Fig 1.
Changes in the incidence of invasive pneumococcal disease (IPD) from 2004 through 2013 among all ages and vaccine status in the United States based on CDC Active Bacterial Core (ABC) surveillance data [6].
Rates of IPD expressed as cases per 100,000 population are on the y-axis, and calendar year of surveillance are on the x-axis. The green line represents the IPD rate for the five most common non-PCV13 serotypes; purple line, the rate for serotype 3 only; orange line, the rate of (PCV13 –serotype 3) serotypes containing (1, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F, and 23F). Further information can be obtained from the CDC’s ABC surveillance (www.cdc.gov/abcs/).
Fig 2.
A) Rooted maximum likelihood phylogeny of S. pneumoniae serotype 3 CC180 isolates (n = 301) inferred from an alignment of 15,327 SNPs present in the core genome. Phylogeny was out-group rooted using strain AP200 (accession # CP002121) a serotype 11A, ST62 S. pneumoniae invasive isolate, which was found to be immediately basal to the CC180 clade in a global phylogeny of pneumococcal reference genomes and therefore represented the closest out-group. Bootstrap values from 100 replicates are labelled on major clades and epidemiologically relevant sub-clades. Additional values can be found in the supplemental phylogeny. Mean pairwise within- and between-clade SNP difference are presented in the grey shaded box with colors corresponding to the phylogeny. B.) World map illustrating sampled countries and regions with respective proportion of isolates belonging to Clade I-α, I-β, and II. Countries are colored according to region and pie charts represent the proportion of isolates belonging to major serotype 3 clades. The size of the pie chart is scaled to the proportion of strains sampled from each region. Mapping was performed using R package maps v3.3.0. C) Proportion of clade membership by three-year collection window. The proportion of clade membership by region over-time is displayed on the top of the figure. Pie charts are scaled by the number of isolates sampled from a geographic region by time window (i.e., column). The overall proportion of clades for each time window is presented on the bottom of the figure.
Table 1.
Clade composition of CC180 serotype 3 isolates by geographic location.
Percentages are based on 295 isolates with available collection date and country.
Fig 3.
Effective population size (Ne) comparison of serotype 3 CC180 Clades I-α and II.
Ne values were estimated using BEAST 1.8.4, enforcing a GMRF SkyGrid demographic model and relaxed molecular clock. The Ne of Clades I-α and II have been increasing with Clade II in particular, increasing exponentially base on phylodynamic model comparison.
Fig 4.
Maximum clade credibility, time-scaled phylogeny of CC180 serotype 3 (PMEN31) Clade II isolates estimated using the structured coalescent model implemented in BEAST2.
Branches are colored by geographic region and thickness is scaled values of posterior probabilities for geographic migration. Posterior probabilities for internal branches were all <0.30, precluding assessment of ancestral geographic dispersion.
Fig 5.
Phylogeny, polymorphic protein antigen variants, and antibiotic resistance.
Midpoint rooted maximum-likelihood phylogeny corresponding to Fig 2A with geographic region of isolation represented as colored tip shapes. Clades I-α, I-β, and II are shaded consistent with Fig 2A. Bootstrap values from 100 replicates are provided for major clades and sub-clades (see supplemental phylogeny for all values). Strains used in opsonophagocytic killing assay (asterisk) and surface charge experiments (double dagger) are indicated on the phylogeny. Corresponding protein antigen variants for SP1294, NanA, StrH, PspC, and PspA are illustrated on the left half of the heatmap. Eight other protein antigens are excluded due to a lack of variation in the sample. The presence and absence of AMR-associated genes are illustrated on the right half of the heatmap. The last column of the heatmap indicates genotypic antibiotic resistance that was confirmed by phenotypic testing (broth dilution or disk diffusion).
Fig 6.
PCV implementation by country and year based on IVAC data from 1999–2014.
For each country, the year is shaded based on PCV-7, PCV-10, and PCV-13 vaccine usage. Boxes marked with a dot designate the year in which a Clade II isolate was first observed. *Malaysia and China have yet to introduce PCV as part of their national immunization program. Pneumococcal vaccine use in China varies regionally. Data from China reflects PCV use in Hong Kong where serotype 3 isolates were sampled. **PCV has been available in Thailand as an optional vaccine through the National Vaccine Program since 2006. However, uptake is <5% in children under 5. ***In Poland, PCV was available to parents through private pay. Population based vaccination, without catch-up campaign, was introduced in 2017 using PCV10.
Table 2.
Recombination among CC180 lineages.
The number of polymorphisms introduced via recombination to mutation (r/m) and the number of recombination events to polymorphisms introduced by mutation (ρ/Θ) are reported for internal and terminal branches of dominant clades. Recombination occurring on internal branches is considered ancestral while terminal branches represent recent events. The overall recombination rates are also reported.
Fig 7.
Circos plot of recent and ancestral recombination events inferred among CC180 serotype 3 isolates.
Moving from the inner ring outward, rings show a histogram of unique recombination events occurring among isolates belonging to Clades I-β, I-α, and II, respectively, followed by a heatmap of cumulative recombination events. Finally, the outermost ring displays annotations of notable genes and genomic regions corresponding to the location on the OXC141 reference genome.