Understanding the mechanism of fibrillization of hnRNPA1 from phase-separated condensates.

Abstract

Stress granules are membrane-less cytosolic organelles that form through phase separation of RNA-binding proteins and translationally stalled mRNAs. Persistent stress granules have been implicated in the pathogenesis of neurodegenerative disorders such as amyotrophic lateral sclerosis and frontotemporal dementia, in which certain stress granule proteins form the hallmark fibrillar inclusions in affected cells. Heterogeneous nuclear ribonucleoprotein 1 (hnRNPA1) is one of these stress granule proteins, whose C-terminal low-complexity domain (LCD) is sufficient for mediating both phase separation and fibril formation. Disease mutations in the LCD enhance its fibril formation in vitro, and phase separation further enhances fibrillization, but the underlying mechanism linking the two processes is still poorly understood. Proposed mechanisms include enhanced nucleation from the dense phase and further enhancement by disease mutants if they have a higher driving force for phase separation. Here we find that disease mutants of hnRNPA1-LCD that enhance fibrillization show reduced driving force for phase-separation. A combination of bulk fluorescence and microscopy-based assays reveal that fibrillization is nucleated at the interface of condensates. Subsequent fibril growth, in contrast, is slowed down in the interior of WT hnRNPA1-LCD condensates, but not appreciably so in case of disease mutants. Our data supports a competition model between sticker-sticker interactions that drive phase separation and zipper-zipper interactions that drive fibril formation inside the dense phase environment. Stickers are uniformly distributed over the sequence and form frequent crosslinks, decreasing the probability of zippers to productively interact with each other, and thus weakening fibrillization. In contrast, the stronger zippers in disease mutants can effectively outcompete sticker interactions and enhance fibrillization. Our results suggest a potential protective role for phase separation in preventing fibril formation, whereas disease mutations turn stress granules into crucibles for fibrillization.