Despite a long history of vaccination with high coverage, resurgence of pertussis has been observed in the USA, Europe, Asia, South America, and Australia. Moreover, changes in the virulence factors of Bordetella pertussis, the causative agent of pertussis, have been reported in these regions. Recently, the emergence of strains with a novel allele (ptxP3) for the B. pertussis toxin promoter was described . The ptxP3 strains were associated with pertussis resurgence in The Netherlands and Australia, and produced more pertussis toxin (Ptx) than ptxP1 strains, which were dominant earlier in Europe [1–4].
In Finland, the whole cell pertussis vaccine was introduced in 1952. Acellular pertussis vaccines replaced the whole cell ones in the Finnish vaccination programme in 2005. The vaccination coverage for four doses has been >90% since the 1970s . However, the disease still remains endemic in this country. In 1999 and 2003–2004, there were two nationwide epidemics of pertussis, with incidence rates of 17.8 and 24.3–31.3 per 100 000, respectively (Fig. 1) (Infectious Diseases Register, National Institute for Health and Welfare, Helsinki, Finland). Since 1995, only laboratory-confirmed cases based on culture, PCR or ELISA serology have been reported in Finland. During the period 1995–2010, up to 9% of reported cases were culture-positive.
The purposes of this study were to develop an allele-specific real-time PCR for the detection of ptxP3 strains and, with the rapid method, to assess whether the ptxP3 strains have emerged in Finland, where vaccination against pertussis has been used for 60 years.
The isolates tested were from the B. pertussis strain collection of the Pertussis Reference Laboratory of the National Institute for Health and Welfare, Turku, Finland. The total number of isolates received was 695 from 1991 to 2010. From these, 495 (median number of tested isolates/year, 15; range, 4–142) were randomly selected and typed for ptxP3. In addition, 29 isolates recovered from 1953 to 1982 were included.
We developed an allele-specific real-time PCR for genotyping of ptxP3 strains. The template used in the real-time PCR was a bacterial suspension heated at 95°C for 30 min. The sequences of primers were 5′-TTCGGCGCAAAGTCGCGCGAGA-3′ (forward) and 5′-GCGTTTTGATGGTGCCTATT-3′ (reverse). The forward primer was designed to specifically bind ptxP3. To ensure no amplification from other ptxP alleles, a mismatch was added to the second last nucleotide (G) at the 3′-end of the forward primer. The reaction volume was 20 μL. The real-time PCR was run on a Light Cycler I machine (Roche Diagnostics, Mannheim, Germany). The LightCycler programme for amplification was as follows: denaturation for 600 s at 94°C, followed by 30 cycles of 5 s at 95°C, 5 s at 63°C, and 10 s at 72°C. The programme for melting curve analysis was as follows: 0 s at 65°C, 30 s at 55°C, and 0 s at 94°C. Cooling was performed at 40°C for 30 s. LightCycler FastStart DNA Master SYBR Green I (Roche Diagnostics, Penzberg, Germany) was used. Both positive (ptxP3) and negative (ptxP1) controls were included in each run. Specific amplification was only observed for ptxP3 with a specific melting temperature of c. 84°C (mean, 83.75°C, standard deviation, 0.54°C). Intra-assay variation was calculated when a positive control was tested in 10 different runs. The real-time PCR results were identical to those obtained by sequencing when 50 Finnish strains randomly selected from 1992 to 2004 were tested by both methods. Only ptxP1 and ptxP3 strains were identified among the 50 strains tested by sequencing.
When all B. pertussis isolates from 1953 to 1982 were tested, ptxP3 was not detected. The prevalence of the ptxP1 and ptxP3 strains isolated between 1991 and 2010 is shown in Fig. 2. In Finland, the ptxP3 strain was first detected in 1994, and has become predominant since 2003. An increase in the frequency of ptxP3 strains was found to coincide with nationwide pertussis epidemics in 1999 and 2003–2004 (Fig. 1). Since 2007, almost all circulating strains have been ptxP3.
Figure 2. Prevalence of the ptxP1 and ptxP3 strains isolated between 1991 and 2010 in Finland. All Finnish isolates from 1953 to 1990 tested were ptxP1, and are not shown. From 1991 to 2010, 495 strains were typed for ptxP3.
Download figure to PowerPoint
The allele-specific real-time PCR proved to be reliable and accurate, as shown by the correct identification of ptxP3 alleles. The advantage of this approach over sequencing is that the result can be obtained within an hour. Its disadvantage is that novel genotypes can be missed, as well as minor ptxP alleles other than ptxP1 or ptxP3. The minor alleles are not a major issue, as they were reported previously at low frequencies of 9% and 1%, in 1935–1990 and 1991–2004, respectively, when 1248 isolates collected from Europe, the Americas, Asia and Africa were analysed .
Our results show that ptxP3 strains have become the predominant type in Finland, and that they have almost totally replaced ptxP1 strains. Although ptxP3 strains were reported to be associated with pertussis resurgence in The Netherlands and Australia [1,4], in Finland ptxP3 strains cannot be linked directly to pertussis resurgence. The finding in this present study is in line with those observed in Sweden and Denmark, where the emergence of ptxP3 strains has not been found to be associated with the increased incidence of pertussis [2,6].
The allele-specific real-time PCR method has been proven to be specific and rapid for the identification of ptxP3 strains, and can be used to screen large numbers of B. pertussis isolates.