Search bioRxiv⌕ Search

Biology subjects

Vasiljevic, J.

Publications and source records attributed to Vasiljevic, J..

5 recordsLinked to original sources

Droplet-assisted folding of long regulatory RNAs

Long regulatory RNA regions orchestrate complex cellular processes, including gene expression and epigenetic modifications. How these RNAs dynamically fold and refold in response to cellular signals remains poorly understood. Given that RNAs interact with ubiquitous RNA-binding proteins (RBPs) prone to form biomolecular condensates, we explore how protein droplets interacting along an RNA impact its folding process. Attached droplets prevent premature folding by competing with RNA:RNA interactions. When droplets dissolve due to cellular signals, capillary effects cause the RNA to collapse while refolding. We test this process of condensate-guided RNA folding by adapting established RNA secondary structure predictors to mimic various folding pathways and supplement this with coarse-grained simulations. We find that interactions with transient droplets robustly leads to the formation of long-range RNA contacts, which are otherwise hard to achieve. Our results compare favorably with available experimental data. We propose that this strategy, which we call droplet-assisted RNA folding, represents a previously unexplored mechanism for shaping RNA structures. Given the widespread propensity of RBPs to form condensates, this process could play a fundamental role in the structural organization, conditional reshaping, and functional regulation of long regulatory RNAs. SignificanceComplex non-coding RNA regions (e.g. lncRNAs, 3UTRs, etc.) perform several important functions in higher organisms. However, the longer an RNA, the more likely it is to misfold. How can such regulatory regions of more than 1000 nts be reliably folded, considering that premature co-transcriptional folding favors local interactions? We present a mechanism by which biomolecular condensates of RNA-binding proteins (RBPs) assist and control the folding of complex RNAs. Here, droplets of RBPs nucleate around RNA binding motifs, prevent premature misfolding through their RNA chaperone function, and allow robust folding of initially distant regions through coordinated droplet dissolution and detachment. This mechanism provides an alternative perspective on how long RNAs acquire their structures in a context that is functionally dynamic.

molecular biology↗

Post-transcriptional regulation of insulin mRNA storage by G3BP1/2+ condensates in beta cells

Hyperglycemia upregulates insulin translation in pancreatic beta cells. Several RNA- binding proteins involved in this process have been identified, including G3BP1, a stress granule marker downregulated in islets of subjects with type 2 diabetes. We show that in mouse insulinoma MIN6-K8 cells exposed to fasting glucose levels G3BP1 and its paralog G3BP2 colocalize to cytosolic condensates with eIF3b and Ins1/2 mRNA. Upon glucose stimulation, the condensates dissolve and G3BP1/2, eIF3b, and insulin mRNAs redistribute throughout the cytosol. Intriguingly, G3BP1+ condensates in MIN6-K8 cells differ from sodium arsenate-induced stress granules in regards to eIF2 and AMPK phosphorylation. Knockout of G3BP1 or G3BP2 prevented condensate assembly, but only G3BP1 deletion decreased the levels of Ins1/2 mRNA and proinsulin and impaired polysome formation. Like glucose, other insulin secretagogues such as Exendin-4 and palmitate, but not high KCl, prompted the dissolution of G3BP1+ condensates. G3BP1+/Ins mRNA+ condensates were also present in mouse and human beta cells from normoglycemic donors. Hence, G3BP1+ condensates represent a glucose-regulated compartment for the physiological storage and protection of insulin mRNA in resting beta cells.

molecular biology↗

Chrysalis: decoding tissue compartments in spatial transcriptomics with archetypal analysis

Dissecting tissue compartments in spatial transcriptomics (ST) remains challenging due to limited spatial resolution and dependence on single-cell reference data. We present Chrysalis, a novel method to rapidly detect tissue compartments through spatially variable gene (SVG) detection and archetypal analysis without external references. We applied Chrysalis on ST datasets originating from various species, tissues and technologies and demonstrated state-of-the-art performance in identifying cellular niches.

bioinformatics↗

Insulin mRNA is stored in RNA granules in resting beta cells

The glucose-stimulated biosynthesis of insulin in pancreatic islet beta cells is post-transcriptionally regulated. Several RNA-binding proteins (RBPs) that regulate Insulin mRNA stability and translation also bind mRNAs coding for other insulin secretory granule (ISG) proteins. However, an overview of these interactions and their glucose-induced remodelling is still missing. Here we identify two distinct sets of RBPs that were preferentially pulled down with the 5-UTRs of mouse Ins1, Ins2, spliced Ins2, Ica512/Ptprn and Pc2/Pcsk2 mRNAs from extracts of either resting or stimulated mouse insulinoma MIN6 cells. Among RBPs binding to all tested transcripts in resting conditions was hnRNP A2/B1. Hnrnpa2b1 KO MIN6 cells contained lower levels of Ins1 mRNA, proinsulin and insulin, and had reduced insulin secretion. In resting cells, both hnRNP A2/B1 and Insulin mRNAs localized to stress granules, which dissolved upon glucose stimulation. Insulin mRNA-positive RNA granules were also found in human pancreatic beta cells in situ. Our results suggest that resting beta cells store mRNAs for insulin secretory granule proteins in stress granules through specific RNA protein interactions. Glucose stimulation remodels these interactions, releasing the transcripts, and another set of RBPs coordinates their translation.

cell biology↗

Purification of age-distinct insulin secretory granules through antigen restriction using the CLIP-tag

Endocrine cells employ regulated exocytosis of secretory granules to secrete hormones and neurotransmitters. The competence of secretory granules for exocytosis depends on spatio-temporal variables such as proximity to the plasma membrane and age, with newly-generated granules being preferentially released. Despite recent advances, we still lack a comprehensive view about the molecular composition of insulin granules and how this may change over the lifetime of the organelles. Here we report a strategy for the purification of insulin secretory granules of distinct age from insulinoma INS-1 cells. Tagging the granule-resident protein phogrin with a cleavable CLIP tag, we obtained intact fractions of age-distinct granules for proteomic and lipidomic analyses. We find that the lipid composition changes over time along with the physical properties of the membrane.

cell biology↗