The phenylalanine-and-glycine repeats of NUP98 oncofusions form condensates that selectively partition transcriptional coactivators

Liquid-liquid phase separation in normal hematopoiesis and hematological diseases

Liquid-liquid phase separation (LLPS) is an emerging research field in cellular biology. LLPS-driven biomolecular condensates act as reaction chambers and regulatory hubs for critical processes, including chromatin architecture, gene expression, and metabolism. The dysregulation of these processes frequently impedes the proper execution of physiological functions. Current research indicates that abnormal phase separation plays a significant role in the pathogenesis of diseases and aging. This review briefly overviews the fundamental concepts and research methods related to phase separation. We also summarize studies concerning its physiological functions, particularly emphasizing its role in hematopoiesis. We further discuss how abnormal phase separation can lead to hematological disorders, specifically summarizing its involvement in the pathogenesis of leukemia. Despite recent advancements, elucidating LLPS mechanisms in hematopoiesis remains challenging due to the intricate interplay between biomolecular condensates and cellular function. Future research efforts aiming to reveal the role of LLPS in hematological diseases hold promise for novel therapeutic interventions and a deeper understanding of hematopoietic processes.

The sotos syndrome gene Nsd1 safeguards developmental gene enhancers poised for transcription by maintaining the precise deposition of histone methylation

Germline haploinsufficiency of NSD1 is implicated as the etiology of Sotos syndrome; however, the underlying mechanism remains far from being clear. Here, we use mouse embryonic stem cell (mESC) differentiation as a model system to address this question. We found Nsd1 to be indispensable for the faithful differentiation of mESCs into three primary germ layers, particularly, meso-endodermal cell lineages related to the development of the heart and the skeletal system. Time-course transcriptomic profiling following the mESC differentiation revealed that Nsd1 not only facilitates the basal expression but also permits the differentiation-accompanied rapid induction of a suite of meso-endoderm lineage-specifying transcription factor genes such as T and Gata4. Mechanistically, Nsd1 directly occupies putative distal enhancers of the lineage transcription factor genes under the pluripotent cell state, where it deposits H3K36me2 to antagonize the excessive H3K27me3 and maintains the basal H3K27ac level, thereby safeguarding these gene enhancers at a primed state that responds readily to differentiation cues. In agreement, gene rescue assays using the Nsd1 KO mESCs showed that the H3K36me2 catalysis by Nsd1 requires several functional modules within Nsd1 (namely, PHD1-4, PWWP2, and SET) to a similar degree. Disruption of either one of these Nsd1 modules severely abrogated H3K36me2 in mESCs and significantly impaired appropriate induction of developmental genes upon mESC differentiation. Altogether, our study provides novel molecular insight into how the NSD1 perturbation derails normal development and causes the disease.

RNA polymerase II partitioning is a shared feature of diverse oncofusion condensates

Condensates regulate transcription by selectively compartmentalizing biomolecules, yet the rules of specificity and their relationship to function remain enigmatic. To identify rules linked to function, we leverage the genetic selection bias of condensate-promoting oncofusions. Focusing on the three most frequent oncofusions driving translocation renal cell carcinoma, we find that they promote the formation of condensates that activate transcription by gain-of-function RNA polymerase II partitioning through a shared signature of elevated π and π-interacting residues and depletion of aliphatic residues. This signature is shared among a broad set of DNA-binding oncofusions associated with diverse cancers. We find that this signature is necessary and sufficient for RNA polymerase II partitioning, gene activation, and cancer cell phenotypes. Our results reveal that dysregulated condensate specificity is a shared molecular mechanism of diverse oncofusions, highlighting the functional role of condensate composition and the power of disease genetics in investigating relationships between condensate specificity and function.

Deregulated enhancer-promoter communication in cancer through altered nuclear architecture

Enhancers are critical regulators of gene expression. Structural variations in cancer genomes can lead to enhancer hijacking, where oncogenes are activated by mistargeted enhancer activity. Novel enhancer-promoter interactions may also arise through chromosomal rearrangements that create extrachromosomal DNA elements. Additionally, fusion proteins and other mutation-induced alterations in protein properties can lead to the aberrant assembly of proteins into large complexes on the size scale of 0.1-1 μm termed onco-condensates. Transcription factors and co-activators accumulate with cis-regulatory elements in these structures, driving oncogenic programs. Here, we review current evidence of how altered genome architecture and macromolecular assembly result in deregulated enhancer-promoter communication. We discuss emerging strategies to exploit these mechanisms for clinical applications.