Pateamine A-sensitive ribosome profiling reveals the scope of translation in mouse embryonic stem cells.

Alexandra Popa,Pascal Barbry,Kevin Lebrigand,Rainer Waldmann

doi:10.1186/s12864-016-2384-0

Abstract

BackgroundOpen reading frames are common in long noncoding RNAs (lncRNAs) and 5’UTRs of protein coding transcripts (uORFs). The question of whether those ORFs are translated was recently addressed by several groups using ribosome profiling. Most of those studies concluded that certain lncRNAs and uORFs are translated, essentially based on computational analysis of ribosome footprints. However, major discrepancies remain on the scope of translation and the translational status of individual ORFs. In consequence, further criteria are required to reliably identify translated ORFs from ribosome profiling data.ResultsWe examined the effect of the translation inhibitors pateamine A, harringtonine and puromycin on murine ES cell ribosome footprints. We found that pateamine A, a drug that targets eIF4A, allows a far more accurate identification of translated sequences than previously used drugs and computational scoring schemes. Our data show that at least one third but less than two thirds of ES cell lncRNAs are translated. We also identified translated uORFs in hundreds of annotated coding transcripts including key pluripotency transcripts, such as dicer, lin28, trim71, and ctcf.ConclusionPateamine A inhibition data clearly increase the precision of the detection of translated ORFs in ribosome profiling experiments. Our data show that translation of lncRNAs and uORFs in murine ES cells is rather common although less pervasive than previously suggested. The observation of translated uORFs in several key pluripotency transcripts suggests that translational regulation by uORFs might be part of the network that defines mammalian stem cell identity.Electronic supplementary materialThe online version of this article (doi:10.1186/s12864-016-2384-0) contains supplementary material, which is available to authorized users.

Highlights

Open reading frames are common in long noncoding RNAs and 5’UTRs of protein coding transcripts
Since inhibitors of Eif4E, the CAPbinding protein, were not available, we examined the effect of pateamine A, a drug that targets eIF4A [17], a component of the translation initiation complex Eif4F, on ribosome footprints
We show that inhibition of translation with DMDAPatA allows a reliable identification of translated ORFs in ribosome profiling data even for rare transcripts for which previously used computational scoring often perform badly

Summary

Introduction

Open reading frames are common in long noncoding RNAs (lncRNAs) and 5’UTRs of protein coding transcripts (uORFs). The question of whether those ORFs are translated was recently addressed by several groups using ribosome profiling. Most of those studies concluded that certain lncRNAs and uORFs are translated, essentially based on computational analysis of ribosome footprints. Most lncRNAs have ORFs, raising the question of whether those transcripts are translated. ORFs are common in 5’ untranslated regions (5’ UTRs) of protein coding transcripts. They are found in 40 – 50 % of human or rodent 5’ UTRs [5].

Methods

Results

Discussion

Conclusion