Fatty Acid Desaturation Is a Metabolic Vulnerability of Acute Myeloid Leukemia Cells

Acute Myeloid Leukemia (AML) cells - like many other cancer cells - exploit lipid metabolic pathways to support their high demands for energy and biomass, while maintaining an immature phenotype. Though lipid synthesis and catabolism may be equally important to sustain AML cell survival, much of the focus in the field has been on the oxidation of fatty acids (FAs).

To help close this gap, we queried publicly available data to identify FA biosynthetic enzymes that may be essential for AML cells. We found that the expression of Fatty Acid Desaturase 1 (FADS1) - a key regulator of highly unsaturated fatty acid (HUFA) biosynthesis - significantly correlates with poor outcomes in AML and is enriched in subtypes associated with chemotherapy resistance and intermediate-to-poor prognoses.

Through an shRNA-mediated loss-of-function approach, we have demonstrated that FADS1 downregulation slows leukemia cell proliferation in vitro and impedes disease propagation in vivo. Interestingly, pharmacological inhibition of FADS1 selectively impairs AML cell colony formation while sparing normal hematopoietic stem and progenitor cells. FADS1-deficient cells display increased expression of mature myeloid markers, distinct morphology, and superior phagocytosis capabilities. These results suggest that disruption of HUFA biosynthesis pushes AML cells towards differentiation, overcoming a block that is a hallmark of malignant hematopoiesis.

Next, we employed mass spectrometry to examine complex lipids and found that FADS1 knockdown (FADS1 KD) cells accumulate lipids containing very long-chain fatty acids (VLCFAs - i.e., ≥ 22 carbons). These FA species are mostly saturated and monounsaturated and such accumulation is found across several lipid classes such as diglycerides, triglycerides, and phospholipids. To determine the source of VLCFAs upon FADS1 KD, we measured the expression of several genes involved in lipid synthesis and catabolism but found no change that could explain our observations. Strikingly, the mRNA and protein expression of CD36, a scavenger receptor that mediates FA import, is increased 4-fold in FADS1-deficient cells. We speculate that AML cells become heavily reliant on FA import after FADS1 KD due to an inability to assemble building blocks used for proliferation. The greater influx of FA then directly boosts the activity of elongase enzymes by increasing substrate availability.

Lipid remodeling in FADS1-deficient cells goes beyond changes in FA tails affecting the overall levels of membrane structural lipids. Namely, the levels of sphingomyelin decrease at the expense of ceramides, while the proportions of phosphatidylcholine and phosphatidylinositol increase relative to total lipid pools. Because membrane lipid composition can influence cellular function, we performed an RNA-Seq analysis to identify molecular pathways that may be responsible for the anti-leukemia effects of FADS1 downmodulation.

This analysis revealed that FADS1 KD dramatically upregulates pathways related to plasma membrane function such as GTPase and integrin binding, and membrane microdomain. Interestingly, FADS1-deficient cells activate type I interferon signaling cascades. This upregulation of inflammatory pathways seems to be caused by increased toll-like receptor and STING signaling, which are non-redundant. Our data corroborate previous findings showing that TLR stimulation promotes differentiation and growth inhibition in AML cells.

Taken together, our data supports the hypothesis that FADS1 reinforces the differentiation blockade and molecular pathogenesis of AML, and that targeting the biosynthesis of highly unsaturated fatty acids represents an exploitable vulnerability in this aggressive type of blood cancer.

No relevant conflicts of interest to declare.