariant chain (Ii), and H2-DM molecule levels in DCs. The moderate signal generated through TLR4 leads to modest induction of IL-12 and IL-6, therefore additionally enhancing IL-6 production.