The properties of spontaneous mutations in the opportunistic pathogen Pseudomonas aeruginosa

Abstract

Background: Natural genetic variation ultimately arises from the process of mutation. Knowledge of how the raw material for evolution is produced is necessary for a full understanding of several fundamental evolutionary concepts. We performed a mutation accumulation experiment with wild-type and mismatch-repair deficient, mutator lines of the pathogenic bacterium Pseudomonas aeruginosa, and used whole-genome sequencing to reveal the genome-wide rate, spectrum, distribution, leading/lagging bias, and context-dependency of spontaneous mutations.

Results: Wild-type base-pair mutation and indel rates were ~10(-10) and ~10(-11) per nucleotide per generation, respectively, and deficiencies in the mismatch-repair system caused rates to increase by over two orders of magnitude. A universal bias towards AT was observed in wild-type lines, but was reversed in mutator lines to a bias towards GC. Biases for which types of mutations occurred during replication of the leading versus lagging strand were detected reciprocally in both replichores. The distribution of mutations along the chromosome was non-random, with peaks near the terminus of replication and at positions intermediate to the replication origin and terminus. A similar distribution bias was observed along the chromosome in natural populations of P. aeruginosa. Site-specific mutation rates were higher when the focal nucleotide was immediately flanked by C:G pairings.

Conclusions: Whole-genome sequencing of mutation accumulation lines allowed the comprehensive identification of mutations and revealed what factors of molecular and genomic architecture affect the mutational process. Our study provides a more complete view of how several mechanisms of mutation, mutation repair, and bias act simultaneously to produce the raw material for evolution.

Fig. 1

Distribution of number of BPMs and indels accumulated in wild-type MA lines

Fig. 2

Normalized mutation rates for each mutation class. Error bars indicate standard error

Fig. 3

Indel frequencies and mutation rates in homopolymers in MMR-deficient MA lines

Fig. 4

Frequencies for transition classes in the left and right replichore (called in reference to the top strand). Data are from the MMR-deficient MA lines. Observed values are significantly different from expected, for all classes (p < 0.0001)

Fig. 5

Frequency of BPMs as a function of distance from origin of replication. Data are from MMR-deficient MA lines pooled

Fig. 6

Diversity in core genes in natural populations of P. aeruginosa. Diversity data are for 2827 core genes from clinical isolates analyzed in ref. 21. a Mean substitutions per codon for core genes, binned by position along replichore (10 % increments, both replichores pooled). Error bars indicate standard error. b Relationship between diversity in core genes in natural isolates and the number mutations observed in MMR-deficient MA lines. Replichores were divided into bins of set proportional sizes and both measures were calculated for each bin. Analyses were performed with different percentages of replichore length in each bin (3 % to 10 %). Significance of positive correlation is indicated on trendlines

Fig. 7

Context-dependent mutation rates in MMR-deficient MA lines. a Heat map of context-dependent mutation rates of focal nucleotides (lagging stand orientation). Focal nucleotides are along the top, whereas 5’-flanking and 3’-flanking nucleotides are on the left and right, respectively. b Effect of the number of immediately flanking (5’ or 3’) C:G pairings on the mutation rate of focal nucleotides. Error bars indicate standard error