Emerging Microbes & Infections 2019, VOL. 8 https://doi.org/10.1080/22221751.2019.1694395 ORIGINAL ARTICLE Population dynamics and antigenic drift of Bordetella pertussis following whole cell vaccine replacement, Barcelona, Spain, 1986–2015 Alba Mir-Cros a,b*, Albert Moreno-Mingorance a,b*, M. Teresa Martín-Gómez a, Gema Codina a , Thais Cornejo-Sánchez a, Mireia Rajadella,b, Diego Van Essoc, Carlos Rodrigo d, Magda Campins e , Mireia Janéf, Tomàs Pumarola a,b, Anna Fàbrega a,g† and Juan José González-López a,b† a Department of Clinical Microbiology, Hospital Universitari Vall d’Hebron, Barcelona, Spain; bDepartment of Microbiology and Genetics, Universitat Autònoma de Barcelona, Barcelona, Spain; cPrimary Care Health Centre Service ’Muntanya’, Catalan Institute of Health, Barcelona, Spain; dDepartment of Paediatrics, Hospital Universitari Vall d’Hebron, Barcelona, Spain; eDepartment of Preventive Medicine and Epidemiology, Hospital Universitari Vall d’Hebron, Barcelona, Spain; fPublic Health Agency of Catalonia, Barcelona, Spain; gFaculty of Health Sciences, University of Vic – Central University of Catalonia (UVic-UCC), Manresa, Spain ABSTRACT Among the factors associated with the resurgence of whooping cough, special emphasis has been given to pathogen adaptation after the introduction of the acellular vaccine (ACV). To assess the impact of the vaccine transition strategy from whole-cell vaccine (WCV) to ACV on population dynamics of Bordetella pertussis in Barcelona (Spain), we studied 339 isolates collected from 1986 to 2015 by PFGE and multi-locus variable-number tandem repeat analysis (MLVA). Additionally, allelic variants for the pertussis toxin and its promoter, pertactin, type 3 fimbriae and fimbrial serotyping were assessed to determine its antigenic drift. A shift was observed in the B. pertussis population as well as in its antigenic profile concurrently with the introduction of ACV in Barcelona. Four out of the five most prevalent PFGE profiles were replaced by new profiles following the ACV introduction. MLVA type 27 was the dominant genotype, and its frequency increased from 25% to 79.3% after WCV replacement. Antigen typing demonstrated the emergence of prn2, ptxP3, fim3-2 and a shift from the fimbriae 3 to the fimbriae 2 serotypes after the ACV introduction. Our findings support the presence of population and antigenic dynamic changes in B. pertussis likely driven by the introduction of ACV. ARTICLE HISTORY Received 20 June 2019; Revised 17 October 2019; Accepted 6 November 2019 KEYWORDS Whooping cough; PFGE; MLVA; antigenic variants; pertussis vaccine Introduction twenty-first century, not only in Spain and Europe but also worldwide. This unfavourable situation has Whooping cough, or pertussis, is an acute human upper led to pertussis becoming the leading vaccine-preven- respiratory tract infection caused by Bordetella pertussis, table disease in industrialized countries, and hence, a a highly communicable airborne Gram-negative cocco- global public health problem [2–5]. bacillus. The pertussis immunization programme was Pertussis normally presents a classical cyclic pattern first introduced in Spain in 1965, with the adminis- with epidemic waves occurring every 3–5 years inter- tration of a whole cell vaccine (WCV). Later, in 1998 spersed with periods of a lower incidence rate. How- an acellular vaccine (ACV) progressively replaced the ever, at present, the number of cases is higher than in WCV vaccination programme with the aim of improv- previous decades [6]. The reemergence of pertussis in ing vaccine safety. Finally, since 2005 ACV is the only Spain is particularly well supported by recent epide- vaccine administered to the Spanish population [1]. In miological data. The incidences reported for the inter- Catalonia, the vaccination programme against pertussis epidemic years from 2001 to 2010 ranged from 0.7 to follows the same programme (Table 1). 1.9 cases/100,000 population. However, despite having Despite extensive vaccination campaigns and high a vaccination coverage of 96.5% in infants younger immunization rates (86% for global primary vacci- than 12 months, the incidence rates observed after nation in 2018), the incidence of whooping cough sig- the last two epidemic waves (2011 and 2015) remained nificantly reemerged during the first decade of the high even during interepidemic periods (>5 cases/ CONTACT Juan José González-López jjgonzal@vhebron.net Department of Clinical Microbiology, Hospital Vall d’Hebron, Pg.Vall d’Hebron 119-129, 08035 Barcelona, Spain; Anna Fàbrega anna.fabrega@vhir.org Department of Clinical Microbiology, Hospital Vall d’Hebron, Pg.Vall d’Hebron 119-129, 08035 Barcelona, Spain *These authors contributed equally to this article. † These senior authors contributed equally to this article. Supplemental data for this article can be accessed https://10.1080/22221751.2019.1694395 © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of Shanghai Shangyixun Cultural Communication Co., Ltd This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 1712 A. Mir-Cros et al. Table 1. Pertussis vaccination programme used in Catalonia (Spain) since 1965. Primary doses Booster doses Vaccine Vaccine Year type Schedule Pertussis components type Schedule Pertussis components 1965 DTPw 2 doses between 3 Inactivated whole cell NA No booster NA months and 3 years 1967 DTPw 3 doses between 3 Inactivated whole cell NA No booster NA months and 3 years 1975 DTPw 3, 5, 7 months Inactivated whole cell NA No booster NA 1996 DTPw 2–3, 4–5, 6–7 months Inactivated whole cell DTPw 15–18 months Inactivated whole cell 1998–1999 DTPw 2, 4, 6 months Inactivated whole cell DTPa 18 months 4–6 years PT, FHA, PRN 2000 DTPw/DTPa 2, 4, 6 months Inactivated whole cell or DTPa 18 months 4–6 years PT, FHA, PRN PT, FHA, PRN 2002 DTPa 2, 4, 6 months PT, FHA, PRN DTPa 18 months 4–6 years PT, FHA, PRN 2011 DTPa 2, 4, 6 months PT, FHA, PRN DTPa/ 18 months (DTPa), 4–6 PT, FHA, PRN or PT, FHA, dTpa years (dTpa) PRN, FIM2, FIM3a 2014 DTPa 2, 4, 6 months PT, FHA, PRN DTPa/ 18 months (DTPa), 6 years PT, FHA, PRN or PT, FHA, dTpa (dTpa) PRN, FIM2, FIM3a b 2014 NA NA NA dTpa between 27 through 36 PT, FHA, PRN or PT, FHA, weeks of pregnancy PRN, FIM2, FIM3 Notes: NA: not applicable; DTPw: diphtheria-tetanus-whole cell pertussis vaccine; DTPa: diphtheria-tetanus-acellular pertussis vaccine; dTpa: diphtheria-teta- nus-accellular pertussis vaccine with reduced antigenic load of diphtheria, tetanus and pertussis; PT: pertussis toxin; FHA: filamentous haemagglutinin; PRN: pertactin; FIM2: type 2 fimbriae; FIM3: type 3 fimbriae. These data have been collected from Campins M, et al ([1] and personal communication). a The composition containing pertussis fimbrial antigens is used in some cases for the fifth dose. b Introduction of maternal pertussis vaccination. 100,000 population). Of note was the increase reported those collected in the metropolitan area of Barcelona during the last epidemic wave in 2015 which showed a before, during and after the WCV replacement by maximum incidence of almost 18 cases/100,000 popu- ACV. This characterization was achieved by molecu- lation [7]. The figures reported in Catalonia have fol- lar epidemiology analysis of clinical isolates together lowed a similar trend albeit with an even more with a study of the vaccine antigen variants of the pronounced progression, with an incidence of 49 pathogen. cases/100,000 population in 2015 [8]. The resurgence of whooping cough may be related to increased awareness and improved diagnostic Materials and methods tools, although the special emphasis has been given Bacterial isolates and study period to waning immunity, particularly associated with the introduction of ACV and pathogen adaptation [9,10]. A total of 339 non-duplicate B. pertussis clinical isolates ACV contains a combination of different antigens were collected at the Hospital Vall d’Hebron including pertussis toxin (PT), pertactin (PRN) and (Barcelona, Spain). All the isolates were recovered filamentous haemagglutinin (FHA), and in some vac- from cultures of nasopharyngeal samples collected cines, type 2 and type 3 fimbriae (FIM2 and FIM3) from patients diagnosed with pertussis, excluding [2]. Nevertheless, the immunologic protection con- isolates of the studies of contacts. The isolates were col- ferred by ACV is not as enduring as that previously lected from patients with different vaccination status: observed for WCV [11]. The current allelomorphic vaccinated, unvaccinated and partially vaccinated profile of the ACV antigens includes ptxA2/ptxA4, (Table S1, Supplementary Information). prn1/prn7, fhaB1, fim2-1 and fim3-1 [12]. The natu- The isolates were collected over 30 years, from 1986 rally-driven antigenic divergence, together with the to 2015. The study period was divided into three parts impact exerted by the ACV vaccine has likely played on the basis of the vaccine type/s used for routine vac- a key role in the selection of new antigenic variants, cination in our setting: (i) period 1 (1986–1997; n = 82 which is in keeping with the increased frequencies of isolates) was defined by the single use of WCV; period the allelic variants ptxA1, prn2, and fim3-2 seen in 2 (1998–2003; n = 82 isolates) was the transition period countries where ACV has been extensively used as WCV was used only for primary vaccination [2,4,13]. Another bacterial adaptation event associated whereas ACV was used in boosters; and period 3 with the resurgence of this pathogen is the selection of (2007–2015, n = 175 isolates), when ACV completely a new PT promoter type. In 2009 the new ptxP3 variant replaced WCV. Additionally, 10 previously character- was characterized in relation to increased toxin pro- ized strains (B3313, B1900, B1706, B1917, B2726, duction levels and was shown to have rapidly replaced B0366, B0549, B0610, B1916 and B3230) were included the previous predominant ptxP1 allele [14,15]. in the present study for comparison in the clonal relat- The main objective of this study was to shed edness studies. These strains have been described as new light on the impact of the introduction of ACV being the most representative clones (PFGE profiles on the evolution and adaptation of B. pertussis iso- and MLVA types) circulating in several European lates over a 30-year period in Spain, particularly countries during the period 1998–2015 [16–18]. Emerging Microbes & Infections 1713 Pulsed-field gel electrophoresis (PFGE) B1900 (FIM3) and B3313 (FIM2 and FIM3) were used as positive controls. The genetic relatedness of all the B. pertussis isolates included in the present study was determined using DNA fingerprint by PFGE, which was performed as Statistical analysis described previously [19]. The DNA fingerprint profiles were analysed with the GelCompare II v.4.6 PFGE profiles and MLVA types diversity were calcu- software (Applied Maths) using the arithmetic lated based on the ratio between the PFGE profiles or UPGMA as a group method with a band tolerance of MLVA types identified by the total number of isolates. 1% and an optimization setting of 1%. Underrepre- The Simpson diversity index (SDI) was calculated based sented PFGE profiles (n < 5 isolates) were grouped as on the formula SDI = 1 − Σn(n − 1)/N(N − 1) where n “others”. indicates the number of individual PFGE profiles or MLVA types identified and N indicates the number of all PFGE profiles or MLVA types observed [18]. Multi-locus variable-number tandem repeat analysis (MLVA) Results Strain relatedness was also studied using MLVA typing analysis as described previously [20]. A subset of repre- Study periods sentative strains belonging to different PFGE profiles All the isolates recovered in the present study (n = 339) and time periods was selected for our MLVA compari- were grouped into three study periods taking into son (n = 72). Accordingly, one isolate was selected from account the gradual WCV replacement: period 1 each of the less representative PFGE profiles (29 PFGE (1986–1997, n = 82) was defined as WCV adminis- profiles, 29 isolates). For the most prevalent PFGE tration alone; period 2 (1998–2003, n = 82) was the profiles (9 PFGE profiles, 43 isolates), 3 isolates were vaccine transition period, and period 3 (2007–2015, selected from each time period (1 from the beginning, n = 175) was characterized as exclusive ACV one from the middle and 1 from the end of each administration. period). In the case of PFGE profiles partially rep- resented in a time period, available isolates were selected. Each locus was amplified by PCR and the Molecular typing: PFGE resulting fragments analysed as described previously The genetic relatedness of the 339 isolates was evalu- [20]. MLVA types (MT) were assigned using the ated by PFGE. The results identified a total of 38 differ- MLVA typing tool (https://www.mlva.net/bpertussis/ ent PFGE profiles (Table S1, Supplementary default.asp). Information). Of these, 19, 10 and 18 PFGE profiles from periods 1, 2 and 3, respectively were identified. Antigenic variants VH2 was the only PFGE profile detected along the entire period (1986–2015), with a remarkable gap in A total of 231 strains belonging to different PFGE the epidemic year 2000 and the following interepi- profiles were included in the characterization of anti- demic years 2001–2002 (Figure 1). Additionally, a genic variants: a subset of representative strains (n = replacement of PFGE profiles was observed across the 56) was selected for periods 1 and 2 whereas all strains three periods, being more notable from period 1 to collected during period 3 were analysed (n = 175). period 2. In period 1 (1986–1997, WCV) the PFGE Among the genes encoding virulence factors included profiles most frequently detected were VH2 (23.2%), in the ACV, the A subunit of the pertussis toxin VH8 (17.1%), VH5 (14.6%), VH12 (12.2%) and VH7 (ptxA), pertactin (prn) and type 3 fimbriae (fim3) (7.3%). The introduction of ACV during period 2 were studied using PCR-based sequencing as described (1998–2003) correlated with the emergence of VH19 previously [21,22]. Additionally, the promoter type of (40.2%) and VH20 (36.6%), which were only margin- pertussis toxin (ptxP) was also studied [14]. ally detected (1.2%) during period 1. Finally, in period 3 (2007–2015, ACV) PFGE profiles VH19 (28.6%), VH26 (24.6%), VH2 (18.3%) and VH22 (8%) were Fimbrial serotyping the most prevalent. During this last period, VH19 Fimbriae 2 (FIM2) and 3 (FIM3) serotyping of all 339 was the most frequent PFGE profile before the epi- B. pertussis isolates included in the present study were demic year 2011 (from 2007 to 2010, 48.6%; from performed by the indirect ELISA method as described 2011 to 2015, 15.2%), whereas progressive replacement previously [23,24]. Monoclonal antibodies (06/124; by PFGE profiles VH26 and VH2 was observed from FIM2, and 06/128; FIM3) were obtained from the 2011 onwards (from 2007 to 2010, 10% and 11.4%, National Institute for Biological Standards and Control respectively; from 2011 to 2015, 34.3% and 22.9%, (NIBSC) and the reference strains Tohama I (FIM2), respectively) (Figure 1). 1714 A. Mir-Cros et al. Figure 1. Temporal distribution of PFGE profiles of B. pertussis circulating in the metropolitan area of Barcelona from 1986 to 2015. WCV: Whole cell vaccine; ACV: Acellular vaccine; Bold-face type is used to indicate epidemic years. The PFGE profiles of clinical isolates obtained a minor profile (4–7%) in Europe throughout the from several European countries from 1998 to 2015 study period. were compared [16–18]. The results showed that the predominant European PFGE profiles BpSR3, Molecular typing: MLVA BpSR5, BpSR10, BpSR11 and BpSR12 were respect- ively indistinguishable from VH26, VH22, VH2, MLVA typing was performed in 72 isolates selected VH19 and VH20 reported in this study, (Figure S2, as being representative of each PFGE profile (28, 16 Supplementary Information). General trends were and 28 isolates from each period, respectively) similarly observed in both geographical areas (Europe (Table S1, Supplementary Information). Overall, 13 and Spain) (Table 2): (i) BpSR11 (VH19) was the pre- different MLVA types (MTs) were found: 11, 4 and dominant profile during the period 1998–2005 (27– 4 MTs were detected during periods 1, 2 and 3, 40%) and showed decreased prevalence in recent respectively. Among these, MT27 was the most preva- years (15–29%); (ii) BpSR10 (VH2) was less prevalent lent type (56.9%) and clearly showed an increasing during the period 1998–2005 but has increased since frequency: 25% during period 1, 68.8% during period 2006 (from 8% to 27% in Europe, from 4% to 18% in 2 and 79.3% in the third period (Figure 2). Less Spain); (iii) BpSR5 (VH22) has shown a similar low representative MTs were MT16, MT28, MT60, prevalence in both studies since its first detection in MT70, MT95, MT101 and MT158, which were 1998 (1–11%); and (iv) BpSR3 (VH26) remained detected in more than one isolate with a prevalence undetected from 1986 to 2003 but rapidly became per period ranging from 6.3% to 18.8%. Among one of the most prevalent types during the period these, MT28 was the only MT detected throughout 2007–2015 (22–29% in both areas). On the contrary, the three periods (7.1%, 6.3% and 7.1%, respectively). the greatest difference was seen for the PFGE profile The remaining MTs were exclusive of period 1 BpSR12 (VH20). It was one of the most prevalent (MT70, MT16 and MT95), exclusive of period 3 types in Spain (37%) from 1998 to 2003, thereafter (MT101) or present in both periods 1 and 2 (MT60 it showed a remarkable decrease (4%) and was only and MT158). Finally, five MTs were represented by Table 2. PFGE profiles of B. pertussis circulating in the metropolitan area of Barcelona from 1986 to 2015 compared to PFGE profiles circulating in European countries from 1998 to 2015. Time periods European Time periods HVH PFGE 1986–1997 1998–2003 2007–2015 Equivalent EUpert I (1998– EUpert II (2004– EUpert III (2007– EUpert IV (2012– profile n = 82 n = 82 n = 175 PFGE profilea 2001) n = 102 2005) n = 154 2009) n = 140 2015) n = 265 VH2 23 4 18 BpSR10 8 10 21 27 VH19 0 40 29 BpSR11 27 30 13 15 VH20 1 37 4 BpSR12 4 7 4 4 VH22 0 1 8 BpSR5 6 8 11 5 VH26 0 0 25 BpSR3 0 8 22 29 Total 24 82 84 Total 45 63 71 80 Note: Values are %. a These data have been collected from EUpert studies [16–18]. Emerging Microbes & Infections 1715 Figure 2. MLVA types of B. pertussis circulating in the metropolitan area of Barcelona from 1986 to 2015. WCV: Whole cell vaccine; ACV: Acellular vaccine. single isolates: MT32, MT133, MT135 and MT146 the most frequently studied vaccine components from period 1 and MT30 from period 3 (3.6% indi- (PT, PRN and FIM3) among the B. pertussis popu- vidual prevalence per period). lation from Barcelona over the last 30 years (Figure 3). The promoter type of the pertussis toxin (ptxP) was also studied. A total of 231 isolates were analysed Comparative genetic diversity (40 belonging to period 1, 16 to period 2, and 175 The comparison of the genetic diversity among the to period 3). three study periods based on differences in percentages Regarding the ptxA gene, our results revealed that all of PFGE profiles and MTs showed a gradual decrease the isolates analysed carried the ptxA1 allele (Figure 3 over time (Table 3). Thus, the diversity of the PFGE (A)). In contrast, an important shift was observed con- profiles identified decreased from 23.2% in period 1, cerning the promoter type. PtxP1 was the most preva- to 12.2% in period 2, and to 10.3% in period 3. A simi- lent type while WCV was administered (period 1, lar and more evident decrease was seen for the MLVA- 87.5%; period 2, 12.5%) but became undetectable types, from 39.3% in period 1, to 25% in period 2, and when ACV was the only type of vaccine used in to 14.3% in period 3. The Simpson’s diversity indexes Spain (period 3) (Figure 3(B)). An inverse correlation calculated for each study period showed no differences was seen for the new ptxP3 variant associated with in the transition from WCV to ACV on analysing the higher PT production levels. It was first detected in PFGE profiles (0.88 period 1, 0.70 period 2 and 0.82 1996 (5%, period 1) and rapidly became the predomi- period 3), whereas differences were noted when nant allele (87.5% and 97.7% for periods 2 and 3, MLVA was taken into consideration (0.89 period 1, respectively). This predominance was seen in all the 0.52 period 2, and 0.33 period 3). PFGE profiles analysed with the exception of VH5, a PFGE profile almost exclusive of period 1, as the only VH5 isolate collected in 2000 and analysed here har- ACV impact on the evolution of antigen variants boured the ptxP1 variant. Other less prevalent promo- and fimbrial serotype ter types (2.3–2.5%) were observed in period 1 (ptxP5, ptxP10 and ptxP11) and period 3 (ptxP15). Allelic variant analysis was carried out in a selection The pertactin gene allele distribution proved to be of isolates to assess the prevalence and evolution of heterogeneous during period 1. Specifically, prn1 and Table 3. Percentage of PFGE profiles and MLVA types prn3 were the most frequently detected (42.5% and identified and Simpson diversity index calculated for PFGE 20%, respectively), while a lower frequency of prn2 and MLVA in the three study periods. was detected (3.7%). Nonetheless, the prn2 allele Time periods became the most predominant after the introduction Genetic diversity 1986– 1998– 2007– of ACV, reaching a prevalence of 87.5% in period 2 Method analysis 1997 2003 2015 and 97.7% in period 3 (Figure 3(C)). PFGE Diversity (%) 23.2 12.2 10.3 Simpson diversity index 0.88 0.70 0.82 Finally, regarding the fim3 genotype, our results MLVA Diversity (%) 39.3 25 14.3 showed that all the isolates studied in period 1 carried Simpson diversity index 0.89 0.52 0.33 the fim3-1 variant, whereas a similar prevalence of 1716 A. Mir-Cros et al. Figure 3. Antigen variants detected in B. pertussis circulating in the metropolitan area of Barcelona from 1986 to 2015. (A) Pertussis toxin. (B) Pertussis toxin promoter type. (C) Pertactin. (D) Type 3 fimbriae. P1: exclusive use of whole cell vaccine; P2: transition from whole cell vaccine to acellular vaccine; P3: exclusive use of acellular vaccine. fim3-1 and fim3-2 variants were detected in periods 2 Regarding fimbrial serotyping, overall 72.6% of the (43.8% and 56.3%, respectively) and 3 (46.9% and isolates had FIM3, 20.9% FIM2 and 3.2% FIM2/3. 53.1%, respectively) (Figure 3(D)). Intriguingly, from Additionally, 11 isolates (3.2%) were deficient for 2011 onwards fim3-1 showed a progressive reemer- FIM2 and FIM3. The distribution of the serotypes gence to again become the most predominant variant per periods revealed that there was a significant over fim3-2 (data not shown). Thus, the 2 most preva- increase of FIM3 isolates during the transition from lent allelic combinations detected in the present study WCV to ACV (57.3% during period 1,90.2% during during the most recent years were defined by isolates period 2; P < .05) while FIM2 decreased dramatically carrying the same allelic composition for ptxA1, (31.7% during period 1, 0% during period 2; P < .05). ptxP3 and prn2, while only differing in the allele encod- However, once the ACV was completely introduced ing FIM3 (Figure 4). (period 3), a progressive decrease of FIM3 isolates Figure 4. Temporal evolution of the most prevalent antigenic profiles observed in B. pertussis circulating in the metropolitan area of Barcelona from 1986 to 2015. Period 1: exclusive use of whole cell vaccine; Period 2: transition from whole cell vaccine to acellular vaccine; Period 3: exclusive use of acellular vaccine. Emerging Microbes & Infections 1717 as well as in its antigenic profile concurrent with the introduction of ACV, as noted in other countries. Only isolates showing the VH2 PFGE profile were found to persist throughout the entire study period. Isolates belonging to PFGE profiles VH8, VH5 and VH12, which were predominant during the years in which WCV was the only vaccine administered, were almost completely replaced by the newly appearing profiles VH19, VH20 and VH22, which became domi- nant once ACV remained as the only vaccine in use. Finally, PFGE profile VH26 appeared in 2007 and has become the most prevalent profile since 2011. These results are largely consistent with the most Figure 5. Evolution of fimbrial serotypes in B. pertussis isolates prevalent PFGE profiles detected in Europe within circulating in the metropolitan area of Barcelona from 1986 to the same 30-year time frame [2,16–18]. Our findings 2015. WCV: Whole cell vaccine; ACV: Acellular vaccine. are also in agreement with the predominant PFGE profiles circulating in the US since 2000 [3]. Overall, was detected (71.4%) as long as FIM2 isolates gradually these results reinforce the successful intercountry dis- reemerged (25.7%) (Figure 5). During period 1 (WCV), semination of B. pertussis lineages not only among all FIM2 and FIM3 isolates were associated with a fim3- neighbouring countries but also among overseas terri- 1 genotype, while during period 2 (transition from tories with similar vaccination policies. However, slight WCV to ACV) 46.2% and 53.8% of the FIM3 isolates variations in the prevalence of the distribution of some studied belonged to fim3-1 and fim3-2 genotypes, PFGE profiles were noted on comparing local and respectively. Finally, in period 3 (ACV), all the FIM2 European isolates. Specifically, the PFGE profile isolates had the fim3-1 genotype while both genotypes VH20 was one of the most prevalent types in Barcelona were identified (71.2% fim3-2 and 28.8% fim3-1) for from 1998 to 2003 (37%) but showed a noteworthy FIM3 isolates. There was no correlation among sero- decrease from 2007 to 2015 (4%). On the contrary, types and vaccinated, non-vaccinated and partially vac- its equivalent European PFGE profile (BpSR12) was cinated patients (P > .05). only a minor profile in Europe during the whole Combining the analysis of PFGE profiles and study period (4–7%). This fact highlights the need for fimbrial serotype, most of the FIM2 isolates belonged regional studies to fully understand population to PFGE profiles VH26 and VH2 (42.3% and 18.3%, dynamics within and across territories. Probably, the respectively) while most of the FIM3 isolates belonged introduction of whole-genome sequence in future to PFGE profiles VH19, VH2 and VH20 (32.1%, 15.4% studies may shed new light on the understanding of and 14.2%, respectively). Regarding the most common the evolution and rate of lineage mixing within areas. PFGE profiles, isolates belonging to VH2 and VH26 Next, we assessed whether there are epidemiological included positive isolates for the FIM2 (24.1% and differences between epidemic and inter-epidemic 69.8%, respectively) and also the FIM3 (70.4% and periods. The percentage of minority PFGE profiles 27.9%, respectively) serotypes, whereas PFGE profiles was clearly increased in the latter, particularly for VH19, VH20 and VH22 were almost exclusively period 1 and less pronounced in period 2. It is of inter- FIM3 (95.2%, 92.1% and 100%, respectively). There est that the VH12 PFGE profile was only detected was no significant association between FIM2/3, or during the epidemic years 1989, 1992 and 1997. FIM2 and FIM3 deficient isolates and specific PFGE These results suggest that some epidemic waves profiles (Table S1, Supplementary Information). might be related to particular clones sharing specific characteristics, what could contribute to increased dis- semination potential. Discussion The MLVA results supported the PFGE findings Changes in the B. pertussis population have been regarding temporal trends and dynamic changes in reported worldwide [2–4,13]. Nonetheless, there are the population structure of B. pertussis in Barcelona no data related to Spanish strains. In this study, 339 over time. A trend to a lower diversity of MTs was B. pertussis isolates collected in Barcelona during the observed in concurrence with the switch in vaccines, period 1986–2015 were analysed to assess the molecu- mainly due to the growing dominance of MT27 (25% lar evolution of this pathogen. Moreover, the global during WCV administration from 1986 to 1997, to impact of the vaccine transition strategy from WCV 79.3% once ACV was completely implemented). Simi- to ACV was assessed by integrating data from molecu- lar figures have been reported across Europe as well as lar typing and antigenic variance. The results of this in the US and Australia since the introduction of ACV study show a clear shift in the B. pertussis population [2,13,18,25]. In contrast, the prevalence of MT27 is 1718 A. Mir-Cros et al. marginal in countries in which WCV is still used, or isolates were the most prevalent from 2012 to 2015. has been used until only recently, such as in China Conversely, FIM3 isolates were predominant during (6% in period 2012–2013), Poland (2.9% in period the period 2012–2015 in France, United Kingdom 1959–2013) or the Philippines (not detected) [26–28]. and Sweden, although the ACV contained fimbrial Apart from changes in the bacterial epidemiological components only in the first two [18]. Therefore, fac- profiles, we also assessed the impact of the introduction tors other than the type of fimbrial antigen contained of ACV on the allelic selection of vaccine-related viru- in the vaccine may be involved in the serotype shift. lence determinants. We found that all the isolates col- On the other hand, comparing the association of lected across the entire study period were ptxA1, in fimbrial serotype with PFGE profiles we observed agreement with previous reports in which ptxA1 was that, as previously found in the EUpert III and IV described as being the most prevalent variant at the studies, isolates belonging to VH19 (BpSR11), VH20 end of the 1960s immediately after the introduction (BpSR12) and VH22 (BpSR5) PFGE profiles were of the WCV [9]. Consequently, as the PT variants mostly associated with FIM3 regardless of the period included in the ACV are encoded by ptxA2 and in which they were isolated. However, during the tran- ptxA4 alleles, a protective effect against ptxA1 is no sition from WCV to ACV, all VH2 (BpSR10) and longer expected. VH26 (BpSR3) isolates were FIM3 while a shift On the contrary, two remarkable antigenic shifts towards the FIM2 serotype was observed once the were found: the first was related to the new PT promo- ACV was fully implemented [17,18]. ter variant ptxP3, and the second to the prn2 allele. According to the results of the present study, the Strains carrying the ptxP3 variant have been character- alleles detected in the most prevalent isolates circulating ized by increased toxin production levels and higher in our population since 2011, and at least until 2015, virulence potential [14]. Moreover, such strains have were as follows: ptxA1 for the pertussis toxin, which is been associated with the reemergence of the pathogen regulated by a pxtP3 type promoter, prn2 for pertactin in most parts of the world since the 1990s, concomitant and fim3-1 for serotype 3 fimbriae. The corresponding to a shift towards older age groups [13,14,29]. The shifts in the antigenic profile of the strains can be par- abrupt replacement of ptxP1 by ptxP3 seen in period tially correlated with PFGE profile replacement, par- 2 was paralleled by a similar change in the prevalence ticularly in period 3: the PFGE profile VH19 (mainly of PRN variants, with prn2 becoming dominant after ptxA1, ptxP3, prn2, fim3-2) has been replaced by almost completely displacing prn1. This is in the line VH26 (mainly ptxA1, ptxP3, prn2, fim3-1). These of previous reports showing a similar shift towards a results suggest that the absence of uniformity in vaccine clear dominance of the prn2 allele after the introduc- composition has failed to counteract the dissemination tion of ACV, which includes the allelic variants prn1 of fim3-1 isolates. Dissemination of VH26 could be sup- and prn7 which were most frequently detected before ported by a bacterial advantage provided by this the withdrawal of WCV [9,18]. fimbrial allele or by another virulence trait specific of Lastly, we also completed the antigenic profile of our this clone. Conversely, VH2 (mainly ptxA1, prn2, isolates by the characterization of the FIM3 genotype, fim3-1) was detected throughout three study periods, which is absent in some commercially available and several factors may have influenced its continuous ACVs. To date, two alleles have been reported as detection: no selective pressure on pertactin together being the most prevalent: fim3-1 and fim3-2. The with an inner adaptation ability, highlighting that the results of our study show that all the strains analysed same PFGE profile acquired the widely disseminated from period 1 only carried the fim3-1 variant, whereas ptxP3 variant. Thus, the adaptative plasticity of a substantial decrease was subsequently observed B. pertussis facilitates the diversity and selection of the towards a similar proportion (43.8% for period 2 and most appropriate clones in addition to an evolving abil- 46.9% for period 3). Nonetheless, since 2011 fim3-1 ity to integrate chromosomal changes in the same clone. has shown a progressive reemergence towards the pre- In conclusion, these results support the presence of dominant variant. Similar results have been reported in dynamic changes in the B. pertussis population as well Dutch isolates which showed a transitory increased as in its antigenic profile in Barcelona (Spain) over a prevalence of fim3-2 strains which reverted a few 30-year time period, with a trend towards homogeniz- years later [30]. Although overall most of the isolates ation following the introduction of the ACV. Our were FIM3 (72.6%), on stratifying the results according findings suggest that the use of the ACV has likely to the vaccine administered once ACV was introduced, driven the population changes and antigenic shift of a progressive shift from FIM3 to FIM2 was observed, B. pertussis. even though most of ACV compositions used in our area do not contain any fimbrial antigen (Table 1). Similar results were reported in different European Acknowledgments countries, such as Denmark and Finland, where ACV We are grateful to Marjolein Van Gent from the National does not contain fimbrial components and FIM2 Institute for Public Health and the Environment (RIVM, Emerging Microbes & Infections 1719 Bilthoven, the Netherlands) for kindly providing us the llance/burden/vpd/surveillance_type/passive/pertussis reference European B. pertussis isolates used in this study. /en/. We like to thank Victor Llorens, Adelaida Ferrer and Juliana [6] Kilgore PE, Salim AM, Zervos MJ, et al. Pertussis: Esperalba from Hospital Vall d’Hebron for their contri- microbiology, disease, treatment, and prevention. bution in the creation of the B. pertussis isolates collection Clin Microbiol Rev. 2016;29(3):449–486. and for their technical advice. [7] Instituto de Salud Carlos III. Enfermedades de declaración obligatoria. Informes anuales. 2018 [cited 2019 Apr 15]. https://eng.isciii.es/eng.isciii.es/Que Disclosure statement Hacemos/Servicios/VigilanciaSaludPublicaRENAVE/ EnfermedadesTransmisibles/Paginas/Informes.html. No potential conflict of interest was reported by the authors. [8] Canal Salut. Butlletí Epidemiològic de Catalunya (BEC). 2018 [cited 2019 Apr 15]. http://canalsalut. gencat.cat/ca/actualitat/llista_butlletins/salut_publica/ Funding butlleti_epidemiologic_de_catalunya/. [9] Mooi FR, van der Maas NAT, de Melker HE. Pertussis This work was supported by the “Ministerio de Economía y resurgence: waning immunity and pathogen adap- Competitividad”, “Instituto de Salud Carlos III”, and cofin- tation – two sides of the same coin. Epidemiol Infect. anced by the European Regional Development Fun 2014;142(4):685–694. (ERDF) “A Way to Achieve Europe” (Spanish Network for [10] He Q, Mertsola J. Factors contributing to pertussis Research in Infectious Diseases [grant number RD16/0016/ resurgence. Future Microbiol. 2008;3(3):329–339. 0003], “Fondo de Investigación Sanitaria [grant number [11] Ausiello CM, Cassone A. Acellular pertussis vaccines FIS PI18/ 00703]). AMC is supported by the “Agència de and pertussis resurgence: revise or replace? MBio. Gestió d’Ajuts Universitaris i de Recerca, Generalitat de Cat- 2014;5(3):e01339–e01314. alunya” [grant number 2018FI_B_00855]. [12] Octavia S, Maharjan RP, Sintchenko V, et al. Insight into evolution of Bordetella pertussis from comparative genomic analysis: evidence of vaccine-driven selection. ORCID Mol Biol Evol. 2011;28(1):707–715. Alba Mir-Cros http://orcid.org/0000-0003-4827-9824 [13] Octavia S, Sintchenko V, Gilbert GL, et al. Newly emer- Albert Moreno-Mingorance http://orcid.org/0000-0002- ging clones of Bordetella pertussis carrying prn2 and 3073-5046 ptxP3 alleles implicated in Australian pertussis epi- M. Teresa Martín-Gómez http://orcid.org/0000-0001- demic in 2008–2010. J Infect Dis. 2012;205(8):1220– 8119-8749 1224. Gema Codina http://orcid.org/0000-0002-5313-5086 [14] Mooi FR, van Loo IHM, van Gent M, et al. Bordetella Thais Cornejo-Sánchez http://orcid.org/0000-0002-1350- pertussis strains with increased toxin production 0413 associated with pertussis resurgence. Emerg Infect Carlos Rodrigo http://orcid.org/0000-0003-1140-2585 Dis. 2009;15(8):1206–1213. Magda Campins http://orcid.org/0000-0002-8841-6195 [15] Lam C, Octavia S, Bahrame Z, et al. Selection and Tomàs Pumarola http://orcid.org/0000-0002-5171-7461 emergence of pertussis toxin promoter ptxP3 allele in Anna Fàbrega http://orcid.org/0000-0002-3866-2291 the evolution of Bordetella pertussis. Infect Genet Juan José González-López http://orcid.org/0000-0003- Evol. 2012;12(2):492–495. 2419-5909 [16] Hallander H, Advani A, Riffelmann M, et al. Bordetella pertussis strains circulating in Europe in 1999 to 2004 as determined by pulsed-field gel electrophoresis. J References Clin Microbiol. 2007;45(10):3257–3262. [17] Advani A, Hallander HO, Dalby T, et al. Pulsed-field [1] Campins M, Moreno-Pérez D, Gilde Miguel A, et al. gel electrophoresis analysis of Bordetella pertussis iso- Whooping cough in Spain. Current epidemiology, pre- lates circulating in Europe from 1998 to 2009. J Clin vention and control strategies. Recommendations by Microbiol. 2013;51(2):422–428. the pertussis working group. Enferm Infect Microbiol [18] Barkoff A-M, Mertsola J, Pierard D, et al. Surveillance Clin. 2013;31(4):240–253. of circulating Bordetella pertussis strains in Europe [2] van Gent M, Heuvelman CJ, van der Heide HG, et al. during 1998–2015. J Clin Microbiol. 2018;56(5): Analysis of Bordetella pertussis clinical isolates circulat- e01998–17. ing in European countries during the period 1998– [19] Advani A, Donnelly D, Hallander H. Reference system 2012. Eur J Clin Microbiol Infect Dis. 2015;34 for characterization of Bordetella pertussis pulsed-field (4):821–830. gel electrophoresis profiles. J Clin Microbiol. 2004;42 [3] Cassiday PK, Skoff TH, Jawahir S, et al. Changes in (7):2890–2897. predominance of pulsed-field gel electrophoresis [20] Schouls LM, van der Heide HGJ, Vauterin L, et al. profiles of Bordetella pertussis isolates, United States, Multiple-locus variable-number tandem repeat analy- 2000–2012. Emerg Infect Dis. 2016;22(3):442–448. sis of Dutch Bordetella pertussis strains reveals rapid [4] Miyaji Y, Otsuka N, Toyoizumi-Ajisaka H, et al. genetic changes with clonal expansion during the late Genetic analysis of Bordetella pertussis isolates from 1990s. J Bacteriol. 2004;186(16):5496–5505. the 2008–2010 pertussis epidemic in Japan. PLoS [21] Mooi FR, van Oirschot H, Heuvelman K, et al. One. 2013;8(10):e77165. doi:10.1371/journal.pone. Polymorphism in the Bordetella pertussis virulence 0077165. factors P.69/pertactin and pertussis toxin in The [5] World Health Organization. Immunization, vaccines Netherlands: temporal trends and evidence for vac- and biologicals: pertussis. 2018 [cited 2019 Apr 15]. cine-driven evolution. Infect Immun. 1998;66:670– http://www.who.int/immunization/monitoring_survei 675. 1720 A. Mir-Cros et al. [22] Tsang RSW, Lau AKH, Sill ML, et al. Polymorphisms China in 2012–2013. Eur J Clin Microbiol Infect Dis. of the fimbria fim3 gene of Bordetella pertussis strains 2016;35(7):1211–1214. isolated in Canada. J Clin Microbiol. 2004;42 [27] Mosiej E, Krysztopa-Grzybowska K, Polak M, et al. (11):5364–5367. Multi-locus variable-number tandem repeat analysis [23] Tsang RSW, Sill ML, Advani A, et al. Use of mono- of Bordetella pertussis isolates circulating in Poland in clonal antibodies to serotype Bordetella pertussis iso- the period 1959–2013. J Med Microbiol. 2017;66 lates: comparison of results obtained by indirect (6):753–761. whole-cell enzyme-linked immunosorbent assay and [28] Galit SRL, Otsuka N, Furuse Y, et al. Molecular epide- bacterial microagglutination methods. J Clin miology of Bordetella pertussis in the Philippines in Microbiol. 2005;43(5):2449–2451. 2012–2014. Int J Infect Dis. 2015;35:24–26. [24] Heikkinen E, Xing DK, Ölander R-M, et al. Bordetella [29] Schmidtke AJ, Boney KO, Martin SW, et al. Population pertussis isolates in Finland: serotype and fimbrial diversity among Bordetella pertussis isolates, United expression. BMC Microbiol. 2008;8(1):162. States, 1935–2009. Emerg Infect Dis. 2012;18 [25] Bowden KE, Williams MM, Cassiday PK, et al. Molecular (8):1248–1255. epidemiology of the pertussis epidemic in Washington [30] van Gent M, Bart MJ, van der Heide HGJ, et al. Small State in 2012. J Clin Microbiol. 2014;52(10):3549–3557. mutations in Bordetella pertussis are associated with [26] Du Q, Wang X, Liu Y, et al. Direct molecular typing of selective sweeps. PLoS One. 2012;7(9):e46407. doi:10. Bordetella pertussis from nasopharyngeal specimens in 1371/journal.pone.0046407.