The results from this investigation strongly support mucosal delivery as an efficient approach to harness the adjuvant potential of α-GalCer for priming as well as boosting cellular immune responses to co-administered immunogens. This is due to the repeated activation of NKT cells and DCs achieved after intranasal immunization with α-GalCer as an adjuvant. Meanwhile, systemic immunization by the intravenous route resulted in the unresponsiveness of the NKT cells to booster doses of α-GalCer, a phenomenon known as NKT cell anergy. These results are consistent with our earlier published studies which demonstrated the effectiveness and necessity of α-GalCer for repeated immunization by mucosal routes for the induction of strong cellular immune responses to the co-administered antigen [7
Our studies comparing the intravenous and intranasal routes for delivering α-GalCer revealed similar kinetics of activation of NKT cells and DCs in terms of peak levels of IFN-γ production by NKT cells and DC activation at one day after a single immunization and are consistent with literature reports [5
]. The key finding from our investigation is that a booster immunization employing α-GalCer as an adjuvant by the intravenous and intranasal routes revealed vastly different effects on NKT cells and DCs. While a single intravenous administration of α-GalCer, as demonstrated in this manuscript and reported in the literature, leads NKT cells to become unresponsive in terms of inability to produce cytokines in response to a booster dose of α-GalCer and also an inability to proliferate [5
], our data demonstrates that after booster intranasal administration of α-GalCer, a potent activation of the NKT cells is observed for a second time in the lung, including IFN-γ production and expansion as well as DC activation. This repeated activation of NKT cells and DCs occurs regardless of the timing for the administration of the booster dose (i.e. day 5 or 23), suggesting that immunization by the intranasal route is a potential means to allow repeated dosing of the α-GalCer adjuvant without the induction of NKT cell anergy. A recent report published during the preparation of this manuscript showed delivery of α-GalCer by the intradermal route to be effective in avoiding NKT cell anergy, but mechanistic details are not described [15
Of note, NKT cell activation and proliferation occurs in multiple tissues after primary intranasal administration of α-GalCer, but NKT cells are fully re-activated in the lung after the second intranasal administration of α-GalCer, suggesting that the lung is the major site of α-GalCer presentation after intranasal administration. This was confirmed by the observation that α-GalCer presentation to the DN32.D3 NKT cell clone occurs mainly in the lung and to a lesser extent in the lung-draining lymph node up to 5 days after intranasal administration. However, it is unclear as to how NKT cells and DCs are activated in more distal tissues, such as the spleen and liver, after a primary intranasal immunization with α-GalCer. It is possible that either activated DCs and/or activated NKT cells migrate from the lung after stimulation with α-GalCer, or alternatively the cytokine milieu resulting from NKT cell stimulation with α-GalCer may induce activation of these cell types in other tissues. In this regard it has been reported that a decrease in NKT cell populations in the liver coincided with an increase in the blood NKT cell levels after intraperitoneal immunization with α-GalCer, suggesting potential trafficking of NKT cells [16
It has been observed that multiple administrations of DCs pulsed ex vivo with α-GalCer, as opposed to free α-GalCer, do not induce NKT cell anergy [5
]. On the other hand, it has also been shown that injection of B cells pulsed ex vivo with α-GalCer does induce NKT cell anergy [5
]. Here we have shown that after intranasal administration, CD11c+
cells, not B220+
cells, more efficiently present α-GalCer in the lung, suggesting that the intranasal route of immunization preferentially targets α-GalCer presentation to DCs. Interestingly, Hermans et al. [18
] showed that presentation of both α-GalCer and peptide antigen by the same DC was required for the strong activation of antigen-specific T-cell responses. Futhermore, Ko et al. [14
] showed that the responding DC-presenting antigen in the lung-draining LNs also expresses a CD8α−
phenotype. This suggests that the DCs presenting α-GalCer in the lung should show a similar phenotype, which would be intriguing to pursue in the future.
In addition to the potential influence of the phenotype of cells presenting α-GalCer to induce NKT cell anergy, recently it has been reported that expression levels of the cell surface marker PD-1 on NKT cells may also be an important factor for anergy induction. In T cells, higher levels of PD-1 expression were observed to be associated with functional exhaustion resulting from interactions with either of its ligands, PD-L1 or PD-L2, which are both commonly expressed on APCs including B cells, DCs, and macrophages [19
]. It has also been observed that PD-1 expression is up-regulated on the ‘exhausted’ CD8+
T cells in HIV-infected patients and blocking of the PD-1/PD-L1 interaction could rescue the exhausted T cells in terms of restoring functional properties [22
]. Multiple groups have also shown that PD-1 is up-regulated on NKT cells very early after systemic administration of α-GalCer, and that blockade of the PD-1/PD-L1 interaction can reverse the unresponsiveness of the NKT cells [11
]. We observed that while NKT cells from mice administered with α-GalCer by the intravenous route exhibited high levels of PD-1 expression at day 1 post-immunization, those in mice where α-GalCer was delivered by the intranasal route did not (). Furthermore, PD-1 expression on NKT cells coincided with functional exhaustion and unresponsiveness at 24 h after a second dose of α-GalCer by the intravenous route but not when α-GalCer was delivered by the intranasal route where NKT cells were fully functional in terms of IFN-γ production and expansion ( and ). Thus, in addition to the cell type mediating α-GalCer presentation (i.e. DCs versus B cells), the phenotype of NKT cells in terms of PD-1 expression could be another important factor for the avoidance of NKT cell anergy resulting from mucosal α-GalCer delivery (e.g. intranasal route), as opposed to systemic delivery (e.g. intravenous route). These observed differences between intravenous versus intranasal route of α-GalCer delivery may enable the repeated activation of NKT cells to aid in promoting DC activation which allows α-GalCer to serve as an efficient mucosal adjuvant for inducing immune responses to co-administered antigens. In fact, as shown in a booster dose of α-GalCer administered by the intranasal route resulted in a subsequent increase in antigen-specific immune responses, while a booster dose of α-GalCer administered by the intravenous route did not correspond to an increase in antigen-specific immune responses.
In addition to the differences in terms of NKT cell anergy induction or the lack thereof, our investigation revealed several other differences for NKT cell activation after intravenous versus intranasal administration of α-GalCer. First, the timing of NKT cell activation and expansion appeared to be prolonged after intranasal administration of α-GalCer because the peak levels of NKT cell expansion were observed at day 5 post-immunization in the lung, the main responding tissue for this route of immunization. These results differ from that seen after the intravenous immunization where the NKT cell population peaked at day 3 in all tissues tested. In this regard, Fujii et al. [8
] reported that intravenous administration of DCs pulsed ex vivo with α-GalCer, as opposed to free α-GalCer, which is shown to be a potential approach to avoid anergy to NKT cells, resulted in a prolonged NKT cell response, as measured by IFN-γ production. Second, we observed a decrease in the NKT cell population in the spleen and liver at day 1 after the priming immunization by the intravenous route, which is consistent with the literature reports that attribute the decrease in population to the down-regulation of the TCR as the underlying mechanism, but no such decrease in TCR was observed for NKT cells in mice that received the priming immunization via the intranasal route. Incidentally, Fujii et al. [8
] reported a phenomenon describing NKT cell turnover, a decrease in the NKT cell population on day 1 after α-GalCer administration later found to be due to TCR down-regulation, after administration of free α-GalCer that was “less rapid and severe” when DCs pulsed with α-GalCer were administered.
Antigen-specific cellular immune responses were measured after each dose of the α-GalCer adjuvant and OVA antigen mixture, similar to our previously reported studies with a different antigen [7
]. Both these studies demonstrate that multiple doses of α-GalCer, administered by the intranasal route, are necessary to induce efficient antigen-specific cellular immune responses, regardless of the mouse strain used. In addition to the antigen-specific cellular immune responses, effectiveness of α-GalCer as an adjuvant after intranasal immunization to induce humoral immune responses, in terms of antigen-specific IgA and IgG responses has been described in the literature [24
] and also observed in other unrelated studies in our laboratory (data not shown). Thus, our studies provide mechanistic support for mucosal delivery of α-GalCer adjuvant as an attractive strategy for vaccination regimens.
It is also important to note potential inflammatory effects from the intranasal administration of α-GalCer. Different mouse model studies revealed that intranasal administration of α-GalCer can induce airway infiltration of a combination of eosiniphils, neutrophils, and/or monocytes [25
]. Preliminary studies in our lab showed increase in the percentages of eosinophils but not neutrophils or monocytes (data not shown). However, clinical trials performed by Kunii et al. [4
] showed that administration of α-GalCer by a nasal sub-mucosal route was safe.
Overall, this investigation has shown that α-GalCer can be administered by the intranasal route for primary and booster immunizations to induce cellular immune responses to co-administered antigens, without inducing NKT cell anergy. This is in striking contrast to α-GalCer administration by the intravenous route, in which a single dose leads to NKT cell anergy and a reduction in the ability of the adjuvant to boost adaptive immune responses to co-administered antigen. Thus, our data support the intranasal route of immunization as an attractive route for immunization especially because the ability to deliver multiple doses of the vaccine is essential for most therapeutic applications against infectious diseases and cancer.