Adaptation in a Fibronectin Binding Autolysin of Staphylococcus saprophyticus

Abstract

Human-pathogenic bacteria are found in a variety of niches, including free-living, zoonotic, and microbiome environments. Identifying bacterial adaptations that enable invasive disease is an important means of gaining insight into the molecular basis of pathogenesis and understanding pathogen emergence. Staphylococcus saprophyticus, a leading cause of urinary tract infections, can be found in the environment, food, animals, and the human microbiome. We identified a selective sweep in the gene encoding the Aas adhesin, a key virulence factor that binds host fibronectin. We hypothesize that the mutation under selection (aas_2206A>C) facilitates colonization of the urinary tract, an environment where bacteria are subject to strong shearing forces. The mutation appears to have enabled emergence and expansion of a human-pathogenic lineage of S. saprophyticus. These results demonstrate the power of evolutionary genomic approaches in discovering the genetic basis of virulence and emphasize the pleiotropy and adaptability of bacteria occupying diverse niches. IMPORTANCEStaphylococcus saprophyticus is an important cause of urinary tract infections (UTI) in women; such UTI are common, can be severe, and are associated with significant impacts to public health. In addition to being a cause of human UTI, S. saprophyticus can be found in the environment, in food, and associated with animals. After discovering that UTI strains of S. saprophyticus are for the most part closely related to each other, we sought to determine whether these strains are specially adapted to cause disease in humans. We found evidence suggesting that a mutation in the gene aas is advantageous in the context of human infection. We hypothesize that the mutation allows S. saprophyticus to survive better in the human urinary tract. These results show how bacteria found in the environment can evolve to cause disease.

Keywords: Staphylococcus saprophyticus; adhesins; evolution; positive selection; urinary tract infection.

FIG 1

Maximum likelihood phylogeny of S. saprophyticus. Maximum likelihood phylogenetic analysis was performed in RAxML (92) using a whole-genome alignment with repetitive regions masked. The phylogeny is midpoint rooted, and nodes with bootstrap values of less than 90 are labeled. Branch lengths are scaled by the number of substitutions per site. Tips are colored based on the isolation source (pink, human; blue, animal; green, food; orange, environment). Tips are labeled with the isolate name and detailed source information. S. saprophyticus contains two major clades (clade P and clade E). Within clade P, there is a lineage enriched in human-pathogenic isolates (lineage U [branch labeled “U”]).

FIG 2

Recombination in S. saprophyticus. Recombinant regions in the whole-genome alignment of S. saprophyticus were identified using Gubbins (20). Mobile genetic elements are highlighted in blue on the outer rim. The window with low Tajima’s D and π values is highlighted in pink. Few recombination events are inferred within this region.

FIG 3

Sliding window analysis of diversity and neutrality statistics. Population genetic statistics were calculated for lineage U using EggLib (94). Windows were 50 kb in width with a step size of 10 kb. (A) Tajima’s D. (B) π (green) and θ (blue). The lowest values for Tajima’s D and π are found in the same window (1,760,000 to 1,820,000 bp; arrow).

FIG 4

Extended haplotype homozygosity (EHH) of a single nucleotide polymorphism at position 1811777. EHH values for the ancestral allele are in light blue. EHH values for the derived allele are in dark blue.

FIG 5

Nonsynonymous variant in Aas fibronectin binding repeat. (Top) Domains of Aas protein (adapted from reference . R1ab is the peptide used in the fibronectin and thrombospondin binding experiments. (Bottom) Alignment of a portion of R1 showing amino acid sequences in Aas from selected S. saprophyticus strains. Amino acids are colored based on their propensity to form beta strands (light green, high propensity; light blue, low propensity). The alignment visualization was created in JalView.

FIG 6

Fibronectin and thrombospondin binding to human-associated and ancestral strain Aas R1ab. ELISAs detecting the binding of soluble human fibronectin and thrombospondin to plates coated with Aas R1ab at 3 and 10 µg/ml were performed. Results have been normalized to the percentage of binding of 10 µg/ml glycoprotein to human-associated strain R1ab. Human and bovine fibronectin and human thrombospondin bound to the two constructs equally well.

FIG 7

Site frequency spectrum of lineage U. The ancient genome (Troy) was used as the outgroup to determine the ancestral state. Synonymous, nonsynonymous, and intergenic sites were identified with SnpEff (98). The observed synonymous SFS contained an excess of singletons and high-frequency-derived variants. Both the observed SFS and the SFS predicted by the best-fitting model have an excess of singletons compared to the SFS predicted by the standard neutral model with no population size change.

FIG 8

Effects of internal and external recombination on Tajima’s D values. Bacterial populations with a range of recombination rates were simulated with SimBac. (A) Tajima's D values from simulations of internal recombination rates ranging from 0 to 0.03 in the absence of external recombination. (B) Tajima’s D values from simulations performed with an internal recombination rate of 0.003 (r/m = 1) and external recombination rates ranging from 0 to 0.003. Points are filled according to the upper limit of diversity in external recombinant fragments.

FIG 9

Cartoon of fitted demographic models. The observed synonymous SFS was fitted to 5 demographic models, including constant size, instantaneous population size change, exponential population size change, instantaneous population size change followed by exponential population size change, and two instantaneous population size changes. The parameters for the instantaneous and exponential models are the magnitude of the population size change (ν = N_e/N_ancestral) and the timing of the change (τ = number of generations/N_ancestral). For the models with two population size changes, magnitudes are reported as ν_A = N_b/N_ancestral and ν_b = N_e/N_ancestral.