MDSCs promote tumor growth in the MM microenvironment. Bidirectional interaction between MM cells and MDSCs is demonstrated as MDSC-mediated MM growth; conversely, MM cells induced the development of MDSCs in vitro. (A) MDSC-mediated MM growth within the MM stroma is demonstrated by 3H-thymidine proliferation assay. BMSCs were generated in vitro from patients with MM BMMCs, with or without MDSC depletion, and then cocultured for 24 hours with a panel of BMSC-responsive MM cell lines. MM cell proliferation was measured by 3H-thymidine incorporation assay. Shown is the percentage of MM cell proliferation in cocultures of MM cell-BMSC without MDSC relative to MM cell–BMSC with MDSC. *The data represent the percentage of MM cell proliferation in cultures with BMSCs that have been generated from BM mononuclear cells with or without MDSC depletion (mean ± SD of triplicate cultures with statistical significance P < .05 by Student t test, 2-tailed distribution). Blue columns represent MM cells cultured with BMSCs, and red columns represent MM cells cultured with BMSCs without MDSCs. (B) The direct effect of MDSCs on MM growth was demonstrated in MM cell line–MDSC cocultures by 3H-thymidine incorporation assay. MDSCs were isolated from MM-BMMCs and cultured with MM cell lines for 48 hours. MM cell proliferation is shown relative to MM cells alone. The data represent mean ± SD of triplicate cultures and are representative of 3 different experiments. Statistical significance is indicated (Student t test, 1-tailed distribution, P < .05). (C) MM-induced MDSC development in healthy PBMCs is shown. PBMCs from healthy donors were cultured with a panel of MM cell lines for 6 days, and MDSCs were determined by multiparametric flow cytometry analysis. Representative dot plots are shown: CD33+CD15+ MDSCs (gated box) within the CD11b+CD14-HLA-DR-/low gated cells in healthy PBMCs (left), and healthy PBMCs cultured with MM cell line RPMI8226 (right). The data shown are representative of 3 different experiments. (D) MM cell–induced MDSCs were further characterized by their immune-suppressive activity against autologous healthy T cells. MDSCs were isolated from MM cell (RPMI8226) and healthy PBMC cocultures, and autologous healthy T cells were isolated from healthy PBMCs by FACS sorting. Then MDSCs were cultured with T cells in the presence of CD3/CD28/IL-2 stimulatory factors for 4 days, and T-cell proliferation was measured by 3H-thymidine incorporation. T-cell proliferation in cultures with or without MDSCs is shown. The data represent mean ± SD of triplicate cultures with statistical significance P < .05 (Student t test, 2-tailed distribution).