^* joint first authorship

Abstract

Stress-induced mRNP condensation is conserved across eukaryotes, resulting in stress granule formation under intense stresses, yet the mRNA composition and function of these condensates remain unclear. Exposure of ribosome-free mRNA following stress is thought to cause condensation and stress granule formation through mRNA-sequence-dependent interactions, leading to disproportionate condensation of long mRNAs. Here we show that, in striking contrast, virtually all mRNAs condense in response to multiple stresses in budding yeast with minor length dependence and often without stress granule formation. New transcripts escape mRNP condensation, enabling their selective translation. Inhibiting translation initiation causes formation of mRNP condensates that are distinct from stress granules and P-bodies; these translation-initiation-inhibited condensates (TIICs) are omnipresent, even in unstressed cells. Stress-induced mRNAs are excluded from TIICs due to the timing of their expression, indicating determinants of escape that are independent of sequence. Together, our results reveal a previously undetected level of translation-linked molecular organization and stress-responsive regulation.

About the process…

This paper was a wonderful long-term collaboration with the group of D. Allan Drummond, where Edward was a postdoc from 2010-2015. Data underlying figure 1 was collected by Edward back in 2015, and figuring out how to analyse and interpret it was an extended joint effort. The paper took 10 years in total to complete, with major and impressive contributions from the 3 co-first-authors, Hendrik Glauninger, Jared Bard, and Caitlin Wong Hickernell.

Previously a preprint

This work was preprinted on bioRxiv as Transcriptome-wide mRNP condensation precedes stress granule formation and excludes new mRNAs.