Full text
2,090 characters
· extracted from
oa-doi-fallback
· click to expand
ABSTRACT
Aberrant chromatin-associated condensates have emerged as drivers of transcriptional dysregulation in cancer. Although extensive studies have elucidated intrinsic protein sequence features governing their formation, how extrinsic factors within the chromatin environment modulate their assembly and pathogenic function remains poorly understood. Gain-of-function mutations in the histone acetylation reader ENL, found in pediatric leukemia and Wilms tumor, drive oncogenesis by inducing condensate formation at highly selective genomic loci. Here, we uncover a critical role for locally produced gene transcripts in reinforcing the nucleation, chromatin engagement, and oncogenic activity of ENL mutant condensates. Mutant ENL binds RNA in part through a conserved basic patch within its YEATS domain, and this interaction enhances condensate formation both in vitro and across diverse cellular contexts. Using a chemically inducible condensate displacement and re-nucleation system, we show that blocking ENL–RNA interactions or transcription impairs condensate reformation at endogenous targets. RNA interactions preferentially enhance mutant ENL occupancy at top-bound, condensate-permissive loci, leading to increased transcriptional bursting and robust gene activation at the single-cell and single-allele level. In mouse models, disrupting ENL–RNA interactions reduces condensate formation and oncogenic transcription in hematopoietic stem and progenitor cells, thereby suppressing ENL mutant-driven leukemogenesis. Together, these findings demonstrate that locally produced RNA transcripts can promote locus-specific nucleation of pathogenic condensates on chromatin, which in turn drive persistent and hyperactive transcription of oncogenic targets and lead to tumorigenesis.
Competing Interest Statement
L.W. is a co-inventor on a patent filed (US No. 62/949,160) related to the inhibitor used in this manuscript. All other authors declare no competing interests.
Footnotes
↵13 Lead contact
Figures and text revised to include additional data. Author affiliations updated.
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.