Abstract

There is an increasing body of evidence indicating that the bone marrow microenvironment can generate drug resistance in acute leukemia. The mechanisms underlying this effect have not yet been elucidated; signals triggered by direct contact with extracellular matrix components and by mesenchymal cell (MSC)-secreted factors have been implicated. The protective effect of the microenvironment has been primarily observed for classical chemotherapeutic drugs. Recent reports, however, indicate that this can also occur with molecularly targeted therapies. Thus, it was shown that interleukin (IL)-7 desensitizes BCR-ABL+ leukemic cells to imatinib (Williams RT et al. Genes Dev 2007) and MSC-conditioned media protects BCR-ABL+ cells to imatinib and nilotinib (Weisberg E at al. Mol Cancer Ther, 2008). Several tyrosine kinase inhibitors are in clinical development for the treatment of acute myeloid leukemia (AML). The aim of this study was to determine whether bone marrow MSC affected the sensitivity of AML cells to 3 promising tyrosine kinase inhibitors (sorafenib, sunitinib, and midostaurin) and, if so, to begin to elucidate the underlying mechanisms. Using proliferation assays, we found that 3 AML cells lines (MV4-11, U937, and THP1) were significantly less sensitive to the tyrosine kinase inhibitors when cultured in the presence of bone marrow-derived MSC for 24h before exposure to drugs for 72h. In experiments with MV4-11, IC50 increased from 4.7 nM to 55 nM for sorafenib, from 10 nM to 110 nM for sunitinib, and from 28 nM to 135 nM for midostaurin; in experiments with U937, IC50 increases were 5.1 μM to 11 μM, 6.2 μM to > 10 μM, and 230 to > 1000 nM for each drug; and in experiments with THP1, they were 6.3 μM to 11 μM, 2.2 μM to > 10 μM, and 211 nM to 996 nM. Coculture with MSC also reduced sorafenib- and sunitinib-induced apoptosis by > 60%. Interestingly, drug resistance increased even further after coculturing the cell lines with MSC for 4 weeks or longer: sunitinib had virtually no effect on the proliferation of MV4-11 cells at concentrations of up to 100 nM, and on THP-1 cells at 10 μM. To determine whether the induction of drug resistance was dependent on the direct contact of AML cells with MSC, we tested sensitivity to sorafenib after separating MV4-11cells from MSC with transwell inserts. Under these conditions, the protective effect of MSC was lessened but not abrogated. These results indicated that direct contact with MSC was not an absolute requirement for induction of drug resistance and that MSC-secreted soluble factors might be, at least in part, involved. We therefore determined the soluble factors secreted by MSC using a multiplex assay and tested whether their secretion was augmented by contact with AML cells. MSC secreted IL-6 (230 pg/mL), IL-8 (1880 pg/mL), and VCAM-1 (30 pg/mL). When cocultured with MV4-11, U937 and THP-1 cells for 24h, IL-6 secretion increased 1.3 to 1.8-fold, IL-8 increased 1.5 to 2.6-fold, and VCAM-1 increased 2.2 to 5.6-fold; after 72 of coculture, dramatically elevated levels of IL-6 (2140–3869 pg/mL), IL-8 (4296–8068 pg/mL), and VCAM-1 (5109–6389 pg/mL) were observed. The effects of these and other MSC-derived factors on the sensitivity of AML cells lines and primary AML cells to tyrosine kinase inhibitors are being tested. These results indicate that the anti-AML effect of tyrosine kinase inhibitors is strongly inhibited by bone marrow MSC cells, and support the concept that the microenvironment is an important determinant of resistance to these agents in leukemia. We suggest that the development of agents that interfere with the interaction between AML cells and MSC, and with the molecular mechanisms underlying this protective effect of MSC is a crucial step to improve cure rates.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author