Microrna-Dependent Modulation of Osteogenesis in a 3D in Vitro Bone Marrow Model System of Multiple Myeloma
Michaela R. Reagan,Yuji Mishima,Yong Zhang,Patricia Maiso,Salomon Manier,Yu-Tzu Tai,Priya Dhir,Zhi Ning Lu,Masoumeh Memarzadeh,Antonio Sacco,Yosra Aljawai,Susanna Santos,John Ready,David L. Kaplan,Aldo M. Roccaro,Irene M. Ghobrial
DOI: https://doi.org/10.1182/blood.v122.21.3093.3093
IF: 20.3
2013-01-01
Blood
Abstract:Abstract Introduction Recent evidence indicates that tumor cells are not only influenced by their microenvironment, but are also able to drastically alter their surroundings leading to cancer progression. Multiple Myeloma (MM) involves clonal proliferation of malignant plasma cells within the bone marrow, inhibition of osteoblast function, and increased osteoclast activity leading to osteolytic lesions. Our work aims to understand the bi-directional interactions between MM cells and mesenchymal stromal cells (MSCs), using both 2D and 3D in vitro co-culture bone marrow models. Methods We developed a 3D in vitro model system to better mimic myeloma growth within the bone marrow using human MSCs (hMSCs) and fluorescent-, luciferase-labeled MM cell lines seeded into porous, autofluorescent silk scaffolds. Proliferation and osteogenic differentiation of myeloma patient (MM-) and normal donor (ND-) MSCs cultured with or without MM.1S cells were characterized in 2D culture and 3D scaffolds. Non-destructive bioluminescent imaging and fluorescent confocal imaging were used to observe cell growth and cell-cell interactions within scaffolds. Histology was performed to confirm changes in extracellular matrix (ECM) production and bone tissue formation. microRNA (miRNA) profiling was performed on primary ND- (n=3) and MM-MSCs (n=7) using Nanostring technologies. We analyzed 800 human miRNAs from miRBase v.18 and 230 human cancer-related genes using the nCounter® Human Cancer Reference Kit. Gain-of function studies (miRvana mimics) were performed for miRNAs that were down-modulated in MM vs ND-MSCs, and in the 3D model MSCs co-cultured with MM.1S vs MSCs alone, using lipofectamine. Modulation of osteogenesis was evaluated using alizarin red staining and qRT-PCR for the osteogenic markers: IBSP (integrin-binding sialoprotein), Col1a1 (collagen, type I, alpha 1), RUNX2 (runt related transcription factor 2), ALPL (alkaline phosphatase), OPN (secreted phosphoprotein 1), and BGLAP (bone gamma-carboxyglutamate (gla) protein). Results MM-MSCs presented with a lower proliferation rate compared to ND-MSCs and this phenotype was also observed in ND-MSCs co-cultured in the presence of MM.1S cells compared to ND-MSCs alone. Moreover, significant inhibition of MSC growth was evident when co-cultured with MM.1S cells, using a 3D model (Figure 1), where inhibition of osteogenesis, and ECM production were also documented. Alizarin red staining demonstrated inhibited ability for MM-MSCs to undergo osteogenic differentiation. In addition, MM-MSCs differed from ND-MSCs at the gene and miRNA level. Specifically, CDKN1A and CDKN2A were over-expressed in MM vs. ND-MSCs, (P<0.05; fold change >1.2), thus explaining, at least in part, the decreased proliferation of MM-MSCs vs ND-MSCs. Moreover, down-regulation of specific miRNAs (miRNA-199a, -24-3p, -199a, -15a-5p, -16-5p) was demonstrated in MM- vs ND-MSCs, as well as in ND-MSCs vs ND-MSCs co-cultured with MM.1S, using the 3D model. By over-expressing miRNA-199a, -15a-5p and -16-5p, we were able to increase the osteogenic potential, thus suggesting their role in modulating osteogenesis in MM-MSCs. Conclusions Our 3D platform provides a simple, non-destructive, flexible, and clinically relevant tool to spatially and temporally model myeloma growth within bone. It recapitulates decreased bone formation as seen in MM patients and suggests miR-199a-3p, 15a-5p and 16-5p as novel bone anabolic targets. Disclosures: Tai: Onyx: Consultancy. Ghobrial:Onyx: Advisoryboard Other; BMS: Advisory board, Advisory board Other, Research Funding; Noxxon: Research Funding; Sanofi: Research Funding.