Treatment of chronic myeloid leukemia (CML) with imatinib mesylate and other second/third generation c-Abl specific tyrosine kinase inhibitors (TKIs) has significantly extended patient survival1. However, TKIs primarily target differentiated cells and do not eliminate leukemic stem cells (LSCs)2–4. Therefore, targeting minimal residual disease, to prevent acquired resistance and/or disease relapse requires identification of novel LSC-selective target(s) that can be exploited therapeutically5,6. Given that malignant transformation involves cellular metabolic changes, which may in turn render the transformed cells susceptible to specific assaults in a selective manner7, we searched for such vulnerabilities in CML LSCs. We performed metabolic analyses on both stem cell-enriched (CD34+ and CD34+CD38-) and differentiated (CD34-) patient derived CML cells, and compared their signature with that of normal counterparts. Combining stable isotope-assisted metabolomics with functional assays, we demonstrate that primitive CML cells rely on upregulated oxidative metabolism for their survival. We also show that combination-treatment of imatinib with tigecycline, an antibiotic that inhibits mitochondrial protein translation, selectively eradicates CML LSCs, both in vitro and in a xenotransplantation model of human CML. Our findings provide a strong indication for investigating the employment of TKIs in combination with tigecycline to treat CML patients with minimal residual disease.

In chronic myeloid leukemia (CML), tyrosine kinase inhibitor (TKI) treatment induces autophagy that promotes survival and TKI-resistance in leukemic stem cells (LSCs). In clinical studies hydroxychloroquine (HCQ), the only clinically approved autophagy inhibitor, does not consistently inhibit autophagy in cancer patients, so more potent autophagy inhibitors are needed. We generated a murine model of CML in which autophagic flux can be measured in bone marrow-located LSCs. In parallel, we use cell division tracing, phenotyping of primary CML cells, and a robust xenotransplantation model of human CML, to investigate the effect of Lys05, a highly potent lysosomotropic agent, and PIK-III, a selective inhibitor of VPS34, on the survival and function of LSCs. We demonstrate that long-term haematopoietic stem cells (LT-HSCs: LinSca-1c-kitCD48CD150) isolated from leukemic mice have higher basal autophagy levels compared with non-leukemic LT-HSCs and more mature leukemic cells. Additionally, we present that while HCQ is ineffective, Lys05-mediated autophagy inhibition reduces LSCs quiescence and drives myeloid cell expansion. Furthermore, Lys05 and PIK-III reduced the number of primary CML LSCs and target xenografted LSCs when used in combination with TKI treatment, providing a strong rationale for clinical use of second generation autophagy inhibitors as a novel treatment for CML patients with LSC persistence.

A major drawback of tyrosine kinase inhibitor (TKI) treatment in chronic myeloid leukemia (CML) is that primitive CML cells are able to survive TKI-mediated BCR-ABL inhibition, leading to disease persistence in patients. Investigation of strategies aiming to inhibit alternative survival pathways in CML is therefore critical. We have previously shown that a nonspecific pharmacological inhibition of autophagy potentiates TKI-induced death in Philadelphia chromosome-positive cells. Here we provide further understanding of how specific and pharmacological autophagy inhibition affects nonmitochondrial and mitochondrial energy metabolism and reactive oxygen species (ROS)-mediated differentiation of CML cells and highlight ATG7 (a critical component of the LC3 conjugation system) as a potential specific therapeutic target. By combining extra- and intracellular steady state metabolite measurements by liquid chromatography-mass spectrometry with metabolic flux assays using labeled glucose and functional assays, we demonstrate that knockdown of ATG7 results in decreased glycolysis and increased flux of labeled carbons through the mitochondrial tricarboxylic acid cycle. This leads to increased oxidative phosphorylation and mitochondrial ROS accumulation. Furthermore, following ROS accumulation, CML cells, including primary CML CD34+ progenitor cells, differentiate toward the erythroid lineage. Finally, ATG7 knockdown sensitizes CML progenitor cells to TKI-induced death, without affecting survival of normal cells, suggesting that specific inhibitors of ATG7 in combination with TKI would provide a novel therapeutic approach for CML patients exhibiting persistent disease.

We and others have shown that tyrosine kinase inhibitors (TKIs), such as imatinib, fail to eliminate primitive chronic myeloid leukaemia (CML) stem cells (LSCs), suggesting that combination of TKIs with other targeted agents will be required to eradicate the LSC-pool. Therefore, identification of targetable pathways that selectively maintain CML LSC survival is critical. Metabolic reprogramming is a core feature of cancer cells making them susceptible to manipulation in a selective manner. Indeed, in recent years numerous studies have shown that targeting abnormal aspects of metabolism can be of therapeutic value. The aim of this study was to identify and target metabolic dependencies in CML LSCs using stem cell-enriched (CD34+) primary cells isolated from CML patients and healthy donors. Although CML represents a highly suitable model for cancer stem cell studies, investigation of LSCs metabolism has so far been restricted by technical limitations. Therefore, we have applied improved protocols for metabolomics using liquid chromatography-mass spectrometry (LC-MS) and functional assays. Initially we cultured leukaemic cells in the presence of uniformly-labelled glucose with stable (heavy) 13C isotope (13C6-glucose) and compared isotopic enrichment in CD34+ versus CD34- cells isolated from the same CML patient. Our results showed that CD34+ cells contained an increased proportion of isotopologues with 2 or more labelled carbons in tricarboxylic acid (TCA) cycle metabolites (such as citrate, α-ketoglutarate and malate) when compared with CD34- cells, indicating an increase in flux of glucose through the TCA cycle in more primitive CML cells. In contrast, CML CD34+ cells contained reduced levels of glucose-derived lactate when compared with patient-matched CD34- cells, suggesting that more primitive CML cells utilise mitochondrial metabolism rather than glycolysis to supply their energy demands. In line with this, CML CD34+ cells displayed more than a two-fold increase in their mitochondrial oxygen consumption rates (OCR) when compared with CD34- cells (p≤0.05), confirming that mitochondrial metabolism is enhanced in stem cell-enriched CML cells. Next we traced 13C6-glucose in CD34+ cells from healthy donors and compared this with isotopic enrichment in CML CD34+ cells. This revealed that 13C enrichment in TCA cycle metabolites was significantly higher in CML CD34+ cells when compared with their normal counterparts. This correlated with a significant increase in mitochondrial respiration (p≤0.001) and mitochondrial membrane potential in primitive CML cells, including CD34+38- cells (p≤0.001), suggesting that CML LSCs may be selectively sensitive to inhibition of mitochondrial metabolism. Of clinical significance, we show that the antibiotic tigecycline, an inhibitor of mitochondrial translation, reduced this aberrant oxidative metabolism and selectively induced death in primitive CML cells at a clinically achievable concentration. More precisely, in vitro treatment with tigecycline as a single agent, decreased the number of CML CD34+ cell-derived colonies in comparison with untreated conditions (p≤0.001), and combining tigecycline with imatinib resulted in a 50% decrease in colony number when compared to tigecycline or imatinib alone (p≤0.01). Importantly, this drug combination had no effect on colony formation potential of CD34+ cells derived from healthy donors. Moreover, we show that tigecycline alone, or in combination with imatinib, reduced the number of colonies in a long-term culture-initiating cell assay (p≤0.001) while imatinib, as a single agent, did not have any significant effect. To examine the effect on transplantable CML LSCs in vivo, human CML CD34+ cells were injected into irradiated NSG mice. Following confirmation of engraftment mice were treated with imatinib, tigecycline alone and in combination with imatinib. Remarkably, 4-week combination treatment with tigecycline and imatinib led to near complete elimination of CML LSCs measured by the number of human CD45+ and CD34+38- cells in the bone marrow. We conclude that CML LSC are dependent on oxidative phosphorylation for their survival and tigecycline-mediated inhibition of mitochondrial metabolism, combined with TKI treatment, shows potential as a novel therapeutic strategy to selectively target these cells to enhance cure rates. Disclosures Holyoake: Bristol Myers Squib: Honoraria, Research Funding; Novartis: Honoraria, Research Funding.