Bone Marrow Mesenchymal Stromal Cells Transfer Their Mitochondria to Acute Myeloid Leukaemia Blasts to Support Their Proliferation and Survival

Background

Survival of acute myeloid leukaemia (AML) blasts is established to be heavily dependent on the bone marrow microenvironment, where bone marrow mesenchymal stromal cells (BM-MSCs) are an important cell type. Contrary to the Warburg hypothesis, AML blasts rely on oxidative phosphorylation for survival and have increased mitochondrial levels compared to normal CD34+ progenitors. Current research is being directed at the biology behind how the bone marrow microenvironment supports the proliferation of the disease. With the knowledge that AML blasts have an increased mitochondrial mass and that BM-MSCs have the ability to be mitochondrial donors, we examined the BM-MSC AML blast interaction to determine if the increased mitochondrial mass was a result of inter-cellular mitochondrial transfer.

Methods

Primary AML blasts were obtained from patient bone marrow. Primary AML and normal BM-MSCs were isolated from patients bone marrow, with informed consent and under approval from the UK National Research Ethics Service (LRCEref07/H0310/146), using adherence. BM-MSCs were characterised using flow cytometry for expression of CD90+, CD73+, CD105+ and CD45-. Mitochondrial transfer was assessed in vitro using qPCR and MitoTracker staining based methods. A P⁰ OCI-AML3 cell line was created using a 40-day incubation with ethidium bromide, pyruvate and uridine. In vivo experiments using an NSG primary AML xenograft model were also carried out (in accordance with University of East Anglia ethics review board). For mechanistic determination, BM-MSCs with a mCherry mitochondrial labelled protein were created using a lentivirus. Levels of mitochondrial transfer were assessed by mCherry mitochondrial protein acquisition in the AML during co-culture with the BM-MSCs.

Results

We report that BM-MSCs support AML blast survival via the inter-cellular transfer of mitochondria from 'benign' to malignant cells. To examine this transfer we used primary AML blasts and BM-MSCs derived from patient bone marrow, along with AML cell lines. We found in vitro that primary AML blasts increase their mitochondrial mass, respiratory capacity and ATP production after co-culture with primary BM-MSCs. A P⁰ OCI-AML3 cell line, with mutated mitochondrial DNA (mtDNA), was generated using ethidium bromide treatment allowing mitochondrial transfer to be specifically analysed. mtDNA was restored in this cell line after co-culture with primary BM-MSCs. Further to this mouse mtDNA was detected in the P⁰ OCI-AML3 cells after co-culture with the mouse BM-MSC cell line (M2-10B4). Moreover, mitochondrial transfer was directly observed between primary BM-MSCs and primary AML blasts, visualised by the acquisition of a mCherry labelled mitochondrial protein. This transfer of mitochondria was one directional. Moreover, a reduction of mitochondrial transfer was observed in AML blasts upon the addition of cytochalasin to the co-culture, highlighting that mitochondrial transfer is at least in part facilitated through tunnelling nanotubes (TNTs). Finally, mitochondrial transfer was confirmed in vivo whereby murine mitochondria were transferred to human AML in a mouse xenografts model.

Conclusion

Here we show that the bone marrow microenvironment supports the AML blasts by donating mitochondria, which in turn enhances the oxidative phosphorylation and growth capacity of the blasts. Targeting the microenvironment is predicted to provide novel therapeutic approaches for the treatment of cancer.