Introduction & Aims of the Study: The unique properties of the bone marrow allow for migration and proliferation of multiple myeloma (MM) cells, while also providing the ideal environment for development of quiescent, drug-resistant MM cell clones. Recent studies suggest that bone marrow adipocytes (BMAds) can contribute to myeloma-induced bone disease, and reciprocally, that BMAds support MM cells in vitro. Importantly, most data investigating BMAds have been generated using adipocytes differentiated from bone marrow-derived mesenchymal stromal cells (MSCs) in vitro due to the technical challenges associated with isolating and culturing primary adipocytes. To that end, we investigated the MM-BMAd relationship ex vivo utilizing primary adipocytes (pBMAds), which overcome many of the drawbacks associated with using MSC-derived adipocytes.

Methods: pBMAds were isolated from the top of human bone marrow biopsies after centrifugation and seeded onto one of three in vitro systems. To develop an in vitro model with a tissue-like structure to mimic the bone marrow microenvironment, we developed the first 3D, tissue engineered model utilizing pBMAds derived from human bone marrow. Following adherence of pBMAds, MM cells were added to investigate bidirectional tumor-host signaling. Viable pBMAds were measured utilizing CellTiterGlo and RealTime Glo (bioluminescence), alone and in culture with MM cells (MM.1Sgfp+luc+ and RPMI-8226). MM.1Sgfp+luc+ cell number was determined by luciferin spike-in and measurement of bioluminescence. To investigate the effects of myeloma cells on pBMAds, we used fluorescence and confocal microscopy to image cell morphology, and mass spectrometry proteomics of conditioned media samples to specifically interrogate soluble signals. Gene expression in both cell types was investigated by qPCR.

Results: Utilizing adipocytes isolated from 12 different human donors, we found that pBMAds, which are extremely fragile, can be isolated and stably cultured in 2D for 10 days and in 3D for short-terms (~2 weeks) or long-terms (1 month) in vitro (Figure 1A). We monitored stable pBMAd presence in 2D transwells via Cell TiterGlo and on 3D scaffolds via RealTime Glo up to 9 days post-seeding, and by calcein staining after 10 days in culture. Using bioluminescence in pBMAs (Cell TiterGlo) and MM.1S (luciferin spike-in) we were able to detect viability of both cell types in co-culture. Utilizing mass spectrometry proteomics, in pBMAds alone, we detected secretion of CXCL12 (SDF1; p<0.01), as well as TGFB1 (p<0.01) which has implications for bone remodeling. Interestingly, in conditioned media collected from the co-culture condition, we found increased secretion of the WNT inhibitor DKK3 (p<0.05) and detection of proteins known to be associated with extracellular exosomes including: CNTN1, ACP1, OMD, RPS18, RPL24, PGAM1, LAMA4, CTSB, PLOD2 and ARHGDIA (Figure 1B), suggesting that pBMAds and MM cells likely interact via exosomes.

Conclusions: In sum, we developed three in vitro cell culture systems to study primary bone marrow adipocytes and myeloma cells, which could be adapted to investigate many diseases and biological processes involving the bone marrow, including other bone-homing tumor types. Our use of these models to examine the interaction between MM cells and pBMAds has revealed new roles for pBMAds in the MM bone marrow microenvironment, with implications for MM-induced bone disease and pBMAd-MM bidirectional signaling.

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution