Abstract
We and others have reported that activation of the AMP-activated protein kinase (AMPK) mediates cell death in ALL. AMPK has been reported to interact with chromatin-associated proteins to epigenetically regulate gene expression in response to environmental/energy/metabolic stress, however, little is known about AMPK's epigenetic targets. To identify genome-wide genes regulated by direct association of AMPK to chromatin in response to energy/metabolic stress, we performed ChIP-seq with AMPKα2 and RNA pol II antibodies in Bp-ALL (KASUMI-2) and T-ALL (KE-37) cells treated with or without glucose or AICAR, a well known AMPK activator. ChIP-seq identified 132 and 167 significant AMPKα2 peaks located at gene promoters in AICAR-treated KASUMI-2 and KE-37 cells, respectively. GO terms analysis in KASUMI-2 control cells showed that AMPKα2 target set was enriched for genes involved in regulation of hematopoietic cell differentiation, DNA metabolism, DNA topology, and histone H3-K4 methylation, whereas in AICAR-treated cells it included the histone H3-K4 methylation, protein localization, Notch signaling, and mRNA destabilization. In AICAR-treated cells, we found that AMPKα2 was recruited to regions that were enriched for TAF1 and MYC targets, and to the promoter of the epigenetic regulators KMT2A and SETD1A. Among AMPKα2 targets, we identified a cluster of histone genes in which recruitment of AMPKα2 was significantly decreased at gene promoters. ChIP-seq data were validated by ChIP-qPCR on histone genes H1-2, H1-3, and H4C4, and the occupancy of AMPKα2 on histone gene loci in other Bp- and T-ALL cells confirmed. To correlate the level of gene mRNA expression and recruitment of AMPKα2 to chromatin we performed RNA-seq in KASUMI-2 and KE-37 cells treated ± AICAR for 24h. RNA-seq identified 1342 differential expressed genes in at least one of the cell lines, of which 57 were histone genes and 1285 non-histone genes. RT-qPCR confirmed histone gene mRNA downregulation in AICAR-treated cells. In addition, consistent with an epigenetic role for AMPK in ALL cells in response to energy/metabolic stress, we found that expression of the H3-K4 histone methyltransferase KMT2A and SETD1A genes were upregulated in KASUMI-2 cells treated with AICAR. To further investigate the effect of AMPK on histone gene expression, we generated HEK293 cells expressing single knockout (KO) or double knockout (DKO) of AMPKα1 and AMPKα2 genes using CRISPR/Cas9, and determined the level of selected histone gene mRNA expression using RT-qPCR. Our data show that both AMPKα1 and AMPKα2 participate in regulating histone gene expression, and these findings were confirmed using MEF WT, AMPKα1/KO, AMPKα2/KO, and AMPKα1/α2/DKO cells. These data support our previous findings indicating that AMPK kinase activity was required for histone genes’ regulation in response to metabolic stress. Next, we examined the cluster of non-histone genes altered by AICAR, and found that downregulated genes were enriched in MYC, Myc-associated factor X, and transcription initiation factor TFIID subunit 1 (TAF1) targets, and involved in pathways such as gene expression and protein metabolism. In contrast, the upregulated genes were enriched in RUNX1, SPI1, and GATA1 targets, and involved in pathways related to immune system signaling, regulation by TP53, and apoptosis. Using HEK293 CRISPR AMPKα1/KO, AMPKα2/KO and AMPKα1/α2/DKO cells, we demonstrated that AMPK is also required for MYC expression. In conclusion, our data show that (i) in response to metabolic stress, AMPKα2 is recruited to or released from a chromatin-associated complex to regulate gene transcription in ALL, (ii) AMPKα2 targets include the histone, MYC-target, and the epigenetic regulator KMT2A and SETD1A genes (H3-K4 histone methyltransferases), and (iii) AICAR treatment represses histone gene transcription via an AMPK-dependent mechanism. Further elucidation of AMPK's interactions with members of a putative AMPK/chromatin-associated transcription complex may lead to unique opportunities for epigenetic-based therapeutic interventions to exploit synthetic lethality in relapse/refractory ALL and other hematological malignancies.
Disclosures
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
