PMC

Genetic maps of plasmid pLC and pHC. The plasmids are of comparable genome size and differ in the origin of replication. The mob gene originally found in pBBR1 was nonfunctionalized by truncation during plasmid construction (denoted as Ψ_mob).

Simulated evolution of plasmids and chromosomes under neutrality with mutation (500 loci, mutation rate 2 × 10⁻⁹ per locus per generation). (A) Distribution of the number of mutations after 1,000 generations in 100 replicates. The color of the boxes corresponds to the combination of replicon and population size as in the legend. (B) Joint cumulative distribution function (CDF) of the AF of all mutations present in the replicates after 1,000 generations.

Simulated evolutionary dynamics of mutant alleles on plasmids and chromosomes. AF dynamics over 2,000 generations in 100 replicate simulated populations. Columns are the five combinations of simulated replicon type and population size, and rows are the different combinations of selective coefficient (s) and initial AF (f). Generations are on the x-axis, proportion of AF of replicates are stacked along the y-axis, population AF is color coded. For f = 0.001 and f = 0.0001, the starting frequency of mutant alleles is randomly sampled with the respective frequencies. For f = 0.5, the starting frequency of the mutant allele is exactly 0.5 for the chromosome and exactly 0.5 inside each cell for the plasmids. Note that, the initial frequency of cells with a mutant allele in the population is higher in the plasmid simulations compared with chromosome simulations (see , online, for expected number of mutant cells).

Frequency trajectories of plasmid alleles per cell in the 1,010 plasmid simulations. (A) Data shown for with initial allele frequencies f = 0.001, and f = 0.5. The frequency of mutant cells is calculated as the proportion of cells with at least one mutant plasmid allele within the population. The frequency of fixed mutant cells is the proportion of cells in which all plasmids loci contain the mutant allele (i.e., the plasmid allele is fixed in the cell). Starting with a low initial AF (f = 0.001), the difference between mutant plasmid allele dynamics and mutant cells in the population is most pronounced with s = 0.01. The results of the simulation where s = 0.1 are presented for comparison; in this setting, the dynamics of mutant plasmid alleles and mutant cells are similar. In the simulation of plasmid allele dynamics post balancing selection (f = 0.5), the ancestral and mutant alleles are set to an equal ratio within all cells at the beginning of the simulation. (B) Ancestral allele loss is calculated as the median number of generation where the ancestral allele was lost across all simulated populations.

An illustration of the consequences of segregational drift to plasmid allele dynamics. The illustrated PCN is four and its segregation into daughter cells is balanced (i.e., the daughter cells inherit an equal number of plasmids). In the depicted scenario, one allele emerges on the chromosome (red X) and another allele emerges on the plasmid (magenta) at t₀. Both alleles are neutral (i.e., s = 0). The population size (N) and the frequency of hosts (f_hosts) over three generations are presented. The chromosomal allele is inherited into all daughter cells, such that it is present in the total population. Random segregation of the plasmid genotypes leads to decrease in the number of hosts. At t₃, only a small minority in the population harbors the new plasmid allele. In the absence of selection for the plasmid allele presence, the new plasmid allele is deemed to remain at a very low frequency within the population (or be lost). Note.—This is a simplistic example that does not include loss of host cells due to drift and it does not include consideration of recombination among plasmids.

Parallel evolution of the chromosome. (A) A heatmap representing 1,025 the number of shared mutated genes among the 48 evolved populations. The number of genes mutated in population pair x, y is calculated as the number of genes in which genetic variants (single base substitutions or indels) are observed in both populations. Cells along the diagonal present the number of genes where a mutation was detected in the corresponding population and at least one additional population. Annotation bars at the top and left right present the experimental factors (color coded as shown in [C]). (B) An interaction plot of the main factors in the experiment as calculated from the complete set of allele frequencies of parallel variants (included in , online). Dots represent means and error bars represent the standard error of the mean. ANOVA demonstrated significant effects for plasmid replicon type and temperature along with significant interactions between the three factors used in the experiment (statistics in , online). (C) Color-coded matrix of variants whose AF is significantly different among the main factors or their combination (using ANOVA on aligned rank transformed allele frequencies and FDR; statistics in , online). Intragenic variants (11 synonymous and 22 nonsynonymous) are indicated by the gene symbol and, in case of a single base substitution, coded as [ancestral amino acid][amino acid position][alternative amino acid]. The two insertion mutations are indicated by a gene symbol and coded as [nucleotide position in gene][+ inserted nucleotide]. [*] indicates a nonsense codon. The two intergenic variants are indicated by [i] coded as [ancestral nucleotide][genomic position][alternative nucleotide].