Protein Abundance Control by Non-coding Antisense Transcription.

Florian Huber,Michael Knop,Anton Khmelinskii,Lars M Steinmetz,Daria Bunina,Patrick Theer,Ishaan Gupta,Matthias Meurer

doi:10.1016/j.celrep.2016.05.043

Abstract

SummaryStable unannotated transcripts (SUTs), some of which overlap protein-coding genes in antisense direction, are a class of non-coding RNAs. While case studies have reported important regulatory roles for several of such RNAs, their general impact on protein abundance regulation of the overlapping gene is not known. To test this, we employed seamless gene manipulation to repress antisense SUTs of 162 yeast genes by using a unidirectional transcriptional terminator and a GFP tag. We found that the mere presence of antisense SUTs was not sufficient to influence protein abundance, that observed effects of antisense SUTs correlated with sense transcript start site overlap, and that the effects were generally weak and led to reduced protein levels. Antisense regulated genes showed increased H3K4 di- and trimethylation and had slightly lower than expected noise levels. Our results suggest that the functionality of antisense RNAs has gene and condition-specific components.

Highlights

High-throughput technologies such as tiling arrays and deep sequencing enable genome-wide and strand-specific detection of RNAs and have revealed the pervasive nature of transcription in eukaryotic genomes (Bertone et al, 2004; David et al, 2006; Nagalakshmi et al, 2008), resulting in the identification of many classes of non-coding RNAs
We found that the mere presence of antisense stable unannotated transcripts (SUTs) was not sufficient to influence protein abundance, that observed effects of antisense SUTs correlated with sense transcript start site overlap, and that the effects were generally weak and led to reduced protein levels
Our results suggest that the functionality of antisense RNAs has gene and condition-specific components

Summary

Introduction

High-throughput technologies such as tiling arrays and deep sequencing enable genome-wide and strand-specific detection of RNAs and have revealed the pervasive nature of transcription in eukaryotic genomes (Bertone et al, 2004; David et al, 2006; Nagalakshmi et al, 2008), resulting in the identification of many classes of non-coding RNAs (ncRNAs). In Saccharomyces cerevisiae, ncRNAs typically originate from nucleosome-depleted regions (NDRs), which are frequently associated with bidirectional promoters of protein-coding genes (Neil et al, 2009; Xu et al, 2009) Such pervasive transcription from NDRs is limited by a combination of transcriptome surveillance mechanisms such as transcription attenuation mediated by the Nrd1-Nab3-Sen (NNS) termination complex (Arigo et al, 2006; Schulz et al., 2013), suppression of divergent transcription via histone marks (Churchman and Weissman, 2011; Marquardt et al, 2014; Uprety et al, 2016), or rapid degradation of the resulting transcripts by the exosome (Davis and Ares, 2006; van Dijk et al, 2011; Wyers et al, 2005). Strong regulatory functions of asRNAs in yeast have been shown for several genes, including CDC28 (Nadal-Ribelles et al, 2014), PHO84 (Camblong et al, 2009, 2007; Castelnuovo et al, 2013), PHO5 (Uhler et al, 2007), and IME1 (van Werven et al, 2012), each with individual mechanistic characteristics different from RNAi, as S. cerevisiae lacks a functional RNAi machinery (Drinnenberg et al, 2011, 2009)

Results

Discussion

Conclusion