Bortezomib (BTZ), a proteasome inhibitor that induces accumulation of misfolded proteins and apoptosis in highly secretory cells, has improved survival in multiple myeloma (MM) but fails to achieve cure. It has been postulated that a population of MM cells, phenotypically similar to B-cells, are intrinsically resistant to BTZ, and contribute to minimal residual disease (MRD) and relapse (Matsui et al., 2008).

Bone marrow (BM) stroma expresses CYP26 enzymes and creates a retinoic acid (RA)-low environment that prevents differentiation of normal and malignant cells (Meng et al., 2015). Since RA promotes plasma cell differentiation, Ig secretion (Ertesvag et al., 2007) and thus, ER-stress (Xu et al., 2007), we tested if the BM niche induces BTZ resistance by promoting a B-cell program in MM. Consistent with this, MM cells (H929 and MM1S) cultured under low RA conditions (co-cultured with BM stroma or treated with AGN194310, a pan-RA receptor inhibitor) up regulated B-cell markers (BCL6), and down regulated genes associated with plasma cell differentiation and BTZ sensitivity (XBP1s, Blimp, CHOP) (Fig 1A). In addition, inhibition of stromal CYP26 via R115866 reversed these gene expression changes in stroma co-culture conditions (Fig 1A).

To verify whether a low-RA environment promotes BTZ resistance, we incubated MM cells with BM mesenchymal cells for 5 days. Then, MM cells were separated from stroma, and treated with BTZ. Incubation with BM stroma induced BTZ resistance of MM cells, which was completely overcome by CYP26 inhibition or by the CYP26 resistant retinoid IRX5183 (Fig 1B). Moreover, MM cells pretreated with AGN for 5 days were also resistant to BTZ (Fig 1B). Interestingly, this BTZ resistant phenotype was preserved for up to 48h upon removal of MM cells from stromal co-culture (Fig 1C).

Recent studies suggest that malignant cells remodel their microenvironments to build a more protective niche (Schepers et al., 2013). Consistent with this, we found that stromal CYP26A1 was highly upregulated after co-culture with MM cells. Supernatant from MM cells had similar effects, implicating a soluble factor on this interaction. Since secretion of SHH was implicated in chemotherapy resistance in MM (Liu et al., 2014), we tested if MM-induced CYP26 up-regulation was dependent on SHH signaling. As hypothesized, treatment with the Smo antagonist cyclopamine, or the use of SMO knockout (KO) stromal cells, overcame CYP26A1 up regulation by MM cells (Fig 1D)

To test if cell extrinsic SHH signaling in stromal niche contributes to BTZ resistance by modulating RA pathway in MM cells, we modeled MM-BM stroma interactions in a xenograft setting. Each mouse carried two subcutaneous tumors consisting of Luciferase expressing MM1S cells and SmoFl/Fl BM stroma cells, transduced with either a control vector (WT stroma) or Cre-recombinase (Smo KO stroma) (Fig E). Mice were treated with IRX5183, BTZ, or the combination. Consistent with our in vitro data, tumors with Smo KO stroma showed a significant response to BTZ, while those with WT stroma were refractory, as determined by exponential increase in bioluminescence. However, the combination of IRX+BTZ resulted in a dramatic and equivalent response regardless of the phenotype of the stromal compartment (Fig 2A, 2B). Moreover, tumors with Smo KO stroma, or treated with IRX, showed decrease expression of B-cell markers, and up-regulation of genes associated with plasma cell differentiation and BTZ sensitivity.

In conclusion, we show that the BM niche promotes a B-cell program in MM, characterized by a low ER-stress and BTZ resistance. This niche-induced program is maintained even after displacement from the BM, which may have important implications for the use of mobilization-sensitization strategies. Finally, we propose the existence of a bi-directional crosstalk between MM cells and their respective niches. In our model, MM cells via secretion of SHH modify BM stroma to create RA-low environments, induce BTZ resistance and contribute to MRD.

Figure 1.
Figure 1.
View largeDownload slide
Figure 2.
Figure 2.
View largeDownload slide
Disclosures

No relevant conflicts of interest to declare.

Topics:
bone marrow, bortezomib, multiple myeloma, coculture techniques, neoplasms, stress, tumor cells, malignant, antagonists, chemotherapy regimen, cyclophosphamide, doxorubicin, prednisone, and vincristine

Author notes

*

Asterisk with author names denotes non-ASH members.

© 2015 by The American Society of Hematology
2015
Sign in via your Institution