Mesenchymal Stem Cells Preconditioned with Myeloma Cells from High-Risk Patients Support the Growth of Myeloma Cells from Low-Risk Patients

Introduction:

Each disease stage in myeloma (MM) is associated with parallel changes in both the MM clone and the bone marrow (BM) microenvironment. Mesenchymal cell lineages derived from mesenchymal stem cells (MSCs), including osteoblasts, adipocytes and pericytes play an important role in MM cell growth mediated by the modification of the MM niche in the BM.

The overall goal of the study was to test and identify changes induced in MSCs by high-risk (HR) MM cells that impact MSC function and promote oncogenic pathways capable of supporting low-risk (LR) MM cells.

Methods:

Normal MSCs were either cultured alone ("unconditioned") or co-cultured with MM cells for 5 days. The cultured and co-cultured cells were trypsinized, replated for 40 min, followed by serial washing to remove MM cells from the adherent MSCs. More than 95% of the remaining adherent cells after co-culture were MSCs ("preconditioned"). The unconditioned and preconditioned MSCs or their 24 hrs conditioned media (CM; 50%) were tested for their ability to support the 5-days growth of CD138+ MM cells from LR (n=4) and HR (n=3) patients.

To identify factors altered in MSCs by HR MM cells, the unconditioned and preconditioned MSCs and their serum-free conditioned media (n=4) underwent gene expression profiling and proteomic analysis. Whole bone biopsy gene expression profiles from newly diagnosed patients with MM enrolled in Total Therapy clinical trials were used to correlate the altered expression of factors in preconditioned MSCs with their expression in clinical samples.

Results:

Growth of all MM cells tested was increased by inclusion of MSCs preconditioned with HR MM cells by 2.2± 0.2 (p<0.0004) and by CM from these MSCs by 9.6±2.0 (p<0.006), compared to culture of MM cells in fresh media. In contrast, CM from unconditioned MSCs increased growth of HR MM cells by 2.6±0.6 (p<0.01) fold but had minor effect on growth of LR MM cells. CM from MSCs preconditioned with HR MM cells increased growth of LR and HR MM cells by 5.7±0.1 (p<0.0002) and 2.6±1.2 (p<0.04), compared to culture of MM cells in CM from unconditioned MSCs, respectively. Growth of LR MM cells was higher by 2.9±0.3 fold using CM from MSCs preconditioned with HR MM cells than by using CM from MSCs preconditioned with LR MM cells (p<0.005).

To determine the role of cell-cell contact, we compared the effect of the preconditioned MSCs and their CM on growth of LR and HR MM cells. Growth of LR MM cells (p< 0.003) and HR MM cells (p< 0.005) was higher when cultured in CM than in co-culture with MSCs. These data imply that soluble factors from preconditioned MSCs support MM cell proliferation and that adhesion of MM cells to MSCs may restrain proliferation.

Genes overexpressed in preconditioned MSCs included growth factors (e.g. IL6) and receptors (e.g. EDNRA); genes underexpressed include factors associated with activity of osteoblasts (e.g. ITGBL1) and adipocytes (IGFBP2).

A proteomic analysis showed a reduced level of the secreted factors IGFBP2 and ITGBL1 and increase level of IL6 in CM from MSCs preconditioned with HR MM cells compared to CM from unconditioned MSCs. IGFBP2 mediates local bioavailability of IGF1 and IGF2 and is also involved in bone formation and angiogenesis independently of the IGF axis. ITGBL1 is involved in osteoblastogenesis whereas EDNRA is known to be expressed by pericytes.

Global gene expression profiles from patient material showed that EDNRA and IGFBP2 are not expressed in MM cells but are highly expressed in cultured MSCs compared to hematopoietic cells in buffy coat BM samples. EDNRA is overexpressed (p<0.005) whereas IGFBP2 is underexpressed (p<0.005) in whole BM biopsy samples from MM patients with HR disease compared to patients with LR disease (p<0.005) and in Focal Lesion compared to random BM biopsies taken from the same patients (p<0.0000006 for EDNRA and p<0.02 for IGFBP2).

IHC staining of patients' bone biopsies showed higher numbers of EDNRA+ mesenchymal-like cells in MM (n=10) than MGUS/AMM (n=10, p<0.0003) and in HR MM BM than LR MM BM (p<0.03). IHC analysis also revealed that IGFBP2 is highly expressed by immature adipocytes and that its expression in these cells is reduced in HR MM BM.

Conclusion:

Preconditioning of MSCs is essential for promoting growth of MM cells from LR patients. Factors altered in MSCs by HR MM cells are linked to signaling pathways known to directly stimulate MM cell growth and markers associated with distinct MSC lineages changed in HR MM niche.