The impact of leadered and leaderless gene structures on translation efficiency, transcript stability, and predicted transcription rates inMycobacterium smegmatis

Abstract

16Regulation of gene expression is critical for the pathogen Mycobacterium tuberculosis to tolerate 17 stressors encountered during infection, and for non-pathogenic mycobacteria such as 18Mycobacterium smegmatis to survive stressors encountered in the environment. Unlike better 19 studied models, mycobacteria express ~14% of their genes as leaderless transcripts. However, the 20 impacts of leaderless transcript structures on mRNA half-life and translation efficiency in 21 mycobacteria have not been directly tested. For leadered transcripts, the contributions of 5' UTRs 22 to mRNA half-life and translation efficiency are similarly unknown. In both M. tuberculosis and 23 M. smegmatis, the essential sigma factor, SigA, is encoded by an unstable transcript with a 24 relatively short half-life. We hypothesized that sigA's long 5' UTR caused this instability. To test 25 this, we constructed fluorescence reporters and then measured protein abundance, mRNA 26 abundance, and mRNA half-life. From these data we also calculated relative transcription rates. 27We found that the sigA 5' UTR confers an increased transcription rate, a shorter mRNA half-life, 28 and a decreased translation rate compared to a synthetic 5' UTR commonly used in mycobacterial 29 expression plasmids. Leaderless transcripts produced less protein compared to any of the leadered 30 transcripts. However, translation rates were similar to those of transcripts with the sigA 5' UTR, 31 and the protein levels were instead explained by lower transcript abundance. A global comparison 32 of M. tuberculosis mRNA and protein abundances failed to reveal systematic differences in 33 protein:mRNA ratios for natural leadered and leaderless transcripts, consistent with the idea that 34 variability in translation efficiency among mycobacterial genes is largely driven by factors other 35 than leader status. The variability in mRNA half-life and predicted transcription rate among our 36 constructs could not be explained by their different translation efficiencies, indicating that other 37 factors are responsible for these properties and highlighting the myriad and complex roles played 38 by 5' UTRs and other sequences downstream of transcription start sites. 39 40 41 5' UTRs can also regulate gene expression by altering mRNA turnover rates. This can be a 57 consequence of altered translation rates, as impairments to translation often lead to faster mRNA 58 decay [16][17][18][19][20][21][22]. In other cases, mRNA stability is directly affected by sRNA binding to 5' UTRs or 59 by UTR secondary structure [9,[23][24][25][26][27][28]. In E. coli, the half-life of the short-lived transcript bla can 60 be significantly increased when its native 5' UTR is replaced with the 5' UTR of ompA, a long-61 lived transcript [29][30][31]. Conversely, deletion of ompA's native 5' UTR decreased its half-life by 62 5-fold [30]. The longevity conferred by the ompA 5' UTR was attributed to the presence of a non-63 specific stem-loop as well as the specific RBS sequence [30][31][32]. Secondary ...

Ribosome profiling inMycobacterium tuberculosisreveals robust leaderless translation

“…We have previously demonstrated that a substantial subset of M. tuberculosis mRNAs is transcribed as leaderless transcripts that lack a 5'UTR and associated ribosomal recognition signals, but quantification of translation rates for this pathogen at a genome-wide level was lacking. From the data presented here, we were able to quantify similar reliable rates of translation for leaderless Spearman's r value mRNAs (436 out of 2,598; 17%) and for Shine-Dalgarno mRNAs (486 out of 2,598; 19%); thereby corroborating that M. tuberculosis is able to efficiently initiate global translation of leaderless transcripts during exponential growth even in the absence of a Shine-Dalgarno sequence (Shell et al, 2015;Nguyen et al, 2020). Translation efficiencies (TEs) were calculated by dividing the RPKM of the translatome by the RPKM of the transcriptome and revealed significantly higher median efficiency levels for Shine-Dalgarno genes (Mann-Whitney test, P = 0.0195; Figure 2A), confirming previous observations between these genes and higher levels of expression (Ma et al, 2002).…”

“…In particular there is increasing evidence to show that distinct populations of ribosomes preferentially translate mRNAs with certain features (Xue and Barna, 2012;Byrgazov et al, 2013;Shi et al, 2017;). Untranslated regions (UTRs) of mRNAs can influence the stability and translational efficiency of the transcript either by providing binding sites for regulatory proteins or RNAs, adopting secondary structures such as internal ribosome entry sites or by repressing translation (Hughes, 2006;Morris and Geballe, 2000;Nguyen et al, 2020). Furthermore, short open reading frames (ORFs) within 5' UTRs may inhibit translation by occluding the start codon or may themselves be translated (Ruan et al, 1994;Cao and Geballe, 1995;Werner et al, 1987).…”

“…The adaptive responses of mycobacteria to stress have been extensively analysed by transcriptional profiling in well-defined experimental models (Stewart et al, 2002;Betts et al, 2002;Deb et al, 2009;Rohde et al, 2012;Galagan et al, 2013;Rustad et al, 2014;Namouchi et al, 2016;Aguilar-Ayala et al, 2017;Martini et al, 2019). However, due to the well-established poor correlation between levels of mRNA transcripts and proteins in the cell (Cortes et al, 2017;Haider and Pal, 2013) and differences in mRNA and protein half-lives (Nguyen et al, 2020), it is unclear how the changes that occur in the transcriptome of bacteria that enter a persistent state play out in the proteome. Clearly, there are post-transcriptional control mechanisms that influence protein abundance, but these are currently poorly understood in mycobacteria.…”