Characterization of a recombinant fragment that contains a carbohydrate recognition domain of the filamentous hemagglutinin.

The filamentous hemagglutinin (FHA) of Bordetella pertussis plays an important role in establishing infection by attaching the bacteria to the ciliated respiratory epithelial cells. Expression of DNA encoding residues 1141 to 1279 of FHA in Escherichia coli yields a protein of 18,000 Da that exhibits some of the carbohydrate recognition properties of FHA (S. M. Prasad, Y. Yin, E. Rodzinski, E. I. Tuomanen, and H. R. Masure, Infect. Immun. 61:2780-2785, 1993). We have constructed an E. coli strain that expresses this protein, designated fragment A, in a soluble form at markedly elevated levels. Fragment A could be purified with high purity and yields and was immunogenic in mice. Both fragment A and anti-fragment A sera inhibited the binding of B. pertussis to asialo-GM2 and to rabbit ciliated cells. These observations demonstrate that this fragment of FHA contains a cellular binding domain capable of eliciting functional antibodies.

Evaluation of mixtures of purified Haemophilus influenzae outer membrane proteins in protection against challenge with nontypeable H. influenzae in the chinchilla otitis media model.

Nontypeable Haemophilus influenzae (NTHi) is one of the leading causative agents of bacterial otitis media, and no vaccine has been shown to be effective against it. Three outer membrane lipoproteins of NTHi have been investigated extensively and are leading candidates for inclusion in a vaccine against this organism. Hi-PAL (P6), recombinant PCP (rPCP), and e (P4) proteins are antigenically conserved among NTHi strains and elicit bactericidal and protective antibodies. A genetic fusion of the rPCP and Hi-PAL proteins has also been reported. Mixtures of these proteins were used for active immunization experiments in the chinchilla model of otitis media. Chinchillas were immunized either with a mixture of all three lipoproteins or with the mixture of rPCP-PAL hybrid plus e protein. When these animals were challenged with a NTHi strain injected directly into the middle ears, no protection from infection or disease, as measured by otoscopy, was observed in either group. However, effusion and inflammation measured by tympanometry were significantly reduced in animals immunized with the three lipoproteins. Animals that had been immunized with either whole NTHi cells or total outer membranes and then challenged with the homologous strain were significantly protected from both infection and disease, as determined by tympanometry and otoscopy. Unlike other animals antisera, chinchilla antisera against the purified proteins had no bactericidal activity against NTHi but did fix complement on the cell surface. Thus, the chinchilla immune responses to mixtures of these lipoproteins differ from the immune responses observed in other animal species. Further evaluation of these proteins for their vaccine potential remains to be done.

Biochemical and immunological properties of two forms of pertactin, the 69,000-molecular-weight outer membrane protein of Bordetella pertussis.

Two apparent isoforms of the virulence-associated 69,000-molecular-weight protein pertactin were purified from Bordetella pertussis. Mass spectrometry showed a difference of 2,060 Da, which may result from differential C-terminal cleavage of a larger precursor. Both forms were protective in a mouse model, eliciting bactericidal antibodies and reducing respiratory tract colonization.

A bactericidal monoclonal antibody specific for the lipooligosaccharide of Bordetella pertussis reduces colonization of the respiratory tract of mice after aerosol infection with B. pertussis.

A mouse immunoglobulin G3 monoclonal antibody specific for the core oligosaccharide moiety of the lipooligosaccharide (LOS) of Bordetella pertussis has been shown to have complement-dependent bactericidal activity. This monoclonal antibody exhibits bactericidal activity against strains of B. pertussis that express the LOS A phenotype. In addition this monoclonal antibody was effective in reducing colonization by B. pertussis in both the lungs and tracheas of mice after aerosol infection.

Protective immunogenicity of synthetic peptides selected from the amino acid sequence of Bordetella pertussis toxin subunit S1.

Five synthetic peptides identified as antigenic sites on the S1 subunit of pertussis toxin (PT) were coupled to the diphtheria toxin cross-reactive mutant protein CRM 197 or BSA. All peptide conjugates were immunogenic in animals. Two peptide-CRM-conjugates, corresponding to amino acids 1-17 and 169-186, induced especially high antibody titers against native PT in mice (Balb/c, C57/Black and outbred NMRI) as measured by ELISA. Upon in vivo PT challenge (0.5 microgram toxin) of the NMRI mice both the CRM and BSA conjugates of these two peptides fully protected the mice from PT induced leucocytosis. Some of the protected mice receiving peptide 1-17 conjugate had very low antibody titers against PT but high titers against the peptide as measured by ELISA, showing that PT-ELISA does not always reflect in vivo protection. A booster response against PT was noted upon challenge with PT in mice receiving peptide 1-17 or peptide 169-186 conjugate. Antibodies against peptide 170-186 could not be evaluated by ELISA since conjugates of this peptide (like PT itself) bind to immunoglobulins. They may also caused clustering of CHO cells. Rabbit antiserum to the peptide 1-17-CRM conjugate was highly efficient in inhibiting the ADP-ribosylating activity of PT with bovine transducin as substrate whereas the rabbit antiserum raised against the peptide 169-186-CRM conjugate neutralized the clustering effect of PT on CHO cells. Thus there is no concise correlation between the in vivo protection against PT challenge and the in vitro methods used for measuring antibody levels against PT (neutralization of the enzyme activity, the CHO cell clustering activity and titers in PT-ELISA). The CRM-conjugates of these two peptides constitute the first synthetic pertussis vaccine candidate with the ability to provide a chemically well defined, safe and efficient pertussis vaccine.

Pertussis toxin analog with reduced enzymatic and biological activities is a protective immunogen.

Bordetella pertussis TOX3201 has a 12-base-pair insertion in the S1 subunit gene of pertussis toxin (PTX), which encodes for a 4-amino-acid insertion between residues 107 and 108 of the mature S1 subunit (Black et al., Science 240:656-659, 1988). This mutant strain has been shown to secrete a holotoxin analog of PTX, designated CRM3201, with reduced ADP-ribosyltransferase activity. In the present study, we evaluated the biochemical, biological, and immunoprotective activities of purified CRM3201. Assay of enzymatic activities showed that CRM3201 had 20 to 30% of the ADP-ribosyltransferase activity and 55 to 60% of the NAD glycohydrolase activity of native PTX. CRM3201, however, had only 2 to 6% of the activity of PTX in clustering CHO cells, promoting leukocytosis, inducing histamine sensitization, and potentiating an anaphylactic response to bovine serum albumin. In contrast, activities associated with the B oligomer (binding to fetuin, hemagglutination of goose erythrocytes, and lymphocyte mitogen activity) were comparable to those of native PTX. Injection of BALB/c mice with CRM3201 mixed with Al(OH)3 elicited high titers of antibody to PTX (as measured by enzyme-linked immunosorbent assay), which neutralized a leukocytosis-promoting dose of PTX in these mice and neutralized PTX in a CHO cell assay. Passive transfer of the anti-CRM3201 antibody protected 20-day-old Swiss-Webster mice against a lethal aerosol challenge with B. pertussis 18323. Active immunization with CRM3201 significantly reduced lung colonization in adult BALB/c mice with a B. pertussis respiratory infection. These results demonstrate (i) that the reduced ADP-ribosyltransferase activity of CRM3201 is associated with reductions in certain biological and toxic activities of PTX (the enzymatic and biological activities are not, however, totally concordant); (ii) that CRM3201 possesses a functional B oligomer; and (iii) that CRM3201 can induce toxin-neutralizing antibodies which protect mice against a respiratory challenge with B. pertussis. Our studies with CRM3201 show that recombinant analogs of PTX have the potential to be developed into safe, protective immunogens for use in new acellular pertussis vaccines.

Protective immunogenicity of two synthetic peptides selected from the amino acid sequence of Bordetella pertussis toxin subunit S1.

Two peptides, corresponding to amino acids 1-17 and 169-186 of the amino acid sequence of pertussis toxin (PT) subunit S1, were synthesized and coupled to the diphtheria toxin cross-reactive mutant protein CRM 197 and evaluated for immunogenicity and protective capacity against PT challenge in vivo. The peptide-CRM conjugates induced high antibody titers against native toxin in mice (BALB/c, C57/Black, and outbred NMRI) as measured by ELISA. Upon PT challenge (0.5 microgram of toxin) of the NMRI mice, the CRM conjugates of peptides 1-17 and 169-186 fully protected the mice from PT-induced leukocytosis. Immunization with the corresponding bovine serum albumin conjugates of these two peptides also fully protected mice. Rabbit antiserum to the peptide 1-17-CRM conjugate was highly efficient in inhibiting the ADP-ribosylating activity of PT but did not neutralize the clustering effect of PT on Chinese hamster ovary cells. In contrast, the rabbit antiserum raised against the peptide 169-186-CRM conjugate neutralized the clustering effect of PT on Chinese hamster ovary cells but did not inhibit the enzymatic activity of PT. Peptide 169-186-CRM conjugates mimic the immunoglobulin binding properties of PT and also cause clustering of Chinese hamster ovary cells. The CRM conjugates of these two peptides constitute a synthetic pertussis vaccine candidate with the ability to provide a chemically well-defined, safe, and efficient pertussis vaccine.

Bordetella pertussis filamentous hemagglutinin: evaluation as a protective antigen and colonization factor in a mouse respiratory infection model.

Filamentous hemagglutinin (FHA) is a cell surface protein of Bordetella pertussis which functions as an adhesin for this organism. It is a component of many new acellular pertussis vaccines. The proposed role of FHA in immunity to pertussis is based on animal studies which have produced some conflicting results. To clarify this situation, we reexamined the protective activity of FHA in an adult mouse respiratory infection model. Four-week-old BALB/c mice were immunized with one or two doses of 4 or 8 micrograms of FHA and then aerosol challenged with B. pertussis Tohama I. In control mice receiving tetanus toxoid, the CFU in the lungs increased from 10(5) immediately following infection to greater than 10(6) by days 5 and 9 after challenge. Mice immunized with FHA by the intraperitoneal or intramuscular route had significantly reduced bacterial colonization in the lungs. A decrease in colonization of the trachea was also observed in FHA-immunized mice. Evaluation of antibody responses in these mice revealed high titers of immunoglobulin G (IgG) and IgM to FHA in sera and of IgG to FHA in lung lavage fluids. No IgA to FHA was detected. BALB/c mice were also passively immunized intravenously with either goat or rat antibodies to FHA and then aerosol challenged 24 h later, when anti-FHA antibodies were detected in the respiratory tract. Lung and tracheal colonization was markedly reduced in mice immunized with FHA-specific antibodies compared with those receiving control antibodies. In additional studies, the role of FHA in the colonization of the mouse respiratory tract was evaluated by using strain BP101, an FHA mutant of B. pertussis. FHA was important in the initial colonization of the mouse trachea, but was not required for colonization of the trachea later in the infection. FHA was not a factor in colonization of the lungs. Collectively, these experiments demonstrate (i) that systemic immunization with FHA can provide significant protection against B. pertussis infection in both the lower and upper respiratory tract of mice as defined by the lungs and trachea, respectively; (ii) that this protection is mediated primarily by serum antibodies to FHA, which transudate into respiratory secretions; and (iii) that FHA is an important upper respiratory tract colonization factor. These studies provide further evidence for the role of FHA in pertussis pathogenesis and immunity.

Monoclonal antibodies to pertussis toxin: utilization as probes of toxin function.

Six monoclonal antibodies (MAbs) to pertussis toxin (PT) have been generated and characterized. Five of these MAbs (3CX4, 3C4D, 6D11C, 6FX1, and X2X5) interact with determinants on the catalytic subunit (S1) of PT, and one (6DX3) is specific for subunit S4. The MAbs are divided into three groups based upon their ability to neutralize the effects of PT in a Chinese hamster ovary (CHO) cell assay. Three of the MAbs (3CX4, 3C4D and 6D11C) had high neutralization titers, one MAb (6FX1) displayed weak neutralizing activity, and two MAbs (X2X5 and 6DX3) had no neutralizing ability. The combination of one of the high titer MAbs (3CX4) with the low titer MAb (6FX1) resulted in a synergistic enhancement of neutralizing capability. F(ab')2 fragments prepared from MAb's 3CX4 and X2X5 displayed activities in the CHO-cell assay which were identical to the native MAb's. The ability of the MAbs to neutralize PT in the CHO-cell toxin neutralization assay correlated with their ability to inhibit the in vitro ADP-ribosylation of PT. A competition ELISA method demonstrated that this panel of MAbs recognizes at least four separate epitopes on the PT molecule. Biotin-conjugated MAbs were shown to be useful reagents to probe the interaction of pertussis toxin with fetuin.

Comparison of type 2 and type 6 fimbriae of Bordetella pertussis by using agglutinating monoclonal antibodies.

Two types of fimbriae have been identified on the pathogenic gram-negative organism Bordetella pertussis. Monoclonal antibodies to these fimbriae were produced to better understand the role of fimbriae as serotype-specific agglutinogens and to investigate the antigenic relationship between these fimbriae. Three monoclonal antibodies were identified that specifically agglutinated B. pertussis cells containing the U.S. Reference Factor 2 agglutinogen, and six monoclonal antibodies were produced that agglutinated only those strains containing the U.S. Reference Factor 6 agglutinogen. Indirect immunofluorescence studies and immunogold electron microscopy demonstrated that these monoclonal antibodies bind to an outer membrane component on serotype-specific strains of B. pertussis. All of the monoclonal antibodies reacted with native or partially assembled type-specific fimbriae but not with monomeric fimbrial subunits as indicated by Western blot (immunoblot) analysis. The fimbrial agglutinogens recognized by the monoclonal antibodies were also uniquely reactive with either U.S. Reference Factor 2 or 6 antiserum (Eldering agglutinogen 2 or 6 polyclonal antiserum) in an indirect ELISA. No cross-reactivity of the monoclonal antibodies with the unrelated fimbriae was observed in any of the comparative immunological studies. Some of the monoclonal antibodies agglutinated certain strains of B. bronchiseptica, suggesting that this closely related species can contain antigenically similar fimbriae. These monoclonal antibodies should prove useful for further structural and functional analysis of Bordetella fimbriae and for studies on the role that these antigens play in prevention of infection and disease.

Identification of a 69-kilodalton nonfimbrial protein as an agglutinogen of Bordetella pertussis.

Cells of Bordetella pertussis BP353, a nonfimbriated Eldering serotype 1.3 strain, were used as an immunogen to produce three monoclonal antibodies, BPE3, BPD8, and BPE8, that agglutinated the immunizing cells, as well as certain other nonfimbriated and fimbriated serotype 3-containing B. pertussis strains. The antibodies did not agglutinate serotype 1 or nontypable B. pertussis cells. These monoclonal antibodies specifically detected a 69-kilodalton (kDa) band on Western blots (immunoblots) containing whole B. pertussis cell lysates of Eldering agglutinogen serotypes 1.3, 1.3.6, 1.2.3.4, and 1.2.3.4.6. This 69-kDa antigen was released from the bacteria by cell incubation for 60 min at 60 degrees C, and it was purified by affinity chromatography with a BPE3-agarose affinity matrix. Purified material was used to produce a polyclonal antiserum that agglutinated all nonfimbriated and fimbriated B. pertussis cells containing serotype 3 agglutinogen. Immunogold electron microscopy and indirect immunofluorescence studies demonstrated that it is an outer membrane constituent but nonfimbrial in appearance. BPE3 did not detect purified fimbriae on Western blots, and antibodies to these fimbriae did not bind to the 69-kDa component. Although B. bronchiseptica and B. parapertussis cells were not agglutinated by the monoclonal antibodies, antigenically similar proteins were detected in extracts of the bacteria. These results identify the 69-kDa protein as a nonfimbrial agglutinogen present on all virulent strains of B. pertussis. The monoclonal antibodies described here should be useful for further studies on the structure and function of this protein.

ADP-ribosyltransferase activity of pertussis toxin and immunomodulation by Bordetella pertussis.

Pertussis toxin is produced by the causative agent of whooping cough, Bordetella pertussis, and is an adenosine diphosphate (ADP)-ribosyltransferase capable of covalently modifying and thereby inactivating many eukaryotic G proteins involved in cellular metabolism. The toxin is a principal determinant of virulence in whooping cough and is a primary candidate for an acellular pertussis vaccine, yet it is unclear whether the ADP-ribosyltransferase activity is required for both pathogenic and immunoprotective activities. A B. pertussis strain that produced an assembled pertussis holotoxin with only 1 percent of the ADP-ribosyltransferase activity of the native toxin was constructed and was found to be deficient in pathogenic activities associated with B. pertussis including induction of leukocytosis, potentiation of anaphylaxis, and stimulation of histamine sensitivity. Moreover, this mutant strain failed to function as an adjuvant and was less effective in protecting mice from intracerebral challenge infection. These data suggest that the ADP-ribosyltransferase activity is necessary for both pathogenicity and optimum immunoprotection. These findings bear directly on the design of a nontoxic pertussis vaccine.

Agglutinating monoclonal antibodies that specifically recognize lipooligosaccharide A of Bordetella pertussis.

Monoclonal antibodies that specifically agglutinate strains of Bordetella pertussis having serotype 1 agglutinogen were uniquely reactive with the electrophoretically slow-migrating A form of lipooligosaccharide. These monoclonal antibodies should be useful for the structural analysis of B. pertussis lipooligosaccharide and for the establishment of a better-defined serogroup for Bordetella species.

Purification and characterization of serotype 6 fimbriae from Bordetella pertussis and comparison of their properties with serotype 2 fimbriae.

Fimbriae were removed from Bordetella pertussis (serotype 1.3.6) by mechanical shearing and purified by precipitation with ammonium sulfate, pH-dependent precipitation at pH 7.4, followed by two successive extractions of the precipitated fimbriae with 4 M urea. By electron microscopy, the precipitated fimbriae appeared as aggregated bundles of long, relatively straight filaments which were disaggregated to individual flexuous filaments at pH 10.5. These purified fimbriae were identified as serotype 6 agglutinogens, since antibody to the purified fimbriae agglutinated B. pertussis strains serotyped as 1.3.6, 1.2.3.6, or 1.2.3.4.6 but did not agglutinate strains of serotype 1.2.3.4, 1.2.3, or 1.3. In contrast, antibody to serotype 2 fimbriae only agglutinated B. pertussis strains containing serotype 2 agglutinogen. Purified type 6 and 2 fimbriae were found to be weakly cross-reactive by enzyme-linked immunosorbent assay, using polyclonal antibody to each type of fimbria. In an immunoblot assay, polyclonal antibodies to a 22,000-dalton subunit of fimbriae from B. bronchiseptica reacted strongly with the type 2 fimbrial subunit of B. pertussis, but only weakly with the type 6 subunit. When subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, the protein subunit of the type 6 fimbriae migrated with a molecular weight of 21,500, whereas the type 2 fimbrial subunit had a molecular weight of 22,000. The two types of subunits had similar amino acid compositions and showed amino-terminal sequence homology in 15 of 21 amino acids. The amino-terminal amino acid sequences of the B. pertussis fimbriae were distinct from those reported for fimbriae from other gram-negative bacteria. Neither the type 6 nor the type 2 fimbriae caused hemagglutination when assayed with several types of erythrocytes.

Helical structure of Bordetella pertussis fimbriae.

The helical structures of Bordetella pertussis fimbriae of serotypes 2 and 6 were determined by optical diffraction analysis of electron micrographs of negatively stained paracrystalline bundles of purified fimbriae. The fimbrial structure is based on an axial repeat of 13 nm that contains five repeating units in two complete turns of a single-start helix. This structure was confirmed by direct measurements of mass per unit length for individual fimbriae performed by dark-field scanning transmission electron microscopy of unstained specimens. These data further established that the helically repeating unit is a monomer of fimbrial protein (Mr congruent to 22,000 for type 2 and Mr congruent to 21,500 for type 6). Radial density profiles calculated from the scanning transmission electron micrographs showed that the fimbria has peak density at its center, i.e., no axial channel, consistent with the results of conventional negative-staining electron microscopy. The radial profile gives an outermost diameter of approximately 7.5 nm, although the peripheral density is, on average, diffuse, allowing sufficient intercalation between adjacent fimbriae to give a center-to-center spacing of approximately 5.5 nm in the paracrystals. Despite serological and biochemical differences between type 2 and type 6 fimbriae, the packing arrangements of their fimbrial subunits are identical. From this observation, we infer that the respective subunits may have in common conserved regions whose packing dictates the helical geometry of the fimbria. It is plausible that a similar mechanism may underlie the phenomenon of phase variations in other systems of bacterial fimbriae.

Filamentous hemagglutinin has a major role in mediating adherence of Bordetella pertussis to human WiDr cells.

[35S]methionine-labeled Bordetella pertussis adhered to monolayers of WiDr cells, an epitheliumlike cell line from a human intestinal carcinoma. Adherence was proportional to the density of the WiDr cells and to the concentration of B. pertussis in the assay. Adherence of virulent phase I strains Tohama phase I, 114, and BP338 was much greater than adherence of avirulent strains Tohama phase III and 423 phase IV. Mutants deficient in the production of the filamentous hemagglutinin (FHA) were hemagglutination negative and adhered to WiDr cells much less efficiently than the parent strains. Preincubation of B. pertussis cells with FHA increased their hemagglutination activity and adherence to WiDr cells. Goat antibody to FHA inhibited, in a dose-dependent manner, the adherence of strain Tohama I but not the adherence of FHA-deficient mutant Tohama 325. At similar protein concentrations, normal goat antibody, goat antibody to pertussis toxin, or the Fab fragments of goat antibody to serotype 2 fimbriae had no effect on adherence. Also, an FHA-positive strain without fimbriae showed high adherence, while a fimbriated FHA-deficient mutant adhered poorly. Our data indicate that FHA plays a major role in adherence of B. pertussis to human WiDr cells. Fimbriae do not appear to mediate attachment of B. pertussis to WiDr cells.

Characterization of the filamentous hemagglutinin from Bordetella pertussis by gel electrophoresis.

A highly purified preparation of filamentous hemagglutinin (FHA) from Bordetella pertussis was analyzed for its protein composition by gel electrophoretic methods. In this preparation of FHA the following native species could be detected by polyacrylamide gel electrophoresis (PAGE) at pH 3.2: S1 and S2 (inactive subunits or fragments); two monomers, a major form designated Ia (144K), and a minor form Ib, differing only in net charge; and three oligomeric forms, designated II (213K), III (595K) and IV (1064K). Hemagglutinating activity was associated predominantly with component Ia. PAGE of FHA after derivatization with sodium dodecyl sulfate (SDS) showed there to be three major species, designated A, C and D. According to estimated molecular weight values, A, C and D are likely to correspond to S2, Ia and II respectively. Isolated components II, III and IV yield all three SDS-species upon derivatization with SDS. Both moving boundary electrophoresis and gel electrofocusing showed hemagglutinating FHA to be a basic protein. Its apparent pI is 8.1.