ZBTB7A Loss Promotes Synthesis of Lipids and Ketone Bodies in Myeloid Leukemia

Cancer cells reprogram their metabolism at many levels to meet the high demand for ATP and metabolites necessary for rapid tumor growth. We recently showed that inactivation of ZBTB7A, a transcription factor essential for lineage fate decision, causes upregulation of glycolysis culminating in increased energy supply in myeloid leukemia cells (Redondo Monte et al., 2020, Oncogene). ZBTB7A mutations are specifically associated with AML t(8;21), pointing toward cooperation with the RUNX1-RUNX1T1 fusion. However, the underlying mechanism is still not fully understood.

Besides alteration of glycolysis, we found genes related to fatty acid metabolism upregulated in the K562 ZBTB7A knockout (KO) myeloid leukemia cell line. This led us to hypothesize that the loss of ZBTB7A could provide the cells with additional energy via increased fatty acid oxidation. To further investigate the role of ZBTB7A in fatty acid metabolism we performed metabolic flux (Seahorse, Agilent) and metabolic tracing experiments.

In the palmitate oxidation assay, using a setting where the cells were cultivated in a substrate-limited medium overnight prior to the experiment and deprived of glutamine during the assay, the K562 ZBTB7A KO cells did not exhibit an advantage in oxidizing this substrate for ATP production. Moreover, in this context, the ZBTB7A KO cells responded negatively to the inhibition of the CPT1A enzyme by Etomoxir, suggesting that these cells are not dependent on fatty acid oxidation. Interestingly, RNA-seq revealed that in K562 ZBTB7A KO cells several genes from the electron transport chain (ETC) were downregulated, supporting the idea that these cells do not depend on OXPHOS metabolism.

Metabolic tracing of the 13C6 glucose isotope by mass spectrometry confirmed that K562 ZBTB7A KO cells exhibit upregulated glycolysis and pentose phosphate pathways, as previously reported (Redondo Monte et al., 2020, Oncogene; Liu et al., 2014, Genes Dev.). Conversely, tricarboxylic acid cycle (TCA) metabolites were not different between control and knockout cells. We observed that upregulation of glycolysis resulted in high amounts of acetyl-CoA, which is not entering the OXPHOS metabolism but is used as a precursor for palmitate synthesis instead (Figure 1A). Of note, palmitate is required for the production of several complex lipids in cell membranes, which are highly demanded in cancer cells. Accordingly, the transcript level of FASN, the enzyme that catalyses palmitate synthesis, is upregulated in core binding factor (CBF) AML patients with mutation in ZBTB7A (Figure 1B). Remarkably, the K562 ZBTB7A KO cells preferentially produce acetoacetate (ketone bodies) from the additional acetyl-CoA and secrete it to the medium (Figure 1A). The genes HMGCS1 and ACAT1, encoding enzymes that enable ketone bodies reutilization in lipid synthesis, are downregulated upon ZBTB7A loss, indicating that these cells do not use the ketone bodies. Since ketone bodies not only serve as a substrate for ATP production but also act as signaling mediators, we hypothesize that ketone bodies might fuel the bone marrow microenvironment to support leukemia cells expansion.

In summary, we have advanced the understanding of ZBTB7A loss in leukemia metabolism, showing increased lipid and ketone bodies synthesis at the metabolite level. Lipid synthesis enzymes, such as FASN, FADS1, and FADS2 are attractive drug targets for therapeutic applications, especially in combination with standard therapies and glycolysis inhibitors.