MiR-29b antagonizes the pro-inflammatory tumor-promoting activity of multiple myeloma-educated dendritic cells

Dendritic cells (DCs) have a key role in regulating tumor immunity, tumor cell growth and drug resistance. We hypothesized that multiple myeloma (MM) cells might recruit and reprogram DCs to a tumor-permissive phenotype by changes within their microRNA (miRNA) network. By analyzing six different miRNA-profiling data sets, miR-29b was identified as the only miRNA upregulated in normal mature DCs and significantly downregulated in tumor-associated DCs. This finding was validated in primary DCs co-cultured in vitro with MM cell lines and in primary bone marrow DCs from MM patients. In DCs co-cultured with MM cells, enforced expression of miR-29b counteracted pro-inflammatory pathways, including signal transducer and activator of transcription 3 and nuclear factor-κB, and cytokine/chemokine signaling networks, which correlated with patients’ adverse prognosis and development of bone disease. Moreover, miR-29b downregulated interleukin-23 in vitro and in the SCID-synth-hu in vivo model, and antagonized a Th17 inflammatory response. All together, these effects translated into strong anti-proliferative activity and reduction of genomic instability of MM cells. Our study demonstrates that MM reprograms the DCs functional phenotype by downregulating miR-29b whose reconstitution impairs DCs ability to sustain MM cell growth and survival. These results underscore miR-29b as an innovative and attractive candidate for miRNA-based immune therapy of MM.

amplification of cRNA, the clean up and the fragmentation were performed according to the Affymetrix's procedures. Microarray data were generated by Human transcriptom array 2.0 ST (Affymetrix Inc., Santa Clara, Ca) containing over 6 million distinct probes targeting coding transcripts, exon-exon splice junction and non-coding transcript. Arrays were scanned with an Affymetrix GeneChip Scanner 3000. Raw data produced by the Affymetrix Platform (i.e. CEL files) were processed and RMA normalized using Affymetrix Expression Console (EC). Clustering and fold-change analysis were done by using transcription array console (TAC, Affymetrix).

SCID-synth-hu model
In detail, a three-dimensional (3D) bone-like poly-Σ-caprolactone polymeric scaffold (PCLS), presenting interconnected large (100-300 mm) and small pores (1-10 mm) resembling the micro-architecture of a normal human adult bone, was implanted into a 6-weeks old female SCID mouse and seeding of BMMCs into PCLSs was performed. Briefly, A suspension of 8×105 cells in 500 μl of growth medium was threaded into two ending faces of the cylindrical scaffold. Before surgical implantation into a SCID mouse flank, PCLSs were incubated in complete medium at 37°C in 5% CO2 for 24 h to allow cell adhesion on 3D surfaces. Chloralium hydrate anesthesia (400 mg/kg, 0,15 ml) was used during all surgical procedures. After three weeks, 8×105 BM-dependent INA-6 MM cells were injected in vivo into previously implanted PCLSs. Approximately one month later, when sIL6R became detectable in mice sera, miR-29b or negative control (NC) were injected directly into the scaffold (total of 7 injections, 2 days apart). Neutral lipid emulsion (NLE) (MaxSuppressor in vivo RNA Lancer II, BIOO Scientific, Austin, TX) was used for the administration of synthetic miR-29b or NC, according to the manifacturer's instructions.
In vivo effects induced by miR-29b were then evaluated by immunohistochemistry on retrieved scaffolds at the end of treatments.

Immunohistochemistry
For each case 4 μm-thick serial sections were cut from a representative block of formalin fixed, paraffin-embedded tissue, mounted on acid-cleaned glass slides, and heated at 55 °C for 60 minutes. Slides were, dewaxed with xylene, and processed for hematoxylin and eosin and immunohistochemistry. All the procedures were performed at room temperature. Immunohistochemical evaluation of anti-IL-23/Ki67/CD31 was performed using the LSAB+ System HRP (DakoEnvision System, CA), followed by the addition of 3,3′-diaminobenzidine as a chromogen. Endogenous peroxidase activity was quenched for 5 minutes in 3% hydrogen peroxidase, and the slides were rinsed in wash solution (TBST, 0.05 mol/l Tris Buffered Saline with Tween20). Antigen retrieval was performed with EDTA buffer pH 9 for 30 minutes at 98 °C. Slides were washed three times in phosphate-buffered saline (PBS; pH 7.4) for 5 minutes.
Immunostaining was performed using a purified mouse monoclonal antibody antihuman IL-23 (Abcam, 1:150 dilution), anti-Ki67 (Dako, 1:150 dilution) and anti-human CD31 (Dako, 1:40 dilution) for 1 hour at 25 °C. Sections were finally counterstained with hematoxylin. Negative controls were performed in each run by substituting primary antibodies with antibodies with irrelevant specificity but of the same isotype of the primary antibodies.

Data availability
All the data supporting the findings and results of this work are available upon request from the corresponding author.