This article has been reviewed according to Science X's and . have highlighted the following attributes while ensuring the content's credibility:
fact-checked
peer-reviewed publication
trusted source
proofread
Phase separation by fusion protein FC reveals new mechanism for tumor development in rare cancer

Dynamic remodeling of the endomembrane system is a core biological process essential for maintaining cellular compartmentalization and homeostasis. Proteins enriched in intrinsically disordered regions (IDRs) can drive membrane curvature formation in artificial vesicles through liquid-liquid phase separation (LLPS); however, direct evidence for LLPS-mediated endomembrane remodeling in living cells has been lacking.
In Prof. Li Dong's group from the Institute of Biophysics of the Chinese Academy of Sciences, together with researchers from Tsinghua University, provide the first direct evidence that transmembrane fusion proteins (MFPs) enriched in IDRs can dynamically remodel endomembrane systems via LLPS in living cells using Multi-SIM super-resolution imaging technology.
Utilizing the characteristic fusion protein FUS-CREB3L2 (FC) derived from low-grade fibromyxoid sarcoma (LGFMS) as a model, the researchers systematically elucidated the molecular mechanism by which aberrant phase separation drives membrane remodeling and promotes tumorigenesis, offering a new perspective for understanding the biological functions of interactions between LLPS and endomembrane systems.
The FC protein retains both the IDRs from FUS and the transmembrane domain and DNA-binding domain from CREB3L2.
By constructing an inducible FC expression system, the researchers utilized Multi-SIM imaging to capture over 2,300 time points across six continuous hours, recording for the first time the dynamic process by which FC remodels the endoplasmic reticulum (ER) through LLPS.

The researchers discovered that the remodeled ER membrane structures co-localized with COPII vesicle markers, forming COPII-like compartments with diameters significantly larger than those of classic COPII vesicles.
These compartments sequestered the proteases S1P/S2P, triggering spontaneous intramembrane proteolysis of FC, thereby releasing its transcriptionally active N-terminal fragment (FC-N) into the nucleus.
This mechanism sharply contrasts with the behavior of wild-type CREB3L2, which localizes to the ER without changing its morphology and requires induction by the ER stressor brefeldin A (BFA) to undergo proteolysis at the Golgi apparatus.
Once inside the nucleus, FC-N was found to recruit chromatin regulators SSRP1 and CHD7, forming a transcriptional complex that activates a gene expression program characteristic of LGFMS, ultimately promoting malignant cellular behavior.

This study marks the first identification of a regulatory pathway in which aberrant phase separation of a fusion protein drives ER membrane remodeling and nuclear signaling.
The findings not only deepen the mechanistic understanding of LLPS in cancer biology but also lay a theoretical foundation for developing targeted therapies for LGFMS.
More information: Xinyu Wang et al, Aberrant phase separation drives membranous organelle remodeling and tumorigenesis, Molecular Cell (2025).
Journal information: Molecular Cell
Provided by Chinese Academy of Sciences