Mesenchymal Stromal Cells (MSC) As Key Players in the Tumor Microenvironment Transformation from MGUS to Myeloma

Introduction: MSC, through a complex crosstalk with neighboring cells/factors, can inhibit many effector functions of immune cells, thereby promoting an immunosuppressive state in the tumor microenvironment. Recently, we demonstrated that MSC derived from Multiple Myeloma patients (MM-MSC) are able to convert normal immature myeloid cells in Myeloid Derived Suppressor Cells (MDSC), contributing to immune-escape mechanisms. Since it has been demonstrated a connection between the stimulation of specific Toll-like receptors (TLR) and MSC activation status, including two distinct phenotypes defined MSC1 (TLR4-dependent) or MSC2 (TLR3-dependent), we hypothesized that MM-MSC could be activated to better 'serve' the cancer cells.

Methods: Peripheral blood mononucleated cells (PBMC) isolated from healthy subjects (HC) were cultured with healthy controls (HC-), MGUS-, SMM- or MM-MSC . After 6 days, educated neutrophils (ed-N) were isolated using magnetic microbeads.

Results: Only ed-N isolated from SMM- and MM-MSC co-cultures showed an activation (N2) phenotype with suppressive effects on T cell proliferation in vitro (p<0.001) and up-regulation of Arg1, NOS2 and TNFα. Moreover only SMM- and MM-MSC ed-N increased both tube length and number of branch points when co-cultured with Human Brain Microvascular Endothelial Cells (HBMEC) (p<0.05). Adding Bortezomib, Lenalidomide or Pomalidomide during co-culture, ed-N lost the pro-angiogenic activity but did not lose immunosuppressive ability. Western blotting analysis showed the activation of TLR4/MyD88 pathway in MM-MSC with respect to HC-MSC. To investigate whether TLR signaling could transform healthy MSC in immunosuppressive stromal cells like MM-MSC, we pre-treated HC-MSC with LPS or poly(I:C) as agonists, respectively, for TLR4 and TLR3. After 24 h, HC-MSC were then cultured with HC-PBMC. Only ed-N isolated from co-cultures with HC-MSC pre-treated with LPS showed in vitro N2 phenotype with suppressive effects on T cell proliferation (p<0.001) and pro-angiogenic ability (p<0.05). No effects were observed after TLR3 stimulation.

To examine if plasma cells (PC) play a role in MSC polarization, before performing co-cultures with PBMC, we pre-treated HC-MSC with MM cell lines. Plasmacell pre-treatment drives healthy MSC to activate neutrophils in immunosuppressive and pro-angiogenic cells. Moreover, MM cell lines were able to activate TLR4 pathway on HC-MSC as observed by western blot analysis.

We next investigated in vivo the pro-tumor role of MM-MSC with respect to HC-MSC. After 6 days from implanting of mixtures of fluorescently labeled MM cells and HC- or MM-MSC into immunocompetent zebrafish, animals co-injected with PC and MM-MSC showed enhanced tumor colonization and growth calculated as tumor volume and fluorescence intensity compared with those injected with PC and HC-MSC (p<0.05). Cytofluorimetric analysis confirmed higher localization of human CD138+/CD38+ and human CD90+ cells in the whole kidney marrow of zebrafish (p<0.001).

Conclusion: Tumor microenvironment transformation from MGUS to MM is associated with progressive activation of MSC in pro-inflammatory polarized cells which facilitate MM growth in vivo promoting immune-escape mechanisms and angiogenesis. This transformation is at least partially induced by tumor cells themselves, probably through activation of TLR4 pathway on MSC.