Figures
Abstract
We determined the sequence types of isolates that caused invasive pneumococcal disease (IPD) prior to routine use of pneumococcal conjugate vaccines (PCV) in South Africa. PCV-13 serotypes and 6C isolates collected in 2007 (1 461/2 437, 60%) from patients of all ages as part of on-going, national, laboratory-based surveillance for IPD, were selected for genetic characterization. In addition, all 134 non-PCV isolates from children <2 years were selected for characterization. Sequence type diversity by serotype and age category (children <5 years vs. individuals ≥5 years) was assessed for PCV serotypes using Simpson’s index of diversity. Similar genotypes circulated among isolates from children and adults and the majority of serotypes were heterogeneous. While globally disseminated clones were common among some serotypes (e.g., serotype 1 [clonal complex (CC) 217, 98% of all serotype 1] and 14 [CC230, 43%)]), some were represented mainly by clonal complexes rarely reported elsewhere (e.g., serotype 3 [CC458, 60%] and 19A [CC2062, 83%]). In children <2 years, serotype 15B and 8 were the most common serotypes among non-PCV isolates (16% [22/134] and 15% [20/134] isolates, respectively). Sequence type 7052 and 53 were most common among serotypes 15B and 8 isolates and accounted for 58% (7/12) and 64% (9/14) of the isolates, respectively. Serotype 19F, 14, 19A and 15B had the highest proportions of penicillin non-susceptible isolates. Genotypes rarely reported in other parts of the world but common among some of our serotypes highlight the importance of our data as these genotypes may emerge post PCV introduction.
Citation: Ndlangisa KM, du Plessis M, Wolter N, de Gouveia L, Klugman KP, von Gottberg A, et al. (2014) Population Snapshot of Streptococcus pneumoniae Causing Invasive Disease in South Africa Prior to Introduction of Pneumococcal Conjugate Vaccines. PLoS ONE 9(9): e107666. https://doi.org/10.1371/journal.pone.0107666
Editor: Susanna Esposito, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Università degli Studi di Milano, Italy
Received: April 29, 2014; Accepted: August 12, 2014; Published: September 18, 2014
Copyright: © 2014 Ndlangisa et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper and its Supporting Information files.
Funding: This work was funded by the National Health Laboratory Service, Centers for Disease Control and Prevention (CDC), National Centre for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention (NCHHSTP), Global AIDS Program Cooperative Agreement U62/PSO022901. (The contents are solely the responsibility of the authors and do not necessarily represent the official views of the CDC). Molecular characterization of isolates was funded by Pfizer Vaccines Research. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Anne von Gottberg and Mignon du Plessis have received research funding from Pfizer. In addition, Anne von Gottberg has received research funding from Sanofi Pasteur. Keith P. Klugman has received consultancy fees from Pfizer, Sanofi Pasteur, and Novartis. This work was funded by the National Health Laboratory Service, Centers for Disease Control and Prevention (CDC), National Centre for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention (NCHHSTP), Global AIDS Program Cooperative Agreement U62/PSO022901. (The contents are solely the responsibility of the authors and do not necessarily represent the official views of the CDC). Molecular characterization of isolates was funded by Pfizer Vaccines Research. This does not alter the authors’ adherence to PLOS ONE policies on sharing data and materials.
Introduction
In South Africa, a country with a high prevalence of HIV infection, Streptococcus pneumoniae is an important cause of disease [1]. In 1997–1998, the rate of invasive pneumococcal disease (IPD) among HIV-infected South African children <2 years of age, was estimated to be 42 times higher than among HIV-uninfected children [2]. Moreover, adults infected with HIV are more likely to be infected with pediatric serotypes and antibiotic-resistant pneumococci [1], [3].
The first generation 7-valent pneumococcal polysaccharide-protein conjugate vaccine (PCV-7) has significantly reduced vaccine-serotype IPD among vaccinated children in countries where it has been introduced [4]–[6]. Reduction of disease in unvaccinated children and adults because of herd effect was also reported in those countries [5]. Despite the benefits of PCV-7, an increase in disease due to non-vaccine serotypes, mainly serotype 19A, has been reported in some countries where the vaccine is used routinely [4]–[8]. Other serotypes that increased post PCV-7 introduction include 1, 3, 7F and 22F [4], [5]. While these increases were mainly due to the expansion of pre-existing clones, the emergence of sequence types previously associated with PCV-7 serotypes have also been reported among non-PCV-7 serotypes [9]. In the United States, clonal complex (CC) 320/271, which was associated with multidrug resistant serotype 19F isolates prior to PCV-7 introduction, emerged among serotype 19A isolates following PCV-7 introduction [9].
Natural fluctuations in circulating genotypes in the absence of any known interventions have also been shown to occur [10] and factors other than vaccine pressure can result in proliferation of pneumococcal strains. This was demonstrated by an increase in antibiotic-resistant serotype 19A in Spain, Korea and Israel before routine use of PCV-7 [11]–[13]. Baseline data describing the genetic structure of pneumococci prior to the implementation of new interventions such as conjugate vaccines are therefore required to detect genotype changes potentially resulting from such interventions.
In South Africa, PCV-7 was registered in 2005 but was initially only available in the private healthcare sector. The vaccine was introduced into the childhood immunization programme in April 2009 as a novel 2+1 schedule administered at 6, 14 and 40 weeks of age [14] and was replaced by PCV-13 in June 2011. Limited molecular characterization data are available for pneumococcal isolates from South Africa. We therefore used multilocus sequence typing (MLST) to investigate the clonal composition of invasive pneumococci prior to routine use of pneumococcal conjugate vaccines.
Materials and Methods
National surveillance
Isolates were collected as part of the GERMS-SA (Group for Enteric, Respiratory and Meningeal Disease Surveillance) national, laboratory-based surveillance for IPD in South Africa, initiated in 1999 [15]. Isolates and patient data were submitted to the National Institute for Communicable Diseases from 187 participating laboratories across all nine provinces. A case of IPD was defined as the isolation of S. pneumoniae from a normally sterile site specimen (e.g., blood, cerebrospinal fluid [CSF], pleural fluid, joint fluid). Isolates were received on Dorset transport media (Diagnostic Media Products [DMP], Johannesburg, South Africa) and sub-cultured on arrival at the reference laboratory, on 5% horse blood agar plates (DMP) in the presence of an optochin disc (Mast Diagnostics, Virginia, USA). Pure cultures were stored in 10% skim milk (DMP) at −70°C. In addition, cases included patients with normally sterile site specimens testing positive by PCR, or bacterial latex antigen supported by Gram stain microscopy [16]. Confirmed unreported cases identified through audits of databases in participating laboratories were also included. In 2003, national surveillance was enhanced to include additional data such as patient outcome and HIV serological status from approximately 30 sentinel sites in all nine provinces. Surveillance reporting increased from 2003 onwards, and stabilized by 2005 [17]. Isolates for this study were selected from cases reported in 2007, a stable, pre-vaccine year.
Serotyping and antimicrobial susceptibility testing
Serotypes were determined by the Quellung method using type-specific antisera (Statens Serum Institut, Copenhagen, Denmark) [18]. Isolates were initially screened for susceptibility to oxacillin (for penicillin), tetracycline, chloramphenicol, erythromycin, clindamycin, rifampicin and trimethoprim-sulfamethoxazole by disc diffusion (Mast Diagnostics Group Ltd., Merseyside, UK). Isolates testing non-susceptible on disc screening had minimum inhibitory concentrations (MIC) determined using agar dilution or E-test (AB BIODISK, Solna, Sweden). Interpretation of results was done according to Clinical and Laboratory Standards Institute guidelines and breakpoints [19]. Isolates were considered non-susceptible to penicillin at MICs ≥0.12 mg/L using the oral penicillin meningitis breakpoints. For other antimicrobials, isolates were defined as non-susceptible if they were intermediately or fully resistant to the agent tested, and isolates were classified as multidrug resistant if they were non-susceptible to penicillin plus two or more other classes of antimicrobial agents tested.
Strain selection for MLST
Isolates were selected from cases reported in 2007. PCV serotypes (1, 3, 4, 5, 6A, 6B, 7F, 9 V, 14, 18C, 19A, 19F and 23F) and 6C were grouped by patient age category (children <5 years [n = 927] and older individuals ≥5 years [n = 1 510]). Sampling was conducted as follows: for serotypes represented by ≤100 isolates per serotype, per age category, all isolates were included for characterization. Due to the large numbers of isolates available for certain serotypes (4, 6A, 6B, 14, 19F, 23F) sampling was rationalized by selecting 50% where numbers exceeded 100 (per serotype and per age category) (Table 1). Isolates were sorted by collection date and every second isolate was selected. In addition, all non-PCV serotype isolates causing disease in 2007 among children <2 years of age were analyzed (n = 134).
MLST
DNA was extracted directly from 100 µl of stored culture in 10% skim milk using a MagNA Pure LC 2.0 instrument with the MagNA Pure LC DNA Isolation kit III (Roche, Mannheim, Germany) according to the manufacturer’s instructions. Extracted DNA was eluted into 100 µl of elution buffer and stored at −20°C. If poor quality sequencing results were obtained, DNA was re-extracted by boiling pure overnight cultures suspended in 100 µl of sterile deionized water for 10 minutes at 95°C. MLST was performed as previously described [20], using primer pairs described by the Centers for Diseases Control and Prevention, Atlanta, USA (http://www.cdc.gov/ncidod/biotech/strep/alt-MLST-primers.htm) or primers listed on the MLST website (http://spneumoniae.mlst.net/). MLST genes were sequenced in one direction using the Big Dye Terminator v3.1 cycle sequencing kit (Applied Biosystems [ABI], Foster City, CA, USA), and products were analyzed using the ABI 3500 genetic analyzer. Sequences were aligned and analyzed using Lasergene software v9 (DNASTAR, Wisconsin, USA) and compared with sequences on the global MLST database (http://spneumoniae.mlst.net/) to assign allele numbers and ST. For new alleles, amplicons were sequenced in both directions and trace files were submitted to the MLST database curator. New profiles arising as a result of new alleles were also submitted to the curator for assignment of a sequence type.
Data analysis
The eBURST v3 algorithm available on the MLST website was used to determine relationships between sequence types (ST) and to generate population snapshots [21]. The minimum number of identical loci was set at 0 for group definition. Single- (SLV) and double-locus (DLV) variants were defined as ST’s differing from each other by one and two alleles, respectively. The default bootstrap (1 000) was used to assign the group founder ST. A clonal complex was defined as a group of ST’s sharing at least 6 of 7 identical alleles with another ST in the group. Clonal complexes were named after the group founder ST as predicted by eBURST when using all data available in the MLST database. Singletons were defined as unrelated sequence types that did not belong to any clonal complex. Isolates of the same clonal complex or sequence type but with different serotypes were presumed to be capsular switch variants.
Sequence type diversity by serotype and age category (children <5 years of age and individuals ≥5 years) was assessed using Simpson’s index of diversity (D). The value of D ranges from 0 to 1, and the greater the value the greater the sample diversity. Analysis was performed through an online tool for diversity and partition congruency calculations (http://darwin.phyloviz.net/ComparingPartitions/index.php?link=Tool). For this assessment, only PCV serotypes were included.
Results
National IPD surveillance
A total of 4 733 cases of IPD were reported from January through December 2007, of which, 70% (3 329/4 733) had viable isolates available for further characterization. The majority of cases with viable isolates were diagnosed from positive blood cultures (1 879/3 329, 56%), followed by CSF (1 005/3 329, 30%). In addition, isolates were recovered from both blood and CSF (247/3 329, 7%), pleural fluid (1 44/3 329, 4%) or other normally sterile site specimens (54/3 329, 2%).
Among cases with known gender (97% [3 241/3 329]), females and males were equally represented (1 638/3241, 51% and 1 603/3241, 49%, respectively). Age was known for 96% (3 194/3 329) of cases with viable isolates, and of these, 34% (1 084/3 194) were children <5 years, 10% (333/3 194) were older children 5–17 years and 56% (1 777/3 194) were adults ≥18 years. Overall, PCV serotypes were responsible for 86% (927/1 084), 82% (273/333) and 69% (1 235/1 777) of IPD in children <5 years, older children 5–17 years and adults ≥18 years, respectively. Serotypes 14, 6B and 6A were the most prevalent serotypes among children <5 years accounting for 41% (446/1 084), while 1, 14 and 6B were the most common serotypes among older children 5–17 years (135/333, 41%) and 1, 19A and 4 were prevalent among individuals ≥18 years (505/1 777, 28%) (Figure 1). Among children <2 years, serotype 15B and 8 were the most common serotypes among non-PCV serotypes, and represented 16% (22/134) and 15% (20/134) of the isolates, respectively. The remaining isolates (69%, 92/134) belonged to 32 serotypes. Non-susceptibility to penicillin was detected in 43% (1429/3 329), of which 39% (562/1 429) were multidrug resistant. HIV status was known for 32% (1 064/3 329) of cases with viable isolates, the majority of whom were HIV infected (816/1 064, 77%). Overall, PCV serotypes were responsible for 77% (633/816) and 81% (186/231) of IPD in HIV infected and uninfected cases, respectively (p = 0.002).
‘Other’ indicates serotypes not included in pneumococcal conjugate vaccines (PCV-13 and 6C).
Multilocus sequence typing (MLST)
A total of 1 461 isolates representing PCV serotypes and 6C (1 461/2 437, 60%) were selected for MLST (Table S1), of which complete results were available for 73% (1 064/1 461). Poor sequence data or PCR amplification failure for some of the genes resulted in incomplete MLST results for some isolates, 43% (169/397) of which were from children <5 years. MLST results were available for 33% (269/816) of isolates from HIV infected and 37% (85/231) of isolates from HIV uninfected individuals. All 134 isolates representing non-PCV serotypes, from children <2 years of age were selected for characterization and complete results were obtained for 55% (74/134) of these.
MLST genotypes
Data are summarized in Table 1 and eBURST population snapshots, by serotype, are presented as supporting information (Figures S1 to S14). For PCV serotypes, similar clonal complexes were found among isolates from children and older individuals and there were no differences in sequence type diversities between the two age groups (data not shown). Simpson’s diversity indices were therefore calculated for each serotype using combined sequence type data from isolates of children and adults (Figure 2). The majority of serotypes were heterogeneous, however, serotypes 1 and 5 were the least diverse (D = 0.38 and 0.55, respectively) while 6A, 6B, 18C and 23F were the most diverse serotypes, with D ranging from 0.97 for 18C and 23F to 0.98 for 6A and 6B. The majority of serotype 1 isolates were CC217 (143/146, 98%). CC458, which was the largest clonal complex among serotype 3 isolates, represented 60% (56/93) of the serotype 3 isolates. The main clonal complex among serotype 4 isolates was CC32 (31/81, 38%). All serotype 5 isolates were CC289. For serotype 6A, 3 common clonal complexes were identified, of which the most prevalent, CC473, represented 39% (37/95) of the isolates. Several complexes were identified among serotype 6B isolates, of which CC2421 (26/113, 23%) and CC156 (26/113, 23%) were the most common. CC2185 represented majority of serotype 6C isolates (9/11, 82%). CC218 represented 72% (21/29) of serotype 7F isolates. The major clonal complex (CC4881) among serotype 9 V isolates represented 57% (40/70) of the isolates. Of the clonal complexes identified among serotype 14 isolates, CC230 was the largest and represented 43% (49/115) of the isolates. Serotype 18C isolates were differentiated into 4 main clonal complexes. The largest clonal complex among serotype 18C isolates (CC1016) represented 30% (17/57) of the isolates. The majority of serotype 19A isolates were CC2062 (109/131, 83%). CC320 was the most prevalent clonal complex among serotype 19F isolates (23/46, 50%). For serotype 23F, CC242 (25/59, 42%) and CC156 (12/59, 20%) were the major clonal complexes. For non-PCV serotypes, ST7052 which belong to CC346, was predominant (7/12, 58%) among serotype 15B isolates, while ST53 (CC62) was prevalent (9/14, 64%) among serotype 8 isolates (Table 2).
Simpson’s index of diversity was used to assess diversity of isolates that caused invasive pneumococcal disease in 2007 among individuals of all ages, South Africa.
Antimicrobial susceptibility
For those PCV isolates with complete MLST profiles, 640 belong to clonal complexes/singletons representing at least one penicillin non-susceptible isolate (Table 1). Reduced susceptibility to penicillin was detected in 481 of those 640 isolates, of which 192 were multidrug resistant. Serotype 14, 19A and 19F had the largest proportion of isolates that were non-susceptible to penicillin (95% [109/115], 91% [119/131] and 89% [41/46], respectively) (Table 1 and 3). Non-susceptible isolates were common among serotype 14 and 19F isolates belonging to multiple clonal complexes. Except for one serotype 6B isolate, all other isolates belonging to CC230 (n = 49 for serotype 14 and 1 each for 19F and 23F) were multidrug resistant. All isolates belonging CC320 were also multidrug resistant (n = 23 and 1 isolate for serotype 19F and 23F, respectively).
Of the 134 non-PCV serotype isolates recovered from children <2 years, 16% (21/134) were non-susceptible to penicillin and, of these, 19% (4/21) were multidrug resistant. The majority (14/21, 67%) of penicillin non-susceptible non-PCV isolates belonged to serotype 15B and were also non-susceptible to trimethoprim/sulphamethoxazole (13/14, 93%) but none were multidrug resistant. Of the 74 isolates with MLST results, 9 were non-susceptible to penicillin, 8 of which belonged to serotype 15B (Table 2). A non-typeable isolate represented by ST344 was multidrug resistant.
Discussion
Serotypes with low carriage detection rates such as serotype 1 are normally less diverse while the opposite is true for serotypes that are associated with carriage. Consistent with the literature, our serotype 1 and 5 isolates were the least diverse while the remaining PCV serotypes were more heterogeneous. Several globally disseminated clones described by the pneumococcal molecular epidemiology network (PMEN), were identified among our isolates [22]. Sweden1-27 (CC217) represented nearly all (98%) of our serotype 1 isolates. South Africa6B-8 (ST185), which was identified among our serotype 6B isolates recovered in the 1990’s, continues to circulate [23]. Denmark14-32 (CC230) was the most prevalent clone among serotype 14 isolates. ST271, which is a single-locus variant of the multidrug resistant Taiwan19F-14 (ST236), was prevalent among our serotype 19F isolates.
Some serotypes were represented mainly by clonal complexes that appear to be uncommon in other parts of the world. CC458 was the most common genotype among serotype 3 isolates, while Netherlands3-31 (CC180) only represented 15% of our isolates. Our earlier study on a random selection of South African serotype 3 isolates, collected from 2000 through 2005 [24] also showed the predominance of CC458 which accounted for 54% of serotype 3 isolates tested (84/155), while CC180 represented only 12% (18/155) of the isolates. CC458 has also been documented among serotype 3 isolates from Ghana, Egypt, Israel, Costa Rica and Lithuania [25]–[27]. CC5410, which represented the majority of our serotype 4 isolates, has only been reported in Mozambique [28]. Interestingly, the majority (83%) of our serotype 19A isolates were CC2062, which, to our knowledge, has only been reported from South Africa [28]. In our earlier study, CC2062 was a novel genotype identified among 20% (5/25) of multidrug resistant serotype 19A isolates that caused IPD from 1998 through 2004 [29]. Other than CC156, which represented only 7% (9/131) of our serotype 19A isolates, no other common serotype 19A genotype including CC199 (predominant in the US), CC230 (common in Spain and Kenya) and CC320 (common in Spain, Asia and France and the US), was identified among our serotype 19A isolates [30]–[33]. CC320, which was responsible for the post-conjugate vaccine increase in penicillin-resistant serotype 19A in the US [34], represented half of our serotype 19F isolates all of which were penicillin and multidrug resistant.
Although not always the case, isolates with the same sequence type (or clonal complex) are usually of the same serotype [35]. A new or unusual serotype-sequence type combination is therefore often an indication of capsular switching. Pneumococci are highly transformable and a single in vivo recombination event that resulted in both a change from PCV7 serotype 4 to non-PCV7 19A and penicillin non-susceptibility has previously been documented in the US [36]. We identified clonal complexes among our PCV serotypes that expressed more than one capsular type. Although some of these clonal complexes have previously been reported in more than one serotype, we identified new serotype-ST combinations. CC156 has been identified among isolates belonging to multiple serotypes including 6A, 6B, 6C, 19A, 19F and 23F [28]. Similarly, CC156 was represented among our isolates belonging to multiple serotypes but was more common among serotype 6A and 23F isolates. The majority of our serotype 7F isolates were CC218, which, based on data in the global MLST database at the time of writing, is mainly associated with serotype 12F [28]. Nevertheless, this clonal complex has previously been identified among 7F isolates from other parts of the world including Saudi Arabia, Germany and Chile [28]. CC230 is associated with multiple serotypes but has not been reported among serotype 6B isolates [28], however four of our serotype 6B isolates were CC230. Although ST63 is a globally disseminated serotype 15A clone, it has also been identified among other serotypes including serotype 14 [28]. It is therefore not surprising that more than 20% of our serotype 14 isolates were CC63. We also identified clonal complexes that are normally associated with PCV serotypes among some of our non-PCV isolates. A limitation of our study was that our data only includes isolates collected over one year; hence we were unable to determine how long these unusual serotype-clonal complex combinations have been circulating in South Africa.
Since genotypic changes may also occur temporally, either in the absence of any known pressure or due to other selective pressures such as antibiotic use, we were not able to determine such changes with only one year of MLST data. Changes associated with antibiotic pressure have been reported in South Korea and Israel before they routinely used PCV [11], [12]. In South Korea, an increase in pneumococcal disease caused by antibiotic resistant serotype 19A was associated with expansion of ST320. In Israel, penicillin and multidrug resistant pneumococci recovered from patients with acute otitis media increased significantly from less than 10% to greater than 50% and was associated with the introduction and proliferation of two clones (ST276 [CC230] and ST2928 [CC63]) not previously associated with drug resistance. Denmark and Scotland reported fluctuations in pneumococcal genotypes at times of limited antibiotic and PCV use [10], [37]. An increase in the proportion of serotype 19A disease was reported in Denmark while an increase in serotype 1, associated with ST306, was reported in Scotland [10], [37]. Our South African surveillance data for children <5 years also showed a significant increase in penicillin non-susceptible 19A from 18% in 2003 to 78% in 2008, prior to the introduction of PCV [17]. Although limited genotypic data are available for our serotype 19A isolates, our current and previous studies show that the majority of our 2007 penicillin non-susceptible serotype 19A isolates (100/119, 84%) were CC2062, which has circulated among our multidrug-resistant 19A isolates for at least a decade [29]. The increase in penicillin resistance from 2003 to 2008 is therefore likely due to acquisition of pbp genes from other penicillin non-susceptible strains and requires further investigation.
PMEN was initially established for surveillance of globally disseminated pneumococcal clones that are antibiotic resistant. The majority of South African non-susceptible isolates were related to PMEN clones. All serotype 14 and 23F isolates related to Denmark14-32 were multidrug resistant. All serotype 19F and 23F isolates belonging to CC320, which also represent Taiwan19F-14 (ST236), were also multidrug resistant. The majority (80%) of serotype 14 isolates related to Sweden15A-25 (CC63) were multidrug resistant. The majority (73%) of serotype 6B isolates related to South Africa6B-8 [ST185 (CC156)] were non-susceptible to penicillin.
The biggest limitation of PCV is its inability to protect against pneumococcal disease caused by non-vaccine serotypes and hence the possibility of replacement disease. In our study, serotype 15B was the most common serotype among non-PCV isolates from children <2 years and had the highest proportion of isolates that were non-susceptible to penicillin. Moreover, serotype 15B and 8 were among the top 20 serotypes that caused IPD among South African children in 2003 through 2008, prior to the introduction of PCV [17], and may therefore be poised to fill the niche vacated by PCV serotypes. Increases in the number of serogroup 15 cases have been reported in eight hospitals in the US following routine vaccination with PCV [38]. Although serotypes 7F and 22F were not common among disease-causing isolates in South Africa, they have previously been reported as replacement serotypes elsewhere and therefore need to be closely monitored [4], [5].
We were not able to compare sequence types between HIV and non-HIV patients as HIV status was only available for IPD cases reported from sentinel sites (approximately 30 of the 187 laboratories participating in the surveillance program). As a result, HIV status for the majority of cases was unknown and hence there were low numbers of cases with available viable isolates (n = 816 for HIV infected and n = 231 HIV uninfected cases and, of these, 269 and 85 had MLST data, respectively).
This study represents the baseline genetic structure within pneumococcal serotypes prior to routine PCV use in South Africa. The predominance of genotypes such as CC217 (serotype 1), CC458 (serotype 3), CC5410 (serotype 4) and CC2062 (serotype 19A) that appear to be uncommon in other parts of the world highlights the importance of molecular characterisation of pneumococci from South Africa. Moreover, major sequence types circulating among invasive and carriage isolates from other African countries are also different to major sequence types identified globally [39], [40]. These data highlight the importance of molecular data from the African continent. Comparing pre-PCV to post-PCV genotypes (work currently in progress) will potentially provide insight into the impact of PCV on our pneumococcal population.
Supporting Information
Figure S1.
Population snapshot depicting clonal relationships between sequence types of serotype 1 pneumococci causing invasive disease in South Africa, 2007 (n = 146).
https://doi.org/10.1371/journal.pone.0107666.s001
(PDF)
Figure S2.
Population snapshot depicting clonal relationships between sequence types of serotype 3 pneumococci causing invasive disease in South Africa, 2007 (n = 93).
https://doi.org/10.1371/journal.pone.0107666.s002
(PDF)
Figure S3.
Population snapshot depicting clonal relationships between sequence types of serotype 4 pneumococci causing invasive disease in South Africa, 2007 (n = 81).
https://doi.org/10.1371/journal.pone.0107666.s003
(PDF)
Figure S4.
Population snapshot depicting clonal relationships between sequence types of serotype 5 pneumococci causing invasive disease in South Africa, 2007 (n = 18).
https://doi.org/10.1371/journal.pone.0107666.s004
(PDF)
Figure S5.
Population snapshot depicting clonal relationships between sequence types of serotype 6A pneumococci causing invasive disease in South Africa, 2007 (n = 95).
https://doi.org/10.1371/journal.pone.0107666.s005
(PDF)
Figure S6.
Population snapshot depicting clonal relationships between sequence types of serotype 6B pneumococci causing invasive disease in South Africa, 2007 (n = 113).
https://doi.org/10.1371/journal.pone.0107666.s006
(PDF)
Figure S7.
Population snapshot depicting clonal relationships between sequence types of serotype 6C pneumococci causing invasive disease in South Africa, 2007 (n = 11).
https://doi.org/10.1371/journal.pone.0107666.s007
(PDF)
Figure S8.
Population snapshot depicting clonal relationships between sequence types of serotype 7F pneumococci causing invasive disease in South Africa, 2007 (n = 29).
https://doi.org/10.1371/journal.pone.0107666.s008
(PDF)
Figure S9.
Population snapshot depicting clonal relationships between sequence types of serotype 9 V pneumococci causing invasive disease in South Africa, 2007 (n = 70).
https://doi.org/10.1371/journal.pone.0107666.s009
(PDF)
Figure S10.
Population snapshot depicting clonal relationships between sequence types of serotype 14 pneumococci causing invasive disease in South Africa, 2007 (n = 115).
https://doi.org/10.1371/journal.pone.0107666.s010
(PDF)
Figure S11.
Population snapshot depicting clonal relationships between sequence types of serotype 18C pneumococci causing invasive disease in South Africa, 2007 (n = 57).
https://doi.org/10.1371/journal.pone.0107666.s011
(PDF)
Figure S12.
Population snapshot depicting clonal relationships between sequence types of serotype 19A pneumococci causing invasive disease in South Africa, 2007 (n = 131).
https://doi.org/10.1371/journal.pone.0107666.s012
(PDF)
Figure S13.
Population snapshot depicting clonal relationships between sequence types of serotype 19F pneumococci causing invasive disease in South Africa, 2007 (n = 46).
https://doi.org/10.1371/journal.pone.0107666.s013
(PDF)
Figure S14.
Population snapshot depicting clonal relationships between sequence types of serotype 23F pneumococci causing invasive in South Africa, 2007 (n = 59).
https://doi.org/10.1371/journal.pone.0107666.s014
(PDF)
Table S1.
Pneumococcal conjugate vaccine serotype isolate selection for multilocus sequence typing, by patient age category and serotype, South Africa, 2007.
https://doi.org/10.1371/journal.pone.0107666.s015
(PDF)
Acknowledgments
We thank GERMS-SA (Group for Enteric, Respiratory and Meningeal Surveillance) for coordinating national surveillance of IPD in South Africa. We are grateful to CRDM laboratory staff for their technical input and all clinical and laboratory staff throughout South Africa for submitting case reports and isolates for the GERMS-SA national surveillance program, and all patients whose isolates were used in the study.
We acknowledge the use of the pneumococcal MLST database which is located at Imperial College London and is funded by the Wellcome Trust.
Author Contributions
Conceived and designed the experiments: AVG KPK MDP NW. Performed the experiments: KMN LDG. Analyzed the data: KMN LDG MDP. Contributed reagents/materials/analysis tools: AVG MDP. Contributed to the writing of the manuscript: KMN LDG AVG KPK MDP NW.
References
- 1. Jones N, Huebner R, Khoosal M, Crewe-Brown H, Klugman K (1998) The impact of HIV on Streptococcus pneumoniae bacteraemia in a South African population. AIDS 12: 2177–2184.
- 2. Madhi SA, Petersen K, Madhi A, Wasas A, Klugman KP (2000) Impact of human immunodeficiency virus type 1 on the disease spectrum of Streptococcus pneumoniae in South African children. Pediatr Infect Dis J 19: 1141–1147.
- 3. Feldman C, Glatthaar M, Morar R, Mahomed AG, Kaka S, et al. (1999) Bacteremic pneumococcal pneumonia in HIV-seropositive and HIV-seronegative adults. Chest 116: 107–114.
- 4. Pilishvili T, Lexau C, Farley MM, Hadler J, Harrison LH, et al. (2010) Sustained reductions in invasive pneumococcal disease in the era of conjugate vaccine. J Infect Dis 201: 32–41.
- 5. Miller E, Andrews NJ, Waight PA, Slack MP, George RC (2011) Herd immunity and serotype replacement 4 years after seven-valent pneumococcal conjugate vaccination in England and Wales: an observational cohort study. Lancet Infect Dis 11: 760–768.
- 6. Tyrrell GJ, Lovgren M, Chui N, Minion J, Garg S, et al. (2009) Serotypes and antimicrobial susceptibilities of invasive Streptococcus pneumoniae pre- and post-seven valent pneumococcal conjugate vaccine introduction in Alberta, Canada, 2000–2006. Vaccine 27: 3553–3560.
- 7. Munoz-Almagro C, Jordan I, Gene A, Latorre C, Garcia-Garcia JJ, et al. (2008) Emergence of invasive pneumococcal disease caused by nonvaccine serotypes in the era of 7-valent conjugate vaccine. Clin Infect Dis 46: 174–182.
- 8. Pai R, Moore MR, Pilishvili T, Gertz RE, Whitney CG, et al. (2005) Postvaccine genetic structure of Streptococcus pneumoniae serotype 19A from children in the United States. J Infect Dis 192: 1988–1995.
- 9. Beall BW, Gertz RE, Hulkower RL, Whitney CG, Moore MR, et al. (2011) Shifting Genetic Structure of Invasive Serotype 19A Pneumococci in the United States. J Infect Dis 203: 1360–1368.
- 10. Jefferies JM, Smith AJ, Edwards GF, McMenamin J, Mitchell TJ, et al. (2010) Temporal analysis of invasive pneumococcal clones from Scotland illustrates fluctuations in diversity of serotype and genotype in the absence of pneumococcal conjugate vaccine. J Clin Microbiol 48: 87–96.
- 11. Choi EH, Kim SH, Eun BW, Kim SJ, Kim NH, et al. (2008) Streptococcus pneumoniae Serotype 19A in Children, South Korea. Emerg Infect Dis 14: 275–281.
- 12. Dagan R, Givon-Lavi N, Leibovitz E, Greenberg D, Porat N (2009) Introduction and proliferation of multidrug-resistant Streptococcus pneumoniae serotype 19A clones that cause acute otitis media in an unvaccinated population. J Infect Dis 199: 776–785.
- 13. Fenoll A, Granizo JJ, Aguilar L, Gimenez MJ, Aragoneses-Fenoll L, et al. (2009) Temporal Trends of Invasive Streptococcus pneumoniae Serotypes and Antimicrobial Resistance Patterns in Spain from 1979 to 2007. J Clin Microbiol 47: 1012–1020.
- 14. Madhi SA, Cohen C, von Gottberg A (2012) Introduction of pneumococcal conjugate vaccine into the public immunization program in South Africa: translating research into policy. Vaccine 30 Suppl 3C21–7
- 15. Huebner RE, Klugman KP, Matai U, Eggers R, Hussey G (1999) Laboratory surveillance for Haemophilus influenzae type B, meningococcal, and pneumococcal disease. Haemophilus Surveillance Working Group. S Afr Med J 89: 924–925.
- 16. Carvalho MG, Tondella ML, McCaustland K, Weidlich L, McGee L, et al. (2007) Evaluation and improvement of real-time PCR assays targeting lytA, ply, and psaA genes for detection of pneumococcal DNA. J Clin Microbiol 45: 2460–2466.
- 17. von Gottberg A, Cohen C, de Gouveia L, Meiring S, Quan V, et al. (2013) Epidemiology of invasive pneumococcal disease in the pre-conjugate vaccine era: South Africa, 2003–2008. Vaccine 31: 4200–4208.
- 18. Austrian R (1976) The quellung reaction, a neglected microbiologic technique. Mt Sinai J Med 43: 699–709.
- 19.
Clinical and Laboratory Institute (2007) Performance Standards for Antimicrobial Susceptibility Testing. Seventeenth informational supplement 27: 64, 65, 126–128.
- 20. Enright MC, Spratt BG (1998) A multilocus sequence typing scheme for Streptococcus pneumoniae: identification of clones associated with serious invasive disease. Microbiology 144: 3049–3060.
- 21. Feil EJ, Li BC, Aanensen DM, Hanage WP, Spratt BG (2004) eBURST: inferring patterns of evolutionary descent among clusters of related bacterial genotypes from multilocus sequence typing data. J Bacteriol 186: 1518–1530.
- 22.
Aanensen DM, Spratt BG. Multilocus sequence typing network: Pneumococcal mlecular epidemiology network. Available: http://spneumoniae.mlst.net/pmen/pmen.asp. Accessed 2013 June 26.
- 23. Smith AM, Klugman KP (1997) Three predominant clones identified within penicillin-resistant South African isolates of Streptococcus pneumoniae. Microb Drug Resist 3: 385–389.
- 24. Mothibeli KM, du Plessis M, von Gottberg A, de Gouveia L, Adrian P, et al. (2010) An unusual pneumococcal sequence type is the predominant cause of serotype 3 invasive disease in South Africa. J Clin Microbiol 48: 184–191.
- 25. Leimkugel J, Adams FA, Gagneux S, Pfluger V, Flierl C, et al. (2005) An outbreak of serotype 1 Streptococcus pneumoniae meningitis in northern Ghana with features that are characteristic of Neisseria meningitidis meningitis epidemics. J Infect Dis 192: 192–199.
- 26. Porat N, Soley C, Marengolciene MM, Greenberg D, Givon-Lavi N, et al. (2008) An international serotype 3 clone causing pediatric noninvasive infections in Israel, Costa Rica, and Lithuania. Pediatr Infect Dis J 27: 709–712.
- 27. Wasfy MO, Pimentel G, Abdel-Maksoud M, Russell KL, Barrozo CP, et al. (2005) Antimicrobial susceptibility and serotype distribution of Streptococcus pneumoniae causing meningitis in Egypt, 1998–2003. J Antimicrob Chemother 55: 958–964.
- 28.
Aanensen DM, Spratt BG. Multilocus sequence typing network: Streptococcus pneumoniae advance query form. Available: http://spneumoniae.mlst.net/sql/burstspadvanced.asp. Accessed 2013 June 6.
- 29.
Mothibeli KM, du Plessis M, von Gottberg A, Smith A, Klugman KP (11-27-2006) Molecular Epidemiology of multiply resistant pneumococcal serotype 19A isolates from South Africa. National Institute for Communicable Disease (NICD) Academic day. 27 & 28 November 2006. Poster presentation. Poliomyelitis Research Foundation Training Centre, NICD, Sandringham, South Africa.
- 30. Beall B, McEllistrem MC, Gertz RE Jr, Wedel S, Boxrud DJ, et al. (2006) Pre- and post vaccination clonal compositions of invasive pneumococcal serotypes for isolates collected in the United States in 1999, 2001, and 2002. J Clin Microbiol 44: 999–1017.
- 31. Mahjoub-Messai F, Doit C, Koeck JL, Billard T, Evrard B, et al. (2009) Population snapshot of Streptococcus pneumoniae serotype 19A isolates before and after introduction of seven-valent pneumococcal Vaccination for French children. J Clin Microbiol 47: 837–840.
- 32. Picazo J, Ruiz-Contreras J, Hernandez B, Sanz F, Gutierrez A, et al. (2011) Clonal and clinical profile of Streptococcus pneumoniae serotype 19A causing pediatric invasive infections: a 2-year (2007–2009) laboratory-based surveillance in Madrid. Vaccine 29: 1770–1776.
- 33. Shin J, Baek JY, Kim SH, Song JH, Ko KS (2011) Predominance of ST320 among Streptococcus pneumoniae serotype 19A isolates from 10 Asian countries. J Antimicrob Chemother 66: 1001–1004.
- 34. Moore MR, Gertz RE Jr, Woodbury RL, Barkocy-Gallagher GA, Schaffner W, et al. (2008) Population snapshot of emergent Streptococcus pneumoniae serotype 19A in the United States, 2005. J Infect Dis 197: 1016–1027.
- 35. Brueggemann AB, Griffiths DT, Meats E, Peto T, Crook DW, et al. (2003) Clonal relationships between invasive and carriage Streptococcus pneumoniae and serotype- and clone-specific differences in invasive disease potential. J Infect Dis 187: 1424–1432.
- 36.
Brueggemann AB, Pai R, Crook DW, Beall B (2007) Vaccine Escape Recombinants Emerge after Pneumococcal Vaccination in the United States. PLoS Pathog 3: Epub.
- 37. Harboe ZB, Benfield TL, Valentiner-Branth P, Hjuler T, Lambertsen L, et al. (2010) Temporal trends in invasive pneumococcal disease and pneumococcal serotypes over 7 decades. Clin Infect Dis 50: 329–337.
- 38. Gonzalez BE, Hulten KG, Lamberth L, Kaplan SL, Mason EO Jr (2006) Streptococcus pneumoniae serogroups 15 and 33: an increasing cause of pneumococcal infections in children in the United States after the introduction of the pneumococcal 7-valent conjugate vaccine. Pediatr Infect Dis J 25: 301–305.
- 39. Antonio M, Dada-Adegbola H, Biney E, Awine T, O’Callaghan J, et al. (2008) Molecular epidemiology of pneumococci obtained from Gambian children aged 2–29 months with invasive pneumococcal disease during a trial of a 9-valent pneumococcal conjugate vaccine. BMC Infect Dis 8: 81
- 40. Brueggemann AB, Muroki BM, Kulohoma BW, Karani A, Wanjiru E, et al. (2013) Population Genetic Structure of Streptococcus pneumoniae in Kilifi, Kenya, Prior to the Introduction of Pneumococcal Conjugate Vaccine. PLoS One 8: e81539.