Expression of a C Terminally Truncated Form of Pertussis Toxin S1 Subunit Effectively Induces Protection against Pertussis Toxin following DNA-Based Immunization

Abstract

Four plasmids encoding different C terminally and N terminally truncated pertussis toxin S1 subunits of Bordetella pertussis were constructed and tested for inducibility of protection against pertussis toxin in mice after DNA-based immunization. The region encoding an N-terminal 180-amino-acid fragment of the S1 subunit had the most potent ability to induce protective immunity.

Bordetella pertussis is the primary etiologic agent in pertussis disease. To prevent this infectious disease, whole-cell and acellular pertussis vaccines have been very effective in providing protection against B. pertussis infection in humans (17, 20). Acellular pertussis vaccines contain detoxified pertussis toxin (PT) and filamentous hemagglutinin as the major antigens derived from B. pertussis. Detoxified PT is considered to be essential for acellular pertussis vaccines. In fact, vaccination with detoxified PT alone has been shown to eliminate the burden of pertussis disease in a mass vaccination trial (23, 25). PT is composed of five different subunits, S1, S2, S3, S4, and S5. The S1 subunit catalyzes the ADP-ribosylation of G proteins in the target mammalian cell, while S2 through S5 (termed the B-oligomer) deliver S1 to the target cell (22). The mature S1 is composed of 235 amino acids. The enzymatic domain, associated with ADP-ribosyltransferase activity, is located within the N-terminal 180 amino acids of S1 (termed C180), whereas the remainder of S1 (residues 181 to 235) is involved in high-affinity binding to G proteins and interaction with the B-oligomer (1, 8, 10). The N-terminal 175-amino-acid fragment of S1 shares a structural homology with the enzymatic portion of heat-labile enterotoxin subunit A (21).

Monoclonal antibodies against S1 have been demonstrated to neutralize the leukocyte-promoting and islet-activating activities of PT and to protect against B. pertussis infection (18, 19). The monoclonal antibody 1B7 epitope is localized within the C180 region (4, 7, 16); therefore, a recombinant C180 peptide and an N-terminal 179-amino-acid fragment of S1, as well as full-length S1, have been used in novel pertussis vaccine studies (2, 3, 5, 11-14). Recently, Lee et al. (11) genetically fused an N-terminal 179-amino-acid fragment of S1 to the cholera toxin subunit B. The recombinant fusion protein was immunogenic in mice following intranasal immunization and induced protective immunity against an aerosol challenge with B. pertussis. On the other hand, in a significant recent study, full-length S1 expressed in Mycobacterium bovis BCG was demonstrated to be immunogenic and induced protective immunity against live B. pertussis in a mouse intracerebral-challenge model (15).

We previously reported that immunization with plasmid DNA expressing full-length S1 induced protective immunity against challenge with PT or B. pertussis (9). In the present study, a detailed investigation to determine the protective region of the S1 subunit was performed. Four plasmids encoding different C terminally and N terminally truncated S1 polypeptides were constructed and tested for their ability to induce PT-specific antibody production and protection against PT in mice.

Construction of plasmids encoding C terminally and N terminally truncated S1.

Four DNA fragments encoding C terminally and N terminally truncated S1 polypeptides (Fig. 1A), C200 (residues 1 to 200), C180 (residues 1 to 180), C160 (residues 1 to 160), and N40 (residues 40 to 235), were amplified by PCR with B. pertussis Tohama genomic DNA in the manner described previously (9). The sense primers 5′-CCCAAGCTTGCCACCATGGACGATCCTCCCGCCACC-3′ and 5′-CCCAAGCTTGCCACCATGTCCTGCCAGGTCGGCAGC-3′ were used to amplify the C terminally truncated S1 genes (the C200, C180, and C160 genes) and the N terminally truncated S1 gene (the N40 gene), respectively. The antisense primers 5′-CGGGATCCTTACTAAGCGCCTATCACCGGCGC-3′, 5′-CGGGATCCTTACTACGATGTGTAGGGGTTGGG-3′, 5′-CGGGATCCTTACTACTCCGTGGTCGTGGTCTC-3′, and 5′-CGGGATCCTTACTAGAACGAATACGCGAT-3′ were used to amplify the C200, C180, C160, and N40 genes, respectively. Included in the sense and antisense primer sequences are HindIII and BamHI restriction sites (underlined), respectively. The amplified DNA fragments were digested with HindIII and BamHI and ligated into HindIII- and BamHI-digested mammalian expression vector pcDNA3 (Invitrogen). The plasmid constructs pcDNA/C200, pcDNA/C180, pcDNA/C160, and pcDNA/N40 were transformed to Escherichia coli JM109 and purified by using End-free plasmid preparation kits (QIAGEN). The desired constructs were confirmed by automated DNA sequencing with a model 373A DNA sequencer (Applied Biosystems). A plasmid with a coding full-length S1 (pcDNA/S1) (9) and a plasmid without a coding region (pcDNA) were also used as positive and negative control DNAs, respectively.

FIG. 1.

In vitro expression of plasmids encoding C and N terminally truncated S1. (A) Schematic representation of the C and N terminally truncated forms expressed by plasmids pcDNA/C200, pcDNA/C180, pcDNA/C160, and pcDNA/N40. The white boxes correspond to the ADP-ribosyltransferase catalytic domain, and the hatched boxes correspond to high-affinity binding to G proteins and interaction with the B-oligomer. (B) Expression of mRNAs for truncated S1 in COS-7 cells. The expression of truncated S1 transcripts in COS-7 cells 24 h following transfection with the plasmids was determined by Northern blotting. (C) Expression of truncated S1 polypeptide in COS-7 cells. The expression of truncated S1 in COS-7 cells 48 h following transfection with the plasmids was determined by immunoblotting with sodium dodecyl sulfate-14% polyacrylamide gel electrophoresis. The proteins were detected with mouse polyclonal antiserum raised against detoxified PT. Purified PT (10 ng) was run on the gel as a positive control. Three independent experiments were performed, and typical Northern blot and immunoblot data are shown.

Plasmids encoded truncated S1 in COS-7 cells in vitro.

In vitro expression of the C and N terminally truncated S1 in COS-7 cells was analyzed by means of Northern blotting and immunoblotting. COS-7 cells were transfected with pcDNA/C200, pcDNA/C180, pcDNA/C160, pcDNA/N40, pcDNA/S1, or pcDNA by using SuperFect transfection reagent (QIAGEN) according to the manufacturer's instructions. The transfected cells were harvested 24 h after transfection. Total RNA was extracted by using the acid guanidinium thiocyanate-phenol-chloroform method (6), and 5 μg of total RNA was subjected to Northern blot analysis with a nylon membrane (24). A Northern blot was hybridized with a horseradish peroxidase-labeled full-length S1 gene as a probe by using the ECL direct nucleic acid labeling and detection system (Amersham Pharmacia Biotech). As shown in Fig. 1B, messenger RNAs for full-length S1, C200, and N40 were highly expressed in the pcDNA/S1-, pcDNA/C200-, and pcDNA/N40-transfected cells, whereas mRNAs for C180 and C160 had very low expression. To analyze the protein production of the plasmids, immunoblotting was performed with the cell lysates of the transfected COS-7 cells 48 h following transfection as described previously (9). As shown in Fig. 1C, C180 and full-length S1 polypeptides were highly expressed in the pcDNA/C180- and pcDNA/S1-transfected cell lysate, whereas C200 and C160 polypeptides had very low expression and N40 polypeptide was not detected. These results suggest that the truncated forms of C160 and N40 polypeptides were degraded in COS-7 cells, possibly because of the instability of polypeptides.

Immunogenicity of plasmids encoding truncated S1.

Gold particles (1 μm) were coated with purified plasmid DNA according to the manufacturer's protocol (Bio-Rad). Gold particles (0.5 mg) were coated with 2 μg of plasmid DNA and injected into 5-week-old female BALB/c mice purchased from Charles River Japan. Groups of five mice were immunized intradermally with 2 μg of plasmids pcDNA/C200, pcDNA/C180, pcDNA/C160, or pcDNA/N40 on days 0, 14, and 28 by using a Helios gene gun (Bio-Rad) as described previously (9). Age-matched control mice received immunization with pcDNA or pcDNA/S1. Mice were bled on days 17, 25, 33, and 41 after the primary immunization, and serum anti-PT immunoglobulin G (IgG) antibodies were measured by enzyme-linked immunosorbent assay as described previously (9). As shown in Fig. 2, mice immunized with pcDNA/C200, pcDNA/C180, or pcDNA/S1 produced significant levels of anti-PT IgG antibodies. On days 17 and 25, pcDNA/C180 induced significantly (P < 0.02) higher levels of anti-PT IgG production than were observed for the groups of mice immunized with pcDNA/C200 or pcDNA/S1. In contrast, anti-PT IgG was undetectable in the sera of mice immunized with pcDNA/C160, pcDNA/N40, or pcDNA.

FIG. 2.

Induction of anti-PT antibody production in BALB/c mice immunized with plasmids encoding C and N terminally truncated S1. Five mice were immunized with 2 μg of plasmids on days 0, 14, and 28. Anti-PT IgG titers in the sera of the mice were determined on days 17, 25, 33, and 41 after primary immunization. Data represent the means and standard deviations for groups of mice immunized with pcDNA/C200, pcDNA/C180, pcDNA/C160, pcDNA/N40, pcDNA/S1, and pcDNA. An asterisk indicates a significant (P < 0.02) difference from the level for pcDNA/C200.

Protection against PT challenge in immunized mice.

Protection against PT was investigated for the mice in the experiment shown in Fig. 2. The mice were challenged on day 42 by intraperitoneal injection of 2 μg of PT (Seikagaku Corporation) in 0.5 ml of phosphate-buffered saline. Three days after the challenge, mice were bled from the tail vein and white blood cells (WBCs) were counted with a Coultor counter (Beckman Coultor). As shown in Fig. 3, the average WBC count in the peripheral blood of mice immunized with pcDNA/C180 was significantly (P < 0.05) lower than that in the peripheral blood of mice immunized with pcDNA/S1 or pcDNA/C200. Before the challenge, the average WBC count for all groups of mice was 2.2 × 10⁴/μl (range, [2.0 to 2.5] × 10⁴/μl). After the challenge, the average numbers of WBCs in mice immunized with pcDNA, pcDNA/S1, pcDNA/C200, pcDNA/C180, pcDNA/C160, and pcDNA/N40 increased 3.4-, 1.8-, 1.4-, 0.9-, 3.1-, and 3.2-fold, respectively, compared with those before the challenge. The induction of a high level of antibody production and protection by pcDNA/C180 were confirmed in repeated experiments. The results obtained indicate that immunization with pcDNA/C180 induced highly protective immunity against PT challenge but that immunization with pcDNA/C160 and pcDNA/N40 did not. Since C160 and N40 polypeptides were not expressed in COS-7 cells after transfection with those plasmids (Fig. 1C), C160 and N40 polypeptides might be degraded in the immunized mice. The enzymatic domain of S1, associated with ADP-ribosyltransferase activity, is located in residues 2 to 179, and a recombinant fragment (residues 1 to 123) of S1 is known to lose enzymatic activity (16). Therefore, expression of the enzymatic domain may be essential in the induction of protective immunity with DNA-based immunization.

FIG. 3.

Protection against leukocytosis induced by PT in BALB/c mice immunized with plasmids encoding C and N terminally truncated S1. Forty-two days after primary immunization, mice were challenged intraperitoneally with 2 μg of PT. Three days later, mice were bled and the numbers of WBCs in their peripheral blood were counted. Bars indicate the means for five mice in each group. The average WBC counts for all groups of mice before the challenge with PT was 2.2 × 10⁴ cells/μl. An asterisk indicates a significant (P < 0.05) difference.

Figure 4 shows a correlation between anti-PT IgG titers (Fig. 2) and WBC counts (Fig. 3) in individual mice. In mice immunized with pcDNA/S1, the induction of protection against PT varied within the same group; two of five mice were protected completely against PT challenge, whereas three mice were not. The unprotected mice had low anti-PT IgG titers (0.8 to 10.4 U/ml), while the protected mice had high anti-PT IgG titers (41.7 and 46.9 U/ml). The poor antibody responses in pcDNA/S1-immunized mice were also observed in our previous study (9). The C180 and full-length S1 polypeptides were highly expressed in pcDNA/C180- and pcDNA/S1-transfected cells, whereas C200 polypeptide was expressed at a low level (Fig. 1C), suggesting a lack of correlation between protein expression in COS-7 cells and immune response induced in mice. The lack of correlation observed between protein expression and the induction of immune response is possibly due to the difference in the solubilities of these polypeptides in mammalian cells. The C terminally truncated form of recombinant S1 is known to drastically increase in solubility (1). The high solubility of the C180 polypeptide might result in the stable induction of antibody production in mice.

FIG. 4.

Correlation between anti-PT IgG titers (Fig. 2) and WBC counts (Fig. 3) in individual mice. Anti-PT IgG titers in individual mice on day 41 (Fig. 2) were plotted against WBC counts after PT challenge (Fig. 3).

In the present study, we have demonstrated that gene gun injection of pcDNA/C180 and pcDNA/C200 into mice successfully induced protection against PT even more efficiently than injection of pcDNA/S1. pcDNA/C180 in particular had the most potent ability to induce protective immunity. Our finding indicates that the C-terminal region of S1 (residues 181 to 235) was not essential in inducing protective immunity in DNA-based immunization. Moreover, the existence of the C-terminal region was found to reduce stable induction of protection. Previously, Barry et al. (3) demonstrated that a fusion protein consisting of the N-terminal 179-amino-acid fragment of S1 fused to fragment C of tetanus toxin induced antibody response following intranasal immunization with Salmonella enterica serovar Typhi strain CVD 908 live vector expressing the fusion but that fusion of full-length S1 did not. This observation supports the finding in this study that immunization with pcDNA/C180 resulted in more stable induction of protection against PT than immunization with pcDNA/S1.

In conclusion, we constructed four plasmids encoding C and N terminally truncated S1 subunits, and their ability to induce protection against PT in mice after DNA-based immunization was evaluated. The results obtained clearly demonstrate that the pcDNA/C180-expressing N-terminal 180-amino-acid fragment of S1 was the most effective in the induction of protection against PT. The results obtained in the present study may contribute to the antigen designation in the development of a novel pertussis DNA vaccine.