Investigating the influence of codon composition and configuration on mRNA stability in Saccharomyces cerevisiae

Abstract

Ribosomes decode some codons termed ‘optimal’ with greater speed and efficiency than ‘non-optimal’ codons. The codon composition of mRNAs thus plays a vital role in regulating gene expression by modulating mRNA stability and translation efficiency. The codon stabilisation coefficient (CSC) measures the influence of codons on mRNA stability. Optimal codons have positive CSCs and stabilise mRNAs, while non-optimal codons have negative CSCs and destabilise mRNAs. mRNA stability and translation efficiency were thought to exhibit a negative linear correlation with the proportion of non-optimal codons in an mRNA. While this linear relationship can explain about half the variance in mRNA half-lives, non-linear dependencies between codon composition, mRNA stability, and translation efficiency are becoming increasingly evident. In this thesis, we characterise the role of codon configuration in modulating mRNA stability in S. cerevisiae. Using variants of a tandem EGFP-mScarlet reporter, we find that clustering non-optimal codons together destabilises mRNAs and reduces translation efficiency in most cases, regardless of the length or position of the cluster, independently of mRNA surveillance pathways. We show that short non-optimal codon stretches are only able to induce modest position-dependent effects on stability despite variations in codon composition. In contrast, long non-optimal clusters or short aberrant codon stretches, such as CGA codon repeats, can substantially alter mRNA stabilities based on their position. Not accounting for non-optimal codon clustering, we show that while the proportion of non-optimal codons in mRNAs can alter mRNA half-lives by as much as 7-fold, its influence on translation efficiency is much less dramatic (approximately 1.5-fold). While no particular position in the mRNA consistently stabilises or destabilises mRNAs upon inserting non-optimal codon clusters, placing such clusters near the start of genes increases translation efficiencies compared to when they are placed farther from the start codon. We extend our findings to endogenous genes in yeast, showing that heterogeneities in the CSC profiles of highly optimal genes can destabilise mRNAs by as much as 2-fold. Our results reveal that codon configuration modulates mRNA stability much more intricately than previously realised.