Eur J Clin Microbiol Infect Dis DOI 10.1007/s10096-015-2342-9
ARTICLE
Distribution of pilus islands and alpha-like protein genes of group B Streptococcus colonized in pregnant women in Beijing, China B. Lu & D. Wang & H. Zhou & F. Zhu & D. Li & S. Zhang & Y. Shi & Y. Cui & L. Huang & H. Wu
Received: 14 October 2014 / Accepted: 27 January 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Group B Streptococcus (GBS) is one of the major pathogens of severe newborn sepsis and meningitis. Understanding its regional molecular epidemiology is helpful for regulating efficient prevention practice. A total of 160 GBS strains were collected from colonized pregnant women in six hospital settings in Beijing, China. Polymerase chain reaction (PCR) assays were used to identify the pilus island (PI), alp genes profiling of the alpha-like protein family, and capsular polysaccharide (cps) serotyping. The clonal relationships between strains were investigated using multilocus sequence typing (MLST). All isolates carried at least one pilus island. The most frequently detected pilus island was PI-2a alone (70 isolates, 43.8 %). The most prevalent alp gene was
Electronic supplementary material The online version of this article (doi:10.1007/s10096-015-2342-9) contains supplementary material, which is available to authorized users. B. Lu (*) : F. Zhu : D. Li : S. Zhang : Y. Shi : Y. Cui Department of Laboratory Medicine, Civil Aviation General Hospital, No. 1 Gaojing Street, Chaoyang District, Beijing 100123, People’s Republic of China e-mail: [email protected] D. Wang : H. Zhou National Institute for Communicable Disease Control and Prevention, and State Key Laboratory for Infectious Disease Prevention and Control, Chinese Center for Disease Control and Prevention, Beijing 102206, China D. Wang : H. Zhou Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Hangzhou 310003, China L. Huang Department of Laboratory Medicine, First Hospital, Peking University, Beijing 100034, China H. Wu BGI Tianjin, Tianjin 300308, China
rib (60 isolates, 37.5 %). Moreover, a strong association was noted between alp genes, serotyping, and pilus island profiles. The GBS isolates under study hinted similar molecular epidemical characteristics in Beijing to those reported worldwide, but having their regional distributional features.
Introduction Invasive neonatal disease due to Group B Streptococcus (GBS), colonized asymptomatically in the gastrointestinal and genitourinary tracts of 10–30 % of pregnant women, is a heavy health burden throughout the world [1]. Intrapartum antibiotic prophylaxis has been effectively used to prevent neonatal early-onset GBS diseases, but the incidence of lateonset infections remains almost unchanged [1]. Babies can obtain maternal opsonizing antibodies via the placenta. Therefore, a safe, effective vaccine in women of childbearing age, based on the epidemiological characteristics of regional GBS strains, appears to be a possible solution to prevent mother-to-infant transmission [1, 2]. The protective role in humans of capsular polysaccharide (cps)-specific antibodies of GBS has been confirmed for decades [2]. However, the existence of non-typable isolates makes cps-based vaccine incapable of covering all GBS strains. To date, sortasedependent pilus-like structures have been identified on the surface of GBS. Three pilus island (PI) alleles, PI-1, PI-2a, and PI-2b, have been detected and their proteins explored as promising vaccine candidates and potential virulence factors [3–7]. Furthermore, a series of alpha-like proteins (alp) were demonstrated to play an important role in GBS pathogenesis and also regarded as vaccine candidates [8–10]. These proteins, localized on the surface of GBS, are encoded by genes of alpha-C, epsilon, alp2/3, rib, and alp4, respectively, whose
Eur J Clin Microbiol Infect Dis
distributions constitute the main epidemiological characteristics of the local GBS isolates [11]. The distribution of pilus island and alp genes of GBS has been studied in South Africa, America, Egypt, Ireland, German, and Italy [3, 4, 9, 12–15]. However, scarce data are available from mainland China. Therefore, the aim of our work is to elucidate the distributions of pilus island and alp genes, and their associations with other molecular features of the colonized maternal GBS strains in Beijing, China.
Materials and methods
2013 to May 2014, which is a clinical laboratory center collecting specimens from Beijing Amcare Women’s & Children’s Hospital (DY-1), Beijing HarMoniCare Women and Children’s Hospital (DY-2), Oasis International Hospital (DY-3), Beijing Anhui Yihe Women & Children’s Hospital (DY-4), and Beijing Tongren Hospital (TRH) during 2012. The isolates were sent to the Department of Clinical Microbiology of CAGH for further confirmation by the CAMP test and the VITEK-2 automated microbiology system (bioMérieux, France). Moreover, all the GBS colonies were confirmed by 16S rRNA gene sequencing using the primers BAK11w and BAK2, as we described previously [16].
Bacterial isolates A total of 160 isolates, previously identified as GBS in clinical microbiology laboratories, were recovered from −80 °C storage. The strains under analysis were isolated from vaginal/ rectal swabs of pregnant women at 35–37 weeks of gestation who visited the Civil Aviation General Hospital (CAGH) during the periods 2009 to 2011 and September 2013 to January 2014, Beijing Deyi Shuangxin Clinics during the period June
Table 1 Distribution of clonal complexes (CCs), pilus island (PI), serotypes, and alp genes of 160 group B Streptococcus (GBS) strains carried by pregnant women in Beijing, China
NT non-typable; N negative, CC clonal complex; PI pilus island a
The percentages in parentheses represent isolates/total GBS isolates (160)
No. of strainsa Serotype Ia Ib II III V VI NT alp genes alpha-C rib epsilon alp2/3 alp4 N CCs CC1 CC19 CC17 CC23 CC12 CC4 Singletons (ST 24, ST485, ST486)
Determination of cps serotyping, pilus island genes, and alp genes profile of GBS isolates The ten known GBS capsular serotypes (Ia, Ib, II–IX) were determined by using multiplex polymerase chain reaction (PCR) assays as described previously [17, 18], and the isolates that failed to type were deemed non-typable (NT). PCR assays were used to identify the PI-1, PI-2a, or PI-2b genes using
No. of strains (%)
PI-2a, no. (%)
PI-2b, no. (%)
PI-1 + PI-2a, no. (%)
PI-1 + PI-2b, no. (%)
160 (100)
70 (43.8)
21 (13.1)
67 (41.9)
2 (1.3)
33 (20.6) 18 (11.3) 10 (6.3) 72 (45.0) 23 (14.4) 1 (0.6) 3 (1.9)
23 (14.4) 6 (3.8) 7 (4.4) 22 (13.8) 10 (6.3) – 2 (1.3)
8 (5.0) 1 (0.6) – 11 (6.9) 1 (0.6) – –
2 (1.3) 11 (6.9) 3 (1.9) 37 (23.1) 12 (7.5) 1 (0.6) 1 (0.6)
– – – 2 (1.3) – – –
13 (8.1) 60 (37.5) 40 (25.0) 35 (21.9) 8 (5.0) 4 (2.5)
7 (4.4) 13 (8.1) 27 (16.9) 17 (10.6) 5 (3.1) 1 (0.6)
– 5 (3.1) 9 (5.6) 3 (1.9) 2 (1.3) 2 (1.3)
6 (3.8) 41 (25.6) 4 (2.5) 15 (9.4) – 1 (0.6)
– 1 (0.6) – – 1 (0.6) –
24 (15.0) 56 (35.0) 14 (8.8)
13 (8.1) 19 (11.9) –
1 (0.6) – 10 (6.3)
10 (6.3) 37 (23.1) 2 (1.3)
– – 2 (1.3)
27 (16.9) 28 (17.5) 5 (3.1) 6 (3.8)
23 (14.4) 12 (7.5) 2 (1.3) 1 (0.6)
1 (0.6) 1 (0.6) 3 (1.9) 5 (3.1)
3 (1.9) 15 (9.4) – –
– – – –
Eur J Clin Microbiol Infect Dis
primer pairs as described previously [11, 13]. Surface alp genes were detected by using a multiplex PCR assay [8, 11], and strains that tested negative for any of the known alp genes were considered negative (N).
MLST, phylogenetic, and epidemiological analyses Multilocus sequence typing (MLST) was analyzed by sequencing seven housekeeping genes, as previously described [19]. Then, the sequence types (STs) and allelic profiles were confirmed by querying the MLST database (http://pubmlst. org/sagalactiae). Briefly, PCR fragments for the abovementioned genes were amplified and sequenced. Each isolate was assigned an ST. BURST analysis was conducted to analyze clonal complexes (CCs) using eBURST V3 (http://eburst.mlst.net). The GBS isolates were assigned to one of the CCs if they shared five or more alleles with the predominant ST. An ST not identified in any cluster was assigned as a singleton. BioNumerics software version 5.1 (Applied Maths, Belgium) was used to create minimum spanning trees to illustrate the relationships between MLST, CCs, pilus islands, and alp genes.
Fig. 1 Correlation between clonal complexes (CCs), sequence type (ST), alp genes, and pilus islands (PIs) of the group B Streptococcus (GBS) isolates. Minimum spanning tree analysis of GBS isolates according to ST, demonstrating the relationships between 160 carrier isolates collected from six hospital settings. Note: In the minimum spanning tree, the STs are displayed as circles. The size of each circle indicates the number of isolates within this particular type. The founder ST was defined as the ST with the greatest number of single-locus variants. Pilus islands profiling
Statistical analysis The relationships between CCs and related genes were analyzed using Spearman’s correlation with SPSS 17.0 software. Each pair of variables (genes) was compared by correlation measures. The correlation coefficients were obtained, and the correlation was considered significant at the 0.01 or 0.05 levels.
Results Serotyping and MLST All but three isolates were serotyped. The dominant serotypes were cps III, Ia, V, and Ib, which accounted for 91.3 % of all strains (Table 1). A total of 21 individual STs were distinguished, and 18 STs were clustered in six CCs (Fig. 1), representing 96.3 % of all isolates. The most prevalent complex was CC19, followed by CC12, CC23, CC1, CC17, and CC4, which, except for the minor complex CC4, were also highly prevalent in the GBS-related investigations in other regions [12]. Only 14 CC17 strains were identified, including ST17, ST61, ST64, and ST146, all of which carried serotype III.
and alp genes are represented by different colors in the lower and upper figures, respectively. STs that vary by one allele in their multilocus sequence typing (MLST) profiles (single-locus variants) are arranged in circles around the primary founder ST. Heavy solid lines represent singlelocus variants; light solid lines represent double-locus variants; heavy dotted lines represent triple-locus variants; light dotted lines represent quadruple-locus variants. The major CCs are indicated in the figure
Eur J Clin Microbiol Infect Dis
Eur J Clin Microbiol Infect Dis
Fig. 2
Dendrogram constructed from the multilocus sequence typing (MLST) profiles of seven housekeeping genes of GBS isolates in six hospital settings. CAGH Civil Aviation General Hospital; DY-1 Beijing Amcare Women’s & Children’s Hospital; DY-2 Beijing HarMoniCare Women and Children’s Hospital; DY-3 Oasis International Hospital; DY-4 Beijing Anhui Yihe Women & Children’s Hospital; TRH Beijing Tongren Hospital
positive strains were present in serotype III (55/72, 76.4 %). There was a close association between CCs and alp genes, e.g., the majority of CC1, CC19, CC12, and CC23 isolates had alp2/3 (18/24, 75.0 %), rib (49/56, 87.5 %), alphaC (13/28, 46.4 %), and epsilon (23/27, 85.2 %), respectively (Fig. 1, Table S1).
Pilus islands Discussion Of the 160 isolates of GBS, at least one of the three pilus islands was detected, alone or in combinations (Figs. 1 and 2, Table 1). All strains carried either PI-2a (in 85.6 % of strains) or PI-2b (in 14.4 % of strains). The PI-1 locus was detected in 43.1 % of the strains. The most frequently found pilus island was PI-2a alone, followed by PI-1 + PI-2a, PI-2b alone, and PI1 + PI-2b. There was not an even distribution of pilus islands amongst GBS serotypes. However, a close relationship was observed between certain pilus islands and the cps serotype. Most Ia isolates (23/33, 69.7 %) contained only the PI-2a island alone, whereas the serotype III (37/72, 51.4 %), Ib (11/18, 61.1 %), and V (12/23, 52.2 %) strains mainly contained the combination PI-1 + PI-2a. Moreover, relationships between a specific CC and pilus island, such as CC19 and PI-1 + PI-2a (37/56, 66.1 %), CC23 and PI-2a (23/27, 85.2 %), and CC17 and PI-2b (10/14, 71.4 %), were also noticed (Table S1). PI-2bpositive strains were present in almost all CCs, except for CC19. The PI-1 + PI-2b combination was present in only two strains, both contained in CC17 (Fig. 2). The alp family genes Of all the strains, rib was predominant, accounting for 37.5 % (60/160). Four isolates were negative to any known alp genes. The presence of specific alp genes in relation to the cps serotype was noticed (Table 2). Of the 40 epsilon-positive strains isolated, 30 (75 %) corresponded to serotype Ia; 55 ribTable 2
Correlation between the surface protein genes (alp) and serotypes among the strains of GBS
No. of strainsa Serotype Ia Ib II III V VI NT
No. of strains (%)
alpha-C, no. (%)
rib, no. (%)
epsilon, no. (%)
alp2/3, no. (%)
alp4, no. (%)
N, no. (%)
160 (100 %)
13 (8.1)
60 (37.5)
40 (25.0)
35 (21.9)
8 (5.0)
4 (2.5)
33 (20.6) 18 (11.3) 10 (6.3) 72 (45.0) 23 (14.4) 1 (0.6) 3 (1.9)
– 6 (3.8) 1 (0.6) – 5 (3.1) – 1 (0.6)
– 1 (0.6) 1 (0.6) 55 (34.4) 2 (1.3) – 1 (0.6)
30 (18.8) – – 8 (5.0) 2 (1.3) – –
3 (1.9) 11 (6.9) 7 (4.4) 2 (1.3) 12 (7.5) – –
– – – 5 (3.1) 2 (1.3) – 1 (0.6)
– – 1 (0.6) 2 (1.3) – 1 (0.6) –
NT non-typable; N negative a
Studies of the molecular epidemiology among GBS colonizing strains collected from a variety of geographical settings might facilitate the efficient prevention of invasive neonatal infections [1, 20]. To our knowledge, this is the first GBS surveillance study conducted in mainland China dealing with the regional distributions of pilus islands, alp genes, serotypes, STs, and CCs in GBS isolates. In the current study, pilus islands are conserved among GBS strains. All GBS strains carried at least one of the three pilus islands. Either PI-2a or PI-2b is found; however, this is not the case for PI-1, which was identified in 43.1 % of all strains. The pilus-based vaccine will cover all cps serotypes (including non-typable isolates, which may not be covered by a cps-based vaccine). This is consistent with prior studies [3, 4]. Apart from being promising vaccine antigens against neonatal infections, the pilus island could also mediate the contact and attachment of GBS to host cells and contribute to invasion of GBS into the infected host [7]. Our data on the distributional features of the pilus island are basically in agreement with those of the colonizing and invasive GBS isolates reported in South Africa (541 vaginal-colonizing GBS isolates and 284 invasive isolates), America (289 clinical GBS isolates), and Portugal/Spain (276 collected from pregnant women, 98 noninvasive isolates among non-pregnant adults, and 524 invasive isolates) [3, 4, 13]. Based on this homogeneity of the epidemiological features, a pilus-based vaccine developed for other developed and developing countries will have similar
The percentages in parentheses represent isolates/total GBS isolates (160)
Eur J Clin Microbiol Infect Dis
coverage in Beijing. However, the GBS strains in the present study showed some regional characteristics, which demonstrated that the distribution of the pilus island varied geographically. First, the most prevalent pilus island in Beijing was PI2a alone, which is different from the data of Ireland, where PI1 + PI-2a (which accounted for 48.6 %) was predominant [12]. The majority of GBS isolates (85.6 %) carried PI-2a, which had been suggested to be relevant to pathogenesis due to its participation in biofilm formation [5]. Second, the combination of PI-1 and PI-2b was frequently identified in the colonizing isolates in studies conducted elsewhere [3, 4, 13]; however, only two strains of this combination were confirmed in the current study. Third, the close relationship of serotype III and PI + PI-2a, as well as serotype Ia and PI-2a, was notable, which may be explained by the fact that cps serotypes were genetically linked to the pilus island. It is reported that GBS strains usually carry at least one of the alp genes, but this is not always the case [8, 9, 12, 14]. Apart from having been proposed as potential vaccine components [21], GBS surface proteins have various functions. They contributed to the pathogenesis of GBS disease and induced protective immunity [22]. Our study showed that 2.5 % of isolates were negative for any alp gene. The presence of a particular alp gene in relation to the serotype was noted, though to a lesser degree than previously documented [23, 24]. The current study showed specific serotype–alp gene combinations (e.g., serotype V–alp2/3; serotype III–rib, serotype Ia–epsilon, serotype II–rib, and serotype Ib–alp2/3), which might be explained by the geographic differences [23]. The strains were distinguished into 21 STs and five main CCs (except for CC4), as have been reported across the world [12]. A tight relationship between CC and alp, via pilus island profiling, was detected, although this was not absolute [11]. For example, the isolates of CC19 typically contained rib, while CC1 and CC23 represented alp2/3 and epsilon, respectively [12]. Moreover, the relationship between specific CCs and pilus island, such as CC19 and PI-1 + PI-2a, and CC23 and PI-2a, was observed. Furthermore, CC-17 strains have been suggested to be more virulent than strains of other genotypes, and prevalently associated to neonatal diseases [12, 14]. The distribution of the vaccine candidate in this cluster is extremely important for vaccine development. In our study, the close relationship between CC17 and PI-2b was noted. In summary, our data show that the main molecular characteristics of GBS isolates in Beijing’s pregnant women are basically similar to that reported from other parts of the world [4, 11, 15]. However, a relatively low frequency of PI + PI-2b combination and high prevalence of rib were notable features. The study will contribute to the development of a local GBS vaccine strategy against neonatal infections.
Acknowledgments This study was partially supported by the Beijing Municipal Science & Technology Commission (BMSTC), People’s Republic of China (no. Z141107002514036). Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.
References 1. Verani JR, McGee L, Schrag SJ; Division of Bacterial Diseases, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention (CDC) (2010) Prevention of perinatal group B streptococcal disease—revised guidelines from CDC, 2010. MMWR Recomm Rep 59:1–36 2. Edwards MS (2008) Group B streptococcal conjugate vaccine: a timely concept for which the time has come. Hum Vaccin 4(6): 444–448 3. Margarit I, Rinaudo CD, Galeotti CL, Maione D, Ghezzo C, Buttazzoni E, Rosini R, Runci Y, Mora M, Buccato S, Pagani M, Tresoldi E, Berardi A, Creti R, Baker CJ, Telford JL, Grandi G (2009) Preventing bacterial infections with pilus-based vaccines: the group B streptococcus paradigm. J Infect Dis 199(1):108–115 4. Martins ER, Andreu A, Melo-Cristino J, Ramirez M (2013) Distribution of pilus islands in Streptococcus agalactiae that cause human infections: insights into evolution and implication for vaccine development. Clin Vaccine Immunol 20(2):313–316 5. Rinaudo CD, Rosini R, Galeotti CL, Berti F, Necchi F, Reguzzi V, Ghezzo C, Telford JL, Grandi G, Maione D (2010) Specific involvement of pilus type 2a in biofilm formation in group B Streptococcus. PLoS One 5(2):e9216 6. Nobbs AH, Rosini R, Rinaudo CD, Maione D, Grandi G, Telford JL (2008) Sortase A utilizes an ancillary protein anchor for efficient cell wall anchoring of pili in Streptococcus agalactiae. Infect Immun 76(8):3550–3560 7. Sharma P, Lata H, Arya DK, Kashyap AK, Kumar H, Dua M, Ali A, Johri AK (2013) Role of pilus proteins in adherence and invasion of Streptococcus agalactiae to the lung and cervical epithelial cells. J Biol Chem 288(6):4023–4034 8. Creti R, Fabretti F, Orefici G, von Hunolstein C (2004) Multiplex PCR assay for direct identification of group B streptococcal αprotein-like protein genes. J Clin Microbiol 42(3):1326–1329 9. Shabayek S, Abdalla S, Abouzeid AM (2014) Serotype and surface protein gene distribution of colonizing group B streptococcus in women in Egypt. Epidemiol Infect 142(1):208–210 10. Gilbert GL (2004) Vaccines for other neonatal infections: are group B streptococcal infections vaccine-preventable? Expert Rev Vaccines 3(4):371–374 11. De Francesco MA, Caracciolo S, Gargiulo F, Manca N (2012) Phenotypes, genotypes, serotypes and molecular epidemiology of erythromycin-resistant Streptococcus agalactiae in Italy. Eur J Clin Microbiol Infect Dis 31(8):1741–1747 12. Meehan M, Cunney R, Cafferkey M (2014) Molecular epidemiology of group B streptococci in Ireland reveals a diverse population with evidence of capsular switching. Eur J Clin Microbiol Infect Dis 33(7):1155–1162 13. Madzivhandila M, Adrian PV, Cutland CL, Kuwanda L, Madhi SA; PoPS Trial Team (2013) Distribution of pilus islands of group B streptococcus associated with maternal colonization and invasive disease in South Africa. J Med Microbiol 62(Pt 2):249–253 14. Brimil N, Barthell E, Heindrichs U, Kuhn M, Lütticken R, Spellerberg B (2006) Epidemiology of Streptococcus agalactiae colonization in Germany. Int J Med Microbiol 296(1):39–44
Eur J Clin Microbiol Infect Dis 15. Gherardi G, Imperi M, Baldassarri L, Pataracchia M, Alfarone G, Recchia S, Orefici G, Dicuonzo G, Creti R (2007) Molecular epidemiology and distribution of serotypes, surface proteins, and antibiotic resistance among group B streptococci in Italy. J Clin Microbiol 45(9):2909–2916 16. Binghuai L, Yanli S, Shuchen Z, Fengxia Z, Dong L, Yanchao C (2014) Use of MALDI-TOF mass spectrometry for rapid identification of group B Streptococcus on chromID Strepto B agar. Int J Infect Dis 27:44–48 17. Lu B, Li D, Cui Y, Sui W, Huang L, Lu X (2014) Epidemiology of Group B streptococcus isolated from pregnant women in Beijing, China. Clin Microbiol Infect 20(6):O370–O373 18. Yao K, Poulsen K, Maione D, Rinaudo CD, Baldassarri L, Telford JL, Sørensen UB; Members of the DEVANI Study Group, Kilian M (2013) Capsular gene typing of Streptococcus agalactiae compared to serotyping by latex agglutination. J Clin Microbiol 51(2):503–507 19. Jones N, Bohnsack JF, Takahashi S, Oliver KA, Chan MS, Kunst F, Glaser P, Rusniok C, Crook DW, Harding RM, Bisharat N, Spratt BG
20.
21.
22.
23.
24.
(2003) Multilocus sequence typing system for group B streptococcus. J Clin Microbiol 41(6):2530–2536 Melin P, Efstratiou A (2013) Group B streptococcal epidemiology and vaccine needs in developed countries. Vaccine 31(Suppl 4):D31– D42 Gravekamp C, Kasper DL, Paoletti LC, Madoff LC (1999) Alpha C protein as a carrier for type III capsular polysaccharide and as a protective protein in group B streptococcal vaccines. Infect Immun 67(5):2491–2496 Lachenauer CS, Creti R, Michel JL, Madoff LC (2000) Mosaicism in the alpha-like protein genes of group B streptococci. Proc Natl Acad Sci U S A 97(17):9630–9635 Lindahl G, Stålhammar-Carlemalm M, Areschoug T (2005) Surface proteins of Streptococcus agalactiae and related proteins in other bacterial pathogens. Clin Microbiol Rev 18(1):102–127 Kong F, Gowan S, Martin D, James G, Gilbert GL (2002) Molecular profiles of group B streptococcal surface protein antigen genes: relationship to molecular serotypes. J Clin Microbiol 40(2):620–626
ARTICLE
Distribution of pilus islands and alpha-like protein genes of group B Streptococcus colonized in pregnant women in Beijing, China B. Lu & D. Wang & H. Zhou & F. Zhu & D. Li & S. Zhang & Y. Shi & Y. Cui & L. Huang & H. Wu
Received: 14 October 2014 / Accepted: 27 January 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Group B Streptococcus (GBS) is one of the major pathogens of severe newborn sepsis and meningitis. Understanding its regional molecular epidemiology is helpful for regulating efficient prevention practice. A total of 160 GBS strains were collected from colonized pregnant women in six hospital settings in Beijing, China. Polymerase chain reaction (PCR) assays were used to identify the pilus island (PI), alp genes profiling of the alpha-like protein family, and capsular polysaccharide (cps) serotyping. The clonal relationships between strains were investigated using multilocus sequence typing (MLST). All isolates carried at least one pilus island. The most frequently detected pilus island was PI-2a alone (70 isolates, 43.8 %). The most prevalent alp gene was
Electronic supplementary material The online version of this article (doi:10.1007/s10096-015-2342-9) contains supplementary material, which is available to authorized users. B. Lu (*) : F. Zhu : D. Li : S. Zhang : Y. Shi : Y. Cui Department of Laboratory Medicine, Civil Aviation General Hospital, No. 1 Gaojing Street, Chaoyang District, Beijing 100123, People’s Republic of China e-mail: [email protected] D. Wang : H. Zhou National Institute for Communicable Disease Control and Prevention, and State Key Laboratory for Infectious Disease Prevention and Control, Chinese Center for Disease Control and Prevention, Beijing 102206, China D. Wang : H. Zhou Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Hangzhou 310003, China L. Huang Department of Laboratory Medicine, First Hospital, Peking University, Beijing 100034, China H. Wu BGI Tianjin, Tianjin 300308, China
rib (60 isolates, 37.5 %). Moreover, a strong association was noted between alp genes, serotyping, and pilus island profiles. The GBS isolates under study hinted similar molecular epidemical characteristics in Beijing to those reported worldwide, but having their regional distributional features.
Introduction Invasive neonatal disease due to Group B Streptococcus (GBS), colonized asymptomatically in the gastrointestinal and genitourinary tracts of 10–30 % of pregnant women, is a heavy health burden throughout the world [1]. Intrapartum antibiotic prophylaxis has been effectively used to prevent neonatal early-onset GBS diseases, but the incidence of lateonset infections remains almost unchanged [1]. Babies can obtain maternal opsonizing antibodies via the placenta. Therefore, a safe, effective vaccine in women of childbearing age, based on the epidemiological characteristics of regional GBS strains, appears to be a possible solution to prevent mother-to-infant transmission [1, 2]. The protective role in humans of capsular polysaccharide (cps)-specific antibodies of GBS has been confirmed for decades [2]. However, the existence of non-typable isolates makes cps-based vaccine incapable of covering all GBS strains. To date, sortasedependent pilus-like structures have been identified on the surface of GBS. Three pilus island (PI) alleles, PI-1, PI-2a, and PI-2b, have been detected and their proteins explored as promising vaccine candidates and potential virulence factors [3–7]. Furthermore, a series of alpha-like proteins (alp) were demonstrated to play an important role in GBS pathogenesis and also regarded as vaccine candidates [8–10]. These proteins, localized on the surface of GBS, are encoded by genes of alpha-C, epsilon, alp2/3, rib, and alp4, respectively, whose
Eur J Clin Microbiol Infect Dis
distributions constitute the main epidemiological characteristics of the local GBS isolates [11]. The distribution of pilus island and alp genes of GBS has been studied in South Africa, America, Egypt, Ireland, German, and Italy [3, 4, 9, 12–15]. However, scarce data are available from mainland China. Therefore, the aim of our work is to elucidate the distributions of pilus island and alp genes, and their associations with other molecular features of the colonized maternal GBS strains in Beijing, China.
Materials and methods
2013 to May 2014, which is a clinical laboratory center collecting specimens from Beijing Amcare Women’s & Children’s Hospital (DY-1), Beijing HarMoniCare Women and Children’s Hospital (DY-2), Oasis International Hospital (DY-3), Beijing Anhui Yihe Women & Children’s Hospital (DY-4), and Beijing Tongren Hospital (TRH) during 2012. The isolates were sent to the Department of Clinical Microbiology of CAGH for further confirmation by the CAMP test and the VITEK-2 automated microbiology system (bioMérieux, France). Moreover, all the GBS colonies were confirmed by 16S rRNA gene sequencing using the primers BAK11w and BAK2, as we described previously [16].
Bacterial isolates A total of 160 isolates, previously identified as GBS in clinical microbiology laboratories, were recovered from −80 °C storage. The strains under analysis were isolated from vaginal/ rectal swabs of pregnant women at 35–37 weeks of gestation who visited the Civil Aviation General Hospital (CAGH) during the periods 2009 to 2011 and September 2013 to January 2014, Beijing Deyi Shuangxin Clinics during the period June
Table 1 Distribution of clonal complexes (CCs), pilus island (PI), serotypes, and alp genes of 160 group B Streptococcus (GBS) strains carried by pregnant women in Beijing, China
NT non-typable; N negative, CC clonal complex; PI pilus island a
The percentages in parentheses represent isolates/total GBS isolates (160)
No. of strainsa Serotype Ia Ib II III V VI NT alp genes alpha-C rib epsilon alp2/3 alp4 N CCs CC1 CC19 CC17 CC23 CC12 CC4 Singletons (ST 24, ST485, ST486)
Determination of cps serotyping, pilus island genes, and alp genes profile of GBS isolates The ten known GBS capsular serotypes (Ia, Ib, II–IX) were determined by using multiplex polymerase chain reaction (PCR) assays as described previously [17, 18], and the isolates that failed to type were deemed non-typable (NT). PCR assays were used to identify the PI-1, PI-2a, or PI-2b genes using
No. of strains (%)
PI-2a, no. (%)
PI-2b, no. (%)
PI-1 + PI-2a, no. (%)
PI-1 + PI-2b, no. (%)
160 (100)
70 (43.8)
21 (13.1)
67 (41.9)
2 (1.3)
33 (20.6) 18 (11.3) 10 (6.3) 72 (45.0) 23 (14.4) 1 (0.6) 3 (1.9)
23 (14.4) 6 (3.8) 7 (4.4) 22 (13.8) 10 (6.3) – 2 (1.3)
8 (5.0) 1 (0.6) – 11 (6.9) 1 (0.6) – –
2 (1.3) 11 (6.9) 3 (1.9) 37 (23.1) 12 (7.5) 1 (0.6) 1 (0.6)
– – – 2 (1.3) – – –
13 (8.1) 60 (37.5) 40 (25.0) 35 (21.9) 8 (5.0) 4 (2.5)
7 (4.4) 13 (8.1) 27 (16.9) 17 (10.6) 5 (3.1) 1 (0.6)
– 5 (3.1) 9 (5.6) 3 (1.9) 2 (1.3) 2 (1.3)
6 (3.8) 41 (25.6) 4 (2.5) 15 (9.4) – 1 (0.6)
– 1 (0.6) – – 1 (0.6) –
24 (15.0) 56 (35.0) 14 (8.8)
13 (8.1) 19 (11.9) –
1 (0.6) – 10 (6.3)
10 (6.3) 37 (23.1) 2 (1.3)
– – 2 (1.3)
27 (16.9) 28 (17.5) 5 (3.1) 6 (3.8)
23 (14.4) 12 (7.5) 2 (1.3) 1 (0.6)
1 (0.6) 1 (0.6) 3 (1.9) 5 (3.1)
3 (1.9) 15 (9.4) – –
– – – –
Eur J Clin Microbiol Infect Dis
primer pairs as described previously [11, 13]. Surface alp genes were detected by using a multiplex PCR assay [8, 11], and strains that tested negative for any of the known alp genes were considered negative (N).
MLST, phylogenetic, and epidemiological analyses Multilocus sequence typing (MLST) was analyzed by sequencing seven housekeeping genes, as previously described [19]. Then, the sequence types (STs) and allelic profiles were confirmed by querying the MLST database (http://pubmlst. org/sagalactiae). Briefly, PCR fragments for the abovementioned genes were amplified and sequenced. Each isolate was assigned an ST. BURST analysis was conducted to analyze clonal complexes (CCs) using eBURST V3 (http://eburst.mlst.net). The GBS isolates were assigned to one of the CCs if they shared five or more alleles with the predominant ST. An ST not identified in any cluster was assigned as a singleton. BioNumerics software version 5.1 (Applied Maths, Belgium) was used to create minimum spanning trees to illustrate the relationships between MLST, CCs, pilus islands, and alp genes.
Fig. 1 Correlation between clonal complexes (CCs), sequence type (ST), alp genes, and pilus islands (PIs) of the group B Streptococcus (GBS) isolates. Minimum spanning tree analysis of GBS isolates according to ST, demonstrating the relationships between 160 carrier isolates collected from six hospital settings. Note: In the minimum spanning tree, the STs are displayed as circles. The size of each circle indicates the number of isolates within this particular type. The founder ST was defined as the ST with the greatest number of single-locus variants. Pilus islands profiling
Statistical analysis The relationships between CCs and related genes were analyzed using Spearman’s correlation with SPSS 17.0 software. Each pair of variables (genes) was compared by correlation measures. The correlation coefficients were obtained, and the correlation was considered significant at the 0.01 or 0.05 levels.
Results Serotyping and MLST All but three isolates were serotyped. The dominant serotypes were cps III, Ia, V, and Ib, which accounted for 91.3 % of all strains (Table 1). A total of 21 individual STs were distinguished, and 18 STs were clustered in six CCs (Fig. 1), representing 96.3 % of all isolates. The most prevalent complex was CC19, followed by CC12, CC23, CC1, CC17, and CC4, which, except for the minor complex CC4, were also highly prevalent in the GBS-related investigations in other regions [12]. Only 14 CC17 strains were identified, including ST17, ST61, ST64, and ST146, all of which carried serotype III.
and alp genes are represented by different colors in the lower and upper figures, respectively. STs that vary by one allele in their multilocus sequence typing (MLST) profiles (single-locus variants) are arranged in circles around the primary founder ST. Heavy solid lines represent singlelocus variants; light solid lines represent double-locus variants; heavy dotted lines represent triple-locus variants; light dotted lines represent quadruple-locus variants. The major CCs are indicated in the figure
Eur J Clin Microbiol Infect Dis
Eur J Clin Microbiol Infect Dis
Fig. 2
Dendrogram constructed from the multilocus sequence typing (MLST) profiles of seven housekeeping genes of GBS isolates in six hospital settings. CAGH Civil Aviation General Hospital; DY-1 Beijing Amcare Women’s & Children’s Hospital; DY-2 Beijing HarMoniCare Women and Children’s Hospital; DY-3 Oasis International Hospital; DY-4 Beijing Anhui Yihe Women & Children’s Hospital; TRH Beijing Tongren Hospital
positive strains were present in serotype III (55/72, 76.4 %). There was a close association between CCs and alp genes, e.g., the majority of CC1, CC19, CC12, and CC23 isolates had alp2/3 (18/24, 75.0 %), rib (49/56, 87.5 %), alphaC (13/28, 46.4 %), and epsilon (23/27, 85.2 %), respectively (Fig. 1, Table S1).
Pilus islands Discussion Of the 160 isolates of GBS, at least one of the three pilus islands was detected, alone or in combinations (Figs. 1 and 2, Table 1). All strains carried either PI-2a (in 85.6 % of strains) or PI-2b (in 14.4 % of strains). The PI-1 locus was detected in 43.1 % of the strains. The most frequently found pilus island was PI-2a alone, followed by PI-1 + PI-2a, PI-2b alone, and PI1 + PI-2b. There was not an even distribution of pilus islands amongst GBS serotypes. However, a close relationship was observed between certain pilus islands and the cps serotype. Most Ia isolates (23/33, 69.7 %) contained only the PI-2a island alone, whereas the serotype III (37/72, 51.4 %), Ib (11/18, 61.1 %), and V (12/23, 52.2 %) strains mainly contained the combination PI-1 + PI-2a. Moreover, relationships between a specific CC and pilus island, such as CC19 and PI-1 + PI-2a (37/56, 66.1 %), CC23 and PI-2a (23/27, 85.2 %), and CC17 and PI-2b (10/14, 71.4 %), were also noticed (Table S1). PI-2bpositive strains were present in almost all CCs, except for CC19. The PI-1 + PI-2b combination was present in only two strains, both contained in CC17 (Fig. 2). The alp family genes Of all the strains, rib was predominant, accounting for 37.5 % (60/160). Four isolates were negative to any known alp genes. The presence of specific alp genes in relation to the cps serotype was noticed (Table 2). Of the 40 epsilon-positive strains isolated, 30 (75 %) corresponded to serotype Ia; 55 ribTable 2
Correlation between the surface protein genes (alp) and serotypes among the strains of GBS
No. of strainsa Serotype Ia Ib II III V VI NT
No. of strains (%)
alpha-C, no. (%)
rib, no. (%)
epsilon, no. (%)
alp2/3, no. (%)
alp4, no. (%)
N, no. (%)
160 (100 %)
13 (8.1)
60 (37.5)
40 (25.0)
35 (21.9)
8 (5.0)
4 (2.5)
33 (20.6) 18 (11.3) 10 (6.3) 72 (45.0) 23 (14.4) 1 (0.6) 3 (1.9)
– 6 (3.8) 1 (0.6) – 5 (3.1) – 1 (0.6)
– 1 (0.6) 1 (0.6) 55 (34.4) 2 (1.3) – 1 (0.6)
30 (18.8) – – 8 (5.0) 2 (1.3) – –
3 (1.9) 11 (6.9) 7 (4.4) 2 (1.3) 12 (7.5) – –
– – – 5 (3.1) 2 (1.3) – 1 (0.6)
– – 1 (0.6) 2 (1.3) – 1 (0.6) –
NT non-typable; N negative a
Studies of the molecular epidemiology among GBS colonizing strains collected from a variety of geographical settings might facilitate the efficient prevention of invasive neonatal infections [1, 20]. To our knowledge, this is the first GBS surveillance study conducted in mainland China dealing with the regional distributions of pilus islands, alp genes, serotypes, STs, and CCs in GBS isolates. In the current study, pilus islands are conserved among GBS strains. All GBS strains carried at least one of the three pilus islands. Either PI-2a or PI-2b is found; however, this is not the case for PI-1, which was identified in 43.1 % of all strains. The pilus-based vaccine will cover all cps serotypes (including non-typable isolates, which may not be covered by a cps-based vaccine). This is consistent with prior studies [3, 4]. Apart from being promising vaccine antigens against neonatal infections, the pilus island could also mediate the contact and attachment of GBS to host cells and contribute to invasion of GBS into the infected host [7]. Our data on the distributional features of the pilus island are basically in agreement with those of the colonizing and invasive GBS isolates reported in South Africa (541 vaginal-colonizing GBS isolates and 284 invasive isolates), America (289 clinical GBS isolates), and Portugal/Spain (276 collected from pregnant women, 98 noninvasive isolates among non-pregnant adults, and 524 invasive isolates) [3, 4, 13]. Based on this homogeneity of the epidemiological features, a pilus-based vaccine developed for other developed and developing countries will have similar
The percentages in parentheses represent isolates/total GBS isolates (160)
Eur J Clin Microbiol Infect Dis
coverage in Beijing. However, the GBS strains in the present study showed some regional characteristics, which demonstrated that the distribution of the pilus island varied geographically. First, the most prevalent pilus island in Beijing was PI2a alone, which is different from the data of Ireland, where PI1 + PI-2a (which accounted for 48.6 %) was predominant [12]. The majority of GBS isolates (85.6 %) carried PI-2a, which had been suggested to be relevant to pathogenesis due to its participation in biofilm formation [5]. Second, the combination of PI-1 and PI-2b was frequently identified in the colonizing isolates in studies conducted elsewhere [3, 4, 13]; however, only two strains of this combination were confirmed in the current study. Third, the close relationship of serotype III and PI + PI-2a, as well as serotype Ia and PI-2a, was notable, which may be explained by the fact that cps serotypes were genetically linked to the pilus island. It is reported that GBS strains usually carry at least one of the alp genes, but this is not always the case [8, 9, 12, 14]. Apart from having been proposed as potential vaccine components [21], GBS surface proteins have various functions. They contributed to the pathogenesis of GBS disease and induced protective immunity [22]. Our study showed that 2.5 % of isolates were negative for any alp gene. The presence of a particular alp gene in relation to the serotype was noted, though to a lesser degree than previously documented [23, 24]. The current study showed specific serotype–alp gene combinations (e.g., serotype V–alp2/3; serotype III–rib, serotype Ia–epsilon, serotype II–rib, and serotype Ib–alp2/3), which might be explained by the geographic differences [23]. The strains were distinguished into 21 STs and five main CCs (except for CC4), as have been reported across the world [12]. A tight relationship between CC and alp, via pilus island profiling, was detected, although this was not absolute [11]. For example, the isolates of CC19 typically contained rib, while CC1 and CC23 represented alp2/3 and epsilon, respectively [12]. Moreover, the relationship between specific CCs and pilus island, such as CC19 and PI-1 + PI-2a, and CC23 and PI-2a, was observed. Furthermore, CC-17 strains have been suggested to be more virulent than strains of other genotypes, and prevalently associated to neonatal diseases [12, 14]. The distribution of the vaccine candidate in this cluster is extremely important for vaccine development. In our study, the close relationship between CC17 and PI-2b was noted. In summary, our data show that the main molecular characteristics of GBS isolates in Beijing’s pregnant women are basically similar to that reported from other parts of the world [4, 11, 15]. However, a relatively low frequency of PI + PI-2b combination and high prevalence of rib were notable features. The study will contribute to the development of a local GBS vaccine strategy against neonatal infections.
Acknowledgments This study was partially supported by the Beijing Municipal Science & Technology Commission (BMSTC), People’s Republic of China (no. Z141107002514036). Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.
References 1. Verani JR, McGee L, Schrag SJ; Division of Bacterial Diseases, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention (CDC) (2010) Prevention of perinatal group B streptococcal disease—revised guidelines from CDC, 2010. MMWR Recomm Rep 59:1–36 2. Edwards MS (2008) Group B streptococcal conjugate vaccine: a timely concept for which the time has come. Hum Vaccin 4(6): 444–448 3. Margarit I, Rinaudo CD, Galeotti CL, Maione D, Ghezzo C, Buttazzoni E, Rosini R, Runci Y, Mora M, Buccato S, Pagani M, Tresoldi E, Berardi A, Creti R, Baker CJ, Telford JL, Grandi G (2009) Preventing bacterial infections with pilus-based vaccines: the group B streptococcus paradigm. J Infect Dis 199(1):108–115 4. Martins ER, Andreu A, Melo-Cristino J, Ramirez M (2013) Distribution of pilus islands in Streptococcus agalactiae that cause human infections: insights into evolution and implication for vaccine development. Clin Vaccine Immunol 20(2):313–316 5. Rinaudo CD, Rosini R, Galeotti CL, Berti F, Necchi F, Reguzzi V, Ghezzo C, Telford JL, Grandi G, Maione D (2010) Specific involvement of pilus type 2a in biofilm formation in group B Streptococcus. PLoS One 5(2):e9216 6. Nobbs AH, Rosini R, Rinaudo CD, Maione D, Grandi G, Telford JL (2008) Sortase A utilizes an ancillary protein anchor for efficient cell wall anchoring of pili in Streptococcus agalactiae. Infect Immun 76(8):3550–3560 7. Sharma P, Lata H, Arya DK, Kashyap AK, Kumar H, Dua M, Ali A, Johri AK (2013) Role of pilus proteins in adherence and invasion of Streptococcus agalactiae to the lung and cervical epithelial cells. J Biol Chem 288(6):4023–4034 8. Creti R, Fabretti F, Orefici G, von Hunolstein C (2004) Multiplex PCR assay for direct identification of group B streptococcal αprotein-like protein genes. J Clin Microbiol 42(3):1326–1329 9. Shabayek S, Abdalla S, Abouzeid AM (2014) Serotype and surface protein gene distribution of colonizing group B streptococcus in women in Egypt. Epidemiol Infect 142(1):208–210 10. Gilbert GL (2004) Vaccines for other neonatal infections: are group B streptococcal infections vaccine-preventable? Expert Rev Vaccines 3(4):371–374 11. De Francesco MA, Caracciolo S, Gargiulo F, Manca N (2012) Phenotypes, genotypes, serotypes and molecular epidemiology of erythromycin-resistant Streptococcus agalactiae in Italy. Eur J Clin Microbiol Infect Dis 31(8):1741–1747 12. Meehan M, Cunney R, Cafferkey M (2014) Molecular epidemiology of group B streptococci in Ireland reveals a diverse population with evidence of capsular switching. Eur J Clin Microbiol Infect Dis 33(7):1155–1162 13. Madzivhandila M, Adrian PV, Cutland CL, Kuwanda L, Madhi SA; PoPS Trial Team (2013) Distribution of pilus islands of group B streptococcus associated with maternal colonization and invasive disease in South Africa. J Med Microbiol 62(Pt 2):249–253 14. Brimil N, Barthell E, Heindrichs U, Kuhn M, Lütticken R, Spellerberg B (2006) Epidemiology of Streptococcus agalactiae colonization in Germany. Int J Med Microbiol 296(1):39–44
Eur J Clin Microbiol Infect Dis 15. Gherardi G, Imperi M, Baldassarri L, Pataracchia M, Alfarone G, Recchia S, Orefici G, Dicuonzo G, Creti R (2007) Molecular epidemiology and distribution of serotypes, surface proteins, and antibiotic resistance among group B streptococci in Italy. J Clin Microbiol 45(9):2909–2916 16. Binghuai L, Yanli S, Shuchen Z, Fengxia Z, Dong L, Yanchao C (2014) Use of MALDI-TOF mass spectrometry for rapid identification of group B Streptococcus on chromID Strepto B agar. Int J Infect Dis 27:44–48 17. Lu B, Li D, Cui Y, Sui W, Huang L, Lu X (2014) Epidemiology of Group B streptococcus isolated from pregnant women in Beijing, China. Clin Microbiol Infect 20(6):O370–O373 18. Yao K, Poulsen K, Maione D, Rinaudo CD, Baldassarri L, Telford JL, Sørensen UB; Members of the DEVANI Study Group, Kilian M (2013) Capsular gene typing of Streptococcus agalactiae compared to serotyping by latex agglutination. J Clin Microbiol 51(2):503–507 19. Jones N, Bohnsack JF, Takahashi S, Oliver KA, Chan MS, Kunst F, Glaser P, Rusniok C, Crook DW, Harding RM, Bisharat N, Spratt BG
20.
21.
22.
23.
24.
(2003) Multilocus sequence typing system for group B streptococcus. J Clin Microbiol 41(6):2530–2536 Melin P, Efstratiou A (2013) Group B streptococcal epidemiology and vaccine needs in developed countries. Vaccine 31(Suppl 4):D31– D42 Gravekamp C, Kasper DL, Paoletti LC, Madoff LC (1999) Alpha C protein as a carrier for type III capsular polysaccharide and as a protective protein in group B streptococcal vaccines. Infect Immun 67(5):2491–2496 Lachenauer CS, Creti R, Michel JL, Madoff LC (2000) Mosaicism in the alpha-like protein genes of group B streptococci. Proc Natl Acad Sci U S A 97(17):9630–9635 Lindahl G, Stålhammar-Carlemalm M, Areschoug T (2005) Surface proteins of Streptococcus agalactiae and related proteins in other bacterial pathogens. Clin Microbiol Rev 18(1):102–127 Kong F, Gowan S, Martin D, James G, Gilbert GL (2002) Molecular profiles of group B streptococcal surface protein antigen genes: relationship to molecular serotypes. J Clin Microbiol 40(2):620–626
