As established by Eldering et al. (3), fimbrial antigens of Bordetella pertussis cells were used to raise agglutinating antibodies and have thus been called agglutinogens, which are constitutive parts of two fimbrial proteins, Fim2 and Fim3, and of an outer membrane protein called pertactin (Prn), respectively (1, 2). On the basis of this, monoclonal antibodies to Fim2 and Fim3 were obtained (4), and B. pertussis isolates can be typed as being either Fim2, Fim3, or Fim2,3 (5). Polyclonal antisera to agglutinogens 2 and 3 were available commercially, and a microagglutination procedure for serotyping isolates, which has been used for many years, has been described previously (7). Fimbrial typing of the same isolates done either with polyclonal or with monoclonal antisera, however, has not been compared. Thus, we compared the results obtained with two different sets of polyclonal antisera and one set of monoclonal antisera.
For comparison of polyclonal antisera from source 1 and monoclonal antisera, 42 isolates were randomly selected from isolates of B. pertussis from vaccine trials in Germany. For comparison of polyclonal antisera from source 2 and monoclonal antisera, 20 isolates were randomly selected from isolates of B. pertussis from a vaccine trial in Sweden. For serotyping, isolates were reconstituted on charcoal-horse blood medium (polyclonal antisera) or on Bordet-Gengou medium (monoclonal antisera). Polyclonal antisera (source 1) to agglutinogens 1, 2, and 3 (B. pertussis anti-agglutinogen 1 [no. 89/596], B. pertussis anti-agglutinogen 2 [no. 89/598], and B. pertussis anti-agglutinogen 3 [no. 89/600] were purchased from the National Institute for Biological Standards and Controls, Potters Bar, United Kingdom. Microagglutination with these antisera was performed as described previously (7). The other set of polyclonal antisera (source 2) to agglutinogens 1, 2, and 3 was produced in the former National Bacteriological Laboratory (Stockholm, Sweden) as described elsewhere (8). Bacterial suspensions were serotyped using a standardized slide agglutination technique (8).
Two monoclonal antisera were provided by the Laboratory of Pertussis, Food and Drug Administration, Bethesda, Md.: anti-Fim2 from hybridoma F2B2G8 and anti-Fim3 from hybridoma C10C2D5 (4). Bacterial suspensions were made from Bordet-Gengou agar cultures which had been incubated at 37°C in ambient atmosphere for 48 h. The bacteria were suspended in buffered saline (pH 7.2). Typing was done with two bacterial suspensions, each done in duplicate, with optical densities at 650 nm (OD650s) of 1.0 and 0.5. Fifty-microliter volumes of the bacterial suspension were pipetted into microplates (V bottom; Greiner, Poitiers, France). Fifty-microliter volumes of the respective antiserum were added in a concentration of 4 mg/ml in buffered saline (GIBCO). The contents of each microplate were gently mixed and incubated overnight at +37°C. The microplate was read under an inverted mirror twice by two persons. The validity of the reading was checked by the results of negative and positive controls.
A total of 42 isolates were typed with polyclonal antisera from source 1 and monoclonal antisera (Table 1). Three of 42 isolates were untypeable with the polyclonal antisera (monoclonal antisera, two of Fim2,3 and one of Fim2). Using monoclonal antisera, five isolates showed only weak agglutination with antibodies to Fim2, and these were regarded as Fim2 positive. Of the 39 isolates that gave interpretable results with both types of antisera, only 21 (54%) were typed unequivocally with both polyclonal and monoclonal antisera (2 were typed as Fim2, 7 as Fim2,3, and 12 as Fim3). Of the remaining 18 isolates (46%), three were typed as Fim3 with polyclonal antisera and as Fim2 with monoclonal antisera, and two were typed as Fim2,3 with polyclonal antisera and as Fim2 with monoclonal antisera. Fourteen isolates (36%) were typed as Fim3 with polyclonal antisera and as Fim2,3 with monoclonal antisera.
TABLE 1.
Fimbrial typing with polyclonal antisera from source 1 and monoclonal antisera
| Isolate | Fim type determined witha:
|
|
|---|---|---|
| Polyclonal antisera from source 1 | Monoclonal microagglutination | |
| AI2741b | 1,2 | 2 |
| AI3128 | 1,2 | 2 |
| AI1036 | 1,2,3 | 2 |
| AI1534 | 1,2,3 | 2 |
| AI1201 | 1,2,3 | 2,3 |
| AI1484 | 1,2,3 | 2,3 |
| AI1549 | 1,2,3 | 2,3 |
| AI2107 | 1,2,3 | 2,3 |
| AI2231 | 1,2,3 | 2,3 |
| AI2635a | 1,2,3 | 2,3 |
| AI662a | 1,2,3 | 2,3 |
| AI1265 | 1,3 | 2 |
| AI1761 | 1,3 | 2 |
| AI55 | 1,3 | 3 |
| AI2223 | 1,3 | 3 |
| AI2229 | 1,3 | 3 |
| AI134a | 1,3 | 3 |
| AI1477a | 1,3 | 3 |
| AI1513 | 1,3 | 3 |
| AI2549a | 1,3 | 3 |
| AI2657a | 1,3 | 3 |
| AI3990b | 1,3 | 3 |
| AI6275b | 1,3 | 3 |
| AI638 | 1,3 | 3 |
| AI876 | 1,3 | 3 |
| AI22a | 1,3 | 2,3 |
| AI23a | 1,3 | 2,3 |
| AI744 | 1,3 | 2,3 |
| AI1543 | 1,3 | 2,3 |
| AI1561 | 1,3 | 2,3 |
| AI1586 | 1,3 | 2,3 |
| AI1887a | 1,3 | 2,3 |
| AI2024 | 1,3 | 2,3 |
| AI2029 | 1,3 | 2,3 |
| AI2102 | 1,3 | 2,3 |
| AI2200 | 1,3 | 2,3 |
| AI110 | 1,3 | 2,3 |
| AI4249a | 1,3 | 2,3 |
| AI4312a | 1,3 | 2,3 |
| AI11367 | NT | 2 |
| AI1182 | NT | 2,3 |
| AI1484 | NT | 2,3 |
NT, no definite testing possible. Discrepant results are indicated by boldface type.
A total of 20 isolates were typed with polyclonal antisera from source 2 and monoclonal antisera (Table 2). Seventeen isolates gave identical results with both types of antisera (eight were typed as Fim2, six as Fim3, and three as Fim2,3). One isolate was typed as Fim3 with polyclonal antisera and as Fim2 with monoclonal antisera; one was typed as Fim2 with polyclonal antisera and as Fim3 with monoclonal antisera; and one was typed as Fim2,3 with polyclonal antisera and as Fim3 with monoclonal antisera.
TABLE 2.
Fimbrial typing with polyclonal antisera from source 2 and monoclonal antisera
| Isolate | Fim type determined witha:
|
||
|---|---|---|---|
| Polyclonal antisera from source 2 | Monoclonal microagglutination | Monoclonal IF | |
| N322 | 1,2 | 2 | 2 |
| N288 | 1,2 | 2 | 2 |
| N82 | 1,2 | 2 | 2 |
| N985 | 1,2 | 2 | 2 |
| N538 | 1,2 | 2 | 2 |
| N1128 | 1,2 | 2 | 2 |
| N846 | 1,2 | 2 | 2 |
| N151 | 1,2,3 | 2,3 | 2,3 |
| N1216 | 1,2,3 | 2,3 | 2,3 |
| N224 | 1,2,3 | 3 | 3 |
| N869 | 1,3 | 2 | 2 |
| N229 | 1,3 | 3 | 3 |
| N268 | 1,3 | 3 | 3 |
| N332 | 1,3 | 3 | 3 |
| N322 | 1,3 | 3 | 3 |
| N775 | 1,3 | 3 | 3 |
| N896 | 1,2 | 3 | NT |
| N463 | 1,2 | 2 | NT |
| N728 | 1,2,3 | 2,3 | NT |
| N696 | 1,3 | 3 | NT |
IF, direct immunofluorescence; NT, no definite testing possible. Discrepant results are indicated by boldface type.
Our data show that fimbrial typing does not give unequivocal results when different sources of antibodies are used. Since both sets of polyclonal antisera are no longer available and on the basis of our observations reported here, we suggest that fimbrial typing should be done by using monoclonal antibodies and by adhering to a standard protocol (6) and that results of fimbrial typing should be interpreted with respect to the source of the antibodies used for typing.
REFERENCES
- 1.Ashworth L A E, Irons L I, Dowsett A B. Antigenic relationship between serotype-specific agglutinogens and fimbriae of Bordetella pertussis. Infect Immun. 1982;37:1278–1281. doi: 10.1128/iai.37.3.1278-1281.1982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Brennan M J, Li Z M, Cowell J L, Shahin R D, David J L, Carter P H, Manclark C R. Identification of a 69-kilodalton nonfimbrial protein as an agglutinogen of Bordetella pertussis. Infect Immun. 1988;56:3189–3195. doi: 10.1128/iai.56.12.3189-3195.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Eldering G, Hornbeck C, Baker J. Serological studies on Bordetella pertussis and related species. J Bacteriol. 1957;74:133–136. doi: 10.1128/jb.74.2.133-136.1957. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Li Z M, Brennan M J, David J L, Carter P H, Cowell J L, Manclark C R. Comparison of type 2 and type 6 fimbriae of Bordetella pertussis by using agglutinating monoclonal antibodies. Infect Immun. 1988;56:3184–3188. doi: 10.1128/iai.56.12.3184-3188.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mooi F R, van der Heide H G J, ter Avest A R, Welinder K G, Livey I, van der Zeijs B A M, Gaastra W. Characterization of fimbrial subunits from Bordetella pertussis. Microb Pathog. 1987;3:1–8. doi: 10.1016/0882-4010(87)90054-4. [DOI] [PubMed] [Google Scholar]
- 6.Mooi F R, Hallander H, Wirsing von König C H, Guiso N. Epidemiological typing of Bordetella pertussis isolates: recommendations for a standard methodology. Eur J Clin Microbiol Infect Dis. 2000;19:174–181. doi: 10.1007/s100960050455. [DOI] [PubMed] [Google Scholar]
- 7.Preston N W. Essential immunogens in human pertussis: the role of fimbriae. Dev Biol Stand. 1983;51:137–141. [PubMed] [Google Scholar]
- 8.Tiru M, Askelöf P, Granström M, Hallander H. Bordetella pertussis serotype of clinical isolates in Sweden during 1970–1995 and influence of vaccine efficacy trials. Dev Biol Stand. 1997;89:239–245. [PubMed] [Google Scholar]