Significant Decrease in Pertactin-Deficient Bordetella pertussis Isolates, Japan

Prevalence of pertactin-lacking Bordetella pertussis isolates has been observed worldwide. In Japan, however, we found that the frequency of pertactin-deficient isolates in 2014–2016 (8%) was significantly lower than the frequency in 2005–2007 (41%), 2008–2010 (35%), and 2011–2013 (25%). This reduction was closely associated with changes in genotypes.

B ordetella pertussis, a highly communicable, gram-negative coccobacillus, is the etiologic agent of pertussis (whooping cough), an acute respiratory infection that leads to severe illness in children. Vaccination is the most effective method for preventing and controlling pertussis. In Japan, acellular pertussis vaccines (ACVs) were introduced in 1981. Pertussis toxin and filamentous hemagglutinin derived from B. pertussis are the major antigens in ACVs in Japan, and certain ACVs also contain pertactin and fimbriae (1). Pertactin is believed to play a role in adherence to human epithelial cells (2); however, B. pertussis isolates that lack pertactin production have been identified in several countries where ACVs have been introduced (3-7). In Japan, pertactin-deficient isolates have increased significantly since the early 2000s, resulting in a high prevalence of these isolates (5,8). Recent studies have demonstrated that pertactin-deficient strains could colonize the respiratory tract more effectively than pertactin-producing strains in ACV-vaccinated mice (9,10). Supporting these results, an epidemiologic study suggested that ACV-vaccinated persons are more susceptible to pertactin-deficient strains than to pertactin-producing strains (11). These reports imply that pertactin-deficient strains have increased fitness in humans who have been vaccinated with ACVs and that their expansion may reduce the effectiveness of ACVs. We assessed trends in the frequency of pertactin-deficient isolates in Japan and further investigated their genotypes using multilocus variable-number tandem-repeat analysis (MLVA).

The Study
We studied 232 B. pertussis clinical isolates collected from January 2005 through June 2016 in Japan. All isolates were derived from epidemiologically unrelated cases of pertussis. Pertactin production and MLVA types (MTs) of 111 isolates collected during 2005-2012 were previously determined by immunoblotting and MLVA typing, respectively (5,8). For our study, we extended these analyses to additional isolates collected during 2008-2016 (n = 121).
We determined the temporal trend in the frequency of pertactin-deficient isolates by 3-year periods ( Figure,  Among the 232 B. pertussis isolates, 25 MTs were identified; MT27 and MT186 isolates were predominant, and other MT isolates were found at low frequencies (online Technical Appendix Table 1 (Table). Notably, 45 (83.3%) of 54 MT186 isolates were pertactindeficient, whereas only 2 (1.4%) of 139 MT27 isolates were pertactin-deficient. This finding indicates that pertactin-deficient isolates predominate among the MT186 strain but are rare among the MT27 strain.
The most likely explanation for the prevalence of pertactin-deficient strains is vaccine-driven strain evolution, because pertactin is a component of ACVs. No pertactindeficient isolates have been detected in Denmark, where an ACV that does not contain pertactin is used, and pertactindeficient strains exhibit a selective advantage against ACVinduced immunity (7,(9)(10)(11). In Japan, 5 brands of the diphtheria-tetanus-acellular pertussis vaccine (DTaP) had been used to control pertussis for many years. These DTaPs had different formulations of components; only 3 contained the pertactin antigen (1). When the DTaP vaccine was replaced, however, 2 brands of combined DTaP-inactivated poliovirus (DTaP-IPV) vaccine that did not contain pertactin were introduced in November 2012. Therefore, most Japanese children <4 years of age do not have immunity to pertactin; consequently, the selective pressure for pertactin-deficient strains in the host environment has recently been reduced. This effect may be responsible for the recent decline in pertactin-deficient isolates and further supports the hypothesis that pertactin-deficient strains are selected on the basis of host immunity to pertactin. Notably, a new brand of DTaP-IPV vaccine containing pertactin was also introduced in December 2015 in Japan. If the hypothesis of vaccine-driven evolution is correct, pertactin-deficient isolates should increase again in Japan in the near future. Thus, continued surveillance of pertactin-deficient isolates is of particular value.
We demonstrated that genotypic replacement from MT186 to MT27 has taken place among recent B. pertussis isolates in Japan: MT27 is a triple-locus variant of MT186. MT186 strains carry the pertussis-toxin promoter allele ptxP1, whereas MT27 strains carry the allele ptxP3 (8). B. pertussis strains carrying ptxP3 (i.e., MT27) produce more of several virulence factors than do ptxP1 (i.e., MT186) strains (12,13). The population of MT27 strains carrying ptxP3 has increased worldwide (14,15), although a low frequency of ptxP1 isolates was observed in Japan (8), suggesting that MT27 strains are associated with the recent pertussis resurgence. It is possible, therefore, that the genotypic replacement in the B. pertussis population may have resulted from the expansion of the more virulent ptxP3 (i.e., MT27) strains. In addition, given that pertactin-deficient MT27 isolates are rare, this genotypic replacement  may have contributed to the recent decrease in pertactindeficient isolates in Japan. In Japan, most pertactin-deficient isolates carry a deletion of the prn signal sequence (ΔSS), which has been found primarily in MT186 isolates carrying the prn1 allele (online Technical Appendix Table 1). In other countries, a common prn mutation includes an insertion of IS481 into the prn2 allele (4,6,7). In this study, we identified 2 pertactin-deficient MT27 isolates carrying the prn2 allele, due to the IS481 insertion (245-246::IS481) and the transcriptional down-regulation of the prn gene. These pertactin-deficient isolates were previously identified in Europe and the United States (6,7). One possible explanation for the appearance of pertactin-deficient MT27 isolates is that they were imported from other countries.