P-265: MAF-driven metabolic reprogramming mediates H3K27 hyperacetylation to regulate super enhancer-associated genes

Introduction: Overexpression of transcription factor MAF is found in about 50% of multiple myeloma cases and associated with the prognostically unfavorable t(14;16) translocation subtype. Genetic alterations can modify the epigenome through metabolite availability acting as substrates in histone modifications, but how this translates into specificities in gene regulation is unclear.

Methods: We performed multi-omics analysis with mass spectrometry-based metabolomics, ChIP-sequencing and RNA-sequencing to understand the altered metabolic state and the underlying mechanism. Seahorse and various metabolic assays were used in validation experiments, and functional studies were performed.

Results: Here, we report a novel involvement of MAF in metabolically-driven histone acetylation, including the superenhancer (SE) mark H3K27ac, through altering acetyl-CoA metabolism. To sustain a hyperacetylated chromatin state, MAF acquired the metabolic plasticity to induce high influx of glutamine through upregulating amino acid transporter SLC7A5, feeding metabolites into the tricarboxylic acid (TCA) cycle as acetyl-CoA sources. Systematic loss-of-function studies indicated that metabolic enzymes citrate synthase (CS) and ATP-citrate lyase (ACLY) are central to this process, and blocking citrate export from mitochondrial via CRISPR/Cas9 targeting of SLC25A1 synonymously abolished H3K27ac. Silencing of MAF displayed defective mitochondrial oxidative phosphorylation attributed to reduced metabolic flux through TCA cycle and downregulation of electron transport chain complex I/II expression. ChIP-seq profiling of MAF oncogenic epigenome segregated by promoter- and SE-regulated genes revealed broad H3K27ac signal, H3K4me1 and chromatin accessibility overlapping with MAF-bound regions in cis-regulatory elements. Lastly, we identified novel MAF-regulated common SE genes across t(14;16) subtype by imposing stepwise filtering criteria on our published SE datasets and overlapping with MAF RNA-seq, leading to the prioritization of ZC3H3 for further investigation. Dependency experiments suggested that ZC3H3 is unconditionally required for t(14;16) myeloma cell growth and the robust abrogation of ZC3H3 could be pharmacologically achieved by targeting p300 histone acetyltransferase.

Conclusions: Altogether, we delineated a non-canonical epitranscriptional role of MAF in connection to its altered metabolic state, and suggest metabolic disruptions or epigenetic modifiers as a new direction in t(14;16) myeloma therapy.