Limits...
Global mRNA selection mechanisms for translation initiation.

Costello J, Castelli LM, Rowe W, Kershaw CJ, Talavera D, Mohammad-Qureshi SS, Sims PF, Grant CM, Pavitt GD, Hubbard SJ, Ashe MP - Genome Biol. (2015)

Bottom Line: Components of the closed loop complex are highly relevant for many mRNAs, but some heavily translated mRNAs interact poorly with this machinery.Therefore, alternative, possibly Pab1p-dependent mechanisms likely exist to load ribosomes effectively onto mRNAs.Finally, these studies identify and characterize a complex self-regulatory circuit for the yeast 4E-BPs.

View Article: PubMed Central - PubMed

ABSTRACT

Background: The selection and regulation of individual mRNAs for translation initiation from a competing pool of mRNA are poorly understood processes. The closed loop complex, comprising eIF4E, eIF4G and PABP, and its regulation by 4E-BPs are perceived to be key players. Using RIP-seq, we aimed to evaluate the role in gene regulation of the closed loop complex and 4E-BP regulation across the entire yeast transcriptome.

Results: We find that there are distinct populations of mRNAs with coherent properties: one mRNA pool contains many ribosomal protein mRNAs and is enriched specifically with all of the closed loop translation initiation components. This class likely represents mRNAs that rely heavily on the closed loop complex for protein synthesis. Other heavily translated mRNAs are apparently under-represented with most closed loop components except Pab1p. Combined with data showing a close correlation between Pab1p interaction and levels of translation, these data suggest that Pab1p is important for the translation of these mRNAs in a closed loop independent manner. We also identify a translational regulatory mechanism for the 4E-BPs; these appear to self-regulate by inhibiting translation initiation of their own mRNAs.

Conclusions: Overall, we show that mRNA selection for translation initiation is not as uniformly regimented as previously anticipated. Components of the closed loop complex are highly relevant for many mRNAs, but some heavily translated mRNAs interact poorly with this machinery. Therefore, alternative, possibly Pab1p-dependent mechanisms likely exist to load ribosomes effectively onto mRNAs. Finally, these studies identify and characterize a complex self-regulatory circuit for the yeast 4E-BPs.

Show MeSH
Direct pairwise comparisons between RIP-seq experiments for each of the closed loop components and 4E-BPs. (A) Scatterplots display change in the log2 median fold changes (IP/Total) for each of the six proteins compared with one another. Highlighted in red are those transcripts identified as being significantly different between the two experiments according to edgeR's interaction GLM model at a FDR <0.05. (B) Table depicting the total numbers of transcripts that vary significantly across the pairwise comparisons in (A). The numbers represent transcripts that are over-represented in the IPs of proteins listed in the columns relative to proteins listed in each row.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4302535&req=5

Fig3: Direct pairwise comparisons between RIP-seq experiments for each of the closed loop components and 4E-BPs. (A) Scatterplots display change in the log2 median fold changes (IP/Total) for each of the six proteins compared with one another. Highlighted in red are those transcripts identified as being significantly different between the two experiments according to edgeR's interaction GLM model at a FDR <0.05. (B) Table depicting the total numbers of transcripts that vary significantly across the pairwise comparisons in (A). The numbers represent transcripts that are over-represented in the IPs of proteins listed in the columns relative to proteins listed in each row.

Mentions: In order to provide a quantitative assessment of the variation between different RIP-seq datasets in a pairwise fashion, we have used an interaction model derived from the Generalised Linear Model (GLM) function within the EdgeR software package (Bioconductor) [44,45]. More specifically, we compared the ratio of mRNA levels in the IP samples relative to the level in a total RNA sample (log2(IP/Total)) for each gene in each of the six immunoprecipitation experiments, and examined the pairwise correlations between them (FigureĀ 3A). Here the full profile of mRNA enrichment values is presented, rather than those defined by a statistical cutoff. These data are presented as scatterplots cross-comparing the datasets, highlighting in red the transcripts found to be significantly different between the experiments according to the GLM. Strikingly, these plots emphasize the high correlation observed in the binding profiles of the three members of the eIF4F complex (Pearson correlations of 0.755, 0.753 and 0.812). Likewise, the two 4E-BP binding profiles, for Caf20p and Eap1p, also display a similar high correlation with each other. Notably, while Pab1p displays positive correlations with components of the eIF4F complex, it is the only factor assessed that displays a negative correlation with the profiles from the translational repressors Caf20p and Eap1p.Figure 3


Global mRNA selection mechanisms for translation initiation.

Costello J, Castelli LM, Rowe W, Kershaw CJ, Talavera D, Mohammad-Qureshi SS, Sims PF, Grant CM, Pavitt GD, Hubbard SJ, Ashe MP - Genome Biol. (2015)

Direct pairwise comparisons between RIP-seq experiments for each of the closed loop components and 4E-BPs. (A) Scatterplots display change in the log2 median fold changes (IP/Total) for each of the six proteins compared with one another. Highlighted in red are those transcripts identified as being significantly different between the two experiments according to edgeR's interaction GLM model at a FDR <0.05. (B) Table depicting the total numbers of transcripts that vary significantly across the pairwise comparisons in (A). The numbers represent transcripts that are over-represented in the IPs of proteins listed in the columns relative to proteins listed in each row.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4302535&req=5

Fig3: Direct pairwise comparisons between RIP-seq experiments for each of the closed loop components and 4E-BPs. (A) Scatterplots display change in the log2 median fold changes (IP/Total) for each of the six proteins compared with one another. Highlighted in red are those transcripts identified as being significantly different between the two experiments according to edgeR's interaction GLM model at a FDR <0.05. (B) Table depicting the total numbers of transcripts that vary significantly across the pairwise comparisons in (A). The numbers represent transcripts that are over-represented in the IPs of proteins listed in the columns relative to proteins listed in each row.
Mentions: In order to provide a quantitative assessment of the variation between different RIP-seq datasets in a pairwise fashion, we have used an interaction model derived from the Generalised Linear Model (GLM) function within the EdgeR software package (Bioconductor) [44,45]. More specifically, we compared the ratio of mRNA levels in the IP samples relative to the level in a total RNA sample (log2(IP/Total)) for each gene in each of the six immunoprecipitation experiments, and examined the pairwise correlations between them (FigureĀ 3A). Here the full profile of mRNA enrichment values is presented, rather than those defined by a statistical cutoff. These data are presented as scatterplots cross-comparing the datasets, highlighting in red the transcripts found to be significantly different between the experiments according to the GLM. Strikingly, these plots emphasize the high correlation observed in the binding profiles of the three members of the eIF4F complex (Pearson correlations of 0.755, 0.753 and 0.812). Likewise, the two 4E-BP binding profiles, for Caf20p and Eap1p, also display a similar high correlation with each other. Notably, while Pab1p displays positive correlations with components of the eIF4F complex, it is the only factor assessed that displays a negative correlation with the profiles from the translational repressors Caf20p and Eap1p.Figure 3

Bottom Line: Components of the closed loop complex are highly relevant for many mRNAs, but some heavily translated mRNAs interact poorly with this machinery.Therefore, alternative, possibly Pab1p-dependent mechanisms likely exist to load ribosomes effectively onto mRNAs.Finally, these studies identify and characterize a complex self-regulatory circuit for the yeast 4E-BPs.

View Article: PubMed Central - PubMed

ABSTRACT

Background: The selection and regulation of individual mRNAs for translation initiation from a competing pool of mRNA are poorly understood processes. The closed loop complex, comprising eIF4E, eIF4G and PABP, and its regulation by 4E-BPs are perceived to be key players. Using RIP-seq, we aimed to evaluate the role in gene regulation of the closed loop complex and 4E-BP regulation across the entire yeast transcriptome.

Results: We find that there are distinct populations of mRNAs with coherent properties: one mRNA pool contains many ribosomal protein mRNAs and is enriched specifically with all of the closed loop translation initiation components. This class likely represents mRNAs that rely heavily on the closed loop complex for protein synthesis. Other heavily translated mRNAs are apparently under-represented with most closed loop components except Pab1p. Combined with data showing a close correlation between Pab1p interaction and levels of translation, these data suggest that Pab1p is important for the translation of these mRNAs in a closed loop independent manner. We also identify a translational regulatory mechanism for the 4E-BPs; these appear to self-regulate by inhibiting translation initiation of their own mRNAs.

Conclusions: Overall, we show that mRNA selection for translation initiation is not as uniformly regimented as previously anticipated. Components of the closed loop complex are highly relevant for many mRNAs, but some heavily translated mRNAs interact poorly with this machinery. Therefore, alternative, possibly Pab1p-dependent mechanisms likely exist to load ribosomes effectively onto mRNAs. Finally, these studies identify and characterize a complex self-regulatory circuit for the yeast 4E-BPs.

Show MeSH