Immunomodulatory effects of lenalidomide and pomalidomide on interaction of tumor and bone marrow accessory cells in multiple myeloma

Abstract

The bone marrow (BM) microenvironment consists of extracellular-matrix and the cellular compartment including immune cells. Multiple myeloma (MM) cell and BM accessory cell interaction promotes MM survival via both cell-cell contact and cytokines. Immunomodulatory agents (IMiDs) target not only MM cells, but also MM cell-immune cell interactions and cytokine signaling. Here we examined the in vitro effects of IMiDs on cytokine signaling triggered by interaction of effector cells with MM cells and BM stroma cells. IMiDs diminished interleukin-2, interferonγ, and IL-6 regulator suppressor of cytokine signaling (SOCS)1 expression in immune (CD4T, CD8T, natural-killer T, natural-killer) cells from both BM and PB of MM patients. In addition, coculture of MM cells with healthy PBMCs induced SOCS1 expression in effector cells; conversely, treatment with IMiDs down-regulated the SOCS1 expression. SOCS1 negatively regulates IL-6 signaling and is silenced by hypermethylation in MM cells. To define the mechanism of inhibitory-cytokine signaling in effector cells and MM cells, we next analyzed the interaction of immune cells with MM cells that were epigenetically modified to re-express SOCS1; IMiDs induced more potent CTL responses against SOCS1 re-expressing-MM cells than unmodified MM cells. These data therefore demonstrate that modulation of SOCS1 may enhance immune response and efficacy of IMiDs in MM.

Figures

Figure 1

IMiDs activate effector immune cells and reduce inhibitory immune cells in MM in vitro. PBMCs from patients with MM (MM-PBMC) were stimulated with anti-CD3 Ab and incubated for 6 days in the absence or presence of lenalidomide and pomalidomide. IMiDs effect on effector cell activation was analyzed using flow cytometry and cytokine protein array. (A) Histogram plots by flow cytometric analysis show membrane-expression of positive-costimulatory signaling molecules CD28, ICOS, and ICOSL on gated MM-PBMC CD3+ cells incubated for 3 days in the absence or presence of lenalidomide and pomalidomide. Shadowed profiles indicate isotype-matched control immunoglobulin (Ig) staining. One representative of 3 independent experiments is shown. Statistical significance indicated (t test, 1-tailed distribution, P < .05). (B) IMiDs induced proliferation of effector cells in MM-PBMCs is shown by CFSE-staining flow cytometric analysis. Proliferating effector cells were identified by CFSE costaining in gated CD4 PECy5 (CD4T cells), CD8 PECy7 (CD8 T cells), or CD56 PE/CD8 PECy7 (NKT cells) positive subpopulations. Proliferation is represented by division index of each cell population. One representative of 3 independent experiments is shown. Statistical significance indicated (t test, 1-tailed distribution, P < .05). (C) IMiDs effect on cytokine production was determined in MM-PBMC effector cells by intracytoplasmic cytokine staining flow cytometric analysis. MM-PBMCs were stimulated with anti-CD3 Ab and cultured for 16 hours with or without lenalidomide and pomalidomide, and intracytoplasmic expression (percent positive stained cells) of IL-2 (PE) was shown in gated effector cell subpopulations, as shown in the figure by side scatter dot plots. y-axis represents IL-2 PE staining and x-axis represents CD4T cells (top panel), CD8 T cells (middle panel), and NKT cells (bottom panel). One representative of 3 independent experiments is shown. Statistical significance indicated (t test, 1-tailed distribution, P < .05). (D) Secreted IFNγ, IL-2, and IL-6 were measured by cytokine protein array in the supernatants of anti-CD3 Ab-stimulated MM-BMMCs in the absence or presence of lenalidomide and pomalidomide for 24 hours. Expression of cytokines was determined using ImageJ 1.37v (http://rsb.info.nih.gov/ij/) densitometric analysis. Average and SD of triplicate from 1 representative of 3 independent experiments is shown. Statistical significance indicated (t test, 1-tailed distribution, P < .05). (E) Inhibitory CD4T cells were identified in the MM-BMMCs incubated in the absence or presence of lenalidomide and pomalidomide for 48 hours to 6 days. CD4T cells were gated and further analyzed for membrane CD25 and intracellular FOXP3 coexpression by flow cytometry. Top panel demonstrates CD4+CD25+FOXP3+ regulatory CD4T cells at the end of 6 days culture, and bottom panel represents IL-17 producing CD4 T cells (Th17) at the end of a 48-hour culture. One representative of 3 independent experiments is shown. Statistical significance indicated (t test, 1-tailed distribution, P < .05).

Figure 2

IMiDs regulate SOCS1-dependent cytokine-negative feedback mechanism in effector cells in MM. Flow cytometric analysis of intracellular expression of SOCS1 protein in effector cells from cocultures are shown as representative dot plots. To provide direct tumor cell-effector cell contact and -BM contact, healthy donor's PBMCs were cocultured for 16 hours with either MM cell lines (MM1.S and U266) or with MM cell lines and MM-BMSCs, in the absence or presence of lenalidomide and pomalidomide. Intracytoplasmic expression of SOCS1 (PE) protein was then evaluated by flow cytometry in the gated CD4 PECy5 (CD4T cells; A), CD8 PEcy7 (CD8T cells; B), CD56 PE+CD8 PECy7+ (NKT cells; C), and CD56PE+CD8PECy7− (NK cells). (D) Top panel demonstrates intracellular SOCS1 protein expression in effector cells of healthy PBMC cocultured with allogeneic healthy PBMCs. Middle panel demonstrates intracellular SOCS1 protein expression in effector cells of healthy PBMC cocultured with MM cell line MM1.S, and bottom panel demonstrates intracellular SOCS1 protein expression in effector cells of healthy PBMCs cocultured with MM cell line MM1.S and MM-BMSC. Numbers indicate percentage of positive cells. One representative of 3 independent experiments is shown. Statistical significance indicated (t test, 1-tailed distribution, P < .05).

Figure 3

IMiDs regulate ex vivo induced SOCS1 expression and STAT activation. (A) SOCS1 protein expression was shown by immunoprecipitation-Western blot in effector cells, incubated in the presence of aCD3 Ab and lenalidomide (1μM) or pomalidomide (1μM) for 16 hours, and then stimulated with SOCS1 inducing cytokine IFNγ (100 ng/mL) for 15 minutes to 4 hours (top panel). Expression of SOCS1 protein relative to IgH chain was demonstrated (fold) using ImageJ 1.37v (http://rsb.info.nih.gov/ij/) densitometric analysis (bottom panel). One representative of 3 independent experiments is shown. Intracellular expression of STAT1 pTyr701 and STAT1 (B) and STAT3 pTyr705 and STAT3 (C) are shown in PBMCs from healthy donors (n = 3) by flow cytometric analysis. aCD3 Ab stimulated PBMCs were incu-bated in the absence or presence of lenalidomide and pomalidomide for 16 hours, and then stimulated with IFNγ for 15 minutes. Expression of phosphorylated STATs is shown as a percent expression relative to control baseline levels in aCD3 Ab–stimulated cells. Top panels demonstrate phosphorylated STATs expression in CD4 T cells as representative plots, and bottom panels demonstrate phosphorylated STATs expression in all effector cells as percent expression relative to control baseline levels in aCD3 Ab stimulated cells. Numbers indicate percentage of positive cells. One representative of 3 independent experiments is shown. Statistical significance indicated (t test, 1-tailed distribution, P < .05).

Figure 4

IMiDs epigenetically regulate SOCS1 methylation in MM cells in vitro. (A) Baseline SOCS1 protein expression in MM cell lines and unstimulated CD3+ T cells is shown by Western blot. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) expression was determined as an internal control for loading. Data are demonstrated as a representative of 3 separate experiments. (B) Methylation analysis of SOCS1 in MM cell lines MM1.S and U266 is assessed by MSP. MM1.S and U266 cells were incubated either alone or in combination with IMiDs (lenalidomide and pomalidomide; 1μM), 5-Azacytidine (5-AzaC; 0.5μM), and Trichostatin A (TsA; 0.1μM); sodium bisulfite–modified genomic DNA was then subjected to MSP using MSP primers that specifically recognize unmethylated or methylated SOCS1 gene sequences. U indicates the presence of unmethylated SOCS1 gene (175 bp); and M indicates the presence of methylated SOCS1 gene (160 bp). In vitro unmethylated and methylated DNA were used as controls for methylation. W indicates wild-type SOCS1 and N indicates water as control for PCR (blue line indicates the merged photomicrograps run on the same gel but in 2 lanes). (C) SOCS1 gene transcriptional expression is shown by quantitative real-time PCR in MM1.S and U266 cells. Cells were incubated either alone or in combination with IMiDs (1μM), 5-AzaC (0.5μM), and TsA (0.1μM); SOCS1 gene expression was then analyzed by real time, quantitative PCR using SYBR-Green labeling. Data are demonstrated as the mean fold increase relative to baseline levels (untreated cells). All quantitative RT-PCR data are normalized to expression level of GAPDH mRNA. (D) Methylation analysis of SOCS1 in PBMCs and T cells from healthy donors by MSP. After bisulfite modification, genomic DNA of PBMCs and T cells purified from the same PBMCs was subjected to SOCS1-specific MSP (top panel). U indicates unmethylated (175 bp) and M indicates methylated (160 bp) SOCS1 gene. Bottom panel shows expression of SOCS1 mRNA measured by quantitative RT-PCR in PBMCs and T cells from the same healthy donor. Data are demonstrated as the mean fold increase relative to baseline levels (untreated cells). All quantitative RT-PCR data are normalized to expression level of GAPDH mRNA. One representative of 3 independent experiments is shown. (E) Coimmunoprecipitation analysis of SOCS1 mediated JAK1 ubiquitination is shown in CD3T cells incubated in the absence or presence of lenalidomide and pomalidomide after preincubation with or without bortezomib. Cells were stimulated with IFNγ for 15 minutes to induce SOCS1 expression. SOCS1 protein was immunoprecipitated and blotted with anti-SOCS1 moAb, anti-JAK1 Ab, and ubiquitin moAb. Top panel demonstrates IP:IB bands. Bottom panel demonstrates pixel density values of each band expression relative to IgH chain expression. Fold expression of proteins are determined relative to control. One representative of 3 independent experiments is shown.

Figure 5

IMiDs enhance CTL-mediated cytotoxicity against SOCS1-expressing MM cells. (A) Specific cytotoxic T cell activity against SOCS1 gene re-expressing target MM cell lines MM.1S and U266 were evaluated by 51Cr-release assay. To re-express SOCS1 gene, target MM cell lines were incubated either alone or in combination with AzaC, TsA, lenalidomide, or pomalidomide. MM cell–specific CTLs were generated from healthy donors by restimulation with γ-irradiated MM1.S or U266 apoptotic bodies using dendritic cells. CTLs were pretreated with lenalidomide or pomalidomide for 24 hours before use as effector cells against SOCS1 gene re-expressing MM cells in 51Cr-release assay. x-axis represents ratio of effector (IMiDs-pretreated MM cell–specific CTLs) to target SOCS1 gene re-expressing MM cells. y-axis represents specific lysis (percent) of target cells by effector cells. Top panel shows lenalidomide pretreated effector CTL lytic activity against SOCS1 gene re-expressing MM1.S cells; bottom panel shows pomalidomide-pretreated effector CTL lytic activity against SOCS1 gene re-expressing MM1.S cells. One representative of 3 independent experiments is shown. (B) Proliferation of CD8T cells cultured with the combination of lenalidomide and pomalidomide with AzaC or Tsa compared with lenalidomide or pomalidomide alone was evaluated by CFSE-flow cytometric analysis. x-axis represents CFSE stained proliferating cells, y-axis represents gated CD8+ T cells. Data are shown as histo-gram proliferation plots using Flowjo analysis software. One representative of 3 independent experiments is shown. (C) Direct antitumor activity of lenalidomide and pomalidomide, either alone or in combination with AzaC, was shown by 3[H]thymidine incorporation assay in MM1.S and U266 cells at 96 hours of culture. Data represent mean ± standard deviation of triplicate cultures. (*) indicates statistical significance determined by Student t test, 1-tailed distribution P < .01. One representative of 3 separate experiments is shown.