The N-Terminal Domain of RTX Toxin ApxI of Actinobacillus pleuropneumoniae Elicits Protective Immunity in Mice

Actinobacillus pleuropneumoniae, a member of the family Pasteurellaceae, is the etiological agent of porcine pleuropneumoniae. To date, 15 serotypes have been described which variously secrete four different cytotoxins belonging to the RTX toxin family: ApxI, ApxII, ApxIII, and ApxIV (4, 31). The virulence of A. pleuropneumoniae is multifactorial (7, 10, 11, 13, 16-19, 21, 29); however, studies indicate that virulence is strongly correlated with the production of Apx exotoxins (20, 22, 23, 29, 33), with serovars producing ApxI, together with ApxII, being the most virulent (13, 14, 24). Apx toxins are strongly immunogenic and have been studied as a potential vaccine against porcine pleuropneumonia. This calcium-dependent ApxI (15), which shows hemolytic and cytotoxicity activity, is secreted by the most virulent serotypes, i.e., serotypes 1, 5, 9, 10, and 11 (2). Structurally, all Apx toxins share features, including the N-terminal hydrophobic domain and the glycine-rich, Ca²⁺-binding, tandem nonapeptide repeats in the C-terminal third of the toxin (5, 26). The major antigenic segments of RTX toxin ApxI were studied in view of its impact to generate neutralizing and protective antibodies. To locate the protective epitopes of ApxI, three deletion derivatives covering the N-terminal region (X1F1; amino acids [aa] 40 to 380), the activation domain (X1F2; aa 400 to 650), and the Ca²⁺-binding domain (X1F3, aa 680 to 825) were produced (Fig. 1A) from the apxI gene (accession no. X68595 ) (27) of A. pleuropneumoniae 3906. Fragments X1F1 and X1F2 were expressed as His₆-tagged fusion protein, whereas fragment X1F3 was expressed as glutathione S-transferase fusion protein in Escherichia coli. The sizes and seroreactivity of each recombinant protein were verified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Fig. 1B) and immunoblotting (30) with serum samples obtained from infected animals (Fig. 1C).

A. pleuropneumoniae was cultured in brain heart infusion medium (Difco) supplemented with β-NAD (10 μg/ml; Sigma). Culture supernatant containing ApxI was filtered and adjusted to 40 hemolytic units (HU)/ml as determined by hemolytic assay (1). For the hemolytic-neutralization assay, 100 μl of toxin-containing supernatant (40 HU/ml) was mixed with 10 μl of various serum samples or sequential dilutions and then incubated for 1 h at 37°C, followed by determination of the remaining hemolytic activity. The assay was repeated three times to determine the neutralizing effect of each antiserum according to a protocol described elsewhere (1) with 1% pig erythrocytes. Antiserum produced against the N terminus covered by residues 40 to 380 exhibited ApxI hemolytic-neutralizing activity equivalent to the antiserum collected from a field animal infected with A. pleuropneumoniae serotype 1, suggesting that this domain carried putative determinants for its hemolytic activity (Fig. 2). No correlation was found between immunogenicity and neutralization activities of anti-ApxI antisera (data not shown). Therefore, uneven immunogenicity was unlikely to affect hemolytic-neutralizing activities. This finding correlates with studies that show that these regions are essential for binding to erythrocytes and for pore formation and that deletion mutations at the three major hydrophobic domains I, II, and III of the N-terminal half of RTX toxins abolished the hemolytic activity of ApxI and HlyA (6, 9, 12, 25, 28). Surprisingly, the hemolytic-neutralizing activities were not found in antisera raised against the ApxI activation domain (X1F2) and the calcium-binding domain (X1F3), although the Ca²⁺-binding region is essential for full toxic activity. These two domains could possibly be involved in other functions instead of determining the lysis of erythrocytes. Study of the adenylate cyclase-hemolysin in Bordetella pertussis demonstrated that the C terminus is important for the presentation of the protective epitopes in the modification Ca²⁺-binding domain (3). However, this was not observed in ApxI N terminus, in which the conformation could be independent of the changes occurred in both the activation and Ca²⁺-binding domains (X1F2 and X1F3). ApxI antisera were further characterized in a cytotoxicity-neutralization assay with the cytotoxicity detection kit (Roche Diagnostics, Mannheim, Germany) according to the manufacturer's protocol. Antisera, prepared as described above, were preincubated with ApxIII (at 80 U/ml) secreted by A. pleuropneumoniae serotype 8 before coculture with freshly isolated pig neutrophils (8). No cross-neutralization of cytotoxicity was found between the toxins ApxI and ApxIII, indicating that these toxins utilize different cytopathic mechanisms (Fig. 3).

That A. pleuropneumoniae strain 3906 secretes only ApxI (32) and intraperitoneal injection of mice with its culture supernatant was lethal in mice suggested that the ApxI toxin could be the major virulent factor. To confirm this speculation and to determine whether the anti-X1F1 antiserum was able to compromise the lethal effect of ApxI in mice, groups of five mice were inoculated intraperitoneally with 500 μl of filtered culture supernatant with the hemolytic activity adjusted to its 50% lethal dose (12 HU/ml). The supernatant was individually incubated with 10 μl of undiluted sera from X1F1 antiserum, guinea pig preimmune serum, or field samples (designated 9904, 9906, and 9919) at 37°C for 1 h before injection. The number of mice surviving after 24 h showed that the detrimental effect of toxin in mice was fully neutralized by the X1F1 antiserum but not by the guinea pig preimmune serum or field serum samples. The results explicitly demonstrated that ApxI toxin was the major virulent factor in A. pleuropneumoniae 3906 infection and its that lethality could be neutralized by antiserum raised against its N terminus.

Essentially, all in vitro studies pointed out that the N-terminal portion of ApxI might be carrying the putative protective immunogenic epitopes that could stimulate neutralizing antibody. Nevertheless, whether the efficacy achieved by immunizing the mice with the subunit X1F1 protein alone was sufficient to induce protective immunity was not determined. To address this possibility, groups of five mice were intraperitoneally vaccinated twice at 2-week intervals with 20 μg of partially purified recombinant X1F1 protein emulsified with the adjuvant Montanide (Paris, France) ISA 70 according to the manufacturer's protocol. In control groups, mice received only buffer and/or adjuvant. Vaccinated mice were challenged intraperitoneally with a lethal dose of A. pleuropneumoniae strains, including strain 3906 (biovar II, ApxI), strain 4047 (serotype 1, ApxI/II), strain 13039 (serotype 10, ApxI), or field isolate (serotype 5, ApxI/II), immediately 1 week after the booster immunization. Mice immunized with X1F1 recombinant protein alone were sufficient to confer full protection against infection with A. pleuropneumoniae strains 3906, 4074, and 13039 except with field isolate serotype 5, where 80% of mice survival rate was observed in the vaccinated mice. However, control mice showed marginal protection (20 to 40%) after challenge with similar infectious doses used in vaccinated mice (Table 1). The good level of protection reached in this experiment suggests that the N-terminal covered by residues 40 to 380 of ApxI plays an important role in the molecular mechanism of pathogenicity in which ApxI is in involved. The protection of the N-terminal domain in the mouse study indicate that the N terminus of ApxI could be used as a vaccine candidate against A. pleuropneumoniae infection.

• Open in new tab
• Download powerpoint

FIG. 1.

(A) Structural organization of ApxI and the relative positions of each truncated ApxI fragment. Fragment X1F1 covered the N-terminal and putative transmembrane domains. Fragment X1F2 covered the gene C-mediated activation domain, and fragment X1F3 covered the calcium-binding domain of the glycine-rich repeats. (B and C) Expression and immunoblotting of ApxI deletion derivatives. Panel B shows Coomassie blue staining of partially purified ApxI recombinant proteins indicated by filled arrows. Lanes: 1, X1F1; 2, X1F2; 3, X1F3. Panel C shows that A. pleuropneumoniae-infected swine serum was reactive to all recombinant proteins. Molecular mass markers are shown on the left (in kilodaltons).

• Open in new tab
• Download powerpoint

FIG. 2.

Neutralization of the ApxI hemolytic activity. Hemolytic neutralization assays were performed with native ApxI toxin obtained from A. pleuropneumoniae 3906 preincubated with guinea pig preimmune serum (▪), X1F1 antiserum (⧫), X1F2 antiserum (▵), X1F3 antiserum (▴), and field serum (□). The results show the arithmetic means of the hemolytic neutralization findings from three experiments performed in duplicate ± the standard deviations.

• Open in new tab
• Download powerpoint

FIG. 3.

Neutrophil protection assay with anti-ApxI antisera. Cytotoxic neutralization assays were performed with native ApxIII toxin obtained from A. pleuropneumoniae serotype 8 preincubated with guinea pig preimmune serum (▪), X1F1 antiserum (⧫), X1F2 antiserum (▵), X1F3 antiserum (▴), and field serum from A. pleuropneumoniae serotype 8-infected pigs (□). The results are the arithmetic means of the cytotoxic neutralization findings from three experiments performed in duplicate ± the standard deviations.

• Copyright © 2002 American Society for Microbiology