Analysis of Bordetella pertussis isolates collected in Japan before and after introduction of acellular pertussis vaccines

Vaccine 19 (2001) 3248 – 3252 www.elsevier.com/locate/vaccine

Analysis of Bordetella pertussis isolates collected in Japan before and after introduction of acellular pertussis vaccines Nicole Guiso a,*, Caroline Boursaux-Eude a, Christian Weber a, Sally Z. Hausman c, Hiroko Sato 1, Masaaki Iwaki b, Kazunari Kamachi b, Toshifumi Konda b, Drusilla L. Burns c a

Unite´ des Bordetella, Centre National de Re´fe´rence des Bordetelles, Institut Pasteur, Paris, France Department of Bacterial and Blood Products, National Institute of Infectious Diseases, Tokyo, Japan c Laboratory of Respiratory and Special Pathogens Center for Biologics E6aluation and Research, Food and Drug Administration, Bethesda, MD, USA b

Received 13 July 2000; received in revised form 13 December 2000; accepted 18 December 2000

Abstract Because of recent concern that whole-cell pertussis vaccination can drive antigenic divergence of circulating isolates of Bordetella pertussis, we compared 12 clinical isolates of B. pertussis collected in Japan, the first country to introduce acellular pertussis vaccines, with the vaccine strain. We used pulsed-field gel electrophoresis, sequencing of ptx and prn genes and expression of fimbriae. Most of the isolates collected before or after introduction of acellular vaccine possess similar restriction patterns. They contain ptx genes and prn alleles similar to the vaccine strain and to European isolates collected before the introduction of vaccination. Two recently collected isolates exhibiting a different pulsed-field gel electrophoresis pattern possess ptxS1 and prn alleles similar to the alleles harbored by European isolates circulating currently. Our preliminary results suggest that, if acellular pertussis vaccine-induced antigenic divergence exists, it is likely to be a slow or rare process. © 2001 Elsevier Science Ltd. All rights reserved. Keywords: Bordetella pertussis; Acellular vaccines; Japan

Bordetella pertussis is the causative agent of whooping cough. Until recently, protection against this disease was conferred by whole cell pertussis vaccines composed of suspensions of inactivated B. pertussis which have been used world-wide, for almost 50 years ago. Because of undesirable side effects associated with the whole cell vaccine, a considerable amount of effort was expended to develop a new type of vaccine, known as the acellular pertussis vaccine, composed of a few purified bacterial antigens. The acellular vaccines are better tolerated than whole cell vaccines and have been shown to be efficacious in preventing pertussis in children [5]. Acellular vaccines were first introduced in * Corresponding author. Tel.: + 33-1-45688334; fax: +33-140613533. E-mail address: [email protected] (N. Guiso). 1 Present address: 1-33-4 Kamiuma, Setagaya-ku, Tokyo 154-0011 Japan.

Japan as early as 1981 [16,18], but were only recently introduced into other regions including Europe, North America and eastern Asia. In the past few years, concern has been raised about the possibility that antigenic variability may occur in isolates of B. pertussis that would affect the efficacy of pertussis vaccines in the Netherlands, Finland and Italy [7,9,10]. Divergence was noted between recent clinical isolates and strains used for production of whole cell vaccines, raising the possibility that vaccine use has selected for certain strains that differ antigenically from those in the vaccine. In particular, differences have been noted between isolates in the genes encoding the S1 subunit of pertussis toxin (PT), and pertactin (PRN), both important virulence factors of the bacterium. Others have postulated that this divergence may have contributed to disease burden in the Netherlands [10]. Of course, other factors might contribute to the reported increase in pertussis incidence in the Netherlands in-

0264-410X/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S0264-410X(01)00013-5

N. Guiso et al. / Vaccine 19 (2001) 3248–3252

cluding improved surveillance, changes in case definition, manufacturing changes during vaccine production, or waning vaccine-induced immunity. Introduction of acellular pertussis vaccines heightens concerns about the effects of antigenic divergence since acellular vaccines are composed of only a few critical antigens. All acellular pertussis vaccines contain an inactivated form of PT. In fact, vaccines composed solely of inactivated PT have been shown to confer protection against severe pertussis, underscoring the importance of this antigen [1]. Other antigens that may be included in these vaccines are PRN, filamentous hemagglutinin (FHA), and fimbriae types 2 and 3 [4]. In Japan, manufacturers have utilized the strain Tohama I for production of acellular pertussis vaccines [16,17]. Japanese acellular pertussis vaccines by six manufacturers are basically categorized in two types, one containing inactivated PT and FHA, and the other containing, in addition to these two antigens, PRN and type 2 fimbriae [13]. Acellular pertussis vaccines have effectively controlled pertussis disease in Japan since their introduction in 1981 [6]. If the introduction of acellular pertussis vaccines were to accelerate antigenic divergence, then one should first begin to see such drift in Japan, since acellular pertussis vaccines were used in that country almost 15 years before they were introduced into wide-spread use in other countries. For this reason, we compared the 12 available Japanese clinical isolates, five collected before introduction of acellular pertussis vaccines and seven collected since 1992, after acellular vaccines had been in

3249

use for more than 10 years. Characteristics of all isolates used in this study are provided in Table 1. We first examined the DNA from the 12 Japanese clinical isolates by restriction enzyme analysis (using Xba I, and Spe I) and pulsed-field gel electrophoresis (PFGE) as previously described [8]. The patterns obtained were compared with three reference strains representative of the major types of isolates circulating either before the introduction of whole cell vaccine in Europe, Bp 134, or after the introduction of vaccination, B902 and Fr287 [8]. As shown in Fig. 1A and B, the Japanese clinical isolates appear to be closely related when examined by PFGE. DNA from most of the Japanese isolates (with the exception of 7 and 13) exhibit profiles after restriction that are very similar, but not identical, to that of the reference strain Bp 134. However, as it is indicated in Fig. 1A and B, a specific fragment is observed after restriction of the Japanese isolates DNA by the restriction enzyme Spe I that has never been observed after restriction of DNA from European isolates [8]. A compilation of the data obtained with the PFGE technique was performed using the Neighbor Joining method of clustering [14]. As can be seen in Fig. 1C, 10 of the Japanese isolates were grouped with strain Bp 134. This confirms previous results obtained using an RFLP technique [20]. Five of these 10 isolates were from the pre-acellular pertussis vaccine era and five were collected at least 10 years after the introduction of acellular vaccines. Two isolates, collected recently were grouped with the reference strain B 902.

Table 1 Characteristics of clinical isolates and reference strains Strains or isolates Japanese clinical isolates CA-1 TB-49 SA-4 SK-5 TY-7 9 93004-106 93005-107 95008-109 7 13 18 Reference strains Tohama I Bp 134 B-902 Fr. 287 a

Year of collection

1975 1979 1979 1979 1979 1992 1993 1993 1995 1996 1996 1996 B1953 B1950 1996 1996

Origin

Fimbriae type

prn allele type

Chiba pref.a Tokyoa Shizuoka pref.a Kanagawa pref.a Kanagawa pref.a Kanagawa pref.c Fukui pref.b Fukui pref.b Fukui pref.b Kanagawa pref.c Kanagawa pref.c Kanagawa pref.c

3 3 3 3 3 3 3 3 3 2, 3 3 3

B NDd B NDd B NDd NDd B B A A B

1 1 1 1 1 1 1 1 1 2 2 1

Japan USA Sweden France

2 2, 3 3 3

B B A A

1 1 2 3

Gift from Dr M. Watanabe (Kitasato Institute, Tokyo, Japan). Gift from Drs T. Maegawa and S. Nakamura (Kanazawa University, Ishikawa, Japan). c Gift from Dr T. Kato (St. Marianna University, Kanagawa, Japan). d ND, not determined. b

PtxS1 type

3250

N. Guiso et al. / Vaccine 19 (2001) 3248–3252

Fig. 1. PFGE profiles obtained after digestion of DNA with either Xba I (A) or with Spe I (B) restriction enzymes. The arrow indicates the specific band visualized only after digestion of Japanese isolates with Spe I enzyme-lanes: (1) CA-1; (2) TB 49; (3) Bp 134; (4) SA-4; (5) SK-5; (6) TY-7; (7) 93005-106; (8) 93005-107; (9) B-902; (10) 95008-109; (11) Isolate 9; (12) Isolate 7; (13) Isolate 13; (14) Isolate 18; (15) Fr N°287; (l) markers. (C) Compilation of the data obtained after digestion of isolates DNA with both Xha I and Spe I restriction enzymes using the Neighbor Joining method.

Since variation in the B. pertussis genome may or may not reflect antigenic changes in specific proteins, we next characterized critical proteins PT and PRN produced by the isolates to determine if changes in theses proteins, occurring after introduction of acellular pertussis vaccines, could be detected [9,10]. DNA was extracted and amplified by PCR. Sequencing of ptx genes was conducted by Lark Sequencing Technologies (Houston, TX) whereas prn genes and some ptxS1 genes were sequenced by the ESGS company (Groupe CYBERGENE, Evry, France). PT is a multi-subunit protein that has an A–B structure typical of many bacterial toxins [19]. The S1

subunit (A component) of PT is enzymatically active, whereas the other subunits of the protein (S2, S3, S4, and S5) make up the binding component (B) of the toxin. Both the A and B components are necessary for toxin action and both are capable of eliciting protective antibodies [2,11,15,19].We sequenced the ptx coding region of the vaccine strain and of a total of five isolates to determine whether changes in antigenic structure occurred rapidly after introduction of acellular pertussis vaccines. Two of these isolates, SA-4 and TY-7 were collected before the introduction of acellular pertussis vaccines in Japan. Three other isolates, 93005107, 95008-109, and 18, were collected 10 years or more

N. Guiso et al. / Vaccine 19 (2001) 3248–3252

after the introduction of the acellular pertussis vaccines. We chose these latter isolates because they were each collected in different years. Two of the isolates, 93005107 and 95008-109, were collected in Fukui prefecture, whereas isolate 18 was collected in Kanagawa prefecture, about 300 km distant from Fukui prefecture across the backbone of Japan. Thus, in choosing these particular isolates, we hoped to minimize the possibility that all of the isolates were from a single outbreak and therefore clonal in nature. In fact, PFGE analysis shows differences between these isolates (Fig. 1). For each isolate, we determined the sequence of the entire ptx coding region from nucleotide 609 to nucleotide 3628 of the ptx operon, using the numbering system previously described [12]. We found that these five Japanese clinical isolates had identical ptx sequences and differed in only one nucleotide from the Tohama I sequence (out of a total of 3020 nucleotides). This difference occurred in ptxS2 at position 1488 (a “ g), and would result in a change of serine 18 to a glycine residue. All of the other ptx genes were identical to those of Tohama I. All five of these Japanese isolates, the Tohama vaccine strain, as well as the strain CA-1 express an S1B type similar to the type expressed by European isolates circulating before the introduction of vaccination [8]. Interestingly, the sequences of the ptx genes of isolates SA-4, TY-7, 93005-107, 95008-109 and 18 are identical to the originally published ptx sequence that was determined using strain 165 [12], a strain originally obtained from the Michigan Department of Public Health in the U.S. and which was isolated before 1951. Of note, the ptxS1 gene of strain 165 is of the ptxS1B type [8]. As shown in Fig. 1C, analysis of the PFGE data indicates that two isolates (7 and 13) both collected in 1996 were separated from the 10 others, therefore, we also sequenced the ptxS1 genes of these two isolates. As shown in Table 1 these two isolates express a S1A type which differs from S1B type in one nucleotide at position 1190 (g “ a) in the numbering system mentioned above, accompanied by a change of methionine 194 to an isoleucine residue similar to the type expressed by the European isolates currently circulating [3,7,8]. We next examined the repeated sequences of the prn gene encoding an important adhesin of B. pertussis, PRN. A number of B. pertussis PRN types have been identified [8] that differ in a specific region of the protein containing repeated sequences. We found that all five of the clinical isolates collected before introduction of acellular pertussis vaccines and five of the seven isolates collected after wide-spread use of these vaccines contained a prn1 allele type [8,10]. The vaccine strain, Tohama I, also contains a prn1 allele type [3]. Again, the two Japanese clinical isolates, which express a S1A type, express a prn2 allele type. None of the Japanese

3251

isolates exhibits a prn3 allele type which is the third major type found in Europe. As found previously with European isolates, the ptxS1B gene is associated with prn1 allele type and the ptxS1A gene is associated with the expression of prn type 2 in the Japanese isolates [7–10]. Finally, we characterized each of the Japanese strains to determine which fimbriae type they produced using polyclonal sera obtained by Yuji Sato (NIH Japan, Tokyo) and monoclonal antibodies originally obtained from Michael Brennan (FDA, Bethesda, MD). A microagglutination assay was employed as previously described [8,21], in parallel. As shown in Table 1, all Japanese clinical isolates, whether obtained before or after the introduction of acellular pertussis vaccines produced type 3 fimbriae. One isolate, obtained in 1996, also produced type 2 fimbriae. Of note, the strain Tohama I, used in the production of Japanese acellular pertussis vaccines, produces type 2 fimbriae. While a limited number of clinical isolates were examined in this study, due to lack of availability of such isolates in Japan, several observations are striking. First, the ptx regions from the three isolates collected over 10 years after introduction of acellular pertussis vaccines differed by only one nucleotide (translating into one amino acid change) from the vaccine strain and were identical to the ptx region of isolates collected both in Japan and in other regions of the world before introduction of acellular pertussis vaccines. Secondly, the prn types of five of the seven Japanese clinical isolates collected from 1992 to 1996 did not differ from prn types of the vaccine strain, of isolates collected in Japan before introduction of acellular pertussis vaccines, or of isolates currently circulating in Europe before use of whole cell vaccines. Thirdly, two isolates were found to harbor ptxS1 and prn alleles different from those of the vaccine strain. These isolates could have originated from the use of acellular vaccines for more than 10 years or from importation of these strains from other regions of the world. Finally, all Japanese isolates examined produced type 3 fimbriae regardless of whether they were isolated before or after introduction of acellular pertussis vaccines. Taken together, these results suggest that acellular pertussis vaccine-induced divergence in Japan is not such a strong selective pressure that isolates producing antigens very similar to those of the vaccine strain quickly cease to circulate. Acellular pertussis vaccines produced by different manufactures are used in Japan. They all contain inactivated PT and no striking antigenic variation was observed in this protein expressed by the isolates collected after generalized use of these vaccines. Thus it appears that if vaccine-induced antigenic divergence does occur, it must occur at a slow rate. Nonetheless, because of the low number of isolates examined, it is important to continue to carefully mon-

3252

N. Guiso et al. / Vaccine 19 (2001) 3248–3252

itor circulating isolates of B. pertussis, especially in countries that have introduced acellular pertussis vaccines, in order to insure that divergence that could affect vaccine efficacy is not occurring.

Acknowledgements We thank Mitsuru Watanabe, Tsuneo Maegawa, Shinichi Nakamura and Tatsuo Kato for clinical isolates. This work was supported by grants from the Institut Pasteur Fondation and the National Vaccine Program (USA).

References [1] Ad hoc group for the study of pertussis vaccines. Placebo-controlled trial of two acellular pertussis vaccines in Sweden — protective efficacy and adverse events. [published erratum appears in Lancet 1988 May 28, 1 (8596): 1238], Lancet 1988, 1, 955 – 960. [2] Arciniega JL, Shahin RD, Burnette WN, Bartley TD, Whiteley DW, Mar VL, Burns DL. Contribution of the B oligomer to the protective activity of genetically attenuated pertussis toxin. Infect Immun 1991;59:3407–10. [3] Boursaux-Eude C, Thiberge S, Carletti G, Guiso N. Intranasal murine model of Bordetella pertussis infection: II. Sequence variation and protection induced by a tricomponent acellular vaccine. Vaccine 1999;17:2651–60. [4] Edwards KM. Acellular pertussis vaccines — a solution to the pertussis problem? J Infect Dis 1993;168:15–20. [5] Edwards KM, Decker MD. Acellular pertussis vaccines for infants. N Engl J Med 1996;334:391–2. [6] Kimura M, Kuno-Sakai H. Current epidemiology of pertussis in Japan. Pediat Inf Dis J 1990;9:705–9. [7] Mastrantonio P, Spigaglia P, van Oirschot H, van der Heide HGJ, Heuvelman KP, Stefanelli P, Mooi FR. Anti genic variants in Bordetella pertussis strains isolated from vaccinated and unvaccinated children. Microbiol 1999;145:2069–75. [8] Mooi FR, Hallander H, Wirsing von Ko¨ning CH, Hoet B, Guiso N. Epidemiological typing of Bordetella pertussis isolates: recommendations for a standard methodology. Eur Clin Infect Dis J 2000;13:174 – 81.

.

[9] Mooi FR, Qiushui H, van Oirschot H, Mertsola J. Variation in the Bordetella pertussis virulence factors pertussis toxin and pertactin in vaccine strains and clinical isolates in Finland. Infect Immun 1999;67:3133 – 4. [10] Mooi FR, van Oirschot H, Heuvelman K, van der Heide HGJ, Gaastra W, Willems RJL. Polymorphism in the Bordetella pertussis virulence factors P.69/pertactin and pertussis toxin in the Netherlands: temporal trends and evidence for vaccine-driven evolution. Infect Immun 1998;66:670 – 5. [11] Nencioni L, Pizza MG, Volpini G, De Magistris MT, Giovannoni F, Rappuoli R. Properties of the B oligomer of pertussis toxin. Infect Immun 1991;59:4732 – 4. [12] Nicosia A, Perugini M, Franzini C, Casagli MC, Borri MG, Antoni G, Almoni M, Neri P, Ratti G, Rappuoli R. Cloning and sequencing of the pertussis toxin genes: operon structure and gene duplication. Proc Natl Acad Sci USA 1986;83:4631 – 5. [13] Noble GR, Bernier RH, Esber EC, Hardegree MC, Hinman AR, Klein D, Saah AJ. Acellular and whole-cell pertussis vaccines in Japan. Report of a visit by US scientists. J Am Med Assoc 1987;257:1351– 6. [14] Saitou N, Nei M. A new method for reconstructing phylogenetic tree. Mol Biol Evol 1987;4:406 – 25. [15] Sato H, Sato Y. Protective activities in mice of monoclonal antibodies against pertussis toxin. Infect Immun 1990;58:3369– 74. [16] Sato Y, Kimura M, Fukumi H. Development of a pertussis component vaccine in Japan, Lancet 1984; January 21: 122 – 6. [17] Sato Y, Sato H. Further characterization of Japanese acellular pertussis vaccine prepared in 1988 by 6 Japanese manufacturers. Tokai J Exp Clin Med 1988;13:79 – 88. [18] Sato Y, Sato H. Development of accelular pertussis vaccines. Biologicals 1999;27:61 – 9. [19] Tamura M, Nogimori K, Murai S, Yajima M, Ito K, Katada T, Ui M, Ishii S. Subunit structure of islet-activating protein, pertussis toxin, in conformity with the A – B model. Biochemistry 1982;21:5516 – 22. [20] van der Zee A, Vernooij S, Peeters M, van Embden J, Mooi FR. Dynamics of the population structure of Bordetella pertussis as measured by IS1002-associated RFLP: comparison of pre- and post-vaccination strains and global distribution. Microbiol 1996;142:3479– 85. [21] Zhong Ming L, Brennan MJ, David JL, Carter PH, Cowell JL, Manclark CR. Comparison of type 2 and type 6 fimbriae of Bordetella pertussis by using agglutinating monoclonal antibodies. Infect Immun 1988;56:3184 – 8.