Back

Visualizing TERRA RNA G-quadruplex unfolding in FUS biomolecular condensates

Zheng, T.; Fawzi, N. L.

2025-10-29 biochemistry
10.1101/2025.10.29.685336 bioRxiv
Show abstract

RNA G-quadruplexes (rG4s) are remarkably stable secondary structures with critical regulatory roles in gene expression, RNA metabolism, and telomere maintenance. However, their behavior within cells remains controversial, partly due to challenges in detecting rG4s in complex environments. Here, we use solution NMR spectroscopy to investigate how condensates formed by the low-complexity and RGG domains of the RNA-binding protein FUS affect the structure of TERRA, a highly stable model rG4 RNA. We show that FUS LC-RGG1 interacts with TERRA in dilute solution and that binding perturbs, but does not disrupt, the G-quadruplex structure. When co-phase separated with FUS LC-RGG1, however, NMR signatures of TERRAs folded state disappear, and the remaining observable resonances indicate an unfolded conformation, even in buffer containing potassium where TERRA rG4 is exceptionally stable when outside a condensate. Quantitative comparisons with a mutant form of TERRA, used as a baseline for fully unfolded RNA, suggest that at minimum a third of TERRA RNA becomes unfolded in the condensed phase. Thus, our results demonstrate that condensates can shift the structural ensemble of rG4 towards unfolded species, offering a potential mechanistic explanation for their apparent lack of stability in vivo and revealing how phase-separated environments may actively modulate RNA structure and function. HighlightsO_LINMR spectroscopy directly probes RNA structure inside biomolecular condensate. C_LIO_LIFUS binding in solution perturbs TERRA RNA G-quadruplex. C_LIO_LIA sizable fraction of TERRA is unfolded in FUS condensates. C_LI

Published in Journal of Molecular Biology (predicted rank #12) · training set

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.