Nanotechnology is an innovative method of freely controlling nanometre-sized materials. Recent outbreaks of mucosal infectious diseases have increased the demands for development of mucosal vaccines because they induce both systemic and mucosal antigen-specific immune responses. Here we developed an intranasal vaccine-delivery system with a nanometre-sized hydrogel ('nanogel') consisting of a cationic type of cholesteryl-group-bearing pullulan (cCHP). A non-toxic subunit fragment of Clostridium botulinum type-A neurotoxin BoHc/A administered intranasally with cCHP nanogel (cCHP-BoHc/A) continuously adhered to the nasal epithelium and was effectively taken up by mucosal dendritic cells after its release from the cCHP nanogel. Vigorous botulinum-neurotoxin-A-neutralizing serum IgG and secretory IgA antibody responses were induced without co-administration of mucosal adjuvant. Importantly, intranasally administered cCHP-BoHc/A did not accumulate in the olfactory bulbs or brain. Moreover, intranasally immunized tetanus toxoid with cCHP nanogel induced strong tetanus-toxoid-specific systemic and mucosal immune responses. These results indicate that cCHP nanogel can be used as a universal protein-based antigen-delivery vehicle for adjuvant-free intranasal vaccination.

Capable of inducing antigen-specific immune responses in both systemic and mucosal compartments without the use of syringe and needle, mucosal vaccination is considered ideal for the global control of infectious diseases. In this study, we developed a rice-based oral vaccine expressing cholera toxin B subunit (CTB) under the control of the endosperm-specific expression promoter 2.3-kb glutelin GluB-1 with codon usage optimization for expression in rice seed. An average of 30 g of CTB per seed was stored in the protein bodies, which are storage organelles in rice. When mucosally fed, rice seeds expressing CTB were taken up by the M cells covering the Peyer's patches and induced CTB-specific serum IgG and mucosal IgA antibodies with neutralizing activity. When expressed in rice, CTB was protected from pepsin digestion in vitro. Rice-expressed CTB also remained stable and thus maintained immunogenicity at room temperature for >1.5 years, meaning that antigen-specific mucosal immune responses were induced at much lower doses than were necessary with purified recombinant CTB. Because they require neither refrigeration (cold-chain management) nor a needle, these rice-based mucosal vaccines offer a highly practical and cost-effective strategy for orally vaccinating large populations against mucosal infections, including those that may result from an act of bioterrorism. mucosal immunity ͉ protein body ͉ oral vaccine ͉ IgA ͉ cholera toxin B subunit

4 of both GALT and nasopharynx-associated lymphoid tissue, which act as a major inductive site for Ag-specific mucosal immune responses (1, 2). Recently, we also identified M cells in the small intestinal villous epithelium, at effector sites far from the FAE, suggesting that Ag sampling via villous M cells may be responsible for induction of systemic Ag-specific immune responses, such as IgG production via the oral route (3). Still missing, however, were a characterization of the shared and distinctive traits of Peyer's patches (PPs) and villous M cells and a better understanding of the immunological nature of each.Recent comprehensive gene expression analyses using microdissected FAE or whole cells dissociated from the FAE identified genes specifically expressed by PP M cells (4 -6). Similar data, however, have not been available for villous M cells, in part because sufficient numbers of M cells are difficult to isolate from the surrounding intestinal epithelial cells (IECs). In mice, lectin Ulex europaeus agglutinin-1 (UEA-1) possessing affinity for ␣ (1, 2) fucose has been routinely used for the detection of such M cells (3, 7). UEA-1, however, does not alone suffice to identify M cells because it also reacts to goblet cells (3). Our laboratory has recently succeeded in distinguishing M cells from goblet cells by developing a mAb (NKM 16-2-4 mAb) that specifically reacts to murine PP and villous M cells but not goblet cells and IECs (8). Furthermore, our recent separate studies have demonstrated that oral administration of cholera toxin (CT) as mucosal adjuvant resulted in the induction of NKM 16-2-4 mAb ϩ and UEA-1 ϩ M-like cells, which have pocket structure and Ag uptake ability, in the duodenal villous epithelium (Terahara et al., submitted for publication). These recent advances in our understanding of M cells allowed us to define gene expression profiles capable of distinguishing PP M cells, CT-induced villous M-like cells, and IECs. Materials and Methods AnimalsBALB/c mice were purchased from Japan SLC. These mice were maintained under specific pathogen-free conditions in horizontal flow cabinets in our experimental animal facility at the University of Tokyo. Following a previously established protocol (9, 10), CT (List Biologic Laboratories) was dissolved in PBS (20 g/mouse) and then orally administered to BALB/c mice. Two days after CT administration, mice were used for experiments. All animal experiments were approved by the Animal Care and Use Committee of University of Tokyo. Lectins and Abs for the detection of M cellsThe following fluorescence-conjugated lectins and Abs were used for the identification of PP and villous M cells by FACS and histochemistry: PEconjugated UEA-1 (Biogenesis), rhodamine-conjugated UEA-1 (Vector