Variability in Codon Usage in Coronaviruses Is Mainly Driven by Mutational Bias and Selective Constraints on CpG Dinucleotide

Abstract

The Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the third human-emerged virus of the 21st century from the Coronaviridae family, causing the ongoing coronavirus disease 2019 (COVID-19) pandemic. Due to the high zoonotic potential of coronaviruses, it is critical to unravel their evolutionary history of host species breadth, host-switch potential, adaptation and emergence, to identify viruses posing a pandemic risk in humans. We present here a comprehensive analysis of the composition and codon usage bias of the 82 Orthocoronavirinae members, infecting 47 different avian and mammalian hosts. Our results clearly establish that synonymous codon usage varies widely among viruses, is only weakly dependent on their primary host, and is dominated by mutational bias towards AU-enrichment and by CpG avoidance. Indeed, variation in GC3 explains around 34%, while variation in CpG frequency explains around 14% of total variation in codon usage bias. Further insight on the mutational equilibrium within Orthocoronavirinae revealed that most coronavirus genomes are close to their neutral equilibrium, the exception being the three recently infecting human coronaviruses, which lie further away from the mutational equilibrium than their endemic human coronavirus counterparts. Finally, our results suggest that, while replicating in humans, SARS-CoV-2 is slowly becoming AU-richer, likely until attaining a new mutational equilibrium.

Variation in synonymous nucleotide composition among genomes of sarbecoviruses and consequences for the origin of COVID-19

“…Indeed, the bias against C and G nucleotides observed in third codon-positions, which appeared more marked for PangSar and SCoV2rC ( Table 2 ), was also detected by comparing the frequencies of dinucleotides ( Table 3 ) or four-fold degenerate codons (amino-acids A, G, P, T, and V in Table 4 ). In agreement with several previous studies, my results confirmed therefore that mutational bias is the main force shaping codon usage in sarbecoviruses ( Tort et al, 2020 , Daron and Bravo, 2021 , Simmonds and Ansari, 2021 ). Importantly, the bias in favour of C=>U transitions (over U=>C transitions) has been observed in a wide range of mammalian RNA viruses ( Simmonds and Ansari 2021 ), indicating that it is the result of an asymmetrical mechanism shared by all RNA viruses infecting mammals.…”

“…Several mechanisms have been proposed to account for the cytosine deficiency in the genome of sarbecoviruses, such as cytosine deamination resulting from the action of the host APOBEC3 system ( Milewska et al 2018 ), methylation of CpG dinucleotides (Xia 2020), or the limited availability of cytidine triphosphate (CTP), which is used not only for the viral RNA genome synthesis but also for the synthesis of the virus envelope, as well as translation and glycosylation of viral proteins (Ou et al 2021). My results indicated that CG dinucleotides are less frequent than other dinucleotides at both P23 and P31 sites, confirming therefore the selection against CpG dinucleotide discussed in previous studies ( Daron & Bravo, 2021 ). However, this is obviously not the main mechanism explaining the differences between the eight SNC groups.…”

“…Several studies on the synonymous nucleotide composition (SNC) have detected high levels of variation among coronaviruses. Most of them have concluded that the codon usage is mainly driven by mutational bias towards A+U or U enrichment and selection against CpG dinucleotide ( Kandeel et al, 2020 , Tort et al, 2020 , Daron and Bravo, 2021 , Rice et al, 2021 ). However, these studies generally focused on SARS-CoV-2 features, and the variation among animal sarbecoviruses was generally not fully examined.…”

Evolution and diversity of nucleotide and dinucleotide composition in poxviruses

“…In viruses, CUB can be strongly influenced by overall G+C content [25,42,44,98]. This is indeed the case in poxviruses, as we found an extremely strong correlation between variation along the first component of the PCA (PC1) and the variation in G+C content in the third codon position (GC3) (Fig.…”

Prediction of Recurrent Mutations in SARS-CoV-2 Using Artificial Neural Networks

“…Only 14% and 17% of cytosine and guanine SNVs are, respectively, C > G and G > C transversions. This may reflect the CpG avoidance that has been described for SARS-CoV-2 and other coronaviruses [ 45 , 46 ]. This CpG dinucleotide suppression is thought to be due to the fact that it is evading the zinc-finger antiviral protein (ZAP) that specifically binds CpG dinucleotides in ssRNA and causes its degradation [ 47 , 48 ].…”