Evidence for strong mutation bias towards, and selection against, T/U content in SARS-CoV2: implications for attenuated vaccine design

Abstract

Large-scale re-engineering of synonymous sites is a promising strategy to generate attenuated viruses for vaccines. Attenuation typically relies on de-optimisation of codon pairs and maximization of CpG dinculeotide frequencies. So as to formulate evolutionarily-informed attenuation strategies, that aim to force nucleotide usage against the estimated direction favoured by selection, here we examine available whole-genome sequences of SARS-CoV2 to infer patterns of mutation and selection on synonymous sites. Analysis of mutational profiles indicates a strong mutation bias towards T withconcomitant selection against T. Accounting for dinucleotide effects reinforces this conclusion, observed TT content being a quarter of that expected under neutrality. A significantly different mutational profile at CDS sites that are not 4-fold degenerate is consistent with contemporaneous selection against T mutations more widely. Although selection against CpG dinucleotides is expected to drive synonymous site G+C content below mutational equilibrium, observed G+C content is slightly above equilibrium, possibly because of selection for higher expression. Consistent with gene-specific selection against CpG dinucleotides, we observe systematic differences of CpG content between SARS-CoV2 genes. We propose an evolutionarily informed gene-bespoke approach to attenuation that, unusually, seeks to increase usage of the already most common synonymous codons. Comparable analysis of H1N1 and Ebola finds that GC3 deviated from neutral equilibrium is not a universal feature, cautioning against

No evidence for increased transmissibility from recurrent mutations in SARS-CoV-2

“…As identified by previous studies [30][31][32][33][34][35], we recover evidence of strong mutational biases across the SARS-CoV-2 genome. A remarkably high proportion of CàU changes was observed relative to other types of SNPs and this pattern was observed at both non-homoplasic and homoplasic sites ( Figures S6-S8).…”

“…Notably 66% of the detected mutations comprise nonsynonymous changes of which 38% derive from CàU transitions. This high compositional bias, as also detected in other studies [33][34][35], as well as in other members of the Coronaviridae [30][31][32], suggests that mutations observed in the SARS-CoV-2 genome are not solely the result of errors by the viral RNA polymerase during virus replication [33,34]. One possibility is the action of human RNA editing systems which have been implicated in innate and adaptive immunity.…”

Stability of SARS-CoV-2 Phylogenies

“…There are at least four possible sources of (real or apparent) mutations that recur within independent lineages in a tree, and each makes distinct predictions about the source of recurrent mutations (Table 1). In particular, recent work has shown a strong bias towards C>U mutation in the SARS-CoV-2 genome [21,[42][43][44] . Systematic errors, which usually result from consistent errors in molecular biology techniques or bioinformatic data data processing, need not reflect this bias and are not subject to natural selection.…”

“…Hypermutation rather than positive selection may explain many remaining highly recurrent sites. Previous analyses showed that the rate of C>U mutation is exceptionally high relative to other mutation types in the viral genome [10,25,43,47] . This class of mutations should show increased evidence of recurring multiple times because they experience elevated mutation rates [25] .…”

“…This mutational bias might be driven by APOBEC editing of the viral genome [25,43,44,[47][48][49] . Consistent with previous results [43,44,[47][48][49] , we find that 5'-[U|A]C>U mutation more frequently than 5'-[C|G]C>U (P = 0.0501), but we do not see a similar effect for 3' flanking sites at 5'-C>U[U|A] relative to 5'-C>U[G|C] mutations (P = 0.378). The highly biased spectrum of C>U mutations and the correlation with local sequence context implies that the plus-stranded virus biology may be leading to recurrent C>U mutations [47] .…”