Newly synthesized mRNA selectively escapes translational repression following acute stress

  1. Mostafa Zedan
  2. Alexandra P. Schürch
  3. Stephanie Heinrich
  4. Pablo Aurelio Gómez García
  5. Sarah Khawaja
  6. Karsten Weis

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

When cells encounter environmental stress, they rapidly mount an adaptive response by switching from pro-growth to stress-responsive gene expression programs. It is poorly understood how cells selectively silence pre-existing, pro-growth transcripts, yet efficiently translate transcriptionally-induced stress mRNA, and whether these transcriptional and post-transcriptional responses are coordinated. Here, we show that following acute glucose withdrawal in S. cerevisiae, pre-existing mRNAs are not first degraded to halt protein synthesis, nor are they sequestered away in P-bodies. Rather, their translation is rapidly repressed through a sequence-independent mechanism that differentiates between mRNAs produced before and after stress followed by their decay. Transcriptional induction of endogenous transcripts and reporter mRNAs during stress is sufficient to escape translational repression, while induction prior to stress leads to repression. Our results reveal a timing-controlled coordination of the transcriptional and translational responses in the nucleus and cytoplasm ensuring a rapid and widescale reprogramming of gene expression following environmental stress.

Article activity feed

  1. PREreview

    This Zenodo record is a permanently preserved version of a PREreview. You can view the complete PREreview at https://prereview.org/reviews/11475530.

    In this manuscript, the authors use pulse chase analysis and ribosome footprinting to investigate the transcriptional and translational changes in Saccharomyces during starvation stress and ultimately conclude that transcripts produced after the start of stress response escape translational repression. I believe this manuscript includes valuable observations and introduces an interesting starting point to ultimately understand the molecular mechanism by which certain transcripts evade translational repression. However, the way some of the data was displayed makes interpretation difficult and raises questions about the analysis of the data. I will discuss some concerns below and suggest an orthogonal experiment that would greatly improve the manuscript.

    Major points -

    (1)    I want to make the authors aware of the paper Translational Control during Early Dictyostelium Development: Possible Involvement of Poly(A) Sequences (Palatnik, Wilkins and Jacobson, 1984). This manuscript uses very different methods but reaches similar conclusions to the cited work. Moreover, this work argues the critical issue is the length of the poly(A) tail, which was not examined in this work.  It's important to recognize previously published papers and these two works in relation to each other will enrich each other and the field. Also, this paper might provide some ideas for the researchers moving forward by showing the likely mechanism of discrimination.

    (2)    While ribosome footprinting can be a useful method to indirectly quantify translation, it has the limitation of being dependent on the number of that particular transcript in the cell and does not necessarily quantify production of complete and functional protein. Since the authors have already produced a citrine mRNA reporter (sup. Fig. 2), I suggest quantifying fluorescence as a direct read-out for translation of citrine. This would be an easy experiment that would ultimately solidify the results and improve the overall manuscript.

    (3)    I am concerned about the information displayed in Figure 2A. The volcano plots show many more points that have positive fold change in both the total mRNA and footprinting than expected in a situation where almost all translation is completely shut off and transcription is reduced. Can the authors explain this event or use orthogonal methods to confirm that result?

    (4)    For many of the experiments presented oligo(dT) is used to enrich mRNAs containing polyA tails. Thus, their analysis excludes any transcripts with very short polyA tracts, which will be shortened once transcription is repressed during starvation stress. This could lead to erroneous interpretations due to the exclusion of a portion of deadenylated mRNAs. Ideally, these experiments should be done without polyA selection for unbiased results.

    Minor points –

    (1)    I recommend adding transparency to the points in volcano plots so the reader can better see when points are over each other and better interpret how many data points are being presented.

    (2)    Some figures that are in direct comparison have y-axis at different scales. They should be at the same scale to facilitate interpretation.

    Competing interests

    The authors declare that they have no competing interests.

  2. PREreview

    This Zenodo record is a permanently preserved version of a PREreview. You can view the complete PREreview at https://prereview.org/reviews/11477828.

    In this manuscript, the authors used 4-TU pulse labeling to understand how mRNA translation is regulated during acute stress. They demonstrated that for pre-existing mRNAs, transcripts are translationally repressed before their mRNA gets degraded. In addition, the decision to silence or translate a particular mRNA transcript following glucose depletion is dependent on the timing of its production. Transcripts produced prior to glucose deprivation are repressed while those produced after stress are actively translated.

    Overall, the findings from this manuscript are interesting. Using modern sequencing techniques, it reinforces and supports previous work on translation regulation during cell stress, which demonstrated that newly synthesized mRNAs were preferentially translated during starvation in Dictyostelium due to their longer poly(A) tails (Palatnik et al (1984), Cell). However, this manuscript could benefit greatly by a) determining if the difference between pre- and post-synthesized mRNAs is simply the length of the poly(A) tail, or b) deciphering potential mechanisms on how cells distinguish pre-stress and stress-induced transcripts.

    Major points:

    1.     The authors obtained mRNA's translation efficiency by performing RNA sequencing and ribosome profiling. During RNA sequencing, they poly(A) selected the mRNAs using oligod(T) beads. However, as most mRNAs lose their poly(A) tails rapidly following stress, poly(A) selection will cause an under-representation of many mRNAs and bias the population towards mRNAs with longer tails. Consequently, this could skew the interpretation for translation efficiency. The authors should inspect their RNA sequencing results for poly(A) biases, and if so, they could complement by constructing libraries without prior poly(A) selection and/or performing Nanopore sequencing.

    2.     In Fig. 2A, there is a significant portion of genes that become upregulated 15 min and 30 min after glucose deprivation. This observation is intriguing since most genes are suppressed upon stress and only essential genes are actively translating. It would value add to the existing data if the authors followed up on these upregulated genes at 15 min and 30 min to determine if these are the same set of mRNAs as shown in Fig. 2B or if they belong to a different set of mRNAs that are regulated by other cellular pathways and responses.

    3.     In the supplement, the authors presented another reporter system encoding for citrine, a fluorescent protein. It would make the data more impactful and convincing if the authors showed immunofluorescence or live cell images of the reporter protein after pre- or post-stress induction.

    Minor points:

    1.     Axes in Figure 4E seems to be swapped

    Competing interests

    The authors declare that they have no competing interests.

  3. Version published to 10.1101/2024.04.14.589419v3 on bioRxiv
  4. Version published to 10.1101/2024.04.14.589419v2 on bioRxiv
  5. Version published to 10.1101/2024.04.14.589419v1 on bioRxiv