Lactococcus lactis (L. lactis) MG1363-derived NZ9000 (Mobitec) was cultured overnight in GM17 medium (M17 containing 5% D-glucose). After centrifugation of the culture, the pellet was washed with phosphate-buffered saline (PBS). The pellet was suspended in 1 M H₂SO₄ (pH 1.0) and heated in a water bath at 100°C for 30 min. The heated suspension was centrifuged, the supernatant was removed, the pellet was washed for 3 times with PBS, and then resuspended in PBS to prepare BLPs. BLPs were counted using a hemocytometer and stored at -20°C until use.

The Tir gene fragment (encoding amino acids 257 to 409 of full-length Tir) and the NanoLuc gene fragment was amplified by PCR using C. rodentium ATCC51459 genome DNA and pNL1.1 (Promega) as a template, respectively. The fragments were then cloned into the downstream of the GST gene of the pGEX vector to construct the Tir or NanoLuc expression plasmid (pGEX-Tir and pGEX-NanoLuc). The LysM motif of the AcmA gene of L. lactis NZ9000 was cloned into pGEX-Tir or pGEX-NanoLuc to construct pGEX-Tir-LysM and pGEX-NanoLuc-LysM expressing the Tir- and NanoLuc-LysM fusion proteins, respectively. pGEX-Tir, -Tir-LysM and -NanoLuc-LysM were transformed into E. coli BL21, resulting in BL21/pGEX-Tir, BL21/pGEX-Tir-LysM and BL21/pGEX-NanoLuc-LysM, respectively. BL21/pGEX-Tir, BL21/pGEX-Tir-LysM and BL21/pGEX-NanoLuc-LysM were stimulated with 0.1 mM IPTG to induce the expression of GST-Tir, GST-Tir-LysM and GST-NanoLuc-LysM for 6 h at room temperature, then collected by centrifugation and disrupted with sonication. Debris was removed by centrifugation to obtain soluble fractions containing recombinant GST-Tir, -Tir-LysM and -NanoLuc-LysM. GST-Tir was affinity-purified using Glutathione Sepharose 4B. Purified Tir was dialyzed against PBS and stored at -80°C until use for ELISA and ELISPOT assay.

Tir-LysM- and NanoLuc-LysM-bound BLPs (BLP-Tir and BLP-NanoLuc) were prepared by mixing BLPs with the soluble fraction of BL21/pGEX-Tir-LysM or BL21/pGEX-NanoLuc-LysM containing GST-Tir-LysM and -NanoLuc-LysM, respectively, at room temperature for 30 min. The unbound protein was removed by washing 5 times with PBS. The binding of Tir-LysM to the BLPs was confirmed by confocal laser scanning microscopy after immunostaining with anti-Tir serum (13). The amount of Tir-LysM bound to BLPs was quantified by densitometry analysis of CBB-stained gels. The amount of NanoLuc-LysM bound to BLPs was quantified by luciferase assay. BLP-Tir and BLP-NanoLuc were suspended in PBS and stored at -80°C until use.

TLR2 stimulation was assessed by reporter gene assay using human TLR2-expressing HEK293T cells (kindly provided by Dr. Tsuyoshi Sugiyama). HEK293T cells were transfected with a human TLR2-expressing plasmid, pNF-κB-TA-luc (a NF-κB-luciferase reporter plasmid), and pGL4.73 (constitutive Renila luciferase expression plasmid, internal control) using PEImax transfection reagent (Polysciences). After incubation for 48 h, cells were treated with various concentrations of BLPs or 100 ng/ml of Pam3CSK4 (a synthetic TLR2 ligand) for an additional 24 h. Luciferase activities were measured using the Dual-Glo luciferase assay system (Promega) and GloMax multiplate reader (Promega), according to the manufacturer’s instructions.

All animal experiments were approved by the Institutional Animal Care and Use Committee of Gifu Pharmaceutical University and that of Gifu University. Mice were maintained under normal husbandry conditions in the Animal Facilities of Gifu University.

C57BL/6 mice (6-week-old, female, SLC Japan) were light anesthetized by inhalation with isoflurane and intranasally administered with 20 μl/dose of BLP-Tir (containing 2 × 10⁹ BLPs bound with approximately 20 μg of Tir-LysM, 10 μl/each nasal passage). For oral administration, the same amount of BLP-Tir as used for the intranasal administration was suspended in 200 μl/dose of PBS. Administration was performed 3 times at 11- to 12-day intervals for 3 consecutive days (d0-2, d13-15, d27-29). One week after the final administration (d35), blood, feces, BALF (broncho-alveolar lavage fluid), and nasal lavage fluid were collected to measure Tir-specific antibody titers. Sera were prepared from the blood specimens by centrifugation (800 ×g, 20 min, 4°C) after clotting. Fecal pellets were homogenized (100 mg/ml) in PBS containing a complete protease inhibitor mixture (Roche Diagnostics) and, after centrifugation (20,000 ×g, 10 min, 4°C), the supernatants were collected as BALF. Serum, fecal extracts. The BALF was collected by inserting a cannula into the bronchus, injecting 1 ml of PBS, followed by collection by suction. After centrifugation (800 ×g, 20 min, 4°C), the supernatant was collected as BALF. Serum, fecal extract, BALF and nasal lavage fluid specimens were stored at -20°C until use for measurement of antibody titers.

For measurement of the Tir-specific antibody titers by ELISA, 96-well plates (MaxiSorp, Nunc) were coated with 100 ng/well of the purified GST-Tir. Sera, fecal extracts, and BALF diluted with PBST containing 5% skim milk were then added and incubated for 1 h at 37°C. HRP-conjugated polyclonal goat anti-mouse IgG, IgG1, IgG2b, IgG2c, IgG3 and IgA antibodies (all from Southern Biotech) were added and further incubated for 1 h at 37°C. Plates were developed using o-phenylenediamine substrate, the reaction was stopped by adding H₂SO₄, and OD_492-650 values were then measured.

Single cell suspensions from the spleen were prepared from immunized mice 30 days after the final immunization by mechanical disruption and red blood cell lysing. Cells were then suspended in CTL-test medium (ImmunoSpot). Plates for IFN-γ/IL-4 double color ELISPOT assay (ImmunoSpot) or for IL-17 ELISPOT assay (ImmunoSpot) were coated overnight at 4°C with the cytokine-specific capture antibodies. Freshly isolated single cell suspensions from the spleen (5 × 10⁵ cells/well) were plated with the purified GST-Tir (30 μg/ml). Following incubation for 24 h in an incubator at 37°C and 5% CO₂, the spots for IFNγ/IL-4 or IL-17 were developed according to the manufacturer’s instructions. Spot images were captured using a digital microscope (400-CAM057, SANWA-SUPPLY). The number of Tir-specific spot-forming cells (SFC) was counted using Image J (https://imagej.nih.gov/ij/).

Mice were administered with 2 × 10⁹ BLP-NanoLuc orally or intranasally and euthanized 2 or 6 h post-administration. The NALT (nasal-associated lymphoid tissue), PPs (Peyer's patches), mesenteric lymph nodes (mLNs), and the spleen were excised, washed in PBS, weighed, and then homogenized in 0.2 ml of PBS. To liberate any NanoLuc sequestered within the biological samples, the homogenates were completely disrupted by sonication. The debris was removed by brief centrifugation (10,000 ×g, 1 min, 4°C). The supernatant was mixed with the same volume of Nano-Glo Substrate (Promega), and the luciferase activities were then measured using a GloMax multiplate reader (Promega).

To label BLPs with FITC, 2.7 × 10¹⁰ BLPs were mixed with FITC (0.1 mg/ml) in 40 ml of carbonate buffer (0.1 M, pH9.0) overnight at 4°C with rotation. Unbound FITC was removed by centrifugation (10,000 ×g, 5 min, 4°C) and the labelled BLPs were washed 3 times with PBS. The FITC-labeled BLPs (BLP-FITC) in tissues were quantified according to the method reported by Eyles et al. (14). Mice were administered with 2 × 10⁹ BLP-FITC orally or nasally and euthanized at 2, 6, or 24 h post-administration. Excised tissues (NALT, PPs, mLNs, and spleen) were washed in PBS, weighed, and homogenized in PBS containing 1% paraformaldehyde (PFA). To homogenize the NALT, PPs, and mLNs, 1 ml of PBS/1% PFA was used, and 10 ml of PBS/1% PFA was used for the spleen. To liberate any BLP-FITC sequestered within the various tissues, the homogenates were further disrupted by sonication. The number of BLPs in disrupted tissue suspensions was enumerated by flow cytometer (FACSverse, BD).

Mice were administered with 2 × 10⁹ BLP-FITC orally or nasally and euthanized at 6 or 24 h post-administration. Single cell suspensions from the spleen were prepared as above. To stain dead cells, cells were treated with APC-Cy7-conjugated FVD780 (eBioscience). The cells were then incubated with TruStain FcX™ (anti-mouse CD16/32) Antibody (Biolegend) to block Fc receptor, followed by incubation with PE-conjugated anti-CD11b (M1/70, Biolegend) and APC-conjugated anti-CD11c (N418, Biolegend). The cells were analyzed by FACSverse (BD). At least 2.5 × 10⁵ live singlet cells were used for the analysis.

For transcriptomic analysis, total RNA was extracted from the PPs and NALT of mice at 6 h after the oral or nasal administration with BLP-Tir using RNAiso plus (TAKARA) and further purified using Nucleospin RNA (TAKARA) with DNase I treatment according to the manufacturer’s instructions. RNA library preparation was performed using the NEBNext Ultra RNA Library Prep Kit for Illumina (New England Biolabs) according to the manufacturer’s instructions. After assessing the library quality, sequencing was performed using an Illumina Novaseq 6000 with a read configuration of 150-bp paired-ended reads. A total of at least 20 million reads were generated for each sample. One biological replicate was present for each condition. The DNA Data Bank of Japan accession numbers for the RNA sequencing data reported in this study are DRR403116-DRR403119 (https://www.ddbj.nig.ac.jp/dra/index-e.html). Paired-ended clean reads were aligned to the Mus musculus reference genome (mm10, National Center for Biotechnology Information/University of California, Santa Cruz/Ensembl) using Hisat2 v2.0.5. The featureCounts v1.5.0-p3 program was used to count the read numbers mapped for each gene. Fragments per kilobase per million reads of each gene were calculated based on the length of the gene and read count mapped to the gene. Differential expression analysis between two conditions per group was performed using the DESeq2 R package (1.20.0). Genes with an adjusted P-value <0.05 found by DESeq2 were assigned as differentially expressed. Gene ontology (GO) enrichment analysis of differentially expressed genes (DEGs) was implemented using the clusterProfiler R package. GO terms with a corrected P-value <0.05 were considered significantly enriched by DEGs. mRNA levels of indicated genes were further validated by qRT-PCR using primers listed in Table 1 (n=3). To analyze the kinetics of gene expression changes, total RNA was extracted from the PPs and NALT before and at 6 h, 1 and 4 days after the 3 consecutive days administration of BLP-Tir and subjected to qRT-PCR (n=3).

Data were analyzed by the One-way ANOVA followed by Dunnett’s or Tukey’s post hoc multiple comparison test using GraphPad Prism 8 (GraphPad Software). P values <0.05 were considered significant.

We used Tir from C. rodentium as a model antigen to compare the adjuvant effect of BLPs between the oral and nasal administration routes. To maximize the effect of BLP, Tir was conjugated with BLPs as a Tir-LysM fusion protein. Any antigen that is fused to the well-known peptidoglycan-binding motif (LysM) to form an antigen-LysM fusion protein can strongly bind to BLPs in a non-covalent manner (5).

To examine the binding of BLP and Tir-LysM, BLP alone or BLP-Tir was stained with Tir-specific mouse serum and analyzed by confocal microscopy. Only BLP-Tir was stained with the Tir-specific serum, indicating the binding of BLP and Tir-LysM ( Figure 1A ). To test the binding capacity of BLPs with the Tir-LysM, BLP was incubated with a sufficient amount of Tir-LysM and subjected to SDS-PAGE and CBB staining. Densitometric analysis of the CBB-stained gel showed that the maximum binding capacity of BLPs with Tir-LysM was approximately 20 μg of Tir-LysM per 2 × 10⁹ BLPs ( Figure 1B ). A reporter gene assay using HEK293T-TLR2 reporter cells confirmed that BLPs functioned as a TLR2 ligand ( Figure 1C ).

To test the adjuvant effect of BLPs delivered via different routes, mice were orally or nasally administered with BLP-Tir, and then Tir-specific humoral and cellular immune responses were assessed. Oral immunization with the BLP-Tir induced a significantly higher level of Tir-specific serum IgG production, but not BALF IgG and fecal or BALF IgA ( Figure 2A ). On the other hand, in mice nasally immunized with BLP-Tir, the Tir-specific mucosal (fecal and BALF) IgA titers as well as serum and BALF IgG titers were increased significantly ( Figure 2A ). The Tir-specific serum IgG titer in nasally immunized mice was 1,000-fold higher than that in orally immunized mice ( Figure 2B ). These results showed that BLPs function effectively as a nasal adjuvant, but their effectiveness as an oral adjuvant is relatively low. Next, we examined the type of antigen-specific immune responses, both in terms of humoral and cellular responses, induced after the oral or nasal administration of BLP-Tir. To examine the type of humoral immune responses, IgG subclasses were analyzed. In mice nasally administered BLP-Tir, IgG2b was increased to the greatest degree, followed by IgG1 and IgG2c, with little increase observed for IgG3 ( Figure 2C ). Although there were no significant differences, the same tendency was observed between the mice administrated nasally and orally ( Figure 2C ). These results indicate that the administration route did not significantly affect the type of immune response induced, at least in terms of the systemic humoral response.

Next, cellular immune responses were assessed by ELISPOT assay using splenocytes. Mice were immunized in the same manner as above, and 30 days after the final administration, isolated splenocytes were stimulated ex vivo with recombinant GST-Tir, and the numbers of IFN-γ-, IL-4-, and IL-17-producing cells were measured. The numbers of IFN-γ- and IL-4-producing cells were not increased by either the oral or nasal administration of BLP-Tir in comparison with those for non-immunized mice ( Figure 2D ). However, the number of IL-17-producing cells was significantly increased only in the mice administered BLP-Tir via the nasal route ( Figure 2D ). These results indicate that the nasal administration of BLP-Tir mainly induces Th17-type, but not Th1- or Th2-type, cellular immune responses.

Next, we attempted to clarify the reason why the adjuvant effect of BLPs greatly differs between the oral and nasal administration routes. One of possible reasons for the weak adjuvant effect of the orally administered BLP-antigen complex is that the BLPs and/or antigen did not reach the gut-associated lymphoid tissue (GALT), the major induction site of antigen-specific immune responses against orally administered antigens (15). To examine whether the antigen reached the mucosal lymphoid tissue after the oral or nasal administration of a BLP-antigen complex, mice were administered BLP-NanoLuc, a luciferase-bound BLP, via the oral or nasal route, and the luciferase activities in PPs, mLNs, NALT and the spleen were measured. Significant luciferase activity was detected in the PPs and mLNs, but not in the spleen, at 2 h after the oral administration of BLP-NanoLuc ( Figure 3 ). The luciferase activity in the NALT after the nasal administration of BLP-NanoLuc was 1,000 times higher than those observed in the PPs and mLNs after the oral administration of BLP-NanoLuc. Additionally, significant luciferase activity was also detected in the spleen of mice administered BLP-NanoLuc via the nasal route ( Figure 3 ). In mice orally administered BLP-NanoLuc, decreased in luciferase activities were observed in the PPs and mLNs at 6 h post-administration ( Figure 3 ). In contrast, the luciferase activity in the spleen of mice nasally administered BLP-NanoLuc increased over time ( Figure 3 ). These results suggest that the oral administration of a BLP-antigen complex delivers only a small amount of antigen to the GALT in an intact form, but the nasal administration of the same BLP-antigen complex results in the delivery of a greater amount of the antigen to mucosal and systemic lymphoid tissues, with the antigen stayed in the tissues for a longer time.

Next, we compared the number of BLPs reaching the lymphoid tissue after their oral and nasal administration. To this end, FITC-labeled BLPs (BLP-FITC) were orally or nasally administered to mice, and the number of BLP-FITC conjugates contained in each tissue was examined by flow cytometry. The number of BLP-FITC conjugates detected in the PPs at 2, 6 and 24 h after the oral administration of BLP-FITC varied from below the detection limit to over 10³ particles, with no BLP-FITC detectable in the PPs at 24 h post-administration ( Figure 4A ). In the NALT from mice administered nasally, 10⁴-10⁶ BLP-FITC conjugates were detected at 2 h after administration ( Figure 4A ). This represents only 0.1–0.001% of the administered BLP-FITC, but 10–1000 times more than that detected in the PPs after oral administration ( Figure 4A ). Even at 6 or 24 h after the nasal administration, 10³-10⁵ BLP-FITC conjugates were still detected in the NALT ( Figure 4A ). However, only small numbers of BLP-FITC were detected in the mLNs and spleen after the oral administration and in the spleen after the nasal administration of BLP-FITC ( Figure 4A ). These results suggest that unlike conjugated antigens, few BLPs themselves migrate to the lymphatics or into systemic circulation regardless of the route of administration.

Previous reports suggest that nano- or micro-particles are taken up by dendritic cells (DCs) and are then translocated to systemic lymphoid tissues (16, 17). To investigate the possibility of BLP translocation via DCs to the spleen, BLP-FITC was orally or nasally administered to mice. Splenocytes were then isolated at 6 h post-administration, and the cells containing BLP-FITC was measured. However, no cells were observed to have taken up BLP-FITC in the spleen, suggesting that the BLP-FITC detected in the spleen of some mice may exist in cell-free state ( Figure 4B ). On the other hand, a significant increase in the proportions of CD11b⁺CD11c^-, CD11b⁺CD11c^int and CD11b⁺CD11c⁺ subsets were observed in the spleen of mice at 24 h after the nasal administration of BLP-FITC ( Figures 4C, D ). Furthermore, in the nasally administered mice, an increase in spleen weight was also observed ( Figure 4E ). These results suggested that the small numbers of BLPs translocated to the spleen induce an increase in at least the CD11b⁺ cell subsets in the spleen.

To clarify the reaction in the mucosal lymphoid tissue against BLPs, differences in gene expression in the PPs and NALT at 6 h after the oral and nasal administration of BLP-Tir were analyzed by RNA-seq. Oral administration of BLP-Tir induced an increase and decrease in the expression of 329 and 370 genes, respectively, in the PPs ( Figure 5A ; Supplementary Data 1 ). Nasal administration of BLP-Tir induced robust changes in gene expression in the NALT, with the expression of 1,112 genes upregulated and 3,963 genes downregulated ( Figure 5A ; Supplementary Data 2 ). However, no significant GO terms were extracted from the enrichment analysis of the differentially expressed genes (DEGs), with the expression levels of the cytokine gene Il-22, acute phase protein gene Saa1, and antimicrobial peptide genes Defa33, Reg3b, and Reg3g all upregulated in the PPs of mice orally administered with BLP-Tir ( Supplementary Data 1 ). In contrast, GO enrichment analysis of DEGs in the NALT from nasally administered mice resulted in the extraction of many immune response-related GO terms ( Figure 5B ). KEGG analysis also identified the immune response-related functions of the upregulated genes ( Figure 5C ).

To confirm the RNA-seq data and to examine the kinetics of gene expression, mice were administered BLP-Tir via the oral or nasal route for 3 consecutive days, and total RNA was extracted from the PPs or NALT before administration, at 6 h after the first administration, and at 1 and 3 days after the third administration, and then subjected to qRT-PCR ( Figure 6A ). Due to the large variability, no significant increase was observed in the gene expression levels of Saa1 and Reg3g, which were suggested to be upregulated in the PPs after the oral administration of BLP-Tir by RNA-seq analysis ( Figure 6B ). However, the expression levels of these genes clearly increased in some mice, suggesting considerable variability in the PPs among mice in response to orally administered BLP-Tir ( Figure 6B ). These results support the flow cytometric data indicating that a limited/variable number of BLPs reached the PPs after oral administration and suggest that the delivered BLPs elicited weak/transient antimicrobial responses in some mice. Kinetics analysis also showed that gene expression levels of cytokines and chemokines were sharply elevated upon the nasal administration of BLP-Tir, mostly returning to the basal level by 1 day after the 3 consecutive days of administration ( Figure 6B ). Tnf-a, Tgf-b, and Reg3g gene expression levels were not significantly changed in the NALT even after the nasal administration of BLP-Tir ( Figure 6B ). Overall, these results showed that the nasal, but not the oral, administration of BLP-Tir results in the elicitation of strong innate immune responses in the NALT.