ACSL4-Associated Lipid Metabolism Is a Distinct Therapeutic Vulnerability in KMT2A-Rearranged Acute Myeloid Leukemia

Background:

Around 50% of acute myeloid leukemia (AML) patients harbor chromosomal aberrations, such as inversions or translocations. One of the most common chromosomal rearrangements in AML involves the lysine methyltransferase 2A (KMT2A) gene. As KMT2A-rearranged (KMT2Ar) AML is a complex and heterogeneous disease, personalized treatment options and novel therapeutic targets are needed.

Deregulation of lipid metabolism is associated with initiation, maintenance, and treatment response of AML, yet vulnerabilities within the lipidome remain largely undefined. Numerous lipid metabolic enzymes have been identified over the past years as potential targets for treating AML. The enzyme Acyl-CoA synthetase long-chain family member 4 (ACSL4), an activator of polyunsaturated fatty acids, has been reported as both a tumor promoter as well as suppressor in different cancer entities. The role of ACSL4 in AML remains to be studied and could ultimately provide evidence on how to utilize lipid metabolism-associated factors for therapeutic interventions in AML.

Aims:

This study aims to functionally and mechanistically investigate the role of ACSL4 in AML and to explore ACSL4 as a potential novel target for treating AML.

Methods:

We conducted a comprehensive gene-dependency analysis using various publicly available AML CRISPR-Cas9 screening datasets, followed by CRISPR interference (CRISPRi)- and shRNA-mediated knockdown experiments in different AML cell lines. To functionally characterize the identified hit ACSL4, we performed flow cytometry-based apoptosis and cell cycle analyses. Additionally, using shRNAs targeting Acsl4, we carried out colony-forming unit (CFU) assays and in vivo knockdown experiments using an MLL-AF9 mouse model. To understand how ACSL4 contributes to AML pathogenesis, we performed a multiomics approach, including transcriptomics, proteomics, and lipidomics, comparing ACSL4 knockdown versus control cells. Finally, we analyzed patient data from the publicly available TCGA and OHSU cohorts to assess the clinical impact of ACSL4-related changes on patient outcome.

Results:

A comprehensive analysis of various publicly available genome-wide CRISPR-Cas9 datasets revealed an increased dependency of AML cell lines with KMT2A-rearrangements on the fatty acid regulator ACSL4. To validate this KMT2Ar-specific effect, we performed CRISPRi- and shRNA-mediated knockdown experiments of ACSL4 in both KMT2Ar and KMT2A WT AML cell lines, observing decreased proliferation upon ACSL4 knockdown exclusively in KMT2Ar cell lines. Functional consequences of ACSL4 disruption were studied using apoptosis and cell cycle analyses, which revealed an increase in the G0/G1 phase of the cell cycle in KMT2Ar AML, but not in KMT2A WT AML. The knockdown of Acsl4 in a KMT2Ar mouse model resulted in reduced colony-forming capacity ex vivo and significantly delayed leukemia onset in vivo. To understand the molecular differences between ACSL4-dependent and non-dependent AML cells, we performed a multiomics approach, including transcriptomics, proteomics, and lipidomics, comparing the ACSL4 knockdown between ACSL4-dependent and independent AMLs. Integration of the transcriptomics and proteomics data revealed an upregulation of lipid metabolism pathways exclusively in ACSL4-dependent cell lines. From these data, we generated a 12 gene-containing KMT2Ar-ACSL4 dependency signature (KRADS12) and confirmed its association with KMT2Ar and survival in AML patients. Collectively, our findings highlight ACSL4 as a novel vulnerability specifically in KMT2Ar AML.

Summary/Conclusion:

In this study, we present the lipid metabolic enzyme ACSL4 as a selective target in the aggressive KMT2Ar AML subtype. The knockdown of Acsl4 significantly delayed disease onset in a KMT2Ar mouse model. Using a multiomics approach, we defined a signature that can predict AML patient outcome. Overall, ACSL4 could serve as a potential therapeutic target for treating the aggressive KMT2Ar AML subtype.