Mesenchymal Stromal Cells Protect Acute Myeloid Leukemia Cells from Venetoclax-Induced Apoptosis through Contact-Dependent Mechanisms

Abstract

Acute myeloid leukemia (AML) patients have a poor prognosis with standard treatments due to a high relapse rate. Recently, Venetoclax (VEN), the first B-cell lymphoma 2 (BCL-2) selective inhibitor to enter the clinic, has shown a breaking-through effect in combination with demethylating agents in older patients with treatment-naïve AML. However, resistance mechanisms were described, hampering the complete remission of VEN-treated patients. It has been proposed that the bone marrow microenvironment in general and Mesenchymal stromal cells (MSCs), in particular, favor a protective niche that sustains leukemic cell viability and plays a crucial role indetermining drug sensitivity in several hematological disorders, including AML. A previous study demonstrated that VEN, as well as other drugs commonly used in AML therapy, showed a reduced effect in the presence of an MSC-conditioned medium, suggesting that MSC-dependent release of inflammatory signals and other soluble factors could be involved in drug resistance. Our study aims to further investigate the mechanisms of MSC-mediated VEN resistance in AML. We found that primary AML cells isolated from patients at diagnosis and treated with VEN showed different apoptotic rates, indicating an AML cell-specific resistance/sensitivity to the drug. In co-culture experiments, primary MSCs, isolated from healthy donors or AML patients, protected VEN-sensitive AML cells from apoptosis, confirming the critical role of MSCs in drug resistance. To deeply investigate this issue, we performed co-culture experiments using the AML cell line MV4-11, which we demonstrated as sensitive to VEN in a dose-dependent manner. We found that MSCs significantly reduced MV4-11 apoptosis after VEN exposure. The addiction of transwells to the co-culture system abrogated the MSC-protective effect on VEN-induced apoptosis, suggesting the establishment in our setting of a complex interplay between AML cells and MSCs, not only mediated by soluble factors as already demonstrated but also involving direct cell-contact. Furthermore, when we cultured MV4-11 cells in direct contact with MSCs, we were able to distinguish, among leukemic cells, a cell population that remained floating in the supernatant and a cell population that kept adherent to MSCs and could be collected by trypsin detachment only. Interestingly, MSCs protected the adherent but not the floating cells from VEN-induced apoptosis, suggesting a central role of a close adhesion in the MSC-driven resistance to VEN. To gain insight into this mechanism, we sorted adherent vs. floating cells and analyzed and compared Gene Expression Profiling (GEP). Interestingly, the principal component analysis suggested that the floating cell GEP was more similar to the GEP of MV4-11 cells cultured without MSCs compared to adherent cell GEP. We found 159 differential expressed genes (DEGs) in adherent vs. floating MV4-11 cells (FC| ≥ 2 and P ≤ 0.05, according to Transcriptome Analysis Console software guidelines). Enrichment analysis revealed that DEGs were involved in cell adhesion migration and proliferation pathways. Among DEGs, we found that CD90/Thy-1, an adhesion molecule whose expression correlated with unfavorable karyotypes and shorter survival in AML, was significantly up-regulated in adherent MV4-11 cells. Flow cytometry analysis confirmed that the adherent compared to the floating MV4-11 cells, showed a higher expression of CD90 as well as CXCR4 as expected. Our study demonstrated that MSCs protect AML cells from VEN-induced apoptosis through a complex mechanism involving tight adhesion-mediated pathways. Further clarification of these interactions can shed light on putative targets to counteract MSC-mediated drug resistance to VEN treatment.