MiR-16 regulates crosstalk in NF-κB tolerogenic inflammatory signaling between myeloma cells and bone marrow macrophages

Abstract

High levels of circulating miR-16 in the serum of multiple myeloma (MM) patients are independently associated with longer survival. Although the tumor suppressor function of intracellular miR-16 in MM plasma cells (PCs) has been elucidated, its extracellular role in maintaining a nonsupportive cancer microenvironment has not been fully explored. Here, we show that miR-16 is abundantly released by MM cells through extracellular vesicles (EVs) and that differences in its intracellular expression as associated with chromosome 13 deletion (Del13) are correlated to extracellular miR-16 levels. We also demonstrate that EVs isolated from MM patients and from the conditioned media of MM-PCs carrying Del13 more strongly differentiate circulating monocytes to M2-tumor supportive macrophages (TAMs), compared with MM-PCs without this chromosomal aberration. Mechanistically, our data show that miR-16 directly targets the IKKα/β complex of the NF-κB canonical pathway, which is critical not only in supporting MM cell growth, but also in polarizing macrophages toward an M2 phenotype. By using a miR-15a-16-1-KO mouse model, we found that loss of the miR-16 cluster supports polarization to M2 macrophages. Finally, we demonstrate the therapeutic benefit of miR-16 overexpression in potentiating the anti-MM activity by a proteasome inhibitor in the presence of MM-resident bone marrow TAM.

Trial registration: ClinicalTrials.gov NCT01408225.

Keywords: Bone marrow; Cancer; Hematology; Macrophages; Oncology.

Figure 1. EVs and intracellular miR-16 levels are correlated.

(A) Heatmaps showing microRNA expression profile as measured by the NanoString nCounter System in MM cells (RPMI-8226, U266, NCI-H929, MM1.R) (left panel) and in extracellular vesicles (EV) secreted by those cells (right panel). Each column represents 1 sample/cell line with red representing upregulated and blue representing downregulated. Each cell line was run at least in triplicate. Heatmaps were performed using the G-plots package heatmap.2 program, and colored scales were generated using the Z score values. (B) Pie charts showing the percent of the 59 highest intracellular microRNA expression levels and their corresponding EV secreted levels in the 4 cell lines tested. The 12 highest microRNA expression levels among cell lines from miR-16 (blue) to miR-92a (orange) are highlighted in a colored spectrum. (C) miR-16, miR–142-3p, and miR-9 expression levels in EVs released by U266, RPMI-8226, and NCI-H929 MM cell lines. Data are presented as fold change (f.c.) over intracellular microRNA expression for each miRNA. (D) Parallel to C using HS-5 cell line. Values represent the mean ± SD; P values were calculated using ordinary 1-way ANOVA multicomparisons test. Each experiment was performed in triplicate. (E) qPCR showing miR-16 expression in EVs released by Del13 MM cell lines (U266, NCI-H929, RPMI-8226, OPM2, LP-1, L363, MM.1S) and non-Del13 MM cell lines (OCIMY-5, OCIMY-I, MMM.1). Data are presented as 2^-ΔCT values. Values represent the mean ± SD; P values were calculated using 2-tailed unpaired t test. Each experiment was performed in triplicate; the obtained P values are reported.

Figure 2. MiR-16 is downregulated in the BM-MΦ of MM patients (A) Cytokine array showing, under stimulated conditions (i.e., in the presence of single-stranded RNA–mir-25 (ssRNA–miR-25), which stimulates TLR-7 and -8, the levels of NF-κB–induced, M2-associated cytokines (IL-6, IL-8, TNF-α, and VEGF) secreted by CD14⁺ macrophages (BM-MΦ), and CD14^– cells (BM-CD14 neg.) isolated from the BM of MM patients (n = 3).

Data represent mean values ± SD. Cells were incubated with ssRNA–miR-25 for 24 hours prior to detection of cytokines. Cytokine levels were measured in pg/mL using a multiplex cytokine assay. (B) qPCR showing mRNA expression levels of IL-6, IL-8, and TNF-α in BM-MΦ isolated from MM patients (MM) or cancer-free donors (HD) (n = 4/group). P values were calculated using 2-tailed multiple t test. (C) qPCR showing decreased miR-16 expression in BM-MΦ isolated from MM patients as compared with that isolated from HD (n = 7/group). (D) Real-time PCR showing decreased expression of miR-16 in BM-MΦ compared with that in monocytes isolated from the peripheral blood (PB-M) of the same MM patients (n = 3). Data are presented as 2^-ΔCT values. (E) Representative images captured by light microscopy showing PB-M differentiated to M2-like MΦ (M2-MΦ) with macrophage CSF (M-CSF) treatment for 7 days (right panel) as compared with undifferentiated PB-M (UI) (left panel). Magnification ×40. (F) Flow cytometric analysis showing phagocytosis of M2-MΦ in vitro. Phagocytosis assay was performed using latex beads coated with GFP fluorescently labeled IgG antibody. The engulfed fluorescent beads were detected by flow cytometry. Differentiated PB-M to M2-MΦ in the presence of M-CSF showed 84% GFP⁺ cells in contrast to UI cells that were only 4.63%. (G–I) Real-time PCR showing increases in CD163 (G) and IRF4 (H) mRNA expression, as well as a decrease in miR-16 expression (I) in M2-MΦ as compared with UI. CD163 and IRF4 levels were presented as fold changes over UI controls. miR-16 levels were presented as 2^-ΔCT values (n = 3 patients). Data reported in C, D, and G–I represent the mean ± SD; P values were calculated using a 2-tailed unpaired t test.

Figure 3. Extracellular miR-16 impairs MM-EV–induced M2-MΦ.

(A) Representative images captured by light microscopy showing PB-M isolated from a MM patient differentiated to M2-MΦ in the presence of the matched BM acellular fraction (matched BM-ac) (right panel), in contrast to undifferentiated PB-M (UI) (left panel). (B) qPCR showing decreased expression of miR-16 in PB-M isolated from MM patient and differentiated to M2-MΦ in the presence of the matched BM acellular fraction (PB-M + BM-ac), as well as in total MΦ isolated from the BM of the same patient (MM-BM-MΦ), as compared with undifferentiated PB-M using samples obtained from n = 3 MM patients. Values represent the mean ± SD; P values were calculated using ordinary 1-way ANOVA multicomparison. (C) Representative images captured by light microscopy showing differentiation of PB-M obtained from a HD incubated with EV isolated from the BM-acellular fraction of a MM patient (BM-ac) (left panel) or EV-depleted BM-ac (right panel) (n = 4 patients; see Supplemental Figure 2B). (D) Representative images showing phagocytosis of latex beads coated with GFP fluorescently labeled IgG antibody by PB-M differentiated to M2-MΦ when incubated with EV isolated from the BM-ac of a MM patient (+) (right panel), whereas no phagocytosis was observed when PB-M were incubated with the EV-depleted BM-ac (–) (left panel). (E) Flow cytometric analysis showing percent increase in expression of M2-MΦ surface marker (CD163) on PB-M treated with EV isolated from the conditioned media of a MM cell line, NCI-H929, for 7 days (EV) (upper right panel) and compared with cells incubated with EV-depleted acellular fraction (Ctl) (upper left panel). The same effect was seen using EV isolated from another MM cell line (MM.1S) (lower panel). Gating strategy was set using IgG anti-PE antibody isotype control. (F) Flow cytometric analysis showing percent surface expression of CD163 on PB-M differentiated with EV isolated from the conditioned media of MM.1S cells and concomitantly incubated with either ds-miR–16 (middle right panel), ds-miR–223 (lower right panel), or scramble control (Scr) (upper right panel). CD163 percent surface expression on undifferentiated cells (UI) incubated with the microRNAs cited above are also shown (left panels). Gating strategy was set using IgG anti-PE antibody isotype control. (G) Representative images captured by light microscopy showing impairment of PB-M differentiation to M2-MΦ in the presence of ds-miR–16. PB-M were incubated with the EV isolated from MM.1S, along with ds miR-16 (right panel) or Scr control (left panel) for 7 days. All magnifications ×40.

Figure 4. EV isolated from MM cells carrying Del13 strongly induce MΦ polarization.

(A and B) Representative flowsight cytometric analysis overlaid with the respective isotype control (Iso-C) showing comparable CD163 and CD206 percent surface expression on differentiated PB-M isolated from a healthy donor upon treatment with the EV isolated from Del13 cell lines (L363, LP-1, OPM2) (A) or no Del13 cells (OCIMY-5, OCIMY-I, MMM.1) (B) for 4 days. Gating strategy was set using a mix of IgG anti-FITC and IgG anti-APC antibodies isotype control. (C) Bar dot plots showing average CD163 and CD206 percent surface expression on differentiated PB-M treated with EV-Del13 cell lines (red, n = 3) or EV–no-Del13 cell lines (green, n = 3). Each dot plot represents a flowsight cytometric analysis percent reading. Statistical comparisons for each surface marker were performed only between Del13 and no-Del13 cell lines. Values represent the mean ± SD; P values were calculated using the 2-tailed unpaired t test. Percent of untreated PB-M surface markers expression are indicated in blue and were only used as internal control. (D and E) Representative flowsight cytometric analysis overlaid with the respective isotype control (Iso-C) showing comparable CD163 and CD206 percent surface expression on differentiated PB-M isolated from a healthy donor upon treatment with the BM-acellular fraction (BM-ac) of MM patients carrying Del13 (HTB-191, HTB-192, HTB-193) (D) or BM-ac of MM patients with no Del13 (HTB-194, HTB-195, HTB-196) (E) for 4 days. Gating strategy was set using a mix of IgG anti-FITC and IgG anti-APC antibody isotype controls. (F) Bar dot plots showing average CD163 and CD206 percent surface expression on differentiated PB-M treated with BM-ac of MM patients with Del13 (red, n = 3) or no-Del13 (green, n = 3). Percent of untreated PB-M surface markers expression are indicated in blue. Each dot plot represents a flowsight cytometric analysis percent reading. Statistical comparisons for each surface marker were performed only between Del13 and no-Del13 patients. Values represent the mean ± SD; P values were calculated using 2-tailed unpaired t test.

Figure 5. Evaluation of MΦ polarization in a miR–15a/16-1–KO model.

(A and B) Representative flowsight cytometric analysis with overlaid isotype control (Iso-C) showing comparable CD206 percent surface expression on differentiated M0-MΦ isolated from WT or miR-15a/16-1^–/– B6 mouse spleens treated ex vivo with M-CSF (100 U/mL) and IL-4 (20 ng/mL) for 48 hours (A) or 6 days (B). Gating strategy was set using an IgG anti–Alexa Fluor 488 antibody isotype control. (C and D) Bar dot plots showing average CD206 percent surface expression on differentiated MΦ isolated from WT and miR-15a/16-1^–/– mice treated for 48 hours (n = 4 mice/group) (C) or 6 days (n = 3 mice/group) (D). (E and F) Parallel to A and B, except that the surface marker analyzed was Dectin-1 and the gating strategy was set using an IgG anti-APC antibody isotype control. (G and H) Parallel to C and D, except that percents of Dectin-1 expression were recorded. Values represent the mean ± SD; P values were calculated using 2-tailed unpaired t test. P values are reported in the figures.

Figure 6. MiR-16 directly regulates the expression of IKKα/β complex (A and B) Real-time PCR analysis revealing mRNA expression of IKKα and -β complex (A) in MΦ isolated from MM patients (n = 3), and U-937 cell line (B), each transfected with ds miR-16 or Scr control for 48 hours.

(C) Western blot analysis showing protein downregulation of IKKα and -β by miR-16 in U-937 (left panel) and HS-5 cells (right panel) as compared with Scr control after 48 hours of treatment; GAPDH was used as loading control. (D and E) Real-time PCR showing mRNA expression of IKKα (D) and -β (E) in differentiated MΦ isolated from WT or miR-15a/16-1^–/– mouse spleens (n = 4/group; the RNA from each 2 mice/group were pooled for quality purposes and a triplicate PCR reading was done for each pool). In D and E, error bars represent standard deviation. (F) Western blot analysis showing protein levels of IKKβ and IKKα in monocytes/MΦ CD11b fractions isolated from WT or miR-15a/16-1^–/– mice spleens (n = 2 mice, WT; n = 3 mice, miR-15a/16-1^–/–). β-Actin is used as loading control. (G) Real-time PCR showing mRNA downregulation of IKKα expression in MM cell lines (MM.1S and NCI-H929) transfected with ds-miR–16 as compared with Scr control for 48 hours. (H) Western blot analysis showing protein downregulation of IKKα and -β by miR-16 in 3 MM cell lines (U266, MM.1S, and NCI-H929); GAPDH was used as loading control. (I) Luciferase reporter assay revealed direct downregulation of IKKβ 3′UTR transcriptional activity by miR-16. pGL4.11 luciferase vector containing the 3′UTR of IKKβ was transfected in U266 or HS-5 for 18 hours, followed by a second transfection with ds-miR–16 or Scr control for an additional 12 hours. Transfection efficiency was controlled by cotransfection with TK promoter-Renilla vector. Data are presented as percent of Scr control. When not otherwise specified in the legend, the reported P values were calculated using 2-tailed unpaired t test. Each of the reported experiments was performed in triplicate. For C and H, the data presented are representative of 2 independent experiments.

Figure 7. 36.

MiR-16 increases MM cell sensitivity to bortezomib by impairing NF-κB signaling. (A) Western blot analysis showing decrease in P65 nuclear protein in NCI-H929 upon transfection with ds-miR–16 or Scr Control for 48 hours. Histone Deacetylase 1 (HDAC1) was used as nuclear loading control. (B) Single cell flow cytometric analysis showing decrease in nuclear p65 protein expression in MM cells upon transfection with ds-miR–16 as compared with Scr control (upper panel). Percent of normalized P65 frequency in 1 × 10⁴ cells transfected with ds-miR–16 or Scr control are shown (lower panel). (C) Luciferase reporter assay showing decrease in NF-κB transcriptional activity in U-937 cells by miR-16 as compared with Scr control. (D) Parallel to C, except that luciferase activity was assessed in HS-5 cells, no stimulation was induced, and an additional transfection with pGL3 vector containing a trimer of mutated NF-κB binding sites sequence (mut 3×NF-κB) was performed. Data are presented as percent of WT 3×NF-κB Scr control. (E) Cytokine array showing concentrations of cytokines/chemokines IL-6, IL-8, TNF-α, and INF-γ released by BM-MΦ isolated from MM patient upon treatment with ds-miR–16 or Scr sequences encapsulated in liposomes for 48 hours. (F) Luciferase reporter assay showing decrease in NF-κB transcriptional activity in U266 upon transfection with miR-16 as compared with Scr control. Cells were transfected with pGL3 luciferase vector containing a trimer of consensus WT NF-κB binding sites for 6 hours and were then treated with 5 nM bortezomib (bortez), 10 μM BAY11-7082, or vehicle control (veh) for 12 hours. Data are presented as percent of Scr veh control. (G) Representative flowsight showing apoptosis analysis by Annexin-V/PI double staining. Annexin-V/PI staining was performed on MM.1S cells upon transfection with miR-16 or Scr sequences and treated with bortez 2.5 nM or veh DMSO control for 48 hours. The percentages of Annexin-V–positive cells are shown in the upper left quadrants. (H) Luciferase assay showing percent of viable MM.1S GFP/Luc⁺ cells in suspension upon treatment with bortez or DMSO veh control, cocultured with BM-MΦ isolated from a patient for 48 hours. (I) Luciferase assay showing percentage of viable MM.1S GFP/Luc⁺ cells in suspension upon treatment with bortez or DMSO veh control cocultured with BM-MΦ isolated from a MM patient and treated with ds-miR–16 or Scr sequences for 48 hours. Values represent the mean ± SD for each experiment performed in triplicate. When not otherwise specified in the legend, the reported P values were calculated using 2-tailed unpaired t test.

Figure 8. Illustrative diagram showing plasma cells carrying Del13 (PCs Del13) secrete extracellular vesicles (EV) containing cargoes that induce monocyte differentiation toward M2 tumor-supporting M (TAM).

Mechanistically, a lack of significant amounts of miR-16 in EV prevents targeting the IKKα/β complex, resulting in an increase in the NF-κB canonical pathway. Upregulation of the NF-κB pathway leads to enhanced secretion of M2 tumor effector chemokines/cytokines, including IL-6, IL-8, IL-10, and TNF-α, resulting in an increase in TAM levels and in PC clonal expansion. However, normal PCs or PCs carrying WT chromosome 13 (PCs Chr.13 WT) have higher levels of miR-16 in their EV that target the IKKα/β complex, resulting in a decrease in the NF-κB canonical pathway and impairment of monocyte differentiation toward the M2-M subset of the microenvironment and, thus, PC clonal expansion.