Within-host evolution of Staphylococcus aureus during asymptomatic carriage

Abstract

Background: Staphylococcus aureus is a major cause of healthcare associated mortality, but like many important bacterial pathogens, it is a common constituent of the normal human body flora. Around a third of healthy adults are carriers. Recent evidence suggests that evolution of S. aureus during nasal carriage may be associated with progression to invasive disease. However, a more detailed understanding of within-host evolution under natural conditions is required to appreciate the evolutionary and mechanistic reasons why commensal bacteria such as S. aureus cause disease. Therefore we examined in detail the evolutionary dynamics of normal, asymptomatic carriage. Sequencing a total of 131 genomes across 13 singly colonized hosts using the Illumina platform, we investigated diversity, selection, population dynamics and transmission during the short-term evolution of S. aureus.

Principal findings: We characterized the processes by which the raw material for evolution is generated: micro-mutation (point mutation and small insertions/deletions), macro-mutation (large insertions/deletions) and the loss or acquisition of mobile elements (plasmids and bacteriophages). Through an analysis of synonymous, non-synonymous and intergenic mutations we discovered a fitness landscape dominated by purifying selection, with rare examples of adaptive change in genes encoding surface-anchored proteins and an enterotoxin. We found evidence for dramatic, hundred-fold fluctuations in the size of the within-host population over time, which we related to the cycle of colonization and clearance. Using a newly-developed population genetics approach to detect recent transmission among hosts, we revealed evidence for recent transmission between some of our subjects, including a husband and wife both carrying populations of methicillin-resistant S. aureus (MRSA).

Significance: This investigation begins to paint a picture of the within-host evolution of an important bacterial pathogen during its prevailing natural state, asymptomatic carriage. These results also have wider significance as a benchmark for future systematic studies of evolution during invasive S. aureus disease.

Figure 1. Genomic diversity of Staphylococcus aureus in 13 singly-colonized nasal carriers.

For each carriage study participant (A–M) a representation of the maximum likelihood tree is shown relating all colonies isolated and sequenced from that host. Gray circles represent observed genotypes, where area is proportional to sample frequency, and small black circles represent hypothetical intermediate genotypes. Edges (branches) represent mutations, color-coded as follows: synonymous (green), non-synonymous (orange), premature stop (red), non-coding (grey), structural variant (black). Solid edges represent SNPs and dashed edges represent indels. The ordering of mutations along a branch is arbitrary.

Figure 2. Diversity of S. aureus within and between hosts.

Separately for CC22 genomes (A–G) and CC30 genomes (H–M), the clonal frame consensus tree representing the relationships between hosts is shown above a heat map indicating the number of SNP differences between pairs of genomes isolated from the same and different hosts. In the key, cooler colors (closer to blue) represent less divergence and hotter colors (closer to red) represent greater divergence.

Figure 3. Examples of large insertion-deletion polymorphisms within single hosts.

A) 23.9 kb deletion of a Panton-Valentine leukocidin-encoding prophage in four colonies isolated from participant J (contig C618:c65). B) 3.5 kb indel knocking out adhE in three colonies isolated from participant F (contig C608:c44). In both panels, the deleted region is indicated in red. The presence of coding sequences (CDS, dark blue), tRNA (dark red), rRNA (purple) and other features (gray) are indicated by filled rectangles. Sliding windows are shown indicating GC content (black), and positive (green) or negative (purple) GC skew. Positions are indicated relative to the concatenated Velvet assemblies of the host-specific reference genomes. Figures extracted from circular chromosome plot generated using CGView .

Figure 4. Evidence for natural selection on the Staphylococcus aureus genome during asymptomatic carriage.

A) The relative number of synonymous versus non-synonymous SNPs on all branches of the within-host genealogies relating colonies sampled from hosts A-M. Each pie represents a branch in Figure 1, divided into segments according to the proportion of synonymous (green) and non-synonymous (orange) mutations on that branch. The area of the pie is proportional to the number of synonymous and non-synonymous mutations on that branch. The solid line is the uncorrected dN/dS ratio estimated from SNPs within hosts, which was significantly greater than the uncorrected dN/dS ratio estimated from SNPs between hosts (dashed line, McDonald-Kreitman test p<0.001). B) The sample frequency of SNPs, represented by the minor (less frequent) allele. Bars are color-coded according to SNP type: synonymous (green), non-synonymous (orange), nonsense (red) and intergenic (grey). C) The expected and observed number of within-host mutations per gene (solid and hatched bars respectively), combined across participants A–M, Q and R.

Figure 5. Within-host genealogies inferred under a model of nasal population size fluctuations during carriage.

Assuming a model of population growth and decline, the bacterial genealogy within each host A-M was estimated using an extension to BEAST. The maximum clade credibility tree is shown with the tips aligned to the estimated time in the cycle of growth and decline when sampling took place. The shaded area shows the fitted “skyline” for within-host population size (measured as the product of the effective population size and generation length), assuming a mutation rate of 2.7 mutations per megabase per year . Time is arbitrarily measured relative to the trough in population size.

Figure 6. Estimating the number of transmission events from genomic divergence.

We used a simple population genetics model to calculate the probability of the number of mutational differences between two bacterial genomes, conditional on the number of transmission events that have occurred since their most recent common ancestor, under A) slow transmission (0.3 transmissions per year) and B) rapid transmission (1.2 transmissions per year). We employed the model to estimate the Bayesian posterior probability of the number of transmission events conditional on the observed number of mutational differences, under C) slow and D) rapid transmission. E) When we applied the model to CC22 genomes under the slow transmission model, we detected evidence for very recent transmission between some pairs of hosts, including the possibility of direct transmission. In A) and B), the lines are color-coded according to the number of transmission events, indicated by the key. In C) and D), the magnitude of the posterior probability is indicated by the intensity of the shading, as shown by the key. In E), the ten pairs of hosts with most evidence for recent transmission are shown. The colors distinguish transmission pairs.