INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Tetanus continues to be a threat to public health with more than half a million fatalities each year being associated with infection with Clostridium tetani (23). Due to the ubiquitous distribution of the causal agent, vaccination is the most effective medical intervention for protection of the public against this deadly disease. The effectiveness of the vaccine is due, at least in part, to the fact that the sequences of the neurotoxin molecules are conserved among different strains of C. tetani, which permits elicitation of a protective immune response against all C. tetani strains through the use of a single vaccine (23). The effectiveness of the vaccine is limited, however, by the needle-based delivery method currently in use. Effective protection requires injection of three consecutive doses of the tetanus toxoid vaccine (16). Moreover, booster injections must be administered periodically during adulthood to compensate for the age-related decline in antitoxin levels (16). In developing countries, vaccine coverage against this disease is generally low due to failure to follow up as well as a lack of the trained medical personnel and facilities required for administration of the vaccine. In developed countries, although vaccine coverage in childhood is high, there is a general lack of compliance of adults with recommended schedules for booster injections (16). These factors have led to the realization that tetanus vaccination programs would be improved significantly worldwide through the development of low-cost, needle-free vaccines.

Needle-free vaccination requires the development of novel vaccines that can be administered safely and effectively. Several routes of administration are being considered. Both nasal and oral immunizations have been shown to be effective in eliciting an immune response against a number of pathogens. An alternative new modality is noninvasive vaccination onto the skin (NIVS) by topical application of epicutaneous vaccines (8, 10, 11, 17, 18, 22), which would offer a greater safety margin and eliminate the discomfort associated with injections. Prior to our studies, it had not been demonstrated that topical application of epicutaneous vaccines could protect recipients against live pathogenic bacteria in a disease setting. This study was undertaken to determine if administration of a vaccine consisting of an adenovirus (Ad) recombinant (AdCMV-tetC) encoding the immunogenic but atoxic tetanus toxin C fragment (TetC) (16), when inoculated either intranasally or by an epicutaneous patch, can protect animals against a lethal challenge of live C. tetani. We report that administration of a single dose of this vaccine intranasally was 100% protective against intramuscular injection of a lethal amount of C. tetani cells and that administration of a single dose topically was protective in 80% of the mice with the protection rising to 100% when two booster applications were administered consecutively using a patch.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Construction of expression vectors. The TetC fragment was amplified by PCR from plasmid pTET-nir (2) (provided by J. VanCott and J. McGhee, University of Alabama at Birmingham) using the primers 5tetC (5'CGCGGATCCACCATGGAAAATCTGGATTGTTGGGTTG3') and 3tetC (5'CGCGGATCCATCATTTGTCCATCCTTCATC3'). These DNA primers created unique BamHI sites at the ends of the 1.35-kb TetC fragment with a synthetic eukaryotic ribosomal binding site (13) inserted in frame. The PCR-amplified fragment was subsequently cloned into the BamHI site of the pACCMV.PLPA shuttle plasmid (12) under the transcriptional control of the human cytomegalovirus (CMV) early promoter to create the plasmid pAC-tetC. A replication-incompetent E1/E3-defective human Ad serotype 5-derived vector encoding TetC was constructed by homologous recombination between cotransfected pAC-tetC and pJM17 in human 293 cells as described (12). High-titer Ad stocks were prepared by ultracentrifugation over a cesium chloride gradient as described (20). A plasmid expression vector encoding TetC (pCMV-tetC) was constructed by cloning the same TetC fragment into the BamHI site of the plasmid pVR1012 (provided by Vical, Inc., San Diego, Calif.) in the correct orientation under transcriptional control of the human CMV early promoter. The AdCMV-PR8.ha vector encoding the influenza A/PR/8/34 hemagglutinin (HA) was constructed as a control by excising the 1.8-kb BamHI fragment containing the entire coding sequence of HA from the plasmid pDP122B (American Type Culture Collection [ATCC], Manassas, Va.), followed by cloning the HA gene into the BamHI site of pACCMV.PLPA and subsequent homologous recombination with pJM17 in human 293 cells.

Immunization of mice. Young (2 to 3 months old) female BALB/c mice (Jackson Laboratory, Bar Harbor, Maine) were immunized by intranasal inoculation, topical application, intradermal injection, and intramuscular injection of vaccines (Ad-vectored or DNA-based).

Quantitative PCR. Two PCR primer sets were designed to selectively amplify subfragments of the Ad serotype 5 fiber gene. The primers Fb5.3 (5'GCATTGACTTGAAAGAGCCC3') and Fb3.3 (5'AGGAACCATAGCCTTGTTTG) encompass a 560-bp fragment of the fiber gene, whereas the primers Fb5.4 (5'GTAGCAGGAGGACTAAGGAT3') and Fb3.4 (5'TATCCAAGTTGTGGGCTGAG3') encompass a 139-bp subfragment within the 560-bp fragment. The sensitivity of a nested PCR procedure was determined as follows. A PCR mix containing 1 µg of genomic DNA extracted from the abdominal skin of a naïve mouse was spiked with dilutions of purified AdCMV-tetC vector DNA ranging from 1 to 10,000 copies, followed by 40 cycles of amplification with the Fb5.3/Fb3.3 primer pair. After purifying the PCR products with the QIAquick PCR purification kit (Qiagen, Inc., Valencia, Calif.), DNA was subjected to another 40 cycles of amplification with the Fb5.4/Fb3.4 primer pair. Naïve mouse skin DNA without the addition of Ad DNA templates was amplified with identical primer sets as a negative control. Following this amplification protocol, we consistently detected 10 or more copies 100% of the time in several independent experiments. It was therefore concluded that the sensitivity of the assay was 10 copies or less.

ELISA. Titers of anti-TetC immunoglobulin G (IgG) were determined by enzyme-linked immunosorbent assay (ELISA) as described (18) using purified TetC protein (CalBiochem, San Diego, Calif.) as the capture antigen. Briefly, serum samples and peroxidase-conjugated goat anti-mouse IgG (Promega Corp., Madison, Wis.) were incubated sequentially on the plates with extensive washing between incubations. The endpoint was calculated as the dilution of serum producing the same optical density at 490 nm as a 1/100 dilution of preimmune serum. Sera negative at the lowest dilution tested were assigned endpoint titers of 1.

Challenge by live C. tetani. Mice were challenged by footpad injection of 6 × 10³ (50% lethal dose [LD₅₀], 12.5) or 6 × 10⁴ (LD₅₀, 125) C. tetani cells in a volume of 50 µl. Distortion of the backbone due to paralysis was taken as the disease end point. The LD₅₀ was defined as the number of C. tetani cells capable of distorting the backbone of 50% of the mice in a group of 10 animals within a week following footpad injection. A toxigenic strain of C. tetani (ATCC number 9441; ATCC) was cultivated in the ATCC 38 beef liver medium for anaerobes at 37°C under anaerobic gas mixture (80% N₂-10% CO₂-10% H₂). Gram-stained cells were counted under a light microscope using a hemacytometer. Animals either succumbed to the disease or were euthanatized when backbone distortion occurred.

Anti-Ad neutralizing antibody assay. Anti-Ad neutralizing antibodies were measured as cytopathic effect (CPE)-inhibiting antibodies by incubating diluted serum samples with 100 PFU of Ad at 4°C for 1 h, followed by adding the neutralized particles to 10⁵ human 293 cells in a tissue culture well. CPEs were scored from triplicate samples daily by examining the monolayers under a light microscope. The assay was terminated when viruses mixed with a 1/10 dilution of preimmune serum produced 100% CPE (4 days postinfection in this experiment). Sera capable of completely inhibiting CPE (defined as the disruption of confluent monolayers) at the highest dilution tested were assigned endpoint CPE inhibition (CPEI) titers of 1. This assay tests the ability of specific antibodies to prevent Ad from infecting human 293 cells and measures a subset of neutralizing antibodies.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Elicitation of an anti-TetC humoral immune response. The AdCMV-tetC vector was used in these studies to permit administration of a vectored tetanus vaccine in a needle-free manner. The needle-free routes of administration included topical and intranasal inoculations. A DNA-based vaccine (pCMV-tetC) also was administered by these routes as well as by intramuscular injection as another control. We report here that the efficiency of vector absorption into the skin was very low, with approximately 1 in 4,000 particles being taken within 1 h of the topical application (Fig. 1A). Most of the vectors applied to the surface failed to bind to the skin and were subsequently washed away. No adverse effects associated with inoculation were observed in any of the immunized mice during the course of these studies. Absorption of AdCMV-tetC particles following topical application was more effective in eliciting an anti-TetC antibody response than intradermal or intramuscular injection of the same vector at equivalent doses, as determined by seroconversion, which was monitored by analysis of sera from tail bleeds for the production of IgG antibody directed against TetC by ELISA (Fig. 1B).

View larger version (23K): [in this window] [in a new window]   FIG. 1.   Outer layer of skin as immunocompetent tissue with low capacity to absorb vectors from the environment. (A) Determination of the number (#) of vectors absorbed by the skin following topical application. One and 24 h after topical application of 1010 particles of the AdCMV-tetC vector using a patch, the mouse abdominal skin underneath the patch was washed, blotted dry, resected, and immediately homogenized in the DNAZOL solution for DNA extraction. The number of skin-associated vectors per animal was determined quantitatively by amplifying subfragments of the adenoviral fiber gene with the nested PCR procedure in conjunction with limited dilutions, under the assumption that each vector contains a single fiber gene. Three individual animals were analyzed at each time point. The data shown are means ± standard deviations. (B) Comparison of different modes for vaccine administration. AdCMV-tetC particles of various doses were inoculated into mice by topical application and intradermal and intramuscular injection. Animals were immunized only once. Sera were harvested for ELISA-based anti-TetC analysis 3 and 6 weeks postimmunization. NIVS-1, mice immunized by topical application of 1010 Ad particles with approximately 106 particles absorbed by the skin, as determined by the method described in panel A; ID-1, mice immunized by intradermal injection of 106 Ad particles; ID-2, mice immunized by intradermal injection of 108 Ad particles; ID-3, mice immunized by intradermal injection of 1010 Ad particles; IM-1, mice immunized by intramuscular injection of 106 Ad particles; IM-2, mice immunized by intramuscular injection of 108 Ad particles; IM-3, mice immunized by intramuscular injection of 1010 Ad particles. The data shown are GMTs for anti-TetC antibodies (5 animals per group for ID-1 and IM-3; 10 animals per group for NIVS-1, ID-2, ID-3, IM-1, and IM-2). No anti-TetC antibodies were detectable in some groups (both 3 and 6 weeks postimmunization in ID-1; 3 weeks postimmunization in IM-1), and low columns representing negative responses were included for visualization.

Protection against tetanus following live bacterial challenge. Seroconversion does not necessarily confer protection. Therefore, the protective effects of the vaccination regimens were determined by challenge with a toxigenic strain of C. tetani injected in the footpad at a dose of 6 × 10³ cells (LD₅₀, 12.5). On such challenge, unprotected mice inevitably succumbed to tetanus with paralysis of the inoculated paw followed by distortion of the backbone and death. Administration of AdCMV-tetC as a nasal vaccine at a dose of 10⁷ particles conferred complete protection in all animals after a single inoculation or three consecutive inoculations (P < 0.001) (Table 1; Fig. 2A and 3A). Full protection was also achieved after three consecutive applications of 10¹⁰ particles of AdCMV-tetC as an epicutaneous vaccine (P < 0.001), with 80% of the mice being protected (P < 0.001) by a single application (Table 1; Fig. 2B and 3B). All naïve animals that were not immunized (group X), as well as mice immunized with an irrelevant Ad vector (i.e., AdCMV-PR8.ha in groups O and S), were paralyzed severely or succumbed to tetanus within 4 days (Table 1; Fig. 3). Some mice that survived the challenge have been kept alive for up to 4 months and no symptoms of tetanus developed in any of the animals after day 10. Analysis of the anti-TetC IgG levels in the mice indicated that 100% survival was correlated with an anti-TetC titer of >6,400, although some animals with a lower titer could still survive the challenge.

View larger version (37K): [in this window] [in a new window]   FIG. 2.   Dose-response protection against tetanus after bacterial challenge. Animals were immunized by inoculation with Ad-vectored vaccines at escalating doses. Five weeks postimmunization, mice were challenged by footpad injection of a lethal dose (6 × 103) of live C. tetani cells and monitored daily for survival for 14 days. (A) Dose-response protection against tetanus by a single intranasal inoculation of the AdCMV-tetC vector at an escalating dose. Groups B to E, mice were immunized by intranasal inoculation of the AdCMV-tetC vector at an escalating dose (106 to 109 particles; the specific dose for each group is shown in Table 1); group A, mice were immunized by intranasal inoculation of 100 µg of pCMV-tetC DNA. (B) Dose-response protection against tetanus by a single topical application of the AdCMV-tetC vector at an escalating dose. Groups G to J, mice were immunized by topical application of the AdCMV-tetC vector at an escalating dose (107 to 1010 particles; the specific dose for each group is shown in Table 1); group F, mice were immunized by topical application of 100 µg of pCMV-tetC DNA. The data were plotted as percent survival versus number of days after challenge. Numbers in parentheses represent the number of animals for each treatment.

View larger version (29K): [in this window] [in a new window]   FIG. 3.   Effects of booster applications in protecting mice against tetanus after bacterial challenge. Mice were immunized either once or three times at intervals of 3 weeks. Eight weeks after the primary immunization, mice were challenged by footpad injection of 6 × 103 C. tetani cells and monitored daily for survival for 14 days. (A) Mice were immunized by intranasal inoculation with Ad vectors. Group O, intranasal inoculation with AdCMV-PR8.ha three times; group P, intranasal inoculation with AdCMV-tetC once; group Q, intranasal inoculation with AdCMV-tetC three times. (B) Mice were immunized by topical application of Ad vectors. Group S, topical application of AdCMV-PR8.ha three times; group T, topical application of AdCMV-tetC once; group U, topical application of AdCMV-tetC three times; group X, naïve control mice. The data were plotted as described in the legend to Fig. 2. Numbers in parentheses represent the number of animals for each treatment.

View larger version (23K): [in this window] [in a new window]   FIG. 4.   Protective capacity of Ad-vectored vaccines. Five weeks postimmunization, mice were challenged by footpad injection of either 6 × 103 or 6 × 104 C. tetani cells and monitored daily for survival for 14 days. Groups E and M, mice immunized by a single intranasal inoculation with AdCMV-tetC vectors were challenged with 6 × 103 and 6 × 104 C. tetani cells, respectively; groups J and N, mice immunized by a single topical application of AdCMV-tetC vectors were challenged with 6 × 103 and 6 × 104 C. tetani cells, respectively; group L, naïve mice were challenged with 6 × 104 C. tetani cells. The data were plotted as described in the legend to Fig. 2. Numbers in parentheses represent the number of animals for each treatment.

Protection in the context of preexisting immunity to Ad. The possibility that preexisting immunity to Ad could interfere with the ability of an Ad-vectored vaccine to elicit a protective immune response was considered. We investigated this possibility by inoculating the mice intranasally with Ad5 2 weeks prior to the primary administration of the vaccine. Anti-Ad neutralization antibodies were subsequently detected in all mice with exposure to Ad5 (CPEI GMT = 171 for group R and CPEI GMT = 197 for group V) (Table 1).

View larger version (29K): [in this window] [in a new window]   FIG. 5.   Protection in the context of preexisting immunity to Ad. Two groups of mice were inoculated intranasally with 109 particles of Ad5 2 weeks prior to the primary immunization. Eight weeks after immunization with the AdCMV-tetC vector, animals were challenged by footpad injection of 6 × 103 C. tetani cells and monitored daily for survival for 14 days. (A) Protection by intranasal inoculation. Groups R and Q, mice immunized by intranasal inoculation with the AdCMV-tetC vector with and without preexposure to Ad5, respectively. (B) Protection by topical application of a patch. Groups V and U, mice immunized by topical application of the AdCMV-tetC vector with and without preexposure to Ad5, respectively. Groups O, S, and X are control groups described in the legend to Fig. 3. The data were plotted as described in the legend to Fig. 2. Numbers in parentheses represent the number of animals for each treatment.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

A new generation of vaccines is required if the effectiveness of the global vaccine programs is to be enhanced. This demonstration that intranasal inoculation or topical application of an Ad-vectored vaccine results in statistically and preclinically significant protection against tetanus in mice ushers in two new modalities for the needle-free administration of tetanus vaccines. The vaccine used in these studies was a recombinant Ad vector encoding the atoxic C fragment of the tetanus toxin. The use of Ad-vectored vaccines is an attractive strategy due to its efficacy in eliciting an immune response. Another positive aspect of this strategy is the possible negation of the current requirement for an unbroken cold chain, since lyophilized Ad vectors can be reconstituted into active infectious particles (5).

Although all naïve mice and those immunized with irrelevant control vectors were paralyzed within 4 days of challenge, the partially protected animals exhibited symptoms of tetanus at a later date. Death occurred in some partially protected animals as late as 10 days postchallenge (Fig. 5B). This suggests that there is a critical initial time period during infection with C. tetani in which the host immune system is undergoing mobilization prior to achieving a response sufficient to eradicate the pathogen. Although the effectiveness of a tetanus vaccine may depend on its ability to rapidly induce adequate levels of antitoxin, it remains to be seen whether an anti-TetC cellular immune response (16) is involved in the eradication of live C. tetani cells in vivo.

Ad-vectored nasal vaccine elicited high titers of anti-TetC antibodies after a single administration, and these levels were not increased by short-term booster inoculations (groups P and Q). Moreover, as reported previously for rabies (27), the high titers were not suppressed appreciably by preexposure to Ad (groups Q and R). The potency of the intranasal inoculation may reflect the natural tropism of the Ad vector for the airway, where it can transduce a large number of cells and possibly activate the mucosal immune machinery to its full potential.

Although administration of the Ad-vectored vaccine through an epicutaneous patch was capable of causing seroconversion and was effective in protecting the animals from challenge with live C. tetani particularly after boosting, the titers of anti-TetC antibodies were relatively low when compared to those elicited by intranasal inoculations. Several strategies for optimization of the response can be envisioned. As boosting was effective in these studies, the immune response could be enhanced by applying the vaccine patch multiple times or over a large area. The efficacy of epicutaneous administration also may be enhanced by increasing the affinity of the Ad vector for specific cell types within the outer layer of skin (e.g., terminally differentiated keratinocytes, antigen-presenting cells that traffic to the surface, or other cell types along the skin barrier), which may be achieved by inserting a preselected ligand into the adenoviral fiber as previously described (6, 25) in an attempt to augment transduction efficiency in a targeted manner. Such a "skin-binding" vector could be a more effective carrier for epicutaneous vaccines than the current Ad vector due to the overexpression of antigens in specific cell types that are potent immunostimulators.

Emerging evidence suggests that the outer layer of skin may be more immunocompetent than deep tissues (7, 9) (Fig. 1B). It is also conceivable that the immune system may focus its surveillance on an interface that is in frequent contact with environmental pathogens. Overexpression of immunogens from a small number of vectored vaccines in the outer layer of skin may thus elicit a more potent immune response than the inoculation of an equivalent dose into deep tissues. This is borne out by the magnitude of the immune response that was elicited (Table 1 and Fig. 1B) by a relatively low number of vectors absorbed following topical application (Fig. 1A). However, it is unclear whether the leveling off of vaccine potency following topical application of AdCMV-tetC at a dose of 10⁹ particles (groups G to J) (Table 1) was due to saturation of the antigen expression machinery or the antigen presenting capacity within a restricted subset of skin. A skin-targeted vector may amplify antigen expression in the outer layer of skin; however, strategies to mobilize antigen-presenting cells will have to be exploited if antigen presentation should appear to be the limiting step.

The suppression of the efficacy of the Ad-vectored epicutaneous vaccine by preexisting immunity to Ad (Table 1 and Fig. 5) is a problem that must be circumvented during the development of an Ad-vectored vaccine patch. It is likely that the suppression may not be attributed to anti-Ad antibodies that prevent Ad vectors from entering target cells, because a reporter gene (i.e., luciferase) can be effectively expressed in skin from an Ad vector following topical application in mice with preexposure to Ad (data not shown). It is also conceivable that anti-Ad antibodies may not be able to reach the surface of skin in sufficient quantities to neutralize concentrated vectors. A skin-targeted vector may lessen the impact of preexisting immunity to Ad by overexpressing antigens. Alternatively, the capacity for a gutless Ad (3) to mediate long-term transgene expression in immunocompetent animals also may allow an epicutaneous vaccine to elicit a potent immune response in the presence of preexisting anti-Ad immunity if immunosuppression is attributed to rapid elimination of transduced cells by an anti-Ad cellular immune response.

The rapid loss of vector DNA observed in these studies after topical application (Fig. 1A) may be beneficial in preventing the persistence of exogenous DNA. Presumably, the mechanisms that have evolved to degrade or expel skin-associated environmental DNA in order to protect the genomic integrity of the host may contribute to the elimination of the vector DNA following topical application. In contrast to the short half-life of vectored epicutaneous vaccines, DNA-based vaccines have been reported to persist in animals for up to a year after intramuscular injections (26). Conceivably, the degradation of vectored epicutaneous vaccines may contribute to not only a safer vaccine but also a more effective one, as the transient expression of antigens in vivo could foster the longevity of memory T cells by minimizing antigen-induced apoptosis of T lymphocytes (19, 29). If efficacy should be limited by overdegradation, the problem potentially can be circumvented by booster applications as shown in this report (Table 1 and Fig. 3B).

It has been demonstrated previously that topical application of TetC protein in conjunction with cholera toxin is capable of eliciting anti-TetC antibodies in mice (10). It is unclear, however, whether a protein-based epicutaneous vaccine is able to protect animals against bacterial infections. The profile of the immune response elicited by vectored vaccines may be quite different from that induced by their protein-based counterparts since the vector DNA dictates the synthesis of exogenous proteins in animals' own cells after inoculation, and antigens produced in situ can often induce a more solid immunity than inoculation of protein-based vaccines (14). Evidence suggests that an Ad vector encoding TetC could be more effective than the TetC protein as an epicutaneous vaccine since the latter required the coadministration of cholera toxin as an adjuvant (10), whereas topical application of the AdCMV-tetC vector alone was able to protect all of the recipients against challenge with live C. tetani (Fig. 3B). Although topical application of plasmid DNA also has been shown capable of eliciting an antibody response (1, 8, 18), it has not been demonstrated that this modality can evoke any protective immunity. Topical application of naked plasmid DNA, in the absence of an association with Ad or liposome, produced no detectable gene expression in the skin (4, 18), and we report here that naked plasmid DNA was ineffective in eliciting an anti-TetC antibody response or a protective immune response against tetanus following either topical or intranasal inoculations (Table 1 and Fig. 2). The potency of an Ad-vectored vaccine may be attributed to an efficient gene delivery in conjunction with robust transgene expression when compared to its plasmid counterpart (21).

In these studies, intranasal inoculation was more effective than topical application with respect to eliciting an immune response, although this difference may be attributed, in part, to the low absorption efficiency of the skin (Fig. 1A). There are, however, some issues that detract from the practicality of intranasal administration of vaccines. A major concern is the possibility of the elicitation of a severe adverse pulmonary reaction in those individuals who are allergic to components of the vaccine or in individuals with underlying pulmonary disease. Moreover, intranasal administration can irritate the nose, causing responses that result in expulsion of an unpredictable amount of the vaccine.

The recent report that a humoral immune response could be elicited in humans by topical application of a vaccine patch (11) suggests that the outer layer of human skin may be as immunocompetent as that of their animal counterparts. Although an experimental murine tumor could be arrested following topical application of tumor epitope peptides (17), it has not been demonstrated, prior to our studies, that a pathogen of human relevance could be arrested following topical application of any vaccines. We have shown, for the first time, that topical application of a vectored vaccine patch in a noninvasive manner could protect animals against infection by live bacteria in a disease setting.

By expressing antigens in an immunocompetent area without causing tissue damage, nasal and epicutaneous vaccines may emerge as preferred modalities for the inoculation of future vaccines in a simple, effective, economical, painless, and safe manner.