- Email: [email protected]
Selection and emergence of pertussis toxin promoter ptxP3 allele in the evolution of Bordetella pertussis
Infection, Genetics and Evolution 12 (2012) 492–495
Contents lists available at SciVerse ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Short communication
Selection and emergence of pertussis toxin promoter ptxP3 allele in the evolution of Bordetella pertussis Connie Lam a, Sophie Octavia a, Zahra Bahrame a, Vitali Sintchenko b,c, Gwendolyn L. Gilbert b,c, Ruiting Lan a,⇑ a b c
School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, Australia Centre for Infectious Diseases and Microbiology-Public Health, Institute of Clinical Pathology and Medical Research, Westmead Hospital, Westmead, New South Wales, Australia Sydney Emerging Infections and Biosecurity Institute (SEIB) and Sydney Medical School, University of Sydney, Sydney, New South Wales, Australia
a r t i c l e
i n f o
Article history: Received 5 October 2011 Received in revised form 20 December 2011 Accepted 4 January 2012 Available online 24 January 2012 Keywords: Bordetella pertussis Evolution Pertussis toxin promoter Single nucleotide polymorphism
a b s t r a c t Evolutionary studies using single nucleotide polymorphisms (SNPs) have separated Bordetella pertussis isolates into six major clusters, with recent isolates forming cluster I. The expansion of cluster I isolates was characterised by changes in genes encoding antigenic components in acellular vaccines, including pertactin (Prn). Here, we determined the initial emergence of the pertussis toxin promoter allele, ptxP3, from an evolutionary perspective. This allele was previously shown in a study from the Netherlands to be associated with increased pertussis toxin production as a result of a single base mutation in the ptxP. The ptxP region of 313 worldwide isolates was sequenced, including 208 isolates from Australia collected over a 40 year period. Eight alleles were identified, of which only two predominated: ptxP1 and ptxP3. One novel allele was also found. ptxP3 was only found in SNP cluster I of B. pertussis and its emergence is concurrent with the change to the non-vaccine prn2 allele. Our results suggest that the globally distributed cluster I of B. pertussis has the ability to evade vaccine induced selection pressure. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Bordetella pertussis is the cause of pertussis or whooping cough which can be life-threatening disease for young children. Widespread vaccination against B. pertussis was first implemented using a whole cell vaccine (WCV) during the 1950s leading to dramatic reduction in pertussis-related mortality. However, side-effects of the WCV adversely affected uptake of this vaccine and alternative acellular vaccines (ACV) containing only a few antigenic components were developed. Although ACV formulations vary, the antigens used generally include a combination of three to all five of the following: pertussis toxin (PT), pertactin (Prn), filamentous haemagglutinin (FHA) and two types of fimbriae (Fim2 and Fim3) (Godfroid et al., 2005). Despite the high uptake of vaccination programs in most developed countries, the re-emergence of pertussis in highly immunised populations has been of particular concern (Elomaa et al., 2007; Poynten et al., 2004; Tanaka et al., 2003). Several factors may have contributed to the increase in pertussis infections, including waning vaccine immunity and adaptation of the organism to vaccine induced pressures (Mooi et al., 2001).
⇑ Corresponding author. Tel.: +61 2 9385 2095; fax: +61 2 9385 1483. E-mail address: [email protected] (R. Lan). 1567-1348/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.meegid.2012.01.001
We and others have used multilocus variable number tandem repeat analysis (MLVA) and single nucleotide polymorphism (SNP) typing of B. pertussis to clarify its molecular epidemiology and spread (Kurniawan et al., 2010; Octavia et al., 2011; Schouls et al., 2004). SNP typing separated B. pertussis isolates into six major groups, which we named clusters I–VI (Octavia et al., 2011). Cluster I emerged recently and consisted of predominantly MLVA type 27 (MT27) and antigen profiles (AP) 3 and 8, with alleles, prn2-ptxA1-fim2-1-fim3A-fhaB1 and prn2-ptxA1-fim2-1-fim3BfhaB1, respectively. MT27 is prevalent worldwide (Advani et al., 2009; Kurniawan et al., 2010; Litt et al., 2009; Schouls et al., 2004). Our previous analysis of 208 Australian isolates suggested that the introduction of ACV in 1997 coincided with an increase in prevalence of SNP clusters I and IV and a decrease in cluster II in Australia (Octavia et al., 2011). In addition to SNP and antigenic changes, variations have also been reported in ptxP, the promoter region of PT (Mooi et al., 2009). A recent report by Mooi et al. (2009) showed that the resurgence of pertussis in the Netherlands was associated with increased PT production as a result of a single base mutation in ptxP, resulting in a switch from ptxP1 to ptxP3. In this study, we aimed to (a) determine when the ptxP3 allele emerged within the evolutionary history of B. pertussis, using the SNP-based relationships established previously, and (b) determine whether the emergence of ptxP3 was associated with the expansion of cluster I.
C. Lam et al. / Infection, Genetics and Evolution 12 (2012) 492–495
2. Materials and methods 2.1. Selection of bacterial isolates and growth conditions A collection of 313 B. pertussis isolates from 12 countries which were also used in our previous studies was analysed (Kurniawan et al., 2010; Octavia et al., 2011). The majority of isolates were obtained from patients in Australia but the collection contained representative B. pertussis isolates from around the world, including France (eight isolates), Japan (49 isolates), Finland (15 isolates), Hong Kong (11 isolates), Canada (15 isolates), the United States (12 isolates). We also included one earlier isolate each from the Netherlands, Italy, China, Mexico and the UK to represent B. pertussis diversity. The collection spanned important time points in B. pertussis evolution, including the pre-vaccine era (up to 1920s) and the WCV (1950s–1990s) and ACV (1990s to present) periods. It included the most common circulating Pulsed Field Gel Electrophoresis (PFGE) types from Canada, Finland, France, Japan and USA. 2.2. Bacterial growth, DNA extraction and sequencing Bacterial isolates were inoculated onto Charcoal agar (Oxoid) supplemented with 10% horse blood, and cultured at 35 °C for 3– 5 days. DNA was then extracted using the phenol/chloroform method and used for sequencing. The 554-bp ptxP region was amplified by PCR and sequenced using published primers (50 -AATC GTCCTGCTCAACCGCC-30 and 50 -GGTATACGGTGGCGGGAGGA-30 ) (Mooi et al., 2009). The products were then analysed using an Automated DNA Sequence Analyzer ABI3730 (Applied Biosystems) at the sequencing facility of the School of Biotechnology and Biomolecular Sciences, the University of New South Wales. 3. Results and discussion Sequencing of the ptxP promoter of 313 isolates identified eight ptxP alleles, seven of which have been previously described (Mooi et al., 2009). Fig. 1 shows the relationships between the identified ptxP alleles in this collection of isolates, with their respective SNP profiles (SP). The most common alleles were ptxP1 and ptxP3, accounting for more than 95% of the isolates. ptxP1 was present in 73% of isolates (230 out of 313 isolates) and the second most prevalent allele was ptxP3 (73 out of 313 isolates, 23%). NonptxP1/non-ptxP3 alleles that have been reported previously and identified in this set of isolates included: ptxP2, ptxP4, ptxP5, ptxP8 and ptxP9. One ptxP allele not previously described, differed from ptxP1 by a single base change (C to T change at position 137, GenBank accession number JQ029160) and was named ptxP19. Based on the evolutionary relationships between the B. pertussis isolates previously determined using SNPs (Octavia et al., 2011), ptxP1 was found in clusters II–VI whereas ptxP3 was found only in isolates from cluster I. Comparative genome sequencing by Bart et al. (2010), suggested that ptxP2 originated from a distinct lineage that diverged early in B. pertussis evolution. Our findings confirmed this observation as both SP33 and SP34, in a distinct cluster of older isolates (cluster VI in our study – see Fig. 1), contained ptxP2 alleles. All other non-ptxP1/non-ptxP3 alleles were not exclusive to a specific SP and occurred only once. The two isolates carrying the novel allele, ptxP19, (SP26 and SP37) belonged to clusters V and II, respectively (Fig. 1); clearly the change arose as two independent events of either mutation or recombination. Our study show that cluster I has been found across at least seven countries (Australia, Canada, Finland, France, Japan, Hong Kong, and US), representing four continents, indicating the widespread distribution of ptxP3. Isolates with the ptxP3 allele have been reported from many other countries (Kallonen et al., 2011;
493
Mooi et al., 2009). Data from other studies also points to the presence or increased prevalence of ptxP3 isolates based on association of ptxP3 with particular genotypes of other methods. For example, MLVA type MT27 is largely associated with ptxP3, based on this study and that of van Gent et al. (2011). MT27 became predominant in the UK, after the introduction of ACV in 2002 and the MT27 isolates carry prn2 (Litt et al., 2009). Thus, it can be inferred that majority, if not all, also carry ptxP3. A recent SNP based study by van Gent et al. (2011), which included a large number of ptxP3 isolates from the Netherlands and representative isolates from four other countries, also showed that ptxP3 isolates are grouped together and evolved from ptxP1 isolates, similar to our findings. Although there are no representative ptxP3 isolates from the Netherlands in our study, based on the genome sequence data of two Dutch ptxP3 strains (Bart et al., 2010) and our SNPs common with van Gent et al. (2011), the currently circulating ptxP3 isolates in the Netherlands belong to our SNP cluster I. Using SNPs common to both studies (15 common SNPs with two SNPs, BP2249 and BP2366 that can partially be used to assign ptxp3 isolates to cluster I), we were able map most of their ptxP3 SNP types to our cluster I. In addition, an Australian ptxP3 isolate (L584) typed in both studies belongs to their ST10 and our SP12, which are shown to have been the first to diverge among the SNP types carrying ptxP3 (Figs. 1 and 5 of van Gent et al. (2011)). The expansion of cluster I in Australia has been attributed to the introduction of selection pressure from the ACV (Octavia et al., 2011). In this study, we show that ptxP3 is associated with the emergence of cluster I. ptxP3 has been hypothesised to improve binding of the global virulence regulator, BvgA, resulting in the production of more toxin than ptxP1 strains (Mooi et al., 2009); giving the organism a potential selective advantage over ptxP1 strains. The increased levels of PT may lead to a more severe disease, which has been correlated with higher hospitalisation rates of children infected with B. pertussis containing ptxP3 alleles in the Netherlands (Mooi et al., 2009). In Sweden, ptxP3 has not only become the dominant allele, but has also been detected much earlier in regions where only a single component (i.e. PT) pertussis vaccine was used (Advani et al., 2011). This observation further suggests that vaccine-induced selection pressure plays a role in both selection of ptxP3 and expansion of cluster I. In addition to the switch from ptxP1 to ptxP3, the other predominant change observed in cluster I was the change from prn1 to prn2 (Octavia et al., 2011). Prn2 differs from Prn1 by an additional GGFGP motif located near the RGD motif which is postulated to be involved in receptor binding (Leininger et al., 1991). It is likely that conformational changes to Prn2 as a result of the additional GGFGP repeat could contribute to the type specificity of immunological responses (He et al., 2003). This suggestion is also supported by studies of experimental pertussis infection, which have shown that the WCV used in the Netherlands (derived from a strain containing prn1) did not protect against infection with B. pertussis isolates with prn2 allele as efficiently as against isolates with the prn1 allele (King et al., 2001). The ACVs available worldwide differ in their constituents and range from monocomponent (PT only) to five component vaccines (PT, PRN, FHA, FIM2 and FIM3) (Advani et al., 2011; Kallonen et al., 2011). In Australia, the most commonly used vaccine is a 3-component vaccine consisting of PT, PRN and FHA which was introduced in 1997 and replaced the WCV entirely by 2000. A five component ACV with additional FIM2 and FIM3 antigens is also available in Australia but only used in a small proportion of the vaccinations. The alleles for the 3-component ACV are ptxA2, prn1, and fhaB1 derived from Tohama I (Kurniawan et al., 2010). In addition to prn allele mismatch between the ACV and the currently circulating cluster I strains as discussed above, the allelic mismatch of ptxA, between the currently circulating in cluster I allele, ptxA1, and
494
C. Lam et al. / Infection, Genetics and Evolution 12 (2012) 492–495
Year(s) SP15 (1) SP16 (3) SP14 (41)
prn2 and ptxP3
I
SP10 ((1)) SP13 (23) SP41 (1) SP12 (4)
ptxP5 (1 isolate)
SP11 (16) SP42 (1)
SP17 (3) SP37 (34) SP38 (4) SP39 (6) SP9 ((8)) ptxP9 ptxP9,
II
ptxP19 (1 isolate each)
SP6 (2) SP7 (1) SP1 (8) SP8 (1) SP2 ((3)) SP3 (16) SP4 (6) SP5 (8)
III
SP19 (2) SP18 (11) SP20 (1) SP21 (4) SP22 (7) SP40 (5) SP23 (2) SP29 (3)
IV
SP31 (8) SP24 (10) SP30 (25) SP25 ((3)) SP26 (14) SP27 (19) SP28 (2)
ptxP8, P8 ptxP19 P19 (1 isolate each)
V
SP32 (2)
ptxP2 (1 isolate)
SP33 (2) SP34 (1) SP35 (1)
SP36 (1) Tohama 2
VI
Country(s)
2006 2007- 2008 1997-2006
FL AU AU, CA, FL, FR, HK, JP, US
1999 1997- 2005
JP AU, CA, FR, HK, JP, US
2007
JP
1994- 2004
JP, AU
1992- 2004
AU, FL, IT, MX, NL
2004
FL
1995- 2000 1989- 2003 1995- 1998 2000- 2006 1972- 2006
AU AU, FR AU AU AU,, FL
1977- 1978
AU
2006 1973- 2000 1997 1985- 2005 1989- 2001 1995- 2001 1995 1999- 2002
AU AU, FL AU AU AU, JP AU AU
2002 1992- 2008
AU AU
2002 1982
AU AU
1971- 1989
AU
1971- 1989
AU
999 2000 000 1999-
AU
2001- 2002
AU
2001- 2005 2001 1996- 2004 1977- 2004 1946- 1957 1989- 2000 1990 2007 19901996
AU AU AU CN,, US JP AU HK, AU, HK JP JP
1940- 1993
FR, US
1935 1950 1920
US FR UK
1954
JP
Key: ptxP4 ptxP3 p ptxP2 ptxP1
Fig. 1. Distribution of ptxP alleles on the SNP tree. Maximum parsimony tree showing the evolution of Bordetella pertussis based on 65 SNPs was from Octavia et al. (2011). Each branch is represented by a SNP profile (SP) which were grouped into clusters I–VI. The number of isolates sequenced per SP has been included in brackets. The distribution of the ptxP alleles is colour coded with key on the right. The change of prn1 to prn2 and ptxP1 to ptxP3 are marked on the branch. All SPs carried the ptxP1 allele unless otherwise noted on the tree. Country codes are as follows: AU – Australia, CA – Canada, CN – China, FL – Finland, FR – France, HK – Hong Kong, IT – Italy, JP – Japan, MX – Mexico, NL – the Netherlands, UK – United Kingdom, US – United States.
the ACV allele, ptxA2 should also be noted. The increased production of PT as a result of ptxP3 may exacerbate the effect of any antigenic difference between the two ptxA alleles in ACV- induced protection. Therefore, the combination of changes to prn2 and ptxP3, as well as the non-ACV allelic mismatch of ptxA1 in cluster I, may be a significant contributor to the increase in pertussis rates. This factor may play a similar role in pertussis resurgence in countries where the same or similar ACVs are used. 4. Conclusions In conclusion, the resurgence of pertussis in highly vaccinated populations can be, at least in part, explained by genetic changes
that increase the fitness of circulating B. pertussis isolates. The emergence and expansion of cluster I are associated with two important genetic events in B. pertussis evolution: the increased production of pertussis toxin as a result of change to the promoter region (ptxP3) and the change to a non-ACV prn2 allele. This study demonstrated that the ptxP3 allele is associated with cluster I. Combined with previously reported changes to the B. pertussis genome, in particular changes to antigenic determinants, this finding suggests that isolates in this cluster are fitter variants of B. pertussis. These observations highlight the public health relevance of monitoring the global distribution of B. pertussis cluster I isolates in highly immunised populations and may offer important insights for the development of new strategies to reduce the burden of pertussis.
C. Lam et al. / Infection, Genetics and Evolution 12 (2012) 492–495
Acknowledgements We gratefully acknowledge the generous donations of isolates by Dr. Kazunari Kamachi, Department of Bacterial Pathogenesis and Infection Control, National Institute of Infectious Diseases, Tokyo, Japan; Dr. Shane Byrne, Sullivan Nicolaides Pathology, Brisbane, Queensland, Australia; the late Dr. John Tapsall, Prince of Wales Hospital, Sydney, Australia; Dr. Ian Carter, St. George Hospital, Sydney, Australia; Dr. Margret Ip, Chinese University of Hong Kong; Dr. Qiushui He, Pertussis Reference Laboratory, National Public Health Institute, Turku, Finland; Dr. Nicole Guiso, Institut Pasteur, Paris, France; Dr. Raymond Tsang, National Microbiology Laboratory, Public Health Agency of Canada; and Dr. Lucia Tondella, Centers for Disease Control and Prevention, USA. We thank Marjolein van Gent for technical assistance. This research was supported by the National Health and Medical Research Council of Australia. Connie Lam was supported by an Australian Postgraduate Award. References Advani, A., Van der Heide, H.G., Hallander, H.O., Mooi, F.R., 2009. Analysis of Swedish Bordetella pertussis isolates with three typing methods: characterization of an epidemic lineage. J. Microbiol. Methods 78, 297–301. Advani, A., Gustafsson, L., Ahren, C., Mooi, F.R., Hallander, H.O., 2011. Appearance of Fim3 and ptxP3-Bordetella pertussis strains, in two regions of Sweden with different vaccination programs. Vaccine 29, 3438–3442. Bart, M.J., van Gent, M., van der Heide, H.G., Boekhorst, J., Hermans, P., Parkhill, J., Mooi, F.R., 2010. Comparative genomics of prevaccination and modern Bordetella pertussis strains. BMC Genomics 11, 627. Elomaa, A., Advani, A., Donnelly, D., Antila, M., Mertsola, J., He, Q., Hallander, H., 2007. Population dynamics of Bordetella pertussis in Finland and Sweden, neighbouring countries with different vaccination histories. Vaccine 25, 918–926. Godfroid, F., Denoel, P., Poolman, J., 2005. Are vaccination programs and isolate polymorphism linked to pertussis re-emergence? Expert. Rev. Vaccines 4, 757– 778. He, Q., Makinen, J., Berbers, G., Mooi, F.R., Viljanen, M.K., Arvilommi, H., Mertsola, J., 2003. Bordetella pertussis protein pertactin induces type-specific antibodies: one
495
possible explanation for the emergence of antigenic variants? J. Infect. Dis. 187, 1200–1205. Kallonen, T., Grondahl-Yli-Hannuksela, K., Elomaa, A., Lutynska, A., Fry, N.K., Mertsola, J., He, Q., 2011. Differences in the genomic content of Bordetella pertussis isolates before and after introduction of pertussis vaccines in four European countries. Infect. Genet. Evol. 11, 2034–2042. King, A.J., Berbers, G., van Oirschot, H.F., Hoogerhout, P., Knipping, K., Mooi, F.R., 2001. Role of the polymorphic region 1 of the Bordetella pertussis protein pertactin in immunity. Microbiology 147, 2885–2895. Kurniawan, J., Maharjan, R.P., Chan, W.F., Reeves, P.R., Sintchenko, V., Gilbert, G.L., Mooi, F.R., Lan, R., 2010. Bordetella pertussis clones identified by multilocus variable-number tandem-repeat analysis. Emerg. Infect. Dis. 16, 297–300. Leininger, E., Roberts, M., Kenimer, J.G., Charles, I.G., Fairweather, N., Novotny, P., Brennan, M.J., 1991. Pertactin, an Arg-Gly-Asp-containing Bordetella pertussis surface protein that promotes adherence of mammalian cells. Proc. Natl. Acad. Sci. USA 88, 345–349. Litt, D.J., Neal, S.E., Fry, N.K., 2009. Changes in genetic diversity of the Bordetella pertussis population in the United Kingdom between 1920 and 2006 reflect vaccination coverage and emergence of a single dominant clonal type. J. Clin. Microbiol. 47, 680–688. Mooi, F.R., van Loo, I.H., King, A.J., 2001. Adaptation of Bordetella pertussis to vaccination: a cause for its reemergence? Emerg. Infect. Dis. 7, 526–528. Mooi, F.R., van Loo, I.H., van Gent, M., He, Q., Bart, M.J., Heuvelman, K.J., de Greeff, S.C., Diavatopoulos, D., Teunis, P., Nagelkerke, N., Mertsola, J., 2009. Bordetella pertussis strains with increased toxin production associated with pertussis resurgence. Emerg. Infect. Dis. 15, 1206–1213. Octavia, S., Maharjan, R.P., Sintchenko, V., Stevenson, G., Reeves, P.R., Gilbert, G.L., Lan, R., 2011. Insight into evolution of Bordetella pertussis from comparative genomic analysis: evidence of vaccine-driven selection. Mol. Biol. Evol. 28, 707– 715. Poynten, M., McIntyre, P.B., Mooi, F.R., Heuvelman, K.J., Gilbert, G.L., 2004. Temporal trends in circulating Bordetella pertussis strains in Australia. Epidemiol. Infect. 132, 185–193. Schouls, L.M., van der Heide, H.G., Vauterin, L., Vauterin, P., Mooi, F.R., 2004. Multiple-locus variable-number tandem repeat analysis of Dutch Bordetella pertussis strains reveals rapid genetic changes with clonal expansion during the late 1990s. J. Bacteriol. 186, 5496–5505. Tanaka, M., Vitek, C.R., Pascual, F.B., Bisgard, K.M., Tate, J.E., Murphy, T.V., 2003. Trends in pertussis among infants in the United States, 1980–1999. JAMA 290, 2968–2975. van Gent, M., Bart, M.J., van der Heide, H.G., Heuvelman, K.J., Kallonen, T., He, Q., Mertsola, J., Advani, A., Hallander, H.O., Janssens, K., Hermans, P.W., Mooi, F.R., 2011. SNP-based typing: a useful tool to study Bordetella pertussis populations. PLoS ONE 6, e20340.
Contents lists available at SciVerse ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Short communication
Selection and emergence of pertussis toxin promoter ptxP3 allele in the evolution of Bordetella pertussis Connie Lam a, Sophie Octavia a, Zahra Bahrame a, Vitali Sintchenko b,c, Gwendolyn L. Gilbert b,c, Ruiting Lan a,⇑ a b c
School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, Australia Centre for Infectious Diseases and Microbiology-Public Health, Institute of Clinical Pathology and Medical Research, Westmead Hospital, Westmead, New South Wales, Australia Sydney Emerging Infections and Biosecurity Institute (SEIB) and Sydney Medical School, University of Sydney, Sydney, New South Wales, Australia
a r t i c l e
i n f o
Article history: Received 5 October 2011 Received in revised form 20 December 2011 Accepted 4 January 2012 Available online 24 January 2012 Keywords: Bordetella pertussis Evolution Pertussis toxin promoter Single nucleotide polymorphism
a b s t r a c t Evolutionary studies using single nucleotide polymorphisms (SNPs) have separated Bordetella pertussis isolates into six major clusters, with recent isolates forming cluster I. The expansion of cluster I isolates was characterised by changes in genes encoding antigenic components in acellular vaccines, including pertactin (Prn). Here, we determined the initial emergence of the pertussis toxin promoter allele, ptxP3, from an evolutionary perspective. This allele was previously shown in a study from the Netherlands to be associated with increased pertussis toxin production as a result of a single base mutation in the ptxP. The ptxP region of 313 worldwide isolates was sequenced, including 208 isolates from Australia collected over a 40 year period. Eight alleles were identified, of which only two predominated: ptxP1 and ptxP3. One novel allele was also found. ptxP3 was only found in SNP cluster I of B. pertussis and its emergence is concurrent with the change to the non-vaccine prn2 allele. Our results suggest that the globally distributed cluster I of B. pertussis has the ability to evade vaccine induced selection pressure. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Bordetella pertussis is the cause of pertussis or whooping cough which can be life-threatening disease for young children. Widespread vaccination against B. pertussis was first implemented using a whole cell vaccine (WCV) during the 1950s leading to dramatic reduction in pertussis-related mortality. However, side-effects of the WCV adversely affected uptake of this vaccine and alternative acellular vaccines (ACV) containing only a few antigenic components were developed. Although ACV formulations vary, the antigens used generally include a combination of three to all five of the following: pertussis toxin (PT), pertactin (Prn), filamentous haemagglutinin (FHA) and two types of fimbriae (Fim2 and Fim3) (Godfroid et al., 2005). Despite the high uptake of vaccination programs in most developed countries, the re-emergence of pertussis in highly immunised populations has been of particular concern (Elomaa et al., 2007; Poynten et al., 2004; Tanaka et al., 2003). Several factors may have contributed to the increase in pertussis infections, including waning vaccine immunity and adaptation of the organism to vaccine induced pressures (Mooi et al., 2001).
⇑ Corresponding author. Tel.: +61 2 9385 2095; fax: +61 2 9385 1483. E-mail address: [email protected] (R. Lan). 1567-1348/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.meegid.2012.01.001
We and others have used multilocus variable number tandem repeat analysis (MLVA) and single nucleotide polymorphism (SNP) typing of B. pertussis to clarify its molecular epidemiology and spread (Kurniawan et al., 2010; Octavia et al., 2011; Schouls et al., 2004). SNP typing separated B. pertussis isolates into six major groups, which we named clusters I–VI (Octavia et al., 2011). Cluster I emerged recently and consisted of predominantly MLVA type 27 (MT27) and antigen profiles (AP) 3 and 8, with alleles, prn2-ptxA1-fim2-1-fim3A-fhaB1 and prn2-ptxA1-fim2-1-fim3BfhaB1, respectively. MT27 is prevalent worldwide (Advani et al., 2009; Kurniawan et al., 2010; Litt et al., 2009; Schouls et al., 2004). Our previous analysis of 208 Australian isolates suggested that the introduction of ACV in 1997 coincided with an increase in prevalence of SNP clusters I and IV and a decrease in cluster II in Australia (Octavia et al., 2011). In addition to SNP and antigenic changes, variations have also been reported in ptxP, the promoter region of PT (Mooi et al., 2009). A recent report by Mooi et al. (2009) showed that the resurgence of pertussis in the Netherlands was associated with increased PT production as a result of a single base mutation in ptxP, resulting in a switch from ptxP1 to ptxP3. In this study, we aimed to (a) determine when the ptxP3 allele emerged within the evolutionary history of B. pertussis, using the SNP-based relationships established previously, and (b) determine whether the emergence of ptxP3 was associated with the expansion of cluster I.
C. Lam et al. / Infection, Genetics and Evolution 12 (2012) 492–495
2. Materials and methods 2.1. Selection of bacterial isolates and growth conditions A collection of 313 B. pertussis isolates from 12 countries which were also used in our previous studies was analysed (Kurniawan et al., 2010; Octavia et al., 2011). The majority of isolates were obtained from patients in Australia but the collection contained representative B. pertussis isolates from around the world, including France (eight isolates), Japan (49 isolates), Finland (15 isolates), Hong Kong (11 isolates), Canada (15 isolates), the United States (12 isolates). We also included one earlier isolate each from the Netherlands, Italy, China, Mexico and the UK to represent B. pertussis diversity. The collection spanned important time points in B. pertussis evolution, including the pre-vaccine era (up to 1920s) and the WCV (1950s–1990s) and ACV (1990s to present) periods. It included the most common circulating Pulsed Field Gel Electrophoresis (PFGE) types from Canada, Finland, France, Japan and USA. 2.2. Bacterial growth, DNA extraction and sequencing Bacterial isolates were inoculated onto Charcoal agar (Oxoid) supplemented with 10% horse blood, and cultured at 35 °C for 3– 5 days. DNA was then extracted using the phenol/chloroform method and used for sequencing. The 554-bp ptxP region was amplified by PCR and sequenced using published primers (50 -AATC GTCCTGCTCAACCGCC-30 and 50 -GGTATACGGTGGCGGGAGGA-30 ) (Mooi et al., 2009). The products were then analysed using an Automated DNA Sequence Analyzer ABI3730 (Applied Biosystems) at the sequencing facility of the School of Biotechnology and Biomolecular Sciences, the University of New South Wales. 3. Results and discussion Sequencing of the ptxP promoter of 313 isolates identified eight ptxP alleles, seven of which have been previously described (Mooi et al., 2009). Fig. 1 shows the relationships between the identified ptxP alleles in this collection of isolates, with their respective SNP profiles (SP). The most common alleles were ptxP1 and ptxP3, accounting for more than 95% of the isolates. ptxP1 was present in 73% of isolates (230 out of 313 isolates) and the second most prevalent allele was ptxP3 (73 out of 313 isolates, 23%). NonptxP1/non-ptxP3 alleles that have been reported previously and identified in this set of isolates included: ptxP2, ptxP4, ptxP5, ptxP8 and ptxP9. One ptxP allele not previously described, differed from ptxP1 by a single base change (C to T change at position 137, GenBank accession number JQ029160) and was named ptxP19. Based on the evolutionary relationships between the B. pertussis isolates previously determined using SNPs (Octavia et al., 2011), ptxP1 was found in clusters II–VI whereas ptxP3 was found only in isolates from cluster I. Comparative genome sequencing by Bart et al. (2010), suggested that ptxP2 originated from a distinct lineage that diverged early in B. pertussis evolution. Our findings confirmed this observation as both SP33 and SP34, in a distinct cluster of older isolates (cluster VI in our study – see Fig. 1), contained ptxP2 alleles. All other non-ptxP1/non-ptxP3 alleles were not exclusive to a specific SP and occurred only once. The two isolates carrying the novel allele, ptxP19, (SP26 and SP37) belonged to clusters V and II, respectively (Fig. 1); clearly the change arose as two independent events of either mutation or recombination. Our study show that cluster I has been found across at least seven countries (Australia, Canada, Finland, France, Japan, Hong Kong, and US), representing four continents, indicating the widespread distribution of ptxP3. Isolates with the ptxP3 allele have been reported from many other countries (Kallonen et al., 2011;
493
Mooi et al., 2009). Data from other studies also points to the presence or increased prevalence of ptxP3 isolates based on association of ptxP3 with particular genotypes of other methods. For example, MLVA type MT27 is largely associated with ptxP3, based on this study and that of van Gent et al. (2011). MT27 became predominant in the UK, after the introduction of ACV in 2002 and the MT27 isolates carry prn2 (Litt et al., 2009). Thus, it can be inferred that majority, if not all, also carry ptxP3. A recent SNP based study by van Gent et al. (2011), which included a large number of ptxP3 isolates from the Netherlands and representative isolates from four other countries, also showed that ptxP3 isolates are grouped together and evolved from ptxP1 isolates, similar to our findings. Although there are no representative ptxP3 isolates from the Netherlands in our study, based on the genome sequence data of two Dutch ptxP3 strains (Bart et al., 2010) and our SNPs common with van Gent et al. (2011), the currently circulating ptxP3 isolates in the Netherlands belong to our SNP cluster I. Using SNPs common to both studies (15 common SNPs with two SNPs, BP2249 and BP2366 that can partially be used to assign ptxp3 isolates to cluster I), we were able map most of their ptxP3 SNP types to our cluster I. In addition, an Australian ptxP3 isolate (L584) typed in both studies belongs to their ST10 and our SP12, which are shown to have been the first to diverge among the SNP types carrying ptxP3 (Figs. 1 and 5 of van Gent et al. (2011)). The expansion of cluster I in Australia has been attributed to the introduction of selection pressure from the ACV (Octavia et al., 2011). In this study, we show that ptxP3 is associated with the emergence of cluster I. ptxP3 has been hypothesised to improve binding of the global virulence regulator, BvgA, resulting in the production of more toxin than ptxP1 strains (Mooi et al., 2009); giving the organism a potential selective advantage over ptxP1 strains. The increased levels of PT may lead to a more severe disease, which has been correlated with higher hospitalisation rates of children infected with B. pertussis containing ptxP3 alleles in the Netherlands (Mooi et al., 2009). In Sweden, ptxP3 has not only become the dominant allele, but has also been detected much earlier in regions where only a single component (i.e. PT) pertussis vaccine was used (Advani et al., 2011). This observation further suggests that vaccine-induced selection pressure plays a role in both selection of ptxP3 and expansion of cluster I. In addition to the switch from ptxP1 to ptxP3, the other predominant change observed in cluster I was the change from prn1 to prn2 (Octavia et al., 2011). Prn2 differs from Prn1 by an additional GGFGP motif located near the RGD motif which is postulated to be involved in receptor binding (Leininger et al., 1991). It is likely that conformational changes to Prn2 as a result of the additional GGFGP repeat could contribute to the type specificity of immunological responses (He et al., 2003). This suggestion is also supported by studies of experimental pertussis infection, which have shown that the WCV used in the Netherlands (derived from a strain containing prn1) did not protect against infection with B. pertussis isolates with prn2 allele as efficiently as against isolates with the prn1 allele (King et al., 2001). The ACVs available worldwide differ in their constituents and range from monocomponent (PT only) to five component vaccines (PT, PRN, FHA, FIM2 and FIM3) (Advani et al., 2011; Kallonen et al., 2011). In Australia, the most commonly used vaccine is a 3-component vaccine consisting of PT, PRN and FHA which was introduced in 1997 and replaced the WCV entirely by 2000. A five component ACV with additional FIM2 and FIM3 antigens is also available in Australia but only used in a small proportion of the vaccinations. The alleles for the 3-component ACV are ptxA2, prn1, and fhaB1 derived from Tohama I (Kurniawan et al., 2010). In addition to prn allele mismatch between the ACV and the currently circulating cluster I strains as discussed above, the allelic mismatch of ptxA, between the currently circulating in cluster I allele, ptxA1, and
494
C. Lam et al. / Infection, Genetics and Evolution 12 (2012) 492–495
Year(s) SP15 (1) SP16 (3) SP14 (41)
prn2 and ptxP3
I
SP10 ((1)) SP13 (23) SP41 (1) SP12 (4)
ptxP5 (1 isolate)
SP11 (16) SP42 (1)
SP17 (3) SP37 (34) SP38 (4) SP39 (6) SP9 ((8)) ptxP9 ptxP9,
II
ptxP19 (1 isolate each)
SP6 (2) SP7 (1) SP1 (8) SP8 (1) SP2 ((3)) SP3 (16) SP4 (6) SP5 (8)
III
SP19 (2) SP18 (11) SP20 (1) SP21 (4) SP22 (7) SP40 (5) SP23 (2) SP29 (3)
IV
SP31 (8) SP24 (10) SP30 (25) SP25 ((3)) SP26 (14) SP27 (19) SP28 (2)
ptxP8, P8 ptxP19 P19 (1 isolate each)
V
SP32 (2)
ptxP2 (1 isolate)
SP33 (2) SP34 (1) SP35 (1)
SP36 (1) Tohama 2
VI
Country(s)
2006 2007- 2008 1997-2006
FL AU AU, CA, FL, FR, HK, JP, US
1999 1997- 2005
JP AU, CA, FR, HK, JP, US
2007
JP
1994- 2004
JP, AU
1992- 2004
AU, FL, IT, MX, NL
2004
FL
1995- 2000 1989- 2003 1995- 1998 2000- 2006 1972- 2006
AU AU, FR AU AU AU,, FL
1977- 1978
AU
2006 1973- 2000 1997 1985- 2005 1989- 2001 1995- 2001 1995 1999- 2002
AU AU, FL AU AU AU, JP AU AU
2002 1992- 2008
AU AU
2002 1982
AU AU
1971- 1989
AU
1971- 1989
AU
999 2000 000 1999-
AU
2001- 2002
AU
2001- 2005 2001 1996- 2004 1977- 2004 1946- 1957 1989- 2000 1990 2007 19901996
AU AU AU CN,, US JP AU HK, AU, HK JP JP
1940- 1993
FR, US
1935 1950 1920
US FR UK
1954
JP
Key: ptxP4 ptxP3 p ptxP2 ptxP1
Fig. 1. Distribution of ptxP alleles on the SNP tree. Maximum parsimony tree showing the evolution of Bordetella pertussis based on 65 SNPs was from Octavia et al. (2011). Each branch is represented by a SNP profile (SP) which were grouped into clusters I–VI. The number of isolates sequenced per SP has been included in brackets. The distribution of the ptxP alleles is colour coded with key on the right. The change of prn1 to prn2 and ptxP1 to ptxP3 are marked on the branch. All SPs carried the ptxP1 allele unless otherwise noted on the tree. Country codes are as follows: AU – Australia, CA – Canada, CN – China, FL – Finland, FR – France, HK – Hong Kong, IT – Italy, JP – Japan, MX – Mexico, NL – the Netherlands, UK – United Kingdom, US – United States.
the ACV allele, ptxA2 should also be noted. The increased production of PT as a result of ptxP3 may exacerbate the effect of any antigenic difference between the two ptxA alleles in ACV- induced protection. Therefore, the combination of changes to prn2 and ptxP3, as well as the non-ACV allelic mismatch of ptxA1 in cluster I, may be a significant contributor to the increase in pertussis rates. This factor may play a similar role in pertussis resurgence in countries where the same or similar ACVs are used. 4. Conclusions In conclusion, the resurgence of pertussis in highly vaccinated populations can be, at least in part, explained by genetic changes
that increase the fitness of circulating B. pertussis isolates. The emergence and expansion of cluster I are associated with two important genetic events in B. pertussis evolution: the increased production of pertussis toxin as a result of change to the promoter region (ptxP3) and the change to a non-ACV prn2 allele. This study demonstrated that the ptxP3 allele is associated with cluster I. Combined with previously reported changes to the B. pertussis genome, in particular changes to antigenic determinants, this finding suggests that isolates in this cluster are fitter variants of B. pertussis. These observations highlight the public health relevance of monitoring the global distribution of B. pertussis cluster I isolates in highly immunised populations and may offer important insights for the development of new strategies to reduce the burden of pertussis.
C. Lam et al. / Infection, Genetics and Evolution 12 (2012) 492–495
Acknowledgements We gratefully acknowledge the generous donations of isolates by Dr. Kazunari Kamachi, Department of Bacterial Pathogenesis and Infection Control, National Institute of Infectious Diseases, Tokyo, Japan; Dr. Shane Byrne, Sullivan Nicolaides Pathology, Brisbane, Queensland, Australia; the late Dr. John Tapsall, Prince of Wales Hospital, Sydney, Australia; Dr. Ian Carter, St. George Hospital, Sydney, Australia; Dr. Margret Ip, Chinese University of Hong Kong; Dr. Qiushui He, Pertussis Reference Laboratory, National Public Health Institute, Turku, Finland; Dr. Nicole Guiso, Institut Pasteur, Paris, France; Dr. Raymond Tsang, National Microbiology Laboratory, Public Health Agency of Canada; and Dr. Lucia Tondella, Centers for Disease Control and Prevention, USA. We thank Marjolein van Gent for technical assistance. This research was supported by the National Health and Medical Research Council of Australia. Connie Lam was supported by an Australian Postgraduate Award. References Advani, A., Van der Heide, H.G., Hallander, H.O., Mooi, F.R., 2009. Analysis of Swedish Bordetella pertussis isolates with three typing methods: characterization of an epidemic lineage. J. Microbiol. Methods 78, 297–301. Advani, A., Gustafsson, L., Ahren, C., Mooi, F.R., Hallander, H.O., 2011. Appearance of Fim3 and ptxP3-Bordetella pertussis strains, in two regions of Sweden with different vaccination programs. Vaccine 29, 3438–3442. Bart, M.J., van Gent, M., van der Heide, H.G., Boekhorst, J., Hermans, P., Parkhill, J., Mooi, F.R., 2010. Comparative genomics of prevaccination and modern Bordetella pertussis strains. BMC Genomics 11, 627. Elomaa, A., Advani, A., Donnelly, D., Antila, M., Mertsola, J., He, Q., Hallander, H., 2007. Population dynamics of Bordetella pertussis in Finland and Sweden, neighbouring countries with different vaccination histories. Vaccine 25, 918–926. Godfroid, F., Denoel, P., Poolman, J., 2005. Are vaccination programs and isolate polymorphism linked to pertussis re-emergence? Expert. Rev. Vaccines 4, 757– 778. He, Q., Makinen, J., Berbers, G., Mooi, F.R., Viljanen, M.K., Arvilommi, H., Mertsola, J., 2003. Bordetella pertussis protein pertactin induces type-specific antibodies: one
495
possible explanation for the emergence of antigenic variants? J. Infect. Dis. 187, 1200–1205. Kallonen, T., Grondahl-Yli-Hannuksela, K., Elomaa, A., Lutynska, A., Fry, N.K., Mertsola, J., He, Q., 2011. Differences in the genomic content of Bordetella pertussis isolates before and after introduction of pertussis vaccines in four European countries. Infect. Genet. Evol. 11, 2034–2042. King, A.J., Berbers, G., van Oirschot, H.F., Hoogerhout, P., Knipping, K., Mooi, F.R., 2001. Role of the polymorphic region 1 of the Bordetella pertussis protein pertactin in immunity. Microbiology 147, 2885–2895. Kurniawan, J., Maharjan, R.P., Chan, W.F., Reeves, P.R., Sintchenko, V., Gilbert, G.L., Mooi, F.R., Lan, R., 2010. Bordetella pertussis clones identified by multilocus variable-number tandem-repeat analysis. Emerg. Infect. Dis. 16, 297–300. Leininger, E., Roberts, M., Kenimer, J.G., Charles, I.G., Fairweather, N., Novotny, P., Brennan, M.J., 1991. Pertactin, an Arg-Gly-Asp-containing Bordetella pertussis surface protein that promotes adherence of mammalian cells. Proc. Natl. Acad. Sci. USA 88, 345–349. Litt, D.J., Neal, S.E., Fry, N.K., 2009. Changes in genetic diversity of the Bordetella pertussis population in the United Kingdom between 1920 and 2006 reflect vaccination coverage and emergence of a single dominant clonal type. J. Clin. Microbiol. 47, 680–688. Mooi, F.R., van Loo, I.H., King, A.J., 2001. Adaptation of Bordetella pertussis to vaccination: a cause for its reemergence? Emerg. Infect. Dis. 7, 526–528. Mooi, F.R., van Loo, I.H., van Gent, M., He, Q., Bart, M.J., Heuvelman, K.J., de Greeff, S.C., Diavatopoulos, D., Teunis, P., Nagelkerke, N., Mertsola, J., 2009. Bordetella pertussis strains with increased toxin production associated with pertussis resurgence. Emerg. Infect. Dis. 15, 1206–1213. Octavia, S., Maharjan, R.P., Sintchenko, V., Stevenson, G., Reeves, P.R., Gilbert, G.L., Lan, R., 2011. Insight into evolution of Bordetella pertussis from comparative genomic analysis: evidence of vaccine-driven selection. Mol. Biol. Evol. 28, 707– 715. Poynten, M., McIntyre, P.B., Mooi, F.R., Heuvelman, K.J., Gilbert, G.L., 2004. Temporal trends in circulating Bordetella pertussis strains in Australia. Epidemiol. Infect. 132, 185–193. Schouls, L.M., van der Heide, H.G., Vauterin, L., Vauterin, P., Mooi, F.R., 2004. Multiple-locus variable-number tandem repeat analysis of Dutch Bordetella pertussis strains reveals rapid genetic changes with clonal expansion during the late 1990s. J. Bacteriol. 186, 5496–5505. Tanaka, M., Vitek, C.R., Pascual, F.B., Bisgard, K.M., Tate, J.E., Murphy, T.V., 2003. Trends in pertussis among infants in the United States, 1980–1999. JAMA 290, 2968–2975. van Gent, M., Bart, M.J., van der Heide, H.G., Heuvelman, K.J., Kallonen, T., He, Q., Mertsola, J., Advani, A., Hallander, H.O., Janssens, K., Hermans, P.W., Mooi, F.R., 2011. SNP-based typing: a useful tool to study Bordetella pertussis populations. PLoS ONE 6, e20340.