Osteoclasts promote immune suppressive microenvironment in multiple myeloma: therapeutic implication

Abstract

The number and activity of osteoclasts (OCs) are strongly enhanced by myeloma cells, leading to significant bone lesions in patients with multiple myeloma (MM). Mechanisms remain elusive as to whether myeloma-supporting OCs also induce suppressive immune bone marrow (BM) microenvironment. Here, we first show that OCs significantly protect MM cells against T-cell-mediated cytotoxicity via direct inhibition of proliferating CD4(+) and CD8(+) T cells. The immune checkpoint molecules programmed death ligand 1 (PD-L1), Galectin-9, herpesvirus entry mediator (HVEM), and CD200, as well as T-cell metabolism regulators indoleamine 2, 3-dioxygenase (IDO), and CD38 are significantly upregulated during osteoclastogenesis. Importantly, the levels of these molecules, except CD38, are higher in OCs than in MM cells. Anti-PD-L1 monoclonal antibody (mAb) and IDO inhibitor partly overcome OC-inhibited T-cell responses against MM cells, confirming their roles in OC-suppressed MM cell lysis by cytotoxic T cells. In addition, Galectin-9 and a proliferation-induced ligand (APRIL), secreted by OCs, are significantly upregulated during osteoclastogenesis. Galectin-9 specifically induces apoptosis of T cells while sparing monocytes and MM cells. APRIL induces PD-L1 expression in MM cells, providing additional immune inhibition by OCs. Moreover, CD38 is significantly upregulated during osteoclastogenesis. When targeted by an anti-CD38 mAb, suppressive T-cell function by OCs is alleviated, associated with downregulation of HVEM and IDO. Taken together, these results define the expression of multiple immune proteins and cytokines in OCs essential for suppressive MM BM milieu. These results further support the combination of targeting these molecules to improve anti-MM immunity.

Figure 1

OCs protect MM cells against T-cell–mediated cytotoxicity by upregulating expression of multiple co-inhibitory molecules. (A) OCs and MM-specific CTLs were generated from the same healthy donor. CTLs were cocultured with target cells (KMS28-BM) in the absence or presence of OCs with or without the PD-L1 inhibitor (10 μg/mL)/IDO inhibitors (1-methyl-dl-Trp, 1 mM). After 4 hours, cytotoxicity was evaluated by measuring LDH activity in the supernatants. Shown is mean ± SD of the 3 representative independent experiments. (B) Proliferation of T cells stimulated by anti-CD2/CD3/CD28 beads (T:Bead ratio of 1:1) in the absence or presence of autologous OCs for 5 days was measured with CFSE dilution assay. (C-D) CD14⁺ monocytes were cocultured with RANKL and M-CSF for 14 days, and OCs were identified by TRAP staining. OCs were also cocultured with IFN-γ (20 IU/mL) for 12 hours. Protein expression in monocytes and OCs were determined by immunoblotting (C) and immunofluorescence (D). (E) T cells stimulated by anti-CD2/CD3/CD28 beads (T:Bead ratio of 1:1). Expression of PD-1, CD200R, Tim-3, and BTLA were examined by flow cytometry. (F) OCs were cultured from PBMCs or BM mononuclear cells from MM patients without CD14 selection. Levels of inhibitory molecules were significantly higher in OCs than in MM cells. (G) IHC analysis of 2 representative BM specimens from MM patients shows PD-L1 expression (brown) on OCs. Original magnification: ×20 (×100 in insets). (H) The expression of PD-L1, IDO, Galectin-9, and CD200 in CD138⁺ cells, CD138⁻ cells, and OCs from the same patient was determined by immunoblotting and real-time qRT-PCR. *P < .05; **P < .001; by unpaired 2-sided Student t test. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; ICOSL, inducible T-cell co-stimulator ligand; IHC, immunohistochemistry; PB, peripheral blood.

Figure 2

Time-course and dose-dependent analysis of inhibitory molecule expression during OC differentiation. (A-C) CD14⁺ monocytes were stimulated by RANKL and/or M-CSF followed by flow cytometry for PD-L1, HVEM, and CD200 at day 2, day 5, day 10, and day 14. (D) Monocytes were collected from healthy donors and MM patients to examine PD-L1, HVEM, and CD200 by flow cytometry. *P < .05; **P < .001; by unpaired 2-sided Student t test. SSC, side scatter.

Figure 3

Galectin-9 preferentially induces apoptosis of lymphoid T cells but not myeloid cells and MM cell line cells. (A) CD14⁺ monocytes were stimulated with RANKL and M-CSF. Supernatant was collected to measured Galectin-9 by ELISA, (B) serum was obtained from 5 healthy donors, and simultaneous serum and BM plasma were obtained from 10 MM patients. Galectin-9 level was determined by ELISA. (C) PBMCs from healthy donors and 3 MM cells were treated with recombinant Galectin-9 (1 μg/mL) for 12 hours, stained with Annexin V/PI, and analyzed by flow cytometry. Bottom right quadrants represent early apoptotic cells (Annexin V-positive only) and top right quadrant the late apoptotic cells (Annexin V- and PI-positive). *P < .05; **P < .001; by unpaired 2-sided Student t test. PI, propidium iodide.

Figure 4

APRIL induces PD-L1 expression on human MM cell line cells mainly via paracrine mechanism. (A) CD14⁺ monocytes were stimulated with M-CSF and/or RANKL for 14 days. APRIL expression in these cells was examined by real-time qRT-PCR; 18S was used to normalize APRIL expression. (B) Transwell experiments were performed in which MM cells were placed in the upper chamber and medium alone, or OCs were placed in the lower chambers. After 4 days, mRNA was collected from MM cells and subjected to real-time qRT-PCR for APRIL normalized to 18S. Fold increases compared with controls were shown. (C) MM cell lines were cultured with APRIL (200 ng/mL) for 7 days and PD-L1 expression was examined by flow cytometry. (D) MM cell lines were treated with human recombinant APRIL and/or anti-APRIL mAb (200 ng/mL) for 4 hours, and PD-L1 expression was examined by real-time qRT-PCR. (E) Indicated MM cell lines were stimulated with APRIL and PD-L1 expression was examined by real-time qRT-PCR. (F) MM1R and JJN3 cells were cultured with IFN-γ (500 IU/mL, 24 hours), IL-6 (10 ng/mL, 48 hours), and APRIL (200 ng/mL, 7 days). PD-L1 and pMEK1/2 expression was assessed by immunoblotting of cell lysates. *P < .05; **P < .001; by unpaired 2-sided Student t test.

Figure 5

The MEK/ERK pathway plays an important role in APRIL-induced PD-L1 expression in MM cells. (A-B) PD-L1 expression was examined by flow cytometry (A) and immunoblotting (B) in indicated RPMI 8226 transfectants. (C) PD-L1 mRNA expression was examined in BCMA knockdown (MM1R-TRIPZ-BCMA and H929-TRIPZ-BCMA) vs control MM cells by real-time qRT-PCR. Fold changes of PD-L1/18S to control were shown. (D) MM cells were treated with APRIL and/or U0126 molecules (100 nM) for 4 hours. PD-L1 expression was examined using real-time qRT-PCR; (F) PD-L1, pERK, and pMEK was examined by immunoblotting in indicated cells. (E) Serum-starved MM1S cells were pretreated with U0126 (100 nM) followed by APRIL stimulation. Cell lysate was collected and subjected to immunoblotting with indicated Abs. Shown is mean ± SD from 3 independent experiments. *P < .05; **P < .001; by unpaired 2-sided Student t test.

Figure 6

The impact of anti-CD38 Ab on OC differentiation from CD14-purified monocytes ex vivo. (A) CD38 was examined by immunoblotting in indicated cells. (B) CD38 expression was examined by flow cytometry during osteoclastogenesis. (C) Purified CD14⁺ monocytes were cultured with RANKL and M-CSF in 10% FBS RPMI 1640 medium for 7 days, followed by the addition of anti-CD38 mAb (SAR, 1, 10, 100 μg/mL) into the medium for an additional 7 days. At day 14, CD38 expression on OCs was examined by flow cytometry and immunoblotting. (D-E) CD14⁺ monocytes were cultured with RANKL/M-CSF in 10% RPMI medium for 7 days, and then SAR (0.1, 1 μg/mL) or IFN-γ (20 IU/mL) were added for an additional 7 days followed by TRAP staining to determine TRAP⁺ MNC ( >3 nuclei). (F-G) After CD14⁺ monocytes were stimulated with RANKL and M-CSF for 14 days, SAR, activated autologous T cells (Tact), and IDO inhibitor were added. After 3 days, cytotoxicity was evaluated by measuring LDH activity in supernatants. OCs were stained with Annexin V/PI and analyzed by flow cytometry. MNC, multinucleated cells; PI, propidium iodide.

Figure 7

Anti-CD38 Ab restores T-cell response. (A) Proliferation of T cells stimulated by anti-CD2/CD3/CD28 beads (T:Bead ratio of 1:1) in the absence or presence of SAR anti-CD38 Ab (1 μg/mL) for 6 days was measured with CFSE dilution assay. (B-D) After CD14⁺ monocytes were stimulated with recombinant RANKL and M-CSF for 7 days, SAR anti-CD38 Ab was added for 7 days, and cells were examined for HVEM by flow cytometry (B), western blotting (C), and IDO1 by real-time qRT-PCR (D). *P < .05; **P < .001; by unpaired 2-sided Student t test.