Strain variation among Bordetella pertussis isolates in finland, where the whole-cell pertussis vaccine has been used for 50 years.
ABSTRACT Pertussis is an infectious disease of the respiratory tract caused by Bordetella pertussis. Despite the introduction of mass vaccination against pertussis in Finland in 1952, pertussis has remained an endemic disease with regular epidemics. To monitor changes in the Finnish B. pertussis population, 101 isolates selected from 1991 to 2003 and 21 isolates selected from 1953 to 1982 were studied together with two Finnish vaccine strains. The analyses included serotyping of fimbriae (Fim), genotyping of the pertussis toxin S1 subunit (ptxA) and pertactin (prn), and pulsed-field gel electrophoresis (PFGE) after digestion of B. pertussis genomic DNA with XbaI restriction enzyme. Strains isolated before 1977 were found to harbor the same ptxA as the strains used in the Finnish whole-cell pertussis vaccine, and strains isolated before 1982 harbored the same prn as the strains used in the Finnish whole-cell pertussis vaccine. All recent isolates, however, represented genotypes distinct from those of the two vaccine strains. A marked shift of predominant serotype from Fim serotype 2 (Fim2) to Fim3 has been observed since the late 1990s. Temporal changes were seen in the genome of B. pertussis by PFGE analysis. Three PFGE profiles (BpSR1, BpSR11, and BpSR147) were distinguished by their prevalence between 1991 and 2003. The yearly emergence of the three profiles was distributed periodically. Our study stresses the importance of the continuous monitoring of emerging strains of B. pertussis and the need to obtain a better understanding of the relationship of the evolution of B. pertussis in vaccinated populations.
- [show abstract] [hide abstract]
ABSTRACT: Despite widespread vaccination during 30 years, the hypothesis of a resurgence of pertussis in France has been raised by outbreaks and sporadic case reports. No surveillance data were available after 1985. A survey was undertaken in 1993 and 1994 in a pediatric hospital network able to confirm cases; the network (22 hospitals) represents 19.6% of pediatric admissions in France. Case definition included clinical (> or = 21 days of paroxysmal cough), laboratory-confirmed (culture or serology by immunoblot) or epidemiologically confirmed pertussis (documented contact with a laboratory-confirmed case). The pattern of transmission was studied in the household. Vaccine status was obtained from health records. during a 15-month period 560 cases (316 index cases, 244 household contact cases) were reported; 49% of index cases and 20% of contact cases were confirmed by culture and/or serology. Sixty-five percent of index cases were younger than 1 year of age (the incidence in this age group could be estimated to be 95/100000) and 66% were hospitalized for a mean duration of 2 weeks. Infection was acquired from parents (34%) and siblings (46%). Seventy-three percent of index cases were unvaccinated. Although pertussis vaccination coverage is very high in France, the organism is still circulating, affecting, within the pediatric population, mostly non- or incompletely vaccinated infants. These results strongly support the importance of adhering to the immunization schedule and suggest introducing booster dose(s) to prolong vaccine immunity and reduce the exposure to Bordetella pertussis of infants too young to be immunized.The Pediatric Infectious Disease Journal 06/1998; 17(5):412-8. · 3.57 Impact Factor
Article: Pertussis in Poland.[show abstract] [hide abstract]
ABSTRACT: Since 1997, an unexpected 2-5-fold increase in the incidence of pertussis has been reported in Poland in comparison with the previous 10 years, although the introduction of the diphtheria-tetanus-pertussis (DTP) vaccination in 1960 reduced the incidence of pertussis approximately 100-fold in the 1980s. The aim of the study was to analyse all available data on pertussis in Poland to identify the risks associated with its re-emergence. Available data on notification, incidence, mortality, hospitalization, geographical distribution, incidence according to age, and diagnosis of pertussis were collected from national surveillance monographs and statistically evaluated. Analyses performed in the study found two periods of rising and falling trends: in the incidence before and after 1989, respectively. Moreover, after 1989, the age-specific incidence among children aged 0-4 years decreased, and among 5-9, 10-14, and 15-19 year olds increased in comparison to the previous decade. The incidence rate of pertussis among infants was similar in both decades analysed. Clustering of pertussis incidence increase in provinces along a line from North East to South West was observed. As vaccination coverage did not decrease and diagnostics have not been improved since the 1980s, it is possible that waning immunity and the appearance of Bordetella pertussis vaccine escape mutants are involved in the changing pertussis epidemiological parameters. Further monitoring studies, together with improving diagnostics, might allow more precise epidemiological data to be obtained. An additional booster dose of acellular pertussis vaccine at age 6 years has been included in the current vaccination schedule.International Journal of Epidemiology 05/2004; 33(2):358-65. · 6.98 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Whooping cough is presently one of the ten most common causes of death from infectious disease worldwide. Despite a high vaccine uptake, resurgences of this disease have been observed in several countries. Virulence factors of Bordetella pertussis include agglutinogens, fimbriae, P.69/pertactin, pertussis toxin, filamentous haemagglutinin, adenylate cyclase, tracheal cytotoxin, dermonecrotic toxin, lipopolysaccharide, tracheal colonisation factor, serum resistance factor, and type III secretion. Virulence factor expression is regulated by the bvgAS locus, a two-component signal transduction system. The pathophysiologic sequence consists of attachment (fimbriae, P.69/pertactin, tracheal colonisation factor, pertussis toxin, filamentous haemagglutinin), evasion of host defence (adenylate cyclase, pertussis toxin, serum resistance factor), local effects (tracheal cytotoxin), and systemic effects (pertussis toxin). Bordetella pertussis is transmitted by respiratory droplets and causes disease only in humans. Various diagnostic methods are available, including culture, serological methods, and the polymerase chain reaction. Serotyping of isolates to detect agglutinogens 2 and 3 is useful because serotype 1,2 may be associated with higher mortality, and antibodies to these antigens (agglutinins) may be protective in both animals and humans. Immunisation using whole-cell vaccine is effective but is reactogenic. Acellular vaccines containing one to five components are being used increasingly in various countries. Protective immunity to pertussis correlates with high levels of antibody to each of pertactin, fimbriae, and pertussis toxin; however, doubt remains as to the relationship between agglutinogen 3 and fimbria 3, making results of trials investigating these virulence factors difficult to interpret.European Journal of Clinical Microbiology 03/2000; 19(2):77-88. · 3.02 Impact Factor
JOURNAL OF CLINICAL MICROBIOLOGY, Aug. 2005, p. 3681–3687
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Vol. 43, No. 8
Strain Variation among Bordetella pertussis Isolates in Finland, Where
the Whole-Cell Pertussis Vaccine Has Been Used for 50 Years
Annika Elomaa,1,3* Abdolreza Advani,2Declan Donnelly,2Mia Antila,1Jussi Mertsola,3
Hans Hallander,2and Qiushui He1
Pertussis Reference Laboratory, National Public Health Institute, Turku, Finland1; Department of Immunology and Vaccine
Research, Swedish Institute for Infectious Disease Control, Solna, Sweden2; and Department of Pediatrics,
Turku University Hospital, Turku, Finland3
Received 7 December 2004/Returned for modification 19 January 2005/Accepted 12 April 2005
Pertussis is an infectious disease of the respiratory tract caused by Bordetella pertussis. Despite the intro-
duction of mass vaccination against pertussis in Finland in 1952, pertussis has remained an endemic disease
with regular epidemics. To monitor changes in the Finnish B. pertussis population, 101 isolates selected from
1991 to 2003 and 21 isolates selected from 1953 to 1982 were studied together with two Finnish vaccine strains.
The analyses included serotyping of fimbriae (Fim), genotyping of the pertussis toxin S1 subunit (ptxA) and
pertactin (prn), and pulsed-field gel electrophoresis (PFGE) after digestion of B. pertussis genomic DNA with
XbaI restriction enzyme. Strains isolated before 1977 were found to harbor the same ptxA as the strains used
in the Finnish whole-cell pertussis vaccine, and strains isolated before 1982 harbored the same prn as the
strains used in the Finnish whole-cell pertussis vaccine. All recent isolates, however, represented geno-
types distinct from those of the two vaccine strains. A marked shift of predominant serotype from Fim
serotype 2 (Fim2) to Fim3 has been observed since the late 1990s. Temporal changes were seen in the
genome of B. pertussis by PFGE analysis. Three PFGE profiles (BpSR1, BpSR11, and BpSR147) were
distinguished by their prevalence between 1991 and 2003. The yearly emergence of the three profiles was
distributed periodically. Our study stresses the importance of the continuous monitoring of emerging
strains of B. pertussis and the need to obtain a better understanding of the relationship of the evolution
of B. pertussis in vaccinated populations.
Bordetella pertussis, a small gram-negative bacterium, is the
causative agent of the respiratory infection called pertussis. B.
pertussis produces many virulence factors that are responsible
for the clinical features of the disease (9, 20, 26). The virulence
factors of B. pertussis are generally divided into two groups,
adhesins and toxins. The adhesins, such as filamentous hem-
agglutinin (FHA) (22, 23), fimbriae (Fim) (6), and pertactin
(Prn) (10), facilitate attachment of the bacteria to the host.
Toxins, such as pertussis toxin (Ptx) (17) and adenylate cyclase
toxin (ACT) (5), enable the bacteria to evade the immune
system of the host. Several virulence factors, such as FHA,
Fim, Ptx, and Prn, have been used in acellular pertussis vac-
cines (19). Methods have been developed for the typing of the
virulence factors (14); and changes in Fim, Ptx, and Prn have
been found in B. pertussis isolates around the world (4, 8, 15,
18, 24, 25).
Vaccinations against pertussis have been in use for more
than 50 years. Despite the high vaccination coverage rates
among children, the incidence of pertussis has increased in
countries such as Australia, France, The Netherlands, Poland,
and the United States (2, 4, 7, 18, 21, 25). In Finland, vacci-
nation against pertussis has been a part of the National Vac-
cination Program since 1952. The vaccine is produced at the
National Public Health Institute, Helsinki, Finland. Strain
18530 (Prn genotype prn1, Ptx subunit S1 genotype ptxA3, Fim
serotype 3 [Fim3]) has been used since 1962. In 1976, strain
1772 (prn1, ptxA2, Fim2.3) was added to the vaccine, and the
vaccine has not changed since then. The vaccination coverage
rate has been high in Finland. The latest survey carried out by
the National Public Health Institute showed that the rate of
coverage with the four doses of the diphtheria-tetanus-pertus-
sis vaccine was 95.6% among children born in 1999 (16). Still,
pertussis has been endemic in Finland, and there has been a
notable increase in the numbers of pertussis cases during the
last few years: 1,264 (incidence, 24.3/100,000 population) lab-
oratory-confirmed cases were reported in 2003, whereas 315
(6.1/100,000 population) laboratory-confirmed cases were re-
ported in 2001. The laboratory confirmation was done by PCR,
culture, or serology.
The connection between these changes and the reemergence
of pertussis has, however, remained unclear. The changes ob-
served in the virulence factors of B. pertussis may affect the
virulence and immunogenic characteristics of the bacteria.
Furthermore, changes in B. pertussis might be of importance,
especially in the adolescent population with waning immunity,
although this might not have an obvious reduction of the effi-
cacy of pertussis vaccines in younger children.
The aim of this study was to study changes in Finnish B.
pertussis isolates during the 50 years of the whole-cell vaccina-
tion, 1953 to 2003. We analyzed 122 Finnish B. pertussis iso-
lates and 2 Finnish vaccine strains (18530 and 1772). The
analysis included serotyping of Fim and genotyping of ptxA and
prn. Pulsed-field gel electrophoresis (PFGE) analysis of the
isolates was performed. We also wanted to compare circulating
strains with the vaccine strains.
* Corresponding author. Mailing address: National Public Health
Institute, Pertussis Reference Laboratory, Kiinamyllynkatu 13, 20520
Turku, Finland. Phone: 358-2-331 6632. Fax: 358-2-331 6699. E-mail:
MATERIALS AND METHODS
Bacterial strains. Bacterial isolates were selected from the B. pertussis strain
collection of the Pertussis Reference Laboratory of the National Public Health
Institute, Turku, Finland. The strain collection includes 359 Finnish B. pertussis
isolates recovered from 1991 to 2003. The isolates had been sent from the local
laboratories to the Pertussis Reference Laboratory, which is situated in south-
western Finland. The hospital district of southwest Finland has been more active
than other areas in sending the isolates, and thus, 75% of the strains in the
collection are from that district. Of the 359 isolates recovered from 1991 to 2003,
101 isolates were selected on the basis of the following criteria: (i) that they
represented 30% of the isolates collected each year or (ii) that they represented
at least two isolates from each year but (iii) not more than two isolates from each
town per year. The purposes of the selection criteria were to have an extensive
survey of the strains circulating in Finland and exclude the impact of local
outbreaks in the final conclusions. The isolates represent 34 communities from
13 hospital districts (Fig. 1). The isolates selected were from patients from 19
days to 69 years of age, with the median age being 6.5 years. The isolates were
divided by gender rather evenly, as 55% of the isolates were from female patients
and 45% from male patients. The gender of the patient was not reported for
three of the isolates.
In addition to the isolates obtained from 1991 to 2003, isolates from earlier
decades were also included in the study to obtain more information on the
genetic and antigenic changes over a longer time. The numbers of the isolates
from 1953 to 1965, 1977, and 1982 were 7, 10, and 4, respectively. The genders
and the ages of the patients were not reported for these isolates. The yearly
distribution of the isolates is shown in Table 1.
Culture and DNA extraction. The bacteria were cultured on Regan-Lowe
medium containing charcoal agar and 10% defibrinated sheep blood at 35°C for
2 or 3 days. The colonies on the plates were harvested in water for DNA
isolation. Extraction of DNA from the bacterial suspension was performed with
a High Pure PCR template preparation kit (Roche Diagnostics GmbH, Mann-
heim, Germany), according to the manufacturer’s instructions. DNA concentra-
tions were measured with a SmartSpec 3000 spectrophotometer (Bio-Rad, Rich-
mond, Calif.) and adjusted with water to 3 ng/?l. The DNA solutions were stored
Serotyping. The serotypes of six control strains (two Fim2 strains, three Fim3
strains, and one Fim2.3 strain) and two Finnish vaccine strains (Fim3 and
Fim2.3) obtained by both the microtiter plate and the slide agglutination meth-
ods were first compared. The control strains were kindly provided by D. Xing
from the National Institute for Biological Standards and Control (NIBSC),
United Kingdom. As the two methods gave identical typing results, the B.
pertussis isolates of this study were serotyped by slide agglutination. Antibody
reactions were performed with monoclonal antibodies against the major subunit
of Fim. The antibodies against Fim2 and Fim3 were also provided by D. Xing
from NIBSC. The antibodies (for Fim2, monoclonal antibody NIBSC 04/154; for
Fim3, monoclonal antibody NIBSC 04/156) were produced for research purposes
only. They had been produced in mice and were purified partially, and their
concentrations were not declared. Thus, appropriate dilutions for Fim2 and Fim3
antibodies were defined experimentally.
Before serotyping, the bacteria were cultured on Regan-Lowe medium con-
taining charcoal agar and defibrinated sheep blood at 35°C for 2 days. The
agglutination reaction was done on a glass slide with 40 ?l of diluted antibody
solution (dilution of 1:100 in phosphate-buffered saline). If the agglutination
reaction was obtained with the Fim2 antibody, the Fim3 antibody, or both antibod-
ies, the serotype was defined as Fim2, Fim3, or Fim2.3, respectively. If no reaction
was detected, the serotype was defined as untypeable. Autoagglutination was exam-
ined with phosphate-buffered saline in parallel with monoclonal antibodies.
Real-time PCR for genotyping. The isolates were analyzed for the pertussis
toxin and pertactin genes by LightCycler PCR and gel electrophoresis methods,
according to the protocols described by Ma ¨kinen et al. (12, 13).
FIG. 1. Distribution of the B. pertussis isolates among the 22 hospital districts of Finland. Between 1991 and 2003, the strain collection of the
Pertussis Reference Laboratory in the National Public Health Institute of Finland obtained 359 isolates from 13 hospital districts, and 101 isolates
were analyzed in this study. The Pertussis Reference Laboratory is located in the region of the hospital district of southwest Finland.
3682ELOMAA ET AL.J. CLIN. MICROBIOL.
Pertussis toxin S1 subunit. Different alleles of the ptxA gene were distin-
guished by two fluorescent resonance energy transfer (FRET) probe assays (12).
The assays were performed with a LightCycler DNA Master Hybridization
Probes kit (Roche Diagnostics GmbH, Mannheim, Germany). The first assay
differentiates the ptxA1 allele from the ptxA2, ptxA3, and ptxA4 alleles. In the
second probe assay, the ptxA4 type of DNA is differentiated from the ptxA2 and
ptxA3 types. The control strains for ptxA1, ptxA2, ptxA3, and ptxA4 were
PRCB333, 1772, 18530, and 18323, respectively. Negative controls without DNA
were included in each run.
Pertactin. Eight allelic variants (prn1 to prn8) of the prn gene were discrimi-
nated in three steps: allele-specific amplification with SYBR Green dye, hybrid-
ization assay with FRET probes, and gel electrophoresis (13). In the first step,
the prn6 to prn8 alleles are distinguished from the prn1 to prn5 alleles, as no
specific amplification of DNAs with the prn6 to prn8 alleles is observed. The
allele-specific amplification was done with a LightCycler-FastStart DNA Master
SYBR Green I kit (Roche Diagnostics GmbH). The second step, PCR with
FRET probes, distinguishes alleles prn1, prn2 to prn4, and prn5 from each other.
The hybridization probe assay was performed with the LightCycler DNA Master
Hybridization Probes kit (Roche Diagnostics GmbH). The prn2 to prn4 types
were further distinguished by agarose gel electrophoresis of the PCR products
from the hybridization probe assay. The alleles were differentiated by the sizes of
the PCR products. The control strains for different prn allelic variants were 1772
(prn1), PRCB333 (prn2), RUS4 (prn3), PRCB9 (prn4), B935 (prn5), and 18323
PFGE. Isolates were analyzed by the PFGE protocol described by Advani et al.
(1). The PFGE analyses were performed in The Swedish Institute for Infectious
Disease Control, Solna, Sweden. Genomic DNA was digested with the restriction
enzyme XbaI (Amersham Biosciences, Little Chalfont, United Kingdom). The
electrophoresis analyses were performed on a DRIII contour-clamped homoge-
neous electric field apparatus (Bio-Rad). The band patterns obtained were an-
alyzed with BioNumerics, version 3, software (Applied Maths, Sint-Martens-
Latem, Belgium). Different PFGE profiles were defined by one or more band
differences in the DNA band patterns. The clustering method used was the
unweighted pair group with arithmetic clustering (UPGMA) dendrogram type
with the Dice similarity coefficient, 1% optimization, and 1% tolerance. The
nomenclature was based on the defined profiles already observed in Sweden
(BpSR). Profiles assigned BpFINR have been found only among the Finnish
Serotyping. All three serotypes, Fim2, Fim2.3, and Fim3,
were observed among the B. pertussis isolates (Table 1). How-
ever, the proportion of each type has changed over time. In the
1980s and 1990s, the predominant serotype of the isolates was
Fim2. Since 1999, the Fim3 type started to replace Fim2 in the
Finnish isolates. All the isolates from the year 1998 produced
Fim2, but in 2003 the proportion of Fim2 serotype isolates
decreased to 16% (n ? 3). The Fim2.3 type was also seen more
often than it had been in the earlier years.
Genotyping. The ptxA genotype of Finnish pertussis vaccine
strain 18530 was ptxA3. The isolates from 1953 to 1965 and
strain 1772, added to the vaccine in 1976, represented the
ptxA2 genotype. All other isolates represented the ptxA1 ge-
TABLE 1. Characteristics of the Finnish vaccine strains and B. pertussis isolates analyzed in this study
Strain and yr of
No. of isolates
% Isolates with the indicated:
ptxA alleleprn allele (%)Serotypea
Most frequent (%)
1977 10100 901030 20503BpSR23 (80)
19824 1001001001BpSR18 (100)
24 10029674 9249 BpSR1 (38)
231009 8344 918BpSR147 (52)
251004 96 6012 2811BpSR147 (24)
29100973 243 72 11 BpSR11 (38)
aThe serotypes of one isolate recovered in 1994 and two isolates recovered in 1996 were untypeable.
bYears when the strains were introduced in the Finnish whole-cell vaccine.
cThe years 1991 to 2003 are grouped into periods of 3 to 4 years to correlate the incidence peaks in Finland.
VOL. 43, 2005 STRAIN VARIATION IN FINNISH B. PERTUSSIS ISOLATES3683
notype (Table 1). Thus, the change in the prevalent ptxA type
of the circulating B. pertussis strains had already occurred in
In the 1950s, the prevalent prn allele was prn1, which is also
represented in the Finnish vaccine strains (Table 1). The
change from prn1 to prn2 was already seen in the isolates from
1982, as all of them represented prn2. In 1991 and 1992, one-
third of the isolates still represented prn1 allele (N ? 4), but
prn1 was last seen in Finnish isolates in 1999 (n ? 1). All the
recent isolates demonstrated the prn2 allele. The prn3 and prn4
alleles occurred only twice among the isolates tested.
PFGE. A total of 34 different profiles were found among the
strains studied (Table 1 and Fig. 2). The profile of vaccine
strain 18530 (BpFINR13) was not found among the isolates.
Two profiles, BpFINR1 and BpFINR14, were found among
the isolates recovered from 1953 to 1965. Those profiles have
not appeared since then. Vaccine strain 1772, added to the
Finnish pertussis vaccine in 1976, and 8 of the 10 isolates from
1977 represented the profile BpSR23, which was also found in
1 isolate recovered in 1999. The profiles of other isolates from
1977 (BpFINR9 and BpSR46) have not occurred since then.
The isolates from 1982 represented profile BpSR18, which still
appeared occasionally, in five isolates, between 1992 and 2003.
The number of profiles found from 1991 to 2003 was 29.
Most of the profiles appeared in one to seven isolates, but
three profiles were clearly distinguished by their higher prev-
alence. Profiles BpSR1, BpSR11, and BpSR147 were repre-
sented in 14, 15, and 18 isolates from 1991 to 2003 (Table 2).
The yearly emergence of those profiles was not, however,
evenly distributed but periodic. All the BpSR1 profiles ap-
peared from 1991 to 1998, and all BpSR11 profiles appeared
from 1999 to 2003. The appearance of BpSR147 overlapped
with those of BpSR1 and BpSR11, as the isolates with the
BpSR147 profile were found from 1996 to 2000. There were
also 13 BpFINR profiles among the total of 16 isolates recov-
ered from 1991 to 2003. These profiles have not been isolated
in any of the countries within the European Union collabora-
tion (Eupertstrain; France, Germany, The Netherlands, and
The most common PFGE profiles, BpSR1, BpSR11, and
BpSR147, were clearly correlated with serotypes Fim2, Fim3,
and Fim2, respectively (Table 2). Genotype ptxA2 was found in
old isolates, which harbored PFGE profiles BpFINR1 and
BpFINR14. Neither those profiles nor ptxA2 has appeared
since then. Isolates with the prn1 genotype (n ? 26) were
shown to harbor nine PFGE profiles. Seven of the nine profiles
are correlated with prn1. Prn2 was observed in 92 isolates
harboring 24 PFGE profiles, including the most common pro-
files, BpSR1, BpSR11, and BpSR147. Prn3 appeared in two
isolates harboring unique PFGE profiles, BpFINR8 and
BpFINR23, respectively, which were seen only in these iso-
lates. Prn4 was also detected in two isolates, but the PFGE
profiles of those isolates were also found in isolates with the
During the last decade, the number of pertussis cases has
increased in countries with high vaccination coverage rates (4,
7, 18, 21, 25), including Finland. The reason for this reemer-
gence has been of concern. One proposed cause is the anti-
genic variation of B. pertussis, especially in regard to the viru-
lence factors such as Fim, Prn, and Ptx. Furthermore, these
antigens have been included in many acellular vaccines. We
have carried out research among 122 Finnish B. pertussis iso-
lates collected since 1953. The virulence factors analyzed here
were Fim, prn, and ptxA. The isolates were analyzed by PFGE
The results of this study concur with previously published
suggestions on the continuous evolution of B. pertussis and its
virulence factors (4, 24, 25). In Finland, the first known change
in virulence factors had already occurred in the S1 subunit of
pertussis toxin in the 1970s, leading to ptxA1 being the pre-
dominant allele. This transition has been seen in many coun-
tries (4, 7, 11, 18, 24, 25) and is suggested to be vaccine driven
(15), as the strains used in the whole-cell pertussis vaccines do
not contain the ptxA1 variant. The second noticeable change
was seen from 1980 to the 1990s, as the prevalent pertactin
allele shifted from prn1 to prn2. A similar shift has been re-
ported in the United States (4). An increase in the frequency
of the prn3 genotype was not seen in Finland in those days,
which is in contrast to the situation in France and The Neth-
erlands (15, 25), as only two Finnish isolates represented prn3.
In Australia, prn1 changed to prn3, and prn2 did not appear
until the mid-1990s (18). Our results correspond to the shift
observed in the United States (4), indicating that the changes
in ptxA and prn happened through a transition period. The
prevalent genotype, ptxA2 or ptxA3 and prn1, first changed to
ptxA1 and prn1, followed by a second modification to ptxA1 and
prn2. This was the genotype of all Finnish isolates included in
our study from 2003. Ten isolates recovered from 1991 to 1999
were found to harbor prn1. In Italy, the frequency of prn1 was
shown to be higher in an unvaccinated population than in the
vaccinated population (11), which may partially explain the low
number of prn1 isolates in Finland and other countries with
high vaccination coverage rates. We also confirm the findings
of van Loo et al. (24), which indicate that serotypes Fim2,
Fim2.3, and Fim3 have been evident in the B. pertussis popu-
lation throughout the decades and which showed differences
only in the frequencies of the different serotypes. Recently, the
predominant serotype has changed from Fim2 to Fim3, which
has also been the case in Australia (18) and The Netherlands
The PFGE profiles of the Finnish isolates show that the
prevalent PFGE profiles change temporally, as suggested by
Weber et al. (25). The profiles represented in the Finnish
isolates from past decades only occasionally appear in the
recent isolates. All of the BpFINR profiles that were not seen
in the recent Swedish study, based on a large number of iso-
lates (1), and most of the BpSR profiles appear in only a few
isolates within 1 to 3 years. The isolates with the BpFINR
profiles represent unique subtypes that for some reason do not
seem to spread as well as some other strains. Three of the
PFGE profiles among the Finnish isolates were distinguished
by their frequencies. The BpSR1, BpSR11, and BpSR147 pro-
files were found in 14 to 18 isolates within a few years. The
BpSR147 profile was represented in 42% of the isolates during
the years of its occurrence. All other profiles appeared in no
more than four isolates during the period. Thus, few PFGE
profiles are represented in a considerable proportion of circu-
3684ELOMAA ET AL. J. CLIN. MICROBIOL.
lating isolates, as also described earlier (3). Even though the
research material used in our study was limited, the isolates
representing the BpSR1, BpSR11, and BpSR147 profiles were
from three to four hospital districts in Finland and, thus, do not
represent local outbreaks. As the changes in the PFGE profiles
and virulence factors are merged, it is seen that the recent
increase in serotype Fim3 observed since 1999 interfaces with
the appearance of new PFGE profiles. Eight PFGE profiles
were represented among 30 isolates with the Fim3 serotype,
which were detected in 1999 or later, including the BpSR11
FIG. 2. Classification of the 34 PFGE profiles represented in the Finnish B. pertussis isolates and vaccine strains (indicated by “V”) and the
reference strains of PFGE (indicated by asterisks). Classification was performed by using the UPGMA dendrogram type with the Dice similarity
coefficient, 1% optimization, and 1% tolerance.
VOL. 43, 2005 STRAIN VARIATION IN FINNISH B. PERTUSSIS ISOLATES 3685
profile, which was the second most common PFGE profile of
all time in Finland. Among the 30 isolates, 28 were Fim3 and
2 were Fim2.3. The other common PFGE profiles, BpSR1 and
BpSR147, detected during earlier epidemics correlate with se-
rotype Fim2. Weber et al. (25) questioned the connection
between PFGE profiles and antigenic changes. Our results,
however, suggest that the serotype change may be correlated
with the appearance of new PFGE profiles.
Typing of the virulence factors of B. pertussis isolates helps
to validate the idea of the continuous evolution of the bacteria.
However, when additional information is craved, the PFGE
reference system published by Advani et al. (1) is very profit-
able and provides more precise evidence of the changes at the
molecular level. Our results show that ptxA and prn have
changed before the recent reemergence of pertussis. However,
the role of the Fim3 type of strains with new PFGE profiles in
the increased incidence of pertussis should be studied further.
In this study, PFGE analysis of B. pertussis isolates repre-
senting the six decades of experience with whole-cell vaccina-
tion showed the emergence of new PFGE profiles and the
disappearance of the former ones, in parallel with changes in
the virulence factors. In Finland, the whole-cell pertussis vac-
cine has been replaced with an acellular vaccine in 2005. The
effects of acellular vaccines on the circulating B. pertussis
strains should be closely monitored. This study lays a good
background for further monitoring of the circulating B. pertus-
sis isolates in Finland. The exceptionally stable vaccination
history with a high vaccination coverage rate makes Finland a
good location for monitoring of the changes in the B. pertussis
population after the introduction of a new vaccination program
with acellular pertussis vaccines in Finland in 2005.
We thank Maritta Mo ¨ller for technical assistance and Dorothy Xing
from NIBSC for providing the antibodies and control strains for sero-
typing. We also thank the clinical microbiology laboratories in Finland
for sending the B. pertussis isolates to our laboratory over the years.
This work was financially supported by the Academy of Finland, the
Special Governmental Fund for University Hospitals (EVO), and the
European Commission Quality of Life Program (QLK2-CT-2001-
1. Advani, A., D. Donnelly, and H. Hallander. 2004. Reference system for
characterization of Bordetella pertussis pulsed-field gel electrophoresis pro-
files. J. Clin. Microbiol. 42:2890–2897.
2. Baron, S., E. Njamkepo, E. Grimprel, P. Begue, J. C. Desenclos, J. Drucker,
and N. Guiso. 1998. Epidemiology of pertussis in French hospitals in 1993
and 1994: thirty years after a routine use of vaccination. Pediatr. Infect. Dis.
3. Bisgard, K. M., C. D. Christie, S. F. Reising, G. N. Sanden, P. K. Cassiday,
C. Gomersall, W. A. Wattigney, N. E. Roberts, and P. M. Strebel. 2001.
Molecular epidemiology of Bordetella pertussis by pulsed-field gel electro-
phoresis profile: Cincinnati, 1989–1996. J. Infect. Dis. 183:1360–1367.
4. Cassiday, P., G. Sanden, K. Heuvelman, F. Mooi, K. M. Bisgard, and T.
Popovic. 2000. Polymorphism in Bordetella pertussis pertactin and pertussis
toxin virulence factors in the United States, 1935–1999. J. Infect. Dis. 182:
5. Friedman, R. L., R. L. Fiederlein, L. Glasser, and J. N. Galgiani. 1987.
Bordetella pertussis adenylate cyclase: effects of affinity-purified adenylate
cyclase on human polymorphonuclear leukocyte functions. Infect. Immun.
6. Geuijen, C. A. W., R. J. L. Willems, and F. R. Mooi. 1996. The major fimbrial
subunit of Bordetella pertussis binds to sulfated sugars. Infect. Immun. 64:
7. Gzyl, A., E. Augustynowicz, D. Rabczenko, G. Gniadek, and J. Slusarczyk.
2004. Pertussis in Poland. Int. J. Epidemiol. 33:358–365.
8. Hardwick, T. H., P. Cassiday, R. S. Weyant, K. M. Bisgard, and G. N.
Sanden. 2002. Changes in predominance and diversity of genomic subtypes
of Bordetella pertussis isolated in the United States, 1935 to 1999. Emerg.
Infect. Dis. 8:44–49.
9. Kerr, J. R., and R. C. Matthews. 2000. Bordetella pertussis infection: patho-
genesis, diagnosis, management, and the role of protective immunity. Eur.
J. Clin. Microbiol. Infect. Dis. 19:77–88.
10. Leininger, E., M. Roberts, J. G. Kenimer, I. G. Charles, N. Fairweather, P.
Novotny, and M. J. Brennan. 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.
11. Mastrantonio, P., P. Spigaglia, H. van Oirschot, H. G. van der Heide, K.
Heuvelman, P. Stefanelli, and F. R. Mooi. 1999. Antigenic variants in Bor-
detella pertussis strains isolated from vaccinated and unvaccinated children.
Microbiology 145(Pt 8):2069–2075.
12. Ma ¨kinen, J., J. Mertsola, M. K. Viljanen, H. Arvilommi, and Q. He. 2002.
Rapid typing of Bordetella pertussis pertussis toxin gene variants by LightCy-
cler real-time PCR and fluorescence resonance energy transfer hybridization
probe melting curve analysis. J. Clin. Microbiol. 40:2213–2216.
13. Ma ¨kinen, J., M. K. Viljanen, J. Mertsola, H. Arvilommi, and Q. He. 2001.
Rapid identification of Bordetella pertussis pertactin gene variants using
LightCycler real-time polymerase chain reaction combined with melting
curve analysis and gel electrophoresis. Emerg. Infect. Dis. 7:952–958.
14. Mooi, F. R., H. Hallander, C. H. Wirsing von Ko ¨nig, B. Hoet, and N. Guiso.
2002. Epidemiological typing of Bordetella pertussis isolates: recommenda-
tions for a standard methodology. Eur. J. Microbiol. Infect. Dis. 19:174–181.
15. Mooi, F. R., H. van Oirschot, K. Heuvelman, H. G. van der Heide, W.
Gaastra, and R. J. Willems. 1998. Polymorphism in the Bordetella pertussis
virulence factors P.69/pertactin and pertussis toxin in The Netherlands: tem-
poral trends and evidence for vaccine-driven evolution. Infect. Immun. 66:
16. National Public Health Institute, Finland. 14 May 2004, posting date. Vac-
cination coverage. [Online.] http://www.ktl.fi/portal/suomi/osiot/terveyden
17. Pittman, M. 1984. The concept of pertussis as a toxin-mediated disease.
Pediatr. Infect. Dis. 3:467–486.
18. Poynten, M., P. B. McIntyre, F. R. Mooi, K. J. Heuvelman, and G. L. Gilbert.
2004. Temporal trends in circulating Bordetella pertussis strains in Australia.
Epidemiol. Infect. 132:185–193.
19. Sato, Y., and H. Sato. 1999. Development of acellular pertussis vaccines.
20. Smith, A. M., C. A. Guzman, and M. J. Walker. 2001. The virulence factors
of Bordetella pertussis: a matter of control. FEMS Microbiol. Rev. 25:309–
21. Tanaka, M., C. R. Vitek, F. B. Pascual, K. M. Bisgard, J. E. Tate, and T. V.
Murphy. 2003. Trends in pertussis among infants in the United States,
1980–1999. JAMA 290:2968–2975.
22. Urisu, A., J. L. Cowell, and C. R. Manclark. 1986. Filamenteous hemagglu-
tinin has a major role in mediating adherence of Bordetella pertussis to
human WiDr cells. Infect. Immun. 52:695–701.
23. Van den Berg, B. M., H. Beekhuizen, R. J. L. Willems, F. R. Mooi, and R. van
TABLE 2. Characteristics of the three most frequent PFGE profiles among the isolates analyzed in this study
Yr of isolation
% Isolates with the indicated:
ptxA allele prn alleleSerotypea
1231234 Fim2 Fim2.3Fim3
aThe serotypes of one isolate with the BpSR1 profile and two isolates with the BpSR147 profile were untypeable.
3686ELOMAA ET AL.J. CLIN. MICROBIOL.
Furth. 1999. Role of Bordetella pertussis virulence factors in adherence to
epithelial cell lines derived from the human respiratory tract. Infect. Immun.
24. Van Loo, I. H., and F. R. Mooi. 2002. Changes in the Dutch Bordetella
pertussis population in the first 20 years after the introduction of whole-cell
vaccines. Microbiology 148:2011–2018.
25. Weber, C., C. Boursaux-Eude, G. Coralie, V. Caro, and N. Guiso. 2001.
Polymorphism of Bordetella pertussis isolates circulating for the last 10 years
in France, where a single effective whole-cell vaccine has been used for more
than 30 years. J. Clin. Microbiol. 39:4396–4403.
26. Wirsing von Ko ¨nig, C. H., S. Halperin, M. Riffelmann, and N. Guiso. 2002.
Pertussis of adults and infants. Lancet Infect. Dis. 2:744–750.
VOL. 43, 2005 STRAIN VARIATION IN FINNISH B. PERTUSSIS ISOLATES3687