Abstract
A cell-surface protein antigen (PAc) of Streptococcus mutans is considered a virulence factor because it may mediate initial attachment of Streptococcus mutans to tooth surfaces. Thus, inhibiting PAc is predicted to provide protection against caries. To develop vaccines against dental caries, we constructed a DNA vaccine, pCIA-P, which encodes two high-conservative regions of PAc. Expression of the recombinant protein was obtained in eukaryotic cells in vitro and in vivo. In this report, we provide evidence that fewer caries lesions, and high levels of PAc-specific salivary IgA antibody and serum IgG antibody, were observed in gnotobiotic rats following targeted salivary gland (TSG) administration of pCIA-P. This study shows that the recombinant DNA vaccine pCIA-P could induce protective anti-caries immune responses and that TSG immunization is a promising strategy for the inhibition of dental caries.
INTRODUCTION
Cell-surface protein antigen (PAc, also known as antigen B, P1, and Antigen I/II) is a surface fibrillar protein of the cariogenic organism Streptococcus mutans (Okahashi et al., 1989a) and is a virulence factor contributing to the pathogenesis of S. mutans-induced dental caries because of its involvement in initial adherence of the organism to tooth surfaces (Hajishengallis et al., 1992). It has been shown that an alanine-rich region, A-region, and the proline-rich region, P-region, of PAc are of great importance to PAc antigenicity (Okahashi et al., 1989b; Brady et al., 1998). The research suggested that the DNA sequence containing the two regions might be a good candidate for a DNA vaccine against dental caries in humans.
DNA vaccine is a promising new vaccine developed recently. Immunization with an antigen-encoding plasmid has been applied to induce both humoral and cell-mediated immune responses against a growing number of infectious agents, including viruses, bacteria, and parasites (Donnelly et al., 1997). Recent reports about anti-HIV DNA vaccine studies suggest that this novel approach may elicit protective mucosal immunity in mucosal sites such as the gut, rectum, bronchus, and nasopharynx (Sasaki et al., 1998; Klavinskis et al., 1999). A DNA vaccine has some advantages, such as its long-term and stable expression of antigenic protein, the endogenously expressed protein having a conformation similar to that of a natural protein and an antigenicity stronger than that of traditional vaccines. In this study, we constructed a DNA vaccine carrying the A-P fragment of the pac gene and evaluated the gene expression in eukaryotic cells, the systemic and mucosal immune responses to this vaccine, and anti-caries protection in a gnotobiotic rat model.
MATERIALS & METHODS
Plasmid Construction
The A-P fragment, that included the A- and P-regions originating from amino acid residues 222 to 965 of the pac gene encoding the PAc protein of Streptococcus mutans MT8148, was amplified from plasmid pPC41 (Okahashi et al., 1989a) by a polymerase chain-reaction [PCR: Expand High Fidelity PCR System (Boehringer, Mannheim, Germany)]. The recombinant plasmid pCIA-P was verified by restriction digestion and by sequencing of the completely inserted DNA at the Shanghai Sangon company (Shanghai, China) (Fig. 1).
PAc Expression in Cultured Cells
The expression of the recombinant PAc protein was performed in a transient transfection assay with the use of (1,3-di-oleoyloxy-2-(6-carboxy-spermyl)-propyl-amid; Roche Molecular Biochemicals, Mannheim, Germany) according to the manufacturer’s instructions. Briefly, human endothelial cells ECV-304 (3 x 105 cells; purchased from the China Center for Type Culture Collection, CCTCC, Wuhan) were incubated with DNA-liposome complexes for 5 hrs and cultured overnight with complete M199 medium (Invitrogen, Carlsbad, CA, USA). Cells were collected, and the recombinant PAc expression in ECV-304 was performed with an immunohistochemical assay that used avidin-biotin Complex (Sino-American Biotechnology, Henan, China) according to the protocol of the manufacturer. The slides were incubated with anti-PAc IgG antibody (provided by Prof. M.W. Russell, University of Alabama at Birmingham, USA) at 37°C for 30 min and kept at 4°C overnight. Then, the slides were incubated with biotinylated goat anti-rabbit IgG and avidin-biotin peroxidase complex in turn. The slides were colored with DAB (dimethylaminoazobenzene, Boster Co., Wuhan, China) and counterstained with hematoxylin.
Expression in vivo by Immunohistochemistry in situ
Four groups of male Wistar rats (obtained from and maintained by Hubei Medical Laboratory Animal Center, Hubei, China; 18 days old, 9 per group; the animal use protocols had been reviewed and approved by the Review Board of Hubei Medical Laboratory Animal Center) were immunized with 100 μL of pCIA-P plasmid (1 μg/μL) as follows: injected into the quadriceps femoris muscle (Group I) or subcutaneously near the submandibular gland (TSG) (Group II). The immunizations were “boosted” 2 wks later. The control rats were immunized with 100 μL of pCI vector (1 μg/μL) into the quadriceps femoris muscle (Group III) or subcutaneously near the submandibular gland (Group IV). The samples of quadriceps femoris muscles and submandibular gland were collected on day 63, fixed in 4% formalin, embedded in paraffin, and sectioned. The sections were incubated with 3% (v/v) hydrogen peroxide, then incubated with rabbit anti-PAc serum for 2 hrs at 37°C. Next, a biotin-labeled secondary antibody was added and incubated for 20 min at 37°C. The sections were then incubated with streptavidin-biotin-peroxidase complex reagents from a SABC kit (Boster Co., Wuhan, China). The slides were reacted with DAB and counterstained with hematoxylin. Control slides had either no first antibody or no second antibody.
Immunization of Gnotobiotic Rats
Five groups of newborn male Wistar rats (6 per group) were bred and maintained in the same place. These rats were weaned at day 18 and raised on cariogenic diet Keyes 2000 (Navia, 1997). Antibiotics (ampicillin, chloramphenicol, and carbenicillin, 1.0 g/kg diet) were added to the diet on days 20-22, and the animals were then infected with S. mutans Ingbritt on days 24-26. Before and after infection, bacterial samples from occlusal sufaces were examined. Two days after being provided with antibiotics in the diet, the gnotobiotic rats were immunized with 100 μL of pCIA-P plasmid (1 μg/μL) as follows: injection into the quadriceps femoris muscle (Group V); subcutaneous injection near the submandibular gland (TSG) (Group VI) and injection into the buccal mucosa (Group VII). The immunizations were “boosted” 2 wks later. The control (Group VIII) and sham (Group IX) rats were infected via injection into the quadriceps femoris muscle with 100 μL of pCI vector (1 μg/μL) or 0.9% NaCl solution, respectively. On day 63, saliva samples were collected after stimulation of the salivary flow by intraperitoneal injection of 1 mg of pilocarpine (Sigma, St. Louis, MO, USA); blood samples were collected from tails, and the mandibles were removed from individual animals, cleaned, and stained with murexide. The teeth were sectioned and the caries level scored by the Keyes method (Keyes, 1958).
Antibody Analyses
For measurement of anti-PAc IgA and IgG in the saliva and sera of the rats, each well of an ELISA plate was coated with rPAc (10 μg/mL in carbonate buffer, pH 9.6, provided by Prof. M.W. Russell) overnight at 4°C and then blocked with phosphate-buffered saline (PBS) containing 3% bovine serum albumin (BSA). After being washed with phosphate-buffered saline containing 0.1% Tween, pH 9.6 (PBST), 100-μL quantities of diluted saliva or sera were added to each well and incubated for 1.5 hrs at 37°C. After being washed with PBST again, each well received 100 μL of goat anti-rat IgG or goat anti-rat IgA (1:1000; Sigma), incubated for 2 hrs at 37°C, and washed again. Next, a 100-μL quantity of alkaline-phosphatase-conjugated rabbit anti-goat IgG (1:10,000; Sigma) was added to each well and incubated for 5 hrs at 37°C, followed by phosphase substrate (ρ-nitrophenylphosphate) for 30 min at 37°C. Optical density (OD) readings were taken at 405 nm. The end-point titer was defined as the highest dilution with an absorbance ≥ 0.1 over the absorbance of the sham control.
Statistical Analysis
The differences in anti-PAc specific antibody titers and caries protection among the test groups, control groups, and sham groups were determined by analysis of variance.
RESULTS
Expression of Recombinant PAc Protein in Eukaryotic Cells
Many positively staining cells with a brown color in the cytoplasm were found in the pCIA-P transfected ECV-304 cells, but no such specific products could be found in the cells transfected with pCI vector. Analysis of the data indicates that the plasmid pCIA-P can express PAc protein in eukaryotic cells (Fig. 2).
Expression in situ of PAc Protein in Rats
Rats from Group I expressed the PAc protein in an uneven pattern in the sarcoplasm and sarcolemma of muscle fibers (Fig. 3). Group II expressed the PAc protein in both sides of the submandibular gland. It was noticed that PAc expression was diffused, especially in the duct region. Some PAc immunostaining appeared in the duct cavity, acinus, and mucus alveolus. In contrast, no immunoreactive protein was detected in muscles of the pCI- injected groups (Fig. 3).
Immune Response
Fig. 4 summarizes the data from the ELISA assays. The anti-PAc IgG ELISA end-point titers in serum samples of Group V and Group VI rats were significantly higher than those of Group VII, Group VIII, Group IX rats (p < 0.01), but there were no distinct differences between Group V and Group VI rats (p > 0.05). The anti-PAc IgA ELISA end-point titers in saliva samples of Group VI and Group VII rats were significantly higher than those of Group VII, Group VIII, and Group IX rats (p < 0.01). There was no significant difference between Group VI and Group VII (p > 0.05). The anti-PAc IgA levels in serum and anti-PAc IgG levels in saliva of gnotobiotic rats appeared to show no significant differences in any of the immunization groups (data not shown; p > 0.05).
Caries Protection
There were no notably different enamel lesions in all of the groups (p > 0.05), but Group VI and Group VII displayed significantly fewer dentinal slight (Ds) and dentinal moderate (Dm) lesions (p < 0.01) than Group V, Group VIII, and Group IX rats (Fig. 4). Group V rats showed fewer dentinal lesions compared with the controls and the sham group, particularly a significant decrease of Dm lesions (p < 0.05). In addition, no significance in dentinal lesions was found between Group VI and Group VII rats (p > 0.05) (Fig. 4).
DISCUSSION
The high-risk groups still exist, according to previous studies, especially in developing countries, although the prevalence of tooth decay has declined in some countries (Mandel, 1996). Caries immunization may be beneficial for those people. DNA-based immunization is a promising method against disease in humans, since DNA vaccines have advantages over traditional vaccines. Here we report a pioneer study about DNA vaccine against dental caries.
In this study, after rats were immunized with pCIA-P by muscle injection, we found that PAc protein was distributed in muscle fibers in an uneven pattern. This might be because only a few muscle cells took up pCIA-P, or different types of cells had uneven abilities to express foreign proteins (Davis et al., 1994). After rats had been immunized with pCIA-P by TSG, the PAc protein was found to be expressed in duct cells. This is the first report on recombinant PAc protein being expressed in submandibular gland duct cells after DNA immunization. How duct cells express a recombinant protein remains unknown. Kuklin et al. (1997) reported that β-Gal protein could be detected in alveolar epithelial cells and the bronchi as well as in the cervical lymph nodes, after intranasal immunization of mice with plasmid DNA encoding β-galactosidase.
Specific immune defense against cariogenic mutans streptococci is provided largely by salivary secretory IgA antibodies, which are generated by the mucosal immune system. In 1999, Kawabata reported the induction of Porphyromonas gingivalis fimbria-specific IgA and IgG in saliva and serum IgG by TSG (Kawabata et al., 1999). Our study also demonstrated that TSG with pCIA-P could induce the highest salivary anti-PAc IgA antibodies compared with immunization by injection into the muscle and injection into the buccal mucosa (Fig. 4). We found that TSG and the buccal mucosa immunization significantly decreased the dentinal caries levels of rats (Fig. 4). Although Group V rats immunized with pCIA-P in the muscle also had a decrease in dentinal caries compared with the control and sham-treated animals, development of Dm lesions in Group V was higher than that in the TSG and buccal mucosa immunization group. These results demonstrated the important anti-caries role of secretory IgA (Russell et al., 1999). When we inoculated rats subcutaneously near one submandibular gland with pCIA-P, not only was the PAc protein highly expressed, but also the expression of the PAc protein could be observed in duct cells of both submandibular glands. The mucosal immune system can be functionally divided into two sites: inductive and effector. There are many dendritic cells and macrophages in skin tissues near submandibular gland and buccal mucosal tissues. When dendritic cells and macrophages are activated by antigens, they home to distant effector sites, such as the salivary gland, to synthesize and secrete secretory IgA (Torii et al., 1981; Mestecky and McGhee, 1987). Thus, our results may suggest that pCIA-P plasmids might transfect some antigen-presenting cells (APC) such as dendritic cells and macrophages, and then, those cells bring endogenously synthesized antigen to the submandibular gland. How antigen is expressed in the gland is unknown.
Our results also showed that rats inoculated with pCIA-P by different routes displayed numbers of enamel lesions similar to those displayed by the rats of control and sham groups. It seemed that the immune response induced by the anticaries DNA vaccine pCIA-P could not protect rats from enamel caries caused by the strong cariogenic virulence of Streptococcus mutans Ingbritt and a great quantity of sucrose in the diet.
In conclusion, we have constructed the anti-caries DNA vaccine pCIA-P, which expresses the A- and P-regions of the S. mutans PAc protein. We demonstrate that pCIA-P could express recombinant PAc protein in eukaryotic cells and provoke specific immune responses as a novel immunogen. These results suggest that TSG immunization with plasmid DNA might represent a promising genetic immunization strategy against caries.
Construction and verification of plasmid pCIA-P. Recombinant PAc protein could be expressed in eukaryotic cells.
Expression of PAc protein in rats’ tissues after being immunized with pCIA-P by
different routes. The antibody level and caries scores of gnotobiotic rats (6
per group) immunized with various treatments. Data are
expressed as means and standard deviations of antibody level and caries scores
experiments. 
Footnotes
Acknowledgements
The authors thank Prof. Y. Nakano for providing the pPC41 plasmid carrying the pac gene, and Prof. M.W. Russell for providing the PAc protein and anti-PAc serum. We also thank Prof. Y. Abiko, Prof. Z.M. Zheng, and Prof. C. Luisa for editing the English of the manuscript. This study was supported by a grant (No. 39770799) from the Natural Sciences Foundation of China.