Super-enhancer landscape reveals leukemia stem cell reliance on X-box binding protein 1 as a therapeutic vulnerability

• Published in: Science translational medicine Vol. 13; no. 612; p. eabh3462
• Main Authors: Zhou, Jingfeng, Wang, Shubo, Nie, Danian, Lai, Peilong, Li, Yiqing, Li, Yangqiu, Jin, Yanli, Pan, Jingxuan
• Format: Journal Article
• Language: English
• Published: United States 22.09.2021
• Subjects: Animals; Enhancer Elements, Genetic; Humans; Leukemia, Myeloid, Acute; Mice; Protein Binding; Stem Cells; X-Box Binding Protein 1 - genetics
• ISSN: 1946-6242
• DOI: 10.1126/scitranslmed.abh3462

Relapse of patients with chronic myelogenous leukemia (CML) may occur at least partially because leukemia stem cells (LSCs) lack sensitivity to tyrosine kinase inhibitors (TKIs) such as imatinib. The precise regulation of LSC stemness is incompletely understood. Given that traits of LSCs are subject to epigenetic regulation, we hypothesized that LSCs might be dependent on continuous active transcription of genes associated with super-enhancers (SEs), which might, in turn, suggest an opportunity for intervention. In this study, we tested this hypothesis and delineated the SE landscape in LSCs from patients with CML. Disruption of the SE-associated gene transcription by THZ1, a covalent cyclin-dependent kinase 7 (CDK7) inhibitor, efficiently eradicated LSCs in retroviral BCR-ABL–driven CML mice while sparing normal hematopoietic stem cells. Furthermore, we found that X-box binding protein 1 ( ), a substrate of mRNA-splicing endonuclease IRE1α in the unfolded protein response pathway, was an SE-associated oncogene in LSCs. Knockdown of reduced survival and self-renewal capacity in primary CML CD34 cells and eradicated LSCs in CML mice. Selectively blocking generation of the spliced form of Xbp1 by hematopoietic cell–specific conditional knockout suppressed the progression of CML and impaired the leukemogenesis of LSCs in CML mice. Overall, we identified an epigenetic transcriptional program in LSCs, adding to evidence for the theory of “oncogene addiction” and suggesting a potential targeting strategy for CML.