Refining the Ambush Hypothesis: Evidence That GC- and AT-Rich Bacteria Employ Different Frameshift Defence Strategies

Abstract

Stop codons are frequently selected for beyond their regular termination function for error control. The "ambush hypothesis" proposes out-of-frame stop codons (OSCs) terminating frameshifted translations are selected for. Although early indirect evidence was partially supportive, recent evidence suggests OSC frequencies are not exceptional when considering underlying nucleotide content. However, prior null tests fail to control amino acid/codon usages or possible local mutational biases. We therefore return to the issue using bacterial genomes, considering several tests defining and testing against a null. We employ simulation approaches preserving amino acid order but shuffling synonymous codons or preserving codons while shuffling amino acid order. Additionally, we compare codon usage in amino acid pairs, where one codon can but the next, otherwise identical codon, cannot encode an OSC. OSC frequencies exceed expectations typically in AT-rich genomes, the +1 frame and for TGA/TAA but not TAG. With this complex evidence, simply rejecting or accepting the ambush hypothesis is not warranted. We propose a refined post hoc model, whereby AT-rich genomes have more accidental frameshifts, handled by RF2-RF3 complexes (associated with TGA/TAA) and are mostly +1 (or -2) slips. Supporting this, excesses positively correlate with in silico predicted frameshift probabilities. Thus, we propose a more viable framework, whereby genomes broadly adopt one of the two strategies to combat frameshifts: preventing frameshifting (GC-rich) or permitting frameshifts but minimizing impacts when most are caught early (AT-rich). Our refined framework holds promise yet some features, such as the bias of out-of-frame sense codons, remain unexplained.

Fig. 1.

—Correlations between GC content and out-of-frame stop codon excess (Z > 0), when all stop codons are considered together, are significantly negative in each reading frame for coding sequences simulated by random codon shuffling within the CDS. Violin plots emphasize that excesses are biased toward AT-rich genomes.

Fig. 2.

—Correlations between GC content and genome excess of out-of-frame stop codons (Z > 0) are significantly negative (P < 0.01, Spearman’s rank correlation) for all stop codons, in both reading frames, except for +1 TGA (P = 0.348) for the codon shuffle model. Excesses of TAA and TAG are heavily biased toward AT-rich genomes.

Fig. 3.

—Correlations between GC content and out-of-frame stop codon excess (Z > 0), when all stop codons are considered together, are significantly negative (P < 0.01, Spearman’s rank correlation) in each alternative reading frame for coding sequences where synonymous sites are randomized. Violin plots again emphasize a bias towards significant excesses in the AT-rich genomes.

Fig. 4.

—Correlations between GC content and genome excess of out-of-frame stop codons (Z > 0) are significantly negative (P < 0.01, Spearman’s rank correlation) for all stop codons in both alternative reading frames for the synonymous site randomisation model. Excesses of TAA and TAG are heavily biased toward AT-rich genomes, with few genomes exhibiting excesses in the +2 frame.

Fig. 5.

—Log ratios between the A use at synonymous sites of amino acids whose codons when repeated can generate an OSC. Correlations are significantly negative in each case (P < 0.05, Spearman’s rank correlations), suggesting A use at the third site decreases compared with the sixth, as GC mutational biases make encoding OSCs more difficult. When codons are restricted to only A/T ending synonyms, +1 TAA demonstrates a significant positive correlation with GC content (ρ = 0.160, P = 4.827 × 10⁻⁵, Spearman’s rank correlation).

Fig. 6.

—OSC densities are reduced in table 4 genomes when compared with table 11 genomes in each alternative reading frame. Violin plots of the loess regression residuals highlight the reduced residuals for OSC densities in table 4 genomes.

Fig. 7.

—OSC densities for table 4 genomes are reduced for each of the stop codons in the +1 frame. Violin plots of the loess regression residuals confirm the reduced densities of each OSC.

Fig. 8

.—(a) The median probability of frameshifting decreases with increasing GC3 content. (b) Genomes with excesses of OSCs for the synonymous site model tend to have higher +1 frameshift probabilities, suggesting the frequency of OSCs and susceptibility of frameshifting are linked.