Search tips
Search criteria 


Logo of jidLink to Publisher's site
J Infect Dis. 2013 February 1; 207(3): 439–449.
Published online 2012 November 21. doi:  10.1093/infdis/jis703
PMCID: PMC3537446

Pneumococcal Capsular Switching: A Historical Perspective


Background.Changes in serotype prevalence among pneumococcal populations result from both serotype replacement and serotype (capsular) switching. Temporal changes in serotype distributions are well documented, but the contribution of capsular switching to such changes is unknown. Furthermore, it is unclear to what extent vaccine-induced selective pressures drive capsular switching.

Methods.Serotype and multilocus sequence typing data for 426 pneumococci dated from 1937 through 2007 were analyzed. Whole-genome sequence data for a subset of isolates were used to investigate capsular switching events.

Results.We identified 36 independent capsular switch events, 18 of which were explored in detail with whole-genome sequence data. Recombination fragment lengths were estimated for 11 events and ranged from approximately 19.0 kb to ≥58.2 kb. Two events took place no later than 1960, and the imported DNA included the capsular locus and the nearby penicillin-binding protein genes pbp2x and pbp1a.

Conclusions.Capsular switching has been a regular occurrence among pneumococcal populations throughout the past 7 decades. Recombination of large DNA fragments (>30 kb), sometimes including the capsular locus and penicillin-binding protein genes, predated both vaccine introduction and widespread antibiotic use. This type of recombination has likely been an intrinsic feature throughout the history of pneumococcal evolution.

Keywords: Capsule, serotype, switching, pneumococcus

Streptococcus pneumoniae (the “pneumococcus”) is an asymptomatic colonizer of the human nasopharynx and a major cause of otitis media, sinusitis, pneumonia, and meningitis, resulting in approximately 14.5 million annual serious disease episodes globally among young children [1]. The primary virulence determinant is its polysaccharide capsule, of which there are 94 known serotypes [26]. The capsule protects against phagocytosis during invasive pneumococcal disease and may also prevent clearance during nasopharyngeal colonization [7, 8]. Ninety-two capsule types are synthesized and exported through the “Wzy-dependent” pathway, whereby extracellular polymerization of component lipid-linked repeat units is preceded by “flippase-mediated” transfer across the cell membrane [2, 9]. The proteins involved are encoded by the capsule polysaccharide synthesis (cps) genes, located between dexB and aliA on the chromosome. The remaining 2 capsule types are synthesized through independent biochemical pathways.

Sequence analyses of 88 reference cps loci revealed variable lengths (approximately 10–30 kb) and a range of genes specific to capsule production [2]. The serotype-nonspecific genes were present among all Wzy-dependent capsule types, whereas serotype-specific genes were only found among 1 or a subset of types. Unique serotypes, and further diversity within serogroups, evolved through a combination of mutation and interspecies/intraspecies recombination [2, 1013].

Serotype is a key determinant of invasive pneumococcal disease potential and prevalence; certain serotypes are more commonly associated with carriage and others more commonly with invasive pneumococcal disease [14, 15]. A 7-valent pneumococcal conjugate vaccine (PCV7) was introduced in the United States in 2000 [16] and, subsequently, into many other countries. Ten-valent and 13-valent vaccines (PHiD-CV and PCV13, respectively) contain polysaccharides targeted against the original 7 serotypes (4, 6B, 9V, 14, 18C, 19F, and 23F) plus serotypes 1, 5, and 7F (both vaccines) and 3, 6A, and 19A (PCV13 only) and have now replaced PCV7 [17, 18].

Surveillance following PCV7 introduction showed a decline of vaccine type (VT) and an increase of nonvaccine type (NVT) pneumococci in disease and nasopharyngeal carriage [19]. This can be attributed to the phenomena of “serotype replacement,” the expansion of preexisting NVT pneumococci, and/or “serotype switching,” a change of serotype of a single clone by alteration or exchange of its cps locus [20]. (Serotype switching was first described by Griffith in 1928 [21] and was the focus of the transformation studies by Avery and colleagues in 1944 [22].) These effects are not completely independent: capsular switch variants can subsequently expand within a population. Both phenomena can be studied by comparison of the serotypes and genotypes present in populations before and after introduction of pneumococcal vaccines. Pneumococcal genotypes, as defined by multilocus sequence typing (MLST) [23], show serotype-specific associations [14, 24]; any isolate exemplifying a different genotype/serotype combination may represent a capsular switch variant. Such variants usually arise by recombination at the cps locus, and studies of a limited number of such strains indicated that recombination fragment sizes varied from approximately 21.9 kb to approximately 56.5 kb [2528]. In some cases, the fragments also included part or all of the pbp2x and pbp1a genes (2 of the 3 primary penicillin-resistance determining genes, located approximately 8 kb upstream and approximately 7 kb downstream of the cps locus).

Vaccine-induced selective pressure is contributing to the postvaccination changes of serotype epidemiology, but natural fluctuations in serotype prevalence also play a role. Many studies have documented prevaccine temporal changes in relative serotype prevalence: all but 3 included pneumococci isolated no earlier than 1969; only 2 studies provided genotype data (Supplementary Materials). In this study, we used serotyping, MLST, and whole-genome data to study a large, genetically diverse collection of historical and modern pneumococci, with the aim to better understand the mechanisms and role of capsular switching in pneumococcal evolution.


Strains and Genome Sequencing

A global collection of 426 pneumococci recovered during 1937–2007 (Supplementary Materials) were previously serotyped by the Quellung reaction and genotyped by MLST [23] to determine the sequence type (ST). Closely related isolates were assigned to clonal complexes (CCs) by a modified goeBURST method [29] (Supplementary Materials). CCs were named after the predicted founder ST. When no single founder ST could be determined, CCs were named NoneX, where X was the ST of lowest numeric value within the CC. Isolates of the same CC but with different serotypes were presumed to be capsular switch variants.

A total of 96 genetically diverse isolates from our collection were selected for whole-genome sequencing (Supplementary Materials and Table 1) on the Illumina platform as previously described [30]. Seven were excluded as technical failures. Raw sequence data were assembled using Velvet [31], and contigs were deposited in a BIGS database (BIGSdb) [32]. Sequence data were deposited in the European Nucleotide Archive (Supplementary Table 1).

Nucleotide Sequence Analysis

Eight CCs were represented by isolates subjected to whole-genome sequencing and were selected for further cps locus analysis; 7 CCs included capsular switch variants. Within each CC, the ancestral serotype was assumed to be the most common one or that of the oldest isolate. cps sequence alignments for study isolates and published loci were used to investigate serotype changes. Nucleotide sequences of the cps locus, and upstream/downstream flanking sequences where appropriate, were aligned using MUSCLE [33] and imported to MEGA5 [34] for visual inspection of variable sites. When recombination at the cps locus was predicted, potential donor representatives were sought from our whole-genome sequenced isolates, the cps locus reference sequences, and an additional 131 pneumococcal genomes retrieved from GenBank. Putative donors were identified as isolates for which the cps locus differed from that of the recombinant by a maximum of 3 nucleotides (excluding clusters of closely linked substitutions, which were considered to have arisen through recombination within the cps locus itself). MLST data were used to infer the CC most likely represented by the true donor isolate (the same as that of the donor representative). Recombination regions were identified as the minimum region over which the recombinant representative (hereafter, the “recombinant”) differed from the ancestral representative (hereafter, the “ancestor”) and was identical or highly similar (>99.7% sequence identity) to the donor representative (hereafter, the “donor”). Where no suitably matched donor was identified, the maximum recombination region was estimated as the maximum length over which the recombinant differed from the ancestor (Supplementary Materials).


Indication of Serotype Switching

Among the entire collection of 426 pneumococci, 21 of 163 unique CCs were represented by isolates of 2–6 serotypes. At least 36 independent changes of serotype within CCs were represented in our collection, 34 of which predated the introduction of pneumococcal conjugate vaccines (Figure (Figure11).

Figure 1.
Temporal distribution of capsular switch variants among the historical isolate collection. Gray triangles indicate isolation dates for the oldest recombinant representatives of 36 independent capsular switch events identified among the historical pneumococcal ...

Within-CC cps Diversity


Seven of 10 CC1514 (CC15, serotype 14) cps locus sequences were unique (Table (Table1).1). Isolates ICE13 and ICE50 shared wciY locus nucleotide substitutions with 4 CC12414 representatives (isolates 14/9, USA6, Ala289, and ICE594), suggesting a possible recombination event between these CCs. The CC1514 representatives had a 432-bp deletion in the wciY gene, resulting in a predicted 250-amino-acid truncation of the protein. Isolate CGSP14 [35] also had an additional 52-bp deletion within the wciY gene; if this represented a true deletion event rather than a sequencing/assembly error, the resultant protein would be further reduced to 38 amino acids.

Table 1.
Genetic Evolution at the cps Locus Among Isolates Representing the Same Pneumococcal Serotypes and Clonal Complexes (CCs)


Ten CC66 isolates, dated 1952–2005, represented 5 serotypes. The oldest isolate was serotype 7B; the next oldest was a serotype 9N from 1960. An approximately 53.4-kb region of sequence of the serotype 9N isolate, including pbp2x, the cps locus, and pbp1a, differed from the CC667B isolate but was highly similar to a CC37829N isolate dated 1952 (Table (Table22 and Figure Figure2).2). (Although the data strongly suggest that a CC37829N pneumococcus was the donor in this case, it is possible that another clone not represented in this sample was the true donor. Thus, we use the term “CCXX-like” when describing putative donors.) Within the cps locus, a second CC669N isolate differed from the oldest CC669N isolate by only 5 nucleotide substitutions (Table (Table11).

Table 2.
Genetic Changes Leading to Capsular Switching Events Among Pneumococci
Figure 2.
Genetic changes leading to capsular switching within clonal complexes (CCs) 66, 113, and 156/162. Bars represent the cps locus and flanking regions. Filled arrows indicate the relative positions of the dexB (left) and aliA (right) loci. Open arrows indicate ...

Sequences of the 2 CC6619F genomes suggested 2 independent capsular switches, but no potential 19F donors were identified. The putative imports were estimated as the regions over which these genomes clearly differed from those of CC669N; independent recombination breakpoints were indicated (Table (Table22 and Figure Figure22).

There were 3 independent capsular switches among 4 CC6614 representatives. On the basis of sequence similarity, the donors for 2 events were CC1514-like, while the third was CC12414-like. The breakpoints for 1 event could not be determined because of extensive diversity in the cps flanking regions, but breakpoints for the others were estimated (Table (Table22 and Figure Figure2).2). Interestingly, the 5′ breakpoints for 2 serotype 14 switches and one of the aforementioned serotype 19F switches were all 2.2–2.5 kb upstream of dexB (in the ATCC700669 reference genome [36], this corresponds to the clpL locus, which has been shown to be involved in modulating the expression of virulence-related genes [37] and in the development of penicillin nonsusceptibility [38]).

The single CC6623F representative in our collection differed from the CC669N isolates from approximately 2 kb upstream of dexB to approximately 3 kb downstream of aliA, suggesting a potential recombination import of approximately 30.4 kb, but no suitable donor was identified.


Eleven CC113 isolates represented serotypes 18C (n = 6), 35C (n = 2), and 9V, 17F, and 18B (n = 1 each). Serotype 18C was assumed to be the ancestral serotype and was also represented by 1 of the 2 oldest isolates, both dated 1939. The change from serotype 18C to 18B was due to a single nucleotide substitution within the wciX gene [2], while each of the other serotype changes was associated with a recombination event. Both CC11335C isolates likely arose from a single capsular switch event (Table (Table22 and Figure Figure2).2). The serotype 17F change appears to have involved DNA acquisition from a CC57417F-like pneumococcus by a CC11335C pneumococcus. The 5′ region of the CC11317F cps locus was distinct from that of 3 other serotype 17F isolates in our collection but differed by only 1 nucleotide substitution from the CC11335C representatives. The CC11317F genome was highly similar to that of a CC57417F isolate between position 549 of the wze locus and approximately 3 kb downstream of pbp1a, after which the CC11317F genome resembled that of the CC11318C isolates.

There was little to no variation at the cps locus among 5 CC11318C representatives (Table (Table1).1). A sixth isolate differed by 1 nucleotide in the wchA gene and by a putative import of approximately 1.1 kb (marked by 44 nucleotide substitutions) spanning the rmlA and rmlC genes (Figure (Figure33A).

Figure 3.
Variable site maps for the cps loci of serogroups 18 and 7. Rows represent the nucleotide sequences of independent isolates and are labeled as isolate name (year). Nucleotide differences relative to row 1 are shown. Periods indicate an identical base, ...


Twelve of 14 CC124 representatives were serotype 14 and represented 8 unique but highly similar cps sequences (Table (Table1).1). Two isolates were serotypes 9L and 11C, but no suitable donors were identified, and the cps flanking regions of both isolates were highly divergent from those of the CC12414 representatives for ≥30 kb in either direction. Thus, the putative recombination breakpoints could not be estimated.


Seven isolates (dated 1952–2006) representing serotypes 9V, 9A, and 19A were assigned to CC156/162. Four isolates were serotype 9V; the oldest was serotype 9A. The CC156/1629A cps locus differed from that of the oldest CC156/1629V locus by 2 nucleotides [2] (Table (Table22 and Figure Figure2).2). Three of the 4 CC156/1629V cps loci were identical, and the other differed by 1 nucleotide.

Analysis of the cps loci and flanking regions of the 2 CC156/16219A representatives (GenBank accession numbers AGOR01000001–AGOR01000024 and AGQA01000001–AGQA01000007) suggested a single capsular switch event. Approximately 50.4-kb regions of the CC156/16219A genomes were highly similar to those of a CC19919A isolate (Table (Table22 and Figure Figure2)2) and included pbp2x, the cps locus, and pbp1a. The regions directly flanking the putative import did not resemble those of the CC156/1629V representatives, and thus the true recombination breakpoints could not be estimated. However, we deduced that the original import was ≥50.4 kb.


Seven of the CC191 representatives were serotype 7F, and the eighth, and oldest, was serotype 7A. Serotypes 7A and 7F differed by 3 nucleotides (Table (Table22 and Figure Figure4).4). There was a repeated motif (5′-CTA AGA TGA ATA-3′) within the wcwC gene, and the number of repeats differed between isolates (n = 3, 4, or 6). Repeat motifs are difficult to sequence accurately, so further speculations about such changes cannot be made. Apart from the repeat region, the CC1917F cps loci each differed by a maximum of 4 nucleotides (Table (Table11 and Figure Figure33B).

Figure 4.
Genetic changes leading to capsular switching within clonal complexes (CCs) 191, 218, and 574. Bars represent the cps locus and flanking regions. Filled arrows indicate the relative positions of the dexB (left) and aliA (right) loci. Open arrows indicate ...


Eight CC218 isolates represented serotypes 12F (n = 5) and 7F (n = 3). Between approximately 9 kb upstream of pbp2x and approximately 4.8 kb downstream of pbp1a, CC2187F and CC1917F sequences were very similar, although differentiating nucleotides were identified (Table (Table2).2). The clustering of 44 nucleotide substitutions within the cps locus was indicative of recombination events (Figure (Figure33B), but whether these events occurred before or after the cps switch is unknown. Note that approximately 585 bp were missing between pbp2x and dexB in both CC2187F and CC1917F representatives. Table Table11 provides additional details about nucleotide substitutions among the cps loci of CC2187F and CC21812F isolates.


Two CC574 isolates were examined: a serotype 17F isolate from 1952 and a serotype 2 isolate from 1956. The sequences differed from position 326 of dexB to approximately 5 kb downstream of pbp1a. A region of the serotype 2 representative (totaling approximately 35 kb, from approximately 0.6 kb downstream of dexB to approximately 5 kb downstream of pbp1a) was highly similar to that of a CCNone1282 representative dated 1916 (Table (Table22 and Figure Figure4).4). A short, 314-bp region of unknown sequence in the dexB locus was also present.


Within our collection of historical and modern pneumococci, we identified 36 independent capsular switching events. Approximately 94% of the variants were isolated prior to the introduction of PCV7 and were roughly evenly distributed through time (Figure (Figure1).1). The collection was not designed for inferring a capsular switching rate; nevertheless, these data imply that this phenomenon has been a regular occurrence (ie, there is evidence of capsular switching within a diverse range of CCs) every decade throughout the past 7 decades).

Analysis of the cps loci of 10 representatives of CC66 indicated multiple independent changes to the same serotype, supporting the notion that capsular switching may occur regularly among pneumococci. Within this CC, an initial capsular switch from serotype 7B to 9N no later than 1952 was followed by at least 2 independent changes to serotype 19F no later than 1972 and 2005 and by at least 3 independent changes to serotype 14, each no later than the mid-1990s. This is consistent with previous studies that demonstrated multiple changes of serotype within the same CC: CC156/1629V→14 [3941], CC8123F→19F/A [30], and CC6954→19A [25, 28].

We studied 18 capsular switching events in detail. Three of these were presumably the result of nucleotide substitution and/or deletion, a finding consistent with previous work [2]. The remaining 15 events appeared to be due to recombination, and breakpoints for 11 events could be estimated by comparison of the cps loci and flanking sequences of the putative ancestors, donors, and recombinants. The capsular switch recombination fragments identified here (ie, imports of various lengths, inserted at different points around the cps locus, with or without the adjacent pbp sequences) are consistent with fragments detected in previously published studies [25, 42, 43]. Figure Figure55 depicts the exchange of cps loci between CCs, as inferred by our data. Isolates identified as donors from our collection could possibly have represented different CCs than the true donors; however, even if this were true, our conclusions about the number of independent capsular switches and the range of recombination fragment sizes would remain unchanged.

Figure 5.
Inter-clonal complex (CC) cps locus transfers inferred in this study. Circles represent pneumococcal CCs as named. Arrows indicate transfer of cps loci between CCs, as inferred from nucleotide sequence analyses for 15 capsular switching events. Donor ...

Our analyses did not indicate any recombination breakpoint hot spots around the cps locus, which had also not been indicated by any previous studies [25, 27, 28, 30]. It is impossible to know whether the putative imports were acquired through single or multiple recombination events, but the former is most parsimonious. The import lengths were estimated to range from approximately 19.0 to ≥58.2 kb (and apart from 1 example, always included the entire cps locus) and did not show any trends toward increasing/decreasing lengths through time.

The CC21812F→7F and CC667B→9N events were characterized by large (>50 kb) recombination imports and must have taken place no later than 1952 and 1960, respectively. Large-scale recombination between pneumococci has therefore clearly been occurring for decades, although the technology capable of detecting and detailing such events has only recently become available [25, 28, 30, 42]. Additionally, both of these putative imports included the pbp2x + cps + pbp1a loci, as reported for other events [27, 28] and for another newly characterized event in this study (CC156/1629V→19A).

Recently reported in vivo pbp2x ± cps ± pbp1a recombination events were detected soon after widespread vaccination began in the United States. No such recombinants had been reported before in nature, and the imported pbp2x and pbp1a sequences conferred penicillin nonsusceptibility. Thus, it was theorized that vaccine-induced and/or antibiotic-induced selective pressures may play a role in driving these genetic changes [28, 44]. Penicillin was introduced in the 1940s; oral penicillins were not available until the mid-1950s, and their use was initially limited. Consequently, our analyses question the above theory because the CC21812F→7F and CC667B→9N events were both associated with penicillin-susceptible pneumococci and occurred before the widespread use of penicillins and before the introduction of PCV7. Our new data suggest that recombination of the cps locus and flanking regions might be “normal” biological processes, the evolution of which has undoubtedly been influenced by naturally occurring immunity and other selective pressures. Presumably, vaccine-induced immune pressures and/or the pressure of antibiotic use subsequently influence the spread and maintenance of advantageous genes (and/or alleles) by selecting recombinants that are best able to survive. However, the potential negative effects should not be underestimated: the CC6954→19A “vaccine escape” capsular switches in the United States were first recovered from patients with pediatric invasive pneumococcal disease only 3 years after PCV7 introduction and, 2 years later, were the third-most common serotype 19A CC causing invasive pneumococcal disease among all age groups [20]. These strains were penicillin nonsusceptible, owing to the simultaneous acquisition of altered pbp2x and pbp1a genes. Consequently, the increase in prevalence of these strains likely contributed to the increase in pneumococcal penicillin nonsusceptibility in the United States after PCV7 introduction [20, 45].

Another interesting finding was that a CC19919A-like representative was the most probable donor of the 19A cps locus and flanking pbps to the CC156/16219A isolates described in this study. CC19919A representatives were also the donors of the cps locus ± pbps to the vaccine escape progeny [25, 28] and an ST32019→19A cps locus switch [27]. Future studies will attempt to uncover an explanation of why CC19919A representatives appear to be “good” cps locus donors.

Our analyses also revealed recombination at the cps locus that did not result in capsular switching, among isolates belonging to the same ancestral lineages (CC11318C and CC1917F) and different ancestral lineages (CC1917F vs CC2187F and CC1514 vs CC12414). Similar events have been noted within the CC8123F lineage [26] and among serogroup 6 and 19 isolates [1012]. We speculate that this is more likely to be the result of recombination whereby some genes are coincidentally exchanged, rather than the result of exchange that occurred directly in response to selection pressure. It could also be the result of DNA repair mediated by recombination at or near the cps locus. A comparison of the cps loci of CC1514 and CC12414 representatives also indicated a deletion in the CC1514 wciY gene, resulting in a predicted truncation of the putative glycerol phosphotransferase encoded by this gene. Notwithstanding the deletion, the CC1514 isolates were successfully serotyped by the Quellung reaction, indicating successful capsule production. Indeed, an in silico analysis of the predicted cps protein coding regions failed to identify a specific reaction catalyzed by the serotype 14 version of this protein [46].

Given the overall capacity for cps locus recombination, the associations between genotypes and serotypes are puzzling, as is our evidence of a high level of cps sequence conservation within some CCs (eg, CC1917F and CC21812F). Perhaps there is some synergism between serotype and the genetic background of a strain that conveys an advantage to certain combinations over others. This sort of synergistic effect has been invoked to explain changes in the pneumococcal population in South Korea, where there was a prevaccine reduction of multidrug-resistant CC271/32019F pneumococci and replacement by similarly multidrug-resistant CC271/32019A pneumococci over several years [47]. It is also possible that genes, such as those encoding sugar-biosynthesis enzymes, outside the cps locus contribute to capsular expression and facilitate a genotype/serotype association [2].

Our collection of pneumococci provided a unique opportunity to study evolution at the cps locus over approximately 70 years. Capsular switching with/without simultaneous pbp transfer has occurred regularly and prior to both PCV7 introduction and widespread antibiotic use. It is highly likely that the proliferation of newly generated NVT capsular switch variants will continue to be favored by PHiD-CV and PCV13 vaccination programs, as was the case after PCV7 implementation. Penicillin-nonsusceptible variants will have an even greater advantage. Although the magnitude of these selective forces relative to those favoring established genotype/VT associations remains unclear, the implementation of vaccine programs across the globe will most likely favor the intercontinental spread of NVT and penicillin-nonsusceptible pneumococci.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online ( Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.

Supplementary Data:


Acknowledgments.We thank Dalia Denapaite, Irma Ochigava, and Shwan Rachid, for DNA sequencing and MLST analysis; Andries van Tonder, for retrieval of genomic sequence data from GenBank; and the clinicians, microbiologists, and investigators of the Active Bacterial Core surveillance program of the Emerging Infections Program Network, United States.

Financial support.This work was supported by the Wellcome Trust (grant 083511/Z/07/Z to A. B. B. and grant 098051 to J. P. and S. D. B.), Stiftung Rheinland Pfalz für Innovation (to R. H.), and Ciber de Enfermedades Respiratorias (Instituto de Salud Carlos III, Madrid Spain [to J. L.]). A. B. B. is a Wellcome Trust Career Development Fellow.

Potential conflicts of interest.K. P. K. received research funding from and is a consultant for Pfizer Vaccines and is a consultant for GlaxoSmithKline Biologicals and Merck. A. B. B. received research funding from GlaxoSmithKline Biologicals. All other authors report no potential conflicts.

All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.


1. O'Brien KL, Wolfson LJ, Watt JP, et al. Burden of disease caused by Streptococcus pneumoniae in children younger than 5 years: global estimates. Lancet. 2009;374:893–902. [PubMed]
2. Bentley SD, Aanensen DM, Mavroidi A, et al. Genetic analysis of the capsular biosynthetic locus from all 90 pneumococcal serotypes. PLoS Genet. 2006;2:262–9. [PMC free article] [PubMed]
3. Calix JJ, Nahm MH. A new pneumococcal serotype, 11E, has a variably inactivated wcjE gene. J Infect Dis. 2010;202:29–38. [PMC free article] [PubMed]
4. Jin P, Kong F, Xiao M, et al. First report of putative Streptococcus pneumoniae serotype 6D among nasopharyngeal isolates from Fijian children. J Infect Dis. 2009;200:1375–80. [PubMed]
5. Park IH, Pritchard DG, Cartee R, Brandao A, Brandileone MCC, Nahm MH. Discovery of a new capsular serotype (6C) within serogroup 6 of Streptococcus pneumoniae. J Clin Microbiol. 2007;45:1225–33. [PMC free article] [PubMed]
6. Calix JJ, Porambo RJ, Brady AM, et al. Biochemical, genetic and serological characterization of two capsule subtypes among Streptococcus pneumoniae serotype 20 strains: discovery of a new pneumococcal serotype. J Biol Chem. 2012;287:27885–94. [PMC free article] [PubMed]
7. García E, López R. Molecular biology of the capsular genes of Streptococcus pneumoniae. FEMS Microbiol Lett. 1997;149:1–10. [PubMed]
8. Nelson AL, Roche AM, Gould JM, Chim K, Ratner AJ, Weiser JN. Capsule enhances pneumococcal colonization by limiting mucus-mediated clearance. Infect Immun. 2007;75:83–90. [PMC free article] [PubMed]
9. Kolkman MAB, Wakarchuk W, Nuijten PJM, van der Zeijst BAM. Capsular polysaccharide synthesis in Streptococcus pneumoniae serotype 14: molecular analysis of the complete cps locus and identification of genes encoding glycosyltransferases required for the biosynthesis of the tetrasaccharide subunit. Mol Microbiol. 1997;26:197–208. [PubMed]
10. Bratcher PE, Park IH, Oliver MB, et al. Evolution of the capsular gene locus of Streptococcus pneumoniae serogroup 6. Microbiol. 2011;157:189–98. [PMC free article] [PubMed]
11. Elberse K, Witteveen S, van der Heide H, et al. Sequence diversity within the capsular genes of Streptococcus pneumoniae serogroup 6 and 19. PLoS One. 2011;6:e25018. [PMC free article] [PubMed]
12. Mavroidi A, Godoy D, Aanensen DM, Robinson DA, Hollingshead SK, Spratt BG. Evolutionary genetics of the capsular locus of serogroup 6 pneumococci. J Bacteriol. 2004;186:8181–92. [PMC free article] [PubMed]
13. Varvio S, Auranen K, Arjas E, Mãkela PH. Evolution of the capsular regulatory genes in Streptococcus pneumoniae. J Infect Dis. 2009;200:1144–51. [PubMed]
14. Brueggemann AB, Griffiths DT, Meats E, Peto TEA, Crook DW, Spratt BG. Clonal relationships between invasive and carriage Streptococcus pneumoniae and serotype- and clone-specific differences in invasive disease potential. J Infect Dis. 2003;187:1424–32. [PubMed]
15. Hanage WP, Kaijalainen TH, Syrjänen RK, et al. Invasiveness of serotypes and clones of Streptococcus pneumoniae among children in Finland. Infect Immun. 2005;73:431–5. [PMC free article] [PubMed]
16. Black S, Shinefield H, Fireman B, et al. Efficacy, safety and immunogenicity of heptavalent pneumococcal conjugate vaccine in children. Pediatr Infect Dis J. 2000;19:187–95. [PubMed]
17. Esposito S, Tansey S, Thompson A, et al. Safety and immunogenicity of a 13-valent pneumococcal conjugate vaccine compared to those of a 7-valent pneumococcal conjugate vaccine given as a three-dose series with routine vaccines in healthy infants and toddlers. Clin Vaccine Immunol. 2010;17:1017–26. [PMC free article] [PubMed]
18. Prymula R, Schuerman L. 10-valent pneumococcal nontypeable Haemophilus influenzae PD conjugate vaccine: Synflorix™ Expert Rev Vaccines. 2009;8:1479–500. [PubMed]
19. Weinberger DM, Malley R, Lipsitch M. Serotype replacement in disease after pneumococcal vaccination. Lancet. 2011;378:1962–73. [PMC free article] [PubMed]
20. Beall BW, Gertz RE, Hulkower RL, Whitney CG, Moore MR, Brueggemann AB. Shifting genetic structure of invasive serotype 19A pneumococci in the United States. J Infect Dis. 2011;203:1360–8. [PMC free article] [PubMed]
21. Griffith F. The significance of pneumococcal types. J Hyg (Lond) 1928;27:113–59. [PMC free article] [PubMed]
22. Avery OT, Macleod CM, McCarty M. Studies on the chemical nature of the substance inducing transformation of pneumococcal types: induction of transformation by a desoxyribonucleic acid fraction isolated from pneumococcus type III. J Exp Med. 1944;79:137–58. [PMC free article] [PubMed]
23. Enright MC, Spratt BG. A multilocus sequence typing scheme for Streptococcus pneumoniae: identification of clones associated with serious invasive disease. Microbiology. 1998;144:3049–60. [PubMed]
24. Jefferies JMC, Smith A, Clarke SC, Dowson C, Mitchell TJ. Genetic analysis of diverse disease-causing pneumococci indicates high levels of diversity within serotypes and capsule switching. J Clin Microbiol. 2004;42:5681–8. [PMC free article] [PubMed]
25. Golubchik T, Brueggemann AB, Street T, et al. Pneumococcal genome sequencing tracks a vaccine escape variant formed through a multi-fragment recombination event. Nat Genet. 2012;44:352–5. [PMC free article] [PubMed]
26. Marttinen P, Hanage WP, Croucher NJ, et al. Detection of recombination events in bacterial genomes from large population samples. Nucleic Acids Res. 2012;40:e6. [PMC free article] [PubMed]
27. Pillai DR, Shahinas D, Buzina A, et al. Genome-wide dissection of globally emergent multi-drug resistant serotype 19A Streptococcus pneumoniae. BMC Genomics. 2009;10:642. [PMC free article] [PubMed]
28. Brueggemann AB, Pai R, Crook D, Beall B. Vaccine escape recombinants emerge after pneumococcal vaccination in the United States. PLoS Pathog. 2007;3:e168. [PubMed]
29. Francisco AP, Bugalho M, Ramirez M, Carriço JA. Global optimal eBURST analysis of multilocus typing data using a graphic matroid approach. BMC Bioinformatics. 2009;10:152. [PMC free article] [PubMed]
30. Croucher NJ, Harris SR, Fraser C, et al. Rapid pneumococcal evolution in response to clinical interventions. Science. 2011;331:430–4. [PMC free article] [PubMed]
31. Zerbino DR, Birney E. Velvet: algorithms for de novo short read assembly using de Bruijn graphs. Genome Res. 2008;18:821–9. [PubMed]
32. Jolley KA, Maiden MC. BIGSdb: scalable analysis of bacterial genome variation at the population level. BMC Bioinformatics. 2010;10:595. [PMC free article] [PubMed]
33. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32:1792–7. [PMC free article] [PubMed]
34. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. MEGA5: Molecular Evolutionary Genetics Analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol. 2011;28:2731–9. [PMC free article] [PubMed]
35. Ding F, Tang P, Hsu M-H, et al. Genome evolution driven by host adaptations results in a more virulent and antimicrobial-resistant Streptococcus pneumoniae serotype 14. BMC Genet. 2009;10:158. [PMC free article] [PubMed]
36. Croucher NJ, Walker D, Romero P, et al. Role of conjugative elements in the evolution of the multidrug-resistant pandemic clone Streptococcus pneumoniaeSpain23F ST81. J Bacteriol. 2009;191:1480–9. [PMC free article] [PubMed]
37. Kwon H, Kim S, Choi M, et al. Effect of heat shock and mutations in ClpL and ClpP on virulence gene expression in Streptococcus pneumoniae. Infect Immun. 2003;71:3757–65. [PMC free article] [PubMed]
38. Tran TD, Kwon H, Kim E, et al. Decrease in penicillin susceptibility due to heat shock protein ClpL in Streptococcus pneumoniae. Antimicrob Agents Chemother. 2011;55:2714–28. [PMC free article] [PubMed]
39. Coffey TJ, Daniels M, Enright MC, Spratt BG. Serotype 14 variants of the Spanish penicillin-resistant serotype 9V clone of Streptococcus pneumoniae arose by large recombinational replacements of the cpsA-pbp1a region. Microbiology. 1999;145:2023–31. [PubMed]
40. McEllistrem MC, Noller AC, Visweswaran S, Adams JM, Harrison LH. Serotype 14 variants of the France 9V−3 clone from Baltimore, Maryland, can be differentiated by the cpsB gene. J Clin Microbiol. 2004;42:250–6. [PMC free article] [PubMed]
41. Albarracín Orio AG, Cortes PR, Tregnaghi M, Piñas GE. Argentinean Network Pneumococcus Study Group Echenique JR. A new serotype 14 variant of the pneumococcal Spain9V-3 international clone detected in the central region of Argentina. J Med Microbiol. 2008;57:992–9. [PubMed]
42. Hiller NL, Ahmed A, Powell E, et al. Generation of genetic diversity among Streptococcus pneumoniae strains via horizontal gene transfer during a chronic polyclonal paediatric infection. PLoS Pathog. 2010;6:e1001108. [PMC free article] [PubMed]
43. Croucher NJ, Harris SR, Barquist L, Parkhill J, Bentley SD. A high-resolution view of genome-wide pneumococcal transformation. PLoS Pathog. 2012;8:e1002745. [PMC free article] [PubMed]
44. Ansaldi F, Canepa P, de Florentiis D, Bandettini R, Durando P, Icardi G. Increasing incidence of Streptococcus pneumoniae serotype 19A and emergence of two vaccine escape recombinant ST695 strains in Liguria, Italy, 7 years after implementation of the 7-valent conjugated vaccine. Clin Vaccine Immunol. 2011;18:343–5. [PMC free article] [PubMed]
45. Mera RM, Miller LA, Amrine-Madsen H, Sahm DF. The impact of the pneumococcal conjugate vaccine on antimicrobial resistance in the United States since 1996: evidence for a significant rebound by 2007 in many classes of antibiotics. Microb Drug Resist. 2009;15:261–8. [PubMed]
46. Aanensen DM, Mavroldi A, Bentley SD, Reeves PR, Spratt BG. Predicted functions and linkage Specificities of the products of the Streptococcus pneumoniae capsular biosynthetic loci. J Bacteriol. 2007;189:7856–76. [PMC free article] [PubMed]
47. Choi EH, Kim SH, Eun BW, et al. Streptococcus pneumoniae serotype 19A in children, South Korea. Emerg Infect Dis. 2008;14:275–81. [PMC free article] [PubMed]

Articles from The Journal of Infectious Diseases are provided here courtesy of Oxford University Press