α-Galactosylceramide (α-GalCer) is a ligand of invariant Vα14+ NKT cells and is presented by CD1d molecule on APC. NKT cells produce a large amount of Th1 and Th2 cytokines in response to α-GalCer-presented APC. In this study, we assessed whether α-GalCer could act as an effective nasal vaccine adjuvant for mucosal vaccine that would be capable of inducing systemic as well as mucosal immune responses. When α-GalCer was administered with OVA via the intranasal route to C57BL/6 and BALB/c mice, significant OVA-specific mucosal secretory IgA, systemic IgG, and CTL responses were induced with mixed Th1 and Th2 cytokine profiles seen in both strains of mice. Interestingly, as BALB/c mice were intranasally immunized with PR8 hemagglutinin Ag isolated from influenza virus A/PR/8/34 together with α-GalCer, significant protection was afforded against influenza viral infection. When α-GalCer was coimmunized with a replication-deficient live adenovirus to BALB/c mice, it significantly induced both humoral and cellular immune responses. In addition, intranasal administration of OVA with α-GalCer showed complete protection against EG7 tumor challenge in C57BL/6. The adjuvant effects induced by intranasal coadministration with α-GalCer were blocked in CD1d−/− mice, indicating that the immune responses were exclusively mediated by CD1d molecule on APC. Most interestingly, intranasally coadministered α-GalCer activated naive T cells and triggered them to differentiate into functional effector T cells when CFSE-labeled OT-1 cells were adoptively transferred into syngeneic mice. Overall, our results are the first to show that α-GalCer can act as a nasal vaccine adjuvant inducing protective immune responses against viral infections and tumors.
As the main entry site of most environmental Ags, mucosal surfaces, such as those of the respiratory, gastrointestinal, and genital tract, act as the first line of defense against pathogenic Ags. Not surprisingly then, many studies have focused on developing mucosal vaccines capable of effectively inducing both mucosal and systemic immune responses (1, 2, 3, 4). However, mucosal vaccines comprised solely of protein Ags induce only a weak immune response or tolerance. Therefore, a new mucosal vaccine adjuvant must be identified to develop an effective mucosal vaccine (5).
α-Galactosylceramide (α-GalCer),3 a glycolipid originally extracted from marine sponges, is a ligand of the invariant Vα14+ NKT cell and is presented by CD1d molecule on APC (6). Upon activation, NKT cells produce a large amount of both IFN-γ and IL-4, which can modulate the immune responses to certain diseases, especially to autoimmune diseases, or infections (7, 8, 9, 10, 11, 12, 13). Previous studies have reported that activated NKT cells direct immune responses toward Th2 immune responses (14, 15, 16), whereas other studies have shown that activated NKT cells skew immune responses toward Th1 immune responses (17, 18). Recently, it has been shown that the coadministration of α-GalCer with OVA can induce full maturation of dendritic cells (DCs), thereby generating functional Ag-specific Th1 CD4+ T cells and CTLs that are resistant to OVA-expressing tumors (19, 20). In addition, our previous study showed that the administration of α-GalCer triggers the full maturation of mesenteric DCs in vivo, which in turn contributes to the division of T cells in vitro and blocks the tolerance induced by both high and low doses of orally administered OVA (21). These results strongly suggest that α-GalCer could act as an effective and versatile mucosal adjuvant, one that would be capable of biasing the immune response into Th1 and CTL or Th2 immune response.
Several previous studies assessed the potential of α-GalCer as a systemically delivered vaccine adjuvant. Indeed, α-GalCer has been shown to act as an effective adjuvant against infections (22) and tumors (17, 19). Of the available mucosal vaccination routes, the intranasal one is preferred, although adjuvants such as cholera toxin and mutant cholera toxins are known to redirect vaccine protein into olfactory tissues when delivered intranasally (23). Further, the intranasal route requires much lower doses of both adjuvant and coadministered vaccine protein than does oral immunization (24). In this study, α-GalCer was evaluated as a nasal vaccine adjuvant for the induction of humoral as well as cellular immune responses against coadministered Ag in both C57BL/6 and BALB/c mice. We have proven in this study for the first time that α-GalCer is a potent nasal vaccine adjuvant capable of inducing both humoral and cellular immune responses and of affording protection in antiviral and antitumor nasal vaccines.
Materials and Methods
Mice and immunization protocols
Female C57BL/6 and BALB/c mice age 6–8 wk, purchased from the Charles River Breeding Laboratories, as well as CD1d−/− mice were used in this experiment. All mice were housed in the experimental animal facility of the College of Pharmacy, Seoul National University (Seoul, Korea) until use and kept in specific pathogen-free conditions during the entirety of the experiment. Mice were anesthetized by i.p. injection of ketamine (100 mg/kg) and xylazine (10 mg/kg), and immunized intranasally on days 0, 7, and 14 with 20 μl (10 μl/nostril) of PBS-diluted vehicle solution containing 100 μg of OVA alone or together with α-GalCer, unless otherwise noted. In the preliminary experiment, we have examined optimal frequency of nasal immunization (a single, two, or three times) and the result revealed that three times immunization showed sufficient immune responses by α-GalCer nasal vaccine adjuvant (data not shown). In the influenza virus studies, BALB/c mice were used. PR8 hemagglutinin (HA) vaccines (split-product virus vaccines), kindly provided by Dr. S.-I. Tamura (Osaka University, Osaka, Japan), were prepared from influenza virus A/PR/8/34 (PR8, H1N1) by the method of Davenport et al. (25). In the adenovirus studies, BALB/c mice were intranasally immunized with 106 PFU of replication-deficient adenovirus harboring β-galactosidase gene (Ad-LacZ) or together with 0.125 μg of α-GalCer two times at 2-wk interval. Ad-LacZ is the type 5 adenoviral vector, which is replication defective with the El deletion and was kindly donated by ViroMed (26). For the examination of the activation and differentiation of OVA-specific CD8+ T cells into effector cells, 1 × 107 cells of CFSE-labeled OT-1 cells were i.v. transferred into syngeneic mice. One day later, OVA alone or together with 2.0 μg of α-GalCer was administered intranasally. All experiments were done in compliance with relevant laws and institutional guidelines.
Synthesis of α-GalCer
α-GalCer was prepared by coupling of phytosphingosine with hexacosanoic acid followed by the protection/deprotection and galactosylation scheme as described (27) and kindly provided from Dr. S. Kim (Seoul National University, Seoul, Korea). α-GalCer was dissolved in 0.5% Tween 20 in PBS, which is used as a vehicle in all experiments.
One week after the final immunization, serum, nasal wash, and lung wash were collected to assess OVA- and β-galactosidase-specific Ab responses. Nasal wash specimens were collected by gently flushing the nasal passages with 100 μl of sterile PBS (28). To obtain lung wash, bronchoalveolar lavage fluids were taken as previously described (29).
ELISA for Abs
OVA-specific IgG was determined as previously described (29). To assess HA-specific and β-galactosidase-specific Ab responses, ELISA plates with 96-wells were coated with 100 ng of HA or β-galactosidase, respectively. For the assessment of β-galactosidase-specific Ab responses, nasal wash and lung wash were examined at dilutions of 1/5, 1/15, and 1/45, and sera were examined at dilutions of 1/300, 1/900, and 1/2700. For the measurement of IgA, IgG1, and IgG2a titers in other samples, 2-fold serially diluted samples were used and the titers were determined. For the determination of IgA titers, HRP-conjugated goat anti-mouse IgA and the peroxidase substrate tetramethylbenzidine were used. Next, 0.5 N HCl was added to stop the development, and the OD was detected at 450 nm. To determine IgG, IgG1, and IgG2a titers, we used alkaline phosphatase-conjugated goat anti-mouse IgG, IgG1, and IgG2a and the phosphatase substrate, p-nitrophenyl phosphate. Goat anti-mouse IgA, IgG1, and IgG2a were purchased from Southern Biotechnology Associates.
ELISA for cytokine
For the examination of cytokine production, single cells from spleen and lymph nodes were collected by passing through 70-μm nylon mesh and RBC were lysed with ammonium chloride. Cells were cultured at 5 × 106 cells/ml with or without 500 μg/ml OVA for 4 days. Culture supernatants were harvested and the concentrations of IL-4 and IFN-γ were measured by using mouse IL-4 and IFN-γ OptEIA set ELISA kits (BD Pharmingen) according to the manufacturer’s instructions.
Intracellular cytokine staining
Intracellular staining was performed with BD Cytofix/Cytoperm Plus (with GolgiPlug; BD Pharmingen) according to the manufacturer’s instructions. For the measurement of IFN-γ producing CD8+ T cells in BALB/c mice, 2 × 106 cells/ml cells were incubated with 500 μg/ml OVA for 4 days and 1 μl/ml GolgiPlug was added during the last 6 h. For the measurement of epitope-specific cellular immune response in C57BL/6 mice and cytokine producing CD8+ T cells in OT-1 transferred C57BL/6 mice, 2 × 106 cells/ml cells were incubated with 5 μM OVA257–264 peptide and 1 μl/ml GolgiPlug for 6 h. For the measurement of β-galactosidase-specific IFN-γ-producing CD8+ T cells in adenovirus study, 2 × 106 cells/ml spleen cells were incubated with 2.5 μg/ml β-galactosidase for 5 days and 1 μl/ml GolgiPlug was added during the last 6 h. Stainings were performed directly with FITC-conjugated CD3 mAb (clone 145-2C11) and PE-conjugated CD8 mAb (clone 53-6.7; Biolegend), and intracellularly with allophycocyanin-conjugated IL-2 mAb (clone JES6-5H4; Biolegend), and allophycocyanin-conjugated IFN-γ mAb (clone XMG1.2; Biolegend). Stained cells were analyzed with FACSCalibur (BD Biosciences) and CellQuest Pro software (BD Biosciences).
In vivo cytotoxicity assay
Splenocytes from naive C57BL/6 mice were separated, RBC were removed with ammonium chloride and divided into two equal populations, one of which was pulsed with 1 μM OVA257–264 at 37°C for 90 min. The cells were labeled with CFSE (Molecular Probes) at a final concentration of 20 μM for the peptide-pulsed population (CFSEhigh) and 2 μM for the peptide-unpulsed population (CFSElow) at 37°C for 15 min. The two populations were mixed equally and 2 × 107 cells injected i.v. into the immunized mice. After 24 h, single cells from spleen and lymph nodes were isolated and CFSE intensities were analyzed using FACSCalibur and CellQuest Pro software. The percentage of specific killing was calculated as follows: percentage of lysis = [1 − (ratiounprime/ratioprime)] × 100; ratio = percentage CFSElow/percentage CFSEhigh.
Protection assay against EG7
Two weeks after the final immunization, 100 μl of PBS containing 3 × 106 OVA-expressing EG7 cells was injected s.c. into the left flank of the immunized C57BL/6 mice. Fourteen days after the tumor challenge, the mice were sacrificed and palpable tumors were weighed.
Protection assay against influenza virus A/PR/8/34
Two weeks after the final immunization, the immunized BALB/c mice were infected intranasally by dropping 20 μl (10 μl/nostril) of PBS containing influenza virus A/PR/8/34 suspension with 20 LD50 per mouse. Mice were divided into two populations in each group. After 3 days, one population was sacrificed and nasal wash, lung wash, and serum were taken and then PR8 HA-specific Ab responses were examined. The other population was monitored for weight loss and survival rates every other day for 14 days.
Transfer of OT-1 cells
OVA-specific CD8+ T cells (OT-1 cells) were isolated from OT-1 mice using CD8α (Ly-2) magnetic beads (Miltenyi Biotec). These cells (>94% were Vα2-positive) were labeled with 10 μM CFSE at 37°C for 15 min and transferred i.v. into the syngeneic mice.
Data and statistical analysis
Results are expressed as mean ± SE. Statistical analyses were performed upon comparisons made between the control groups and treated groups using the Student t test. Each experiments was repeated at least twice.
Intranasal administration of α-GalCer induced OVA-specific mucosal secretory IgA (S-IgA) and systemic IgG Ab responses in C57BL/6 mice
To clarify mucosal adjuvanticity of α-GalCer, C57BL/6 mice were immunized with α-GalCer plus OVA three times at 1-wk intervals. The mice were sacrificed 1 wk after the last immunization and OVA-specific Ab responses were measured by ELISA. OVA-specific IgA responses in nasal and lung wash were significantly induced in the group receiving 2.0 μg of α-GalCer plus OVA compared with those receiving vehicle or OVA alone (Fig. 1,A). To eliminate the possibility that the IgA responses were due to contamination of the sera, the amount of total IgG in nasal wash was assessed; no total IgG Ab was detected (data not shown). In addition, intranasal administration of different doses of α-GalCer plus OVA elicited significantly higher levels of OVA-specific IgG in sera than did intranasal administration of the vehicle or OVA alone (Fig. 1,B). To determine indirectly whether Th1 or Th2 immune responses predominate in the humoral immune response to OVA, the IgG isotypes in sera were determined and the ratios of IgG1 to IgG2a were calculated. Coadministration of OVA with α-GalCer significantly enhanced both OVA-specific IgG1 and IgG2a levels, indicating that α-GalCer induces both Th1 and Th2 immune responses (Fig. 1 C). These results indicate that α-GalCer is a potent mucosal adjuvant for the induction of Ag-specific mucosal S-IgA and systemic IgG Ab responses and indirectly suggest that α-GalCer induced both Th1 and Th2 immune responses in C57BL/6 mice.
Intranasal administration of α-GalCer enhanced both Th1 and Th2 cytokines in C57BL/6 mice
To assess directly whether α-GalCer skews immune responses toward Th1 or Th2 immune response, mononuclear cells from spleen and cervical lymph node (CLN) were obtained 1 wk after the final immunization and cultured with whole OVA for 4 days. The secretion patterns of IFN-γ and IL-4 in the culture supernatants were examined by sandwich ELISA. Both IFN-γ and IL-4 production were significantly increased in spleen and CLN (Fig. 1,D). A very high level of IFN-γ was induced by a high dose of α-GalCer and levels of IL-4 increased proportionally to the increasing dose of α-GalCer in CLN. To assess the epitope-specific cellular immune response induced by coimmunization of α-GalCer via intranasal route, IFN-γ-producing CD8+ T cells were analyzed after stimulating splenocytes with OVA257–264 peptide. The number of IFN-γ-producing CD8+ T cells was significantly increased in all groups coimmunized with OVA and α-GalCer compared with those in vehicle- or OVA alone-treated groups (p < 0.01, vehicle and OVA vs 0.125, 0.5, and 2.0 μg) (Fig. 1 E). These results demonstrate that intranasal coadministration of α-GalCer induces mixed Th1 and Th2 immune responses in both systemic and mucosal compartments.
α-GalCer induced potent CTL responses in C57BL/6
It has been well established that α-GalCer induces CTL responses when it is administered by the i.v. or oral routes (19, 30). To test whether α-GalCer induces CTL responses when it is coadministered intranasally, a mixture containing equal amounts of OVA257–264-pulsed CFSEhigh and OVA257–264-unpulsed CFSElow splenocytes of syngeneic mice was injected i.v. into the mice 1 wk after the final intranasal immunization. Twenty-four hours later, the specific lysis of peptide-pulsed CFSE-labeled cells in the spleen, mesenteric lymph node (MLN) and CLN was analyzed using flow cytometry. All groups of α-GalCer-treated mice showed potent lytic activities in a dose-dependent manner (Fig. 2, A and B). Interestingly, the cytotoxic activities were induced both mucosally (CLN) and systemically (spleen and MLN). These results show that α-GalCer is a potent nasal vaccine adjuvant for the induction of CTLs in both local and systemic lymphoid organs.
Intranasal administration of α-GalCer adjuvant induced both humoral and cellular immune responses in BALB/c mice
We then examined whether α-GalCer could also be used as a potent nasal vaccine adjuvant in BALB/c mice. In BALB/c mice, intranasal administration of different doses of α-GalCer plus OVA elicited significantly higher levels of OVA-specific IgA responses in nasal wash and lung wash as well as of OVA-specific IgG responses in sera than did intranasal administration of vehicle or OVA alone (Fig. 3, A and B). Additionally, intranasal administration of OVA with α-GalCer significantly enhanced both OVA-specific IgG1 and IgG2a levels (Fig. 3,C). Further, significant levels of both IFN-γ and IL-4 were induced in all α-GalCer-administered groups (Fig. 3 D). Interestingly, the levels of IgG Ab in serum and IL-4 production in spleen were the highest in the mice intranasally receiving 0.5 μg of α-GalCer and then declined as the dose of α-GalCer was increased. Furthermore, mucosal IgA in lung wash and IL-4 production in CLN was inversely proportional to the coadministered dose of α-GalCer. Collectively, it seems likely that a high dose of α-GalCer induces tolerance to coadministered Ag in BALB/c mice.
As there was no known MHC class I-binding peptide of OVA in BALB/c mice, the numbers of IFN-γ-producing CD8+ T cells were examined to test the cytotoxic activity induced by α-GalCer adjuvant in BALB/c mice. As doses of α-GalCer increased, the number of IFN-γ-producing CD8+ T cells decreased (p < 0.05, 0.125 μg vs 0.5 μg; p < 0.01, 0.125 μg and 0.5 μg vs 2.0 μg) (Fig. 3 E). Although the amount of IFN-γ detected by sandwich ELISA did not show dose dependency, the number of IFN-γ-producing CTLs was inversely proportional to the dose of α-GalCer. As a possible explanation for this discrepancy, it should be noted that the amount of IFN-γ detected by sandwich ELISA represents that from all kinds of cells producing IFN-γ, such as CD4+, CD8+ T cells or APCs, whereas the number of effector CTLs detected by FACS represents only CD8+ T cells. Overall, these results indicate that α-GalCer also represents a potent nasal vaccine adjuvanticity in BALB/c mice.
α-GalCer showed a nasal vaccine adjuvanticity for the induction of antiviral immune responses against influenza virus A/PR/8/34 infection
To determine the level of mucosal protection afforded by α-GalCer against viral infection, BALB/c mice were intranasally immunized with PR8 HA Ag alone or together with α-GalCer three times at 1-wk intervals and then intranasally infected with 20 LD50 of influenza virus 2 wk after the final immunization. Three days after the viral infection, PR8 HA-specific Ab responses were measured in the nasal wash, lung wash, and sera. In addition, the weight loss and survival rates of the infected mice were observed every other day. Interestingly, high levels of PR8 HA-specific IgA Abs were detected in the nasal wash, lung wash, and sera of all the α-GalCer-treated groups (Fig. 4,A). Further, high levels of PR8 HA-specific IgG Abs were also detected in sera of all groups of α-GalCer-treated mice (Fig. 4,B). These results indicate that α-GalCer is a potent nasal adjuvant inducing mucosal S-IgA as well as systemic IgG Ab responses against virus-derived Ag. Most interestingly, much greater pathogenesis was observed in the mice immunized without α-GalCer than in the mice coimmunized with α-GalCer, as indicated by survival rate, weight loss, and time to weight loss recovery (Fig. 4,C). Although all mice died within 10 days after the viral infection in the group receiving vehicle alone and 57% of the mice treated with OVA alone died within 14 days, the mice intranasally immunized with α-GalCer plus OVA did not show any significant decrease in survival rate (Fig. 4 C, left). These results indicate that α-GalCer is a potent nasal vaccine adjuvant inducing mucosal S-IgA Abs, systemic IgG Abs, and protection against viral infection.
Intranasal administration of α-GalCer with live virus enhanced immunogenicity against replication-deficient live virus
To examine the adjuvanticity of α-GalCer against live virus, BALB/c mice were intranasally immunized with 106 PFU of replication-deficient live Ad-LacZ alone or together with 0.125 μg of α-GalCer twice at 2-wk interval. One week after the last immunization, β-galactosidase-specific Ab responses were measured in nasal wash, lung wash, and serum. For determination of CTL activities, splenocytes were stimulated with 2.5 μg/ml β-galactosidase for 5 days and IFN-γ-producing CD8+ T cells were assessed by intracellular cytokine staining. Interestingly, β-galactosidase-specific S-IgA and IgG responses were significantly higher in mucosal compartment and sera of α-GalCer-coimmunized group, respectively, compared with those in vehicle or Ad-LacZ alone-treated groups (Fig. 5, A and B). Moreover, the number of IFN-γ-producing CD8+ T cells was significantly enhanced in the α-GalCer-treated group (Fig. 5 C). Taken together, these results indicated that α-GalCer is an effective nasal vaccine adjuvant against the replication-deficient live virus.
α-GalCer showed a nasal vaccine adjuvanticity for the induction of antitumor immune response against EG7 tumor
To assess whether α-GalCer could be used as a nasal vaccine adjuvant for the induction of antitumor activity, C57BL/6 mice were immunized intranasally with OVA alone or together with α-GalCer three times at 1-wk intervals. Two weeks after the final immunization, 3 × 106 cells of EG7 tumor cells were inoculated s.c. into the left flank of the immunized mice. Fourteen days after the inoculation, the mice were sacrificed and palpable tumors were weighed. Palpable tumors were found in all mice receiving the vehicle or OVA alone and in one-third of those to which 0.125 μg of α-GalCer had been coadministered (Fig. 6). The weights of the tumors in the mice intranasally immunized with OVA alone were significantly greater than those in the mice immunized with vehicle alone (p < 0.05). Most interestingly, the formation of tumor masses was completely blocked in the mice intranasally administered with 0.5 and 2.0 μg of α-GalCer plus OVA (Fig. 6). These results indicate that α-GalCer can be used as a potent nasal vaccine adjuvant for the induction of antitumor immune responses.
Nasal adjuvant activity of α-GalCer was exclusively mediated via CD1d molecule
To confirm that the α-GalCer-induced immune responses are mediated by CD1d molecule, we used CD1d−/− C57BL/6 mice, which lack the class I restricting element of NKT cells and therefore lack NKT cells (31). One week after the final intranasal administration, systemic IgG in serum and in vivo CTL activity were examined in wild-type and CD1d−/− C57BL/6 mice. As expected, systemic IgG Ab response was remarkably inhibited in CD1d−/− mice (Fig. 7,A). Further, the lytic activity of CTL was blocked in the draining lymph node and systemic lymphoid organs of CD1d−/− mice (Fig. 7 B). These results clearly demonstrate that the immune responses induced by α-GalCer in this study are exclusively mediated by the CD1d molecule.
Intranasal administration of α-GalCer activates naive T cells and triggers them to differentiate into effector cells
To further clarify the role of α-GalCer on the activation of T cell responses, the expression of surface CD25 molecules was measured in adoptively transferred CFSE-labeled OT-1 cells into the syngeneic mice. One day after the adoptive transfer, the mice were intranasally administered with OVA alone or together with 2.0 μg of α-GalCer. Forty-eight hours later, the expression of CD25 molecules in CLN was analyzed using flow cytometry. A much more substantial population of OT-1 cells expressing CD25 could be detected in the mice intranasally administered with OVA plus α-GalCer compared with those receiving OVA alone, suggesting that α-GalCer nasal adjuvant induced the activation of naive T cells (Fig. 8,A). To verify that the activated T cells were differentiated into functional CTLs, the cells were further stimulated with OVA257–264 peptide for 6 h and then intracellular IL-2 and IFN-γ were measured. Much higher populations of IL-2- and IFN-γ-producing OT-1 cells were observed in the mice intranasally immunized with OVA plus α-GalCer than in those immunized with OVA alone (Fig. 8 B). Collectively, these results demonstrate that intranasal coadministration of α-GalCer induces the activation of naive T cells and that the activated T cells exert a potent effector function.
In this study, we evaluated the adjuvanticity of α-GalCer as a nasal vaccine in both C57BL/6 and BALB/c mice. Whether the Ag used is OVA or PR8 HA, intranasal coadministration of the Ag with α-GalCer induced high levels of systemic IgG and mucosal S-IgA Abs in both C57BL/6 and BALB/c mice. The ratio of IgG1 and IgG2a in serum was not significantly affected by intranasal coadministration with α-GalCer. In addition, ex vivo measurement of cytokine profiles revealed high levels of IFN-γ and IL-4 in mucosal and systemic compartments. Also, Ag-specific CTLs were remarkably induced when Ag was administered together with α-GalCer by the intranasal route. Further, intranasal coadministration with α-GalCer afforded complete protection against influenza viral infection and tumor, and enhanced humoral as well as cellular immune responses against replication-deficient live adenovirus. Taken together, we have proven for the first time that α-GalCer is a potent nasal vaccine adjuvant inducing mixed Th1 and Th2 immune responses and that it can be use as an effective nasal vaccine adjuvant to enhance protection against viral infection and tumors.
A recent study demonstrated that repeated exposure to α-GalCer induced disappearance of NKT cells and increased expression of inhibitory Ly49, resulting in tolerance to Ag (32). We suspect that the amount of α-GalCer may affect the level of immune responses or the degree of tolerance induced to coadministered Ag. To clarify those issues, we evaluated the immune responses by α-GalCer in a dose-dependent manner. Interestingly, the optimal dose of α-GalCer following intranasal administration differed for C57BL/6 and BALB/c mice, demonstrating that dose must be determined based upon strain. Overall, our present results suggest that the precise determination of the optimal dose of α-GalCer as a nasal vaccine adjuvant should be performed in nonhuman primates before clinical trial.
Renegar et al. (33) demonstrated that S-IgA and systemic IgG are important for the protection against viral infection in the murine respiratory tract. Belyakov and Berzofsky (34) indicated that mucosal S-IgA and CTL are responsible for the potential protective local immune responses against HIV infection. In the current study, our results have shown that intranasally delivered α-GalCer adjuvant induced significant levels of mucosal S-IgA, systemic IgG and both systemic and mucosal CTLs, as well as providing complete protection against influenza viral infection. Several other groups have also demonstrated that α-GalCer adjuvant exhibits antitumor activity (19, 20), and that the antitumor effects of α-GalCer are mainly mediated by NK cells in a perforin-independent manner (36), by anti-angiogenesis of IFN-γ secreted by activated NKT cells (37) and by CTLs (35). During the preparation of our manuscript, Silk et al. (30) showed that oral administration of OVA plus α-GalCer provoked antitumor responses. In our study, intranasal administration of OVA plus α-GalCer induced significant IFN-γ production and afforded complete protection against the EG7 tumor. Interestingly, the tumors of the group treated with OVA alone were heavier than those treated with vehicle alone. Previous studies demonstrated that intranasal exposure to protein Ag alone leads to immunological tolerance and CLN is responsible for the induction of the tolerance (38, 39). Importantly, our previous study indicated that the administration of α-GalCer triggered full maturation of mesenteric DC and thereby blocked the induction of oral tolerance (21). In this point of view, the heavier tumor mass in the group treated with OVA alone is due to the induction of tolerance to intranasally administered-OVA Ag, speculating that the tolerance may be blocked by intranasal coadministration with α-GalCer. Taken together, these results demonstrate that α-GalCer is a potent nasal adjuvant inducing complete protection against viral infection and tumors by blocking mucosally induced tolerance to coadministered Ag.
We have demonstrated here that intranasal α-GalCer exerts its adjuvant activity via the CD1d molecule, and activates naive T cells and triggers them to differentiate into potent effector cells. α-GalCer-presented DCs activate NKT cells and, in turn, NKT cells make DCs mature, which is mediated by IFN-γ, TNF-α, or other unknown mediators or mechanisms (30). Oral administration of α-GalCer induces full maturation of mesenteric DCs (21). In addition, when cholera toxin is administered intranasally, plamacytoid DCs induce tolerance and myeloid DCs induce inflammation against coadministered Ag in mice (40). NKT cells are divided into CD4+ and CD4− subpopulations in humans (41). Although CD4− NKT cells selectively produce Th1 cytokines, CD4+ NKT cells potently produce both Th1 and Th2 cytokines. However, to our knowledge, no such studies have been conducted using intranasally administered α-GalCer. Therefore, further study is needed to determine which subpopulations of DC and/or NKT cells are involved in the adjuvanticity of intranasally delivered α-GalCer.
α-GalCer has been used in clinical studies in antitumor immunotherapy, and shown to induce NKT cell memory and then more vigorous immune responses than those observed in control groups (42, 43). In addition, those studies showed that α-GalCer provoked no dose-limiting toxicity (50–4800 μg/m2) and is well tolerated in a dose escalation trial, which indicated that α-GalCer could be used safely. In contrast, a previous study demonstrated that cholera toxin, a GM1-binding molecule, redirects vaccine protein into olfactory tissues when given intranasally (23). In other words, intranasal administration of adjuvant may induce the redirection of coadministered Ag or adjuvant itself into the brain, provoking unwanted effects. With its high m.w. and its hydrophilic group, however, α-GalCer would be hard to redirect to the CNS. In addition, the population of immune cells including APC and T cells in the brain is thought to be so small that it would not be easy for activated NKT cells by α-GalCer to exert their toxic action. In fact, even though mild side effects were induced when α-GalCer was given i.v. in phase I clinical study, these effects were easily prevented by paracetamol pretreatment, and the liver toxicity induced in mice by parenteral administration of α-GalCer was not seen in this clinical study (44). Moreover, no side effects were reported in mice receiving doses up to 2200 μg/kg α-GalCer (43). Taken together, these findings demonstrate that α-GalCer is a safe and effective adjuvant for use in intranasally delivered antiviral and antitumor vaccines.
It is generally accepted that far lower doses are required in nasal drug delivery than in oral delivery because in the former the drug is more rapidly absorbed, has a higher bioavailability and avoids liver first pass effect and metabolism by gastrointestinal tract (24). Actually, much less α-GalCer was required by the intranasal route in this study than was required by the oral route to protect against the EG7 tumor, demonstrating once again that α-GalCer can be used as a safer mucosal adjuvant (30).
Recently, spray formulation of cold-adapted live attenuated vaccine has been used commercially because it is safer and more convenient than live vaccines (45). It seems likely that α-GalCer adjuvant would be suited to such vaccine formulations.
In conclusion, we have proven in this study for the first time that α-GalCer is a potent nasal vaccine adjuvant inducing both humoral and cellular immune responses and is suitable for use in antiviral and antitumor nasal vaccines.
We thank Dr. Shin-Ichi Tamura for kindly providing the PR8 HA Ag, ViroMed Co., Ltd. for providing replication-deficient adenovirus, and Dr. Sanghee Kim for providing α-GalCer. We also thank Dr. Baik L. Seong for his help and cooperation in the virus experiment.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by the Ministry of Science and Technology Nano-Bio Research and Development Program Grant No. M10416220005-04N1622-00510 of the Ministry of Science and Technology, Korea.
Abbreviations in this paper: αGalCer, α-galactosylceramide; DC, dendritic cell; S-IgA, secretory IgA; CLN, cervical lymph node; MLN, mesenteric lymph node; HA, hemagglutinin.