Intranasal Immunization with Zika Virus Envelope Domain III-Flagellin Fusion Protein Elicits Systemic and Mucosal Immune Responses and Protection against Subcutaneous and Intravaginal Virus Challenges

Zika virus (ZIKV) infections in humans are mainly transmitted by the mosquito vectors, but human-to-human sexual transmission is also another important route. Developing a ZIKV mucosal vaccine that can elicit both systemic and mucosal immune responses is of particular interest. In this study, we constructed a recombinant ZIKV envelope DIII (ZDIII) protein genetically fused with Salmonella typhimurium flagellin (FliC-ZDIII) as a novel mucosal antigen for intranasal immunization. The results indicated that the FliC-ZDIII fusion proteins formulated with E. coli heat-labile enterotoxin B subunit (LTIIb-B5) adjuvant greatly increased the ZDIII-specific IgG, IgA, and neutralizing titers in sera, and the ZDIII-specific IgA titers in bronchoalveolar lavage and vaginal fluids. Protective immunity was further assessed by subcutaneous and intravaginal ZIKV challenges. The second-generation FliCΔD3-2ZDIII was shown to result in a reduced titer of anti-FliC IgG antibodies in sera and still retained the same levels of serum IgG, IgA, and neutralizing antibodies and mucosal IgA antibodies without compromising the vaccine antigenicity. Therefore, intranasal immunization with FliCΔD3-2ZDIII fusion proteins formulated with LTIIb-B5 adjuvant elicited the greatest protective immunity against subcutaneous and intravaginal ZIKV challenges. Our findings indicated that the combination of FliCΔD3-2ZDIII fusion proteins and LTIIb-B5 adjuvant for intranasal immunization can be used for developing ZIKV mucosal vaccines.


Introduction
Zika virus (ZIKV) is a small enveloped positive-strand RNA virus belonging to genus Flavivirus, family Flaviviridae [1]. The recent outbreak of ZIKV infections in Brazil in 2014 [2] was blamed for a 20-fold increase in neonatal microcephaly [3]. The ZIKV epidemic has continued to spread throughout South and Central America [4,5]. The World Health Organization declared it a "public health emergency of international concern" in 2016 [6]. Although the majority of ZIKV infections in humans are transmitted by Aedes mosquito vectors, such as A. aegypti and A. albopictus [7,8], ZIKV transmission through sexual partners is another important route for virus transmission in humans [9,10]. Several recent findings incubation with 1 mM IPTG for 4 h. The cells were centrifuged at 6000× g for 10 min at 4 • C. The resulting pellet was resuspended in 40 mL buffer A containing 300 mM Tris, 50 mM NaCl, 10 mM imidazole and 5% glycerol (pH7.2) and was homogenized using an ultrasonic homogenizer at 4 • C. Inclusion bodies were collected via centrifugation at 12,000× g for 10 min at 4 • C and then mixed with 8 M urea. The mixture was added to Toyopearl AF-Chelate-650 resin (Sigma Aldrich, St. Louis, MO, USA) and kept overnight, loaded into a column, and washed with 40 mL buffer A containing 0.5% Triton X-100. The recombinant proteins were eluted in 40% buffer B containing 300 mM Tris, 50 mM NaCl, 500 mM imidazole, and 5% glycerol (pH7.2) and dialyzed with phosphate-buffered saline (PBS). The purified proteins were concentrated through a 3 kD centrifugal filter (Sartorius, Göttingen, Germany) and stored at −20 • C. The molecular weight of the purified protein was checked using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Coomassie blue staining.

ELISA IgG and IgA Titer Assay
Enzyme-linked immunosorbent assay (ELISA) was used to measure ZDIII-specific IgG and IgA in mouse sera, BALFs and VFs. ELISA plates were coated overnight with 2 µg/mL ZDIII proteins at 4 • C and blocked with 1% bovine serum albumin (BSA) at room temperature (RT) for 1 h. Serial dilutions of the serum samples were added to the ELISA plates and incubated for 1 h at RT. After washing three times with PBS containing 0.05% Tween 20 (PBST), horseradish peroxidase (HRP)-conjugated anti-mouse IgG (1:30,000, Genetex, Irvine, CA, USA), or IgA antibodies (1:10,000, Bethyl, Montgomery, TX, USA) were added. The plates were washed with PBST, developed with TMB substrate (BioLegend, San Diego, CA, USA), and the reaction was stopped with 2N H 2 SO 4 . The absorbance was measured using an ELISA reader at 450 nm. The end-point titers were measured as the four-fold absorbance of the negative control. BALF total IgA was measured using an IgA Mouse Uncoated ELISA Kit (Invitrogen, Waltham, MA, USA) in accordance with the manufacturer's instructions.

Plaque Reduction Neutralization Test (PRNT)
Vero cells were seeded overnight at 5 × 10 5 cells/well in 6-well plates at 37 • C. Twofold serial dilutions of mouse serum samples were incubated with 100 plaque-forming units (PFUs) of ZIKV (PRVABC59, ATCC ® VR-1843) for 1 h at 37 • C and then added to 6-well plates. After incubation for 1 h at 37 • C, the infected cells were overlaid with 12% methylcellulose in MEM and incubated at 37 • C for 5-7 days. The cells were fixed and stained with staining buffer containing 2% formalin and 1% crystal violet. The plaques were counted, and the neutralizing antibody titers were determined as the serum dilution causing 50% reduction in plaque number (PRNT 50 ).

T Cell Cytokine ELISA Assay
Splenocytes were seeded at 5 × 10 6 cells/well and 10 6 cells/well, respectively, in 24-wells plates, and then treated with 10 µg/mL recombinant ZDIII proteins for stimulation for 72 h. The culture supernatants were collected and measured for IFN-γ, IL-4, and IL-17A, and IL22 titers using an ELISA MAX kit (BioLegend) according to the manufacturer's instruction (BioLegend). The diluted supernatants were incubated with the capture antibodies for 2 h at RT. The cytokines were detected using specific antibodies for 1 h and interacted with avidin-HRP for 30 min prior to determining coloration and end-point titers.

Subcutaneous Virus Challenge
Immunized mice received 2 mg of MAR1-5A3 (IFNAR1-blocking mAb) via the intraperitoneal route one day before virus challenge with 10 8 PFU of ZIKV via the subcutaneous route. The survival rate and clinical scores were recorded for 14 days. Clinical scores were calculated as follows: normal = 0; ruffled fur = 2; lethargy, pinched, hunched, wasp-waisted = 3; labored breathing, rapid breathing, inactive, neurological = 5; and dead = 10 [34,35]. For the measurement of virus titer in the brain, the challenged mice were sacrificed on day 6, and the brain samples were collected.

Intravaginal Virus Challenge
Immunized mice were administered with 2 mg of medroxyprogesterone acetate (DMPA) 5 days before the virus challenge, and 2 mg of MAR1-5A3 (IFNAR1-blocking mAb) via the intraperitoneal route one day before the virus challenge with 10 8 PFU of ZIKV via the subcutaneous route. The survival rate of the virus challenged mice was recorded for 14 days. On days 1, 3, 5, and 7, the VFs of the mice were collected by pipetting 20 µL of PBS into the vagina of the mice. ZIKV virus titers of the samples were determined via plaque assay.

Statistical Analysis
Statistical analyses were performed using the GraphPad Prism (GraphPad Software, Inc., San Diego, CA, USA). The statistical significance of differences between the groups was assessed using one-way analysis of variance (ANOVA) with Tukey's or Holm-Sidak multiple comparison tests. Differences with a p-value of less than 0.05 (*), 0.01 (**), and 0.001 (***) were considered statistically significant.

Expression, Purification, and Characterization of ZDIII and FliC-ZDIII Fusion Proteins
The cDNAs from the ZDIII gene of the Brazil Paraiba_01/2015 strain (GI:1032590576) and the flagellin gene of Salmonella typhimurium (GI:217062) were synthesized and constructed for encoding recombinant protein expression in E. coli ( Figure 1A). ZDIII and FliC-DIII recombinant proteins were purified using a nickel-chelating affinity column with Toyopearl AF-Chelate-650 resin. The purified recombinant proteins were analyzed using SDS-PAGE gels stained with Coomassie blue (13 kDa for ZDIII, 64 kDa for FliC-ZDIII) ( Figure 1B). The TLR5 activities of ZDIII and FliC-ZDIII recombinant proteins were analyzed in HEK-293A cells that had been transfected with a human TLR5-expression vector and a luciferase reporter vector. The results showed that recombinant FliC and FliC-ZDIII proteins triggered a dose-dependent increase in luciferase reporter TLR-5 activity in 293A cells ( Figure 1C). multiple comparison tests. Differences with a p-value 0.001 (***) were considered statistically significant.

Expression, Purification, and Characterization of ZDII
The cDNAs from the ZDIII gene of the Brazil Para and the flagellin (,) gene of Salmonella typhimurium (GI: structed for encoding recombinant protein expression FliC-DIII recombinant proteins were purified using with Toyopearl AF-Chelate-650 resin. The purified re using SDS-PAGE gels stained with Coomassie blue (1 ZDIII) ( Figure 1B). The TLR5 activities of ZDIII and FliC analyzed in HEK-293A cells that had been transfecte vector and a luciferase reporter vector. The results s FliC-ZDIII proteins triggered a dose-dependent increa tivity in 293A cells ( Figure 1C).   4 linker sequence (GSGSGSGS), were constructed into the pET-22b (+) vector with a C-terminal His-tag for recombinant protein expression. (B) ZDIII and FliC-ZDIII recombinant proteins produced by E. coli and purified using nickel-chelated affinity chromatography. Purified proteins were verified by SDS-PAGE staining with Coomassie blue. The molecular weight marker is indicated as the color ladder. (C) TLR5 signal functional assay was performed by co-culturing the 10-fold serial dilutions of FliC and FliC-ZDIII recombinant proteins with HEK 293A expressing TLR5 receptor and NF-κB reporter vector. The cells were disrupted and treated with neolite luciferase substrate. The TLR5 activity was measured by luciferase activity.
Serum samples were collected at two weeks after the first, second, or third dose immunization and analyzed for the titers of ZDIII-specific IgG and IgA antibodies using ELISA. The results showed that the titers of IgG antibodies and IgA antibodies increased by booster immunizations in the second-and third-dose immune sera ( Figure 2B,C). In the first-dose sera, IgG and IgA titers were barely detected for immunizations with FliC-ZDIII and FliC-ZDIII + LTIIb-B ( Figure 2B,C). Immunizations with the FliC-ZDIII fusion proteins with or without the use of LTIIb-B5 adjuvant greatly increased the elicitations of ZDIII-specific IgG and IgA titers in the second-or third-dose sera. However, no significant differences were observed between FliC-ZIII and FliC-ZDIII + LTIIb-B5 groups. We also measured the neutralizing antibodies (PRNT 50 titers, log2 scale) in the third dose sera. The results indicated that intranasal immunization with ZDIII + LTIIb-B, FliC-ZDIII, and FliC-ZDIII + LTIIb-B5 elicited detectable neutralizing antibodies (PRNT 50 titer), but these titers did not show any statistical differences among all groups ( Figure 2B).
Mucosal vaccination not only triggers the systemic immune responses, such as presence of neutralizing antibodies in sera, but also induces the mucosal immune responses (i.e., presence of secretory IgA) in mucosal fluids. BALFs and VFs were collected after third dose immunizations and analyzed for ZDIII-specific IgA titers using ELISA. The results indicated that intranasal immunization with FliC-ZDIII + LTIIb-B5 elicited approximately 1.5-log IgA titers in BALFs ( Figure 2D) and 1.5-to 2.0-log IgA titers in VFs ( Figure 2E). To further confirm the production secretory IgA in mucosal fluids, we measured the overall total IgA titers. Our results indicated that the differences in the amount of overall IgA titer was significant for the FliC-ZDIII and FliC-ZDIII + LTIIb-B5 immunized groups ( Figure 2D). Immunizations with FliC-ZDIII and FliC-ZDIII + LTIIb-B5 had higher titers of ZDIII-specific IgG antibodies in BALFs than other groups ( Figure 2D). The FliC-ZDIII + LTIIb-B5 immunized group was able to elicit ZDIII-specific IgA antibodies in VFs in mice ( Figure 2E).
To measure T cell responses elicited by intranasal immunizations, the spleens were collected three weeks after the third-dose immunization, seeded in 24-well plates, and then incubated with recombinant ZDIII proteins for stimulation for 72 h. The culture supernatants were collected and measured for IFN-γ, IL-4, IL17A, and IL-22 production using ELISA. The results indicated that the level of IFN-γ cytokine production in the spleen of the FliC-ZDIII + LTIIb-B5 group was significantly higher than those in the ZDIII, ZDIII + LTIIb-B5, and FliC-ZDIII, groups ( Figure 3A). However, the FliC-ZDIII and FliC-ZDIII + LTIIb-B5 immunized groups showed an increased but still relatively low level of IL-4 production, and no detectable levels were found for the ZDIII and ZDIII + LTIIb-B5 groups ( Figure 3B). For IL-17A and IL-22 production, we found that FliC-ZDIII and FliC-ZDIII + LTIIb-B5 had significant titers in the spleen compared to all other immunized groups ( Figure 3C,D). Therefore, intranasal immunization with FliC-ZDIII fusion protein induced potent Th1 and Th17 responses. μg LTIIb-B5 were immunized three doses via the intranasal route. Sera were collected two weeks after each dose immunization. VFs were collected two weeks after the 3rd dose immunization. The spleen and BALFs were collected three weeks after the 3rd dose immunization; (B) ZDIII-specific IgG titers in 1st dose, 2nd dose, and 3rd dose sera; and the ZIKV neutralizing antibody titers (PRNT-50) in 3rd dose sera; (C) ZDIII-specific IgA titers in 1st dose, 2nd dose, and 3rd dose sera; (D) ZDIIIspecific total IgA titers, the overall total IgA titers, and ZDIII-specific total IgG titers in BALFs after the 3rd dose immunization; (E) ZDIII-specific total IgA titers in VFs after the 3rd dose. Statistical test for multiple comparison for all groups except PBS were analyzed using one-way ANOVA with Tukey's or Holm-Sidak's multiple comparison tests. (*, p < 0.05 and ** p < 0.01). Error bars are plotted as standard deviation from the mean value.
Mucosal vaccination not only triggers the systemic immune responses, such as presence of neutralizing antibodies in sera, but also induces the mucosal immune responses (i.e., presence of secretory IgA) in mucosal fluids. BALFs and VFs were collected after LTIIb-B5 were immunized three doses via the intranasal route. Sera were collected two weeks after each dose immunization. VFs were collected two weeks after the 3rd dose immunization. The spleen and BALFs were collected three weeks after the 3rd dose immunization; (B) ZDIII-specific IgG titers in 1st dose, 2nd dose, and 3rd dose sera; and the ZIKV neutralizing antibody titers (PRNT-50) in 3rd dose sera; (C) ZDIII-specific IgA titers in 1st dose, 2nd dose, and 3rd dose sera; (D) ZDIII-specific total IgA titers, the overall total IgA titers, and ZDIII-specific total IgG titers in BALFs after the 3rd dose immunization; (E) ZDIII-specific total IgA titers in VFs after the 3rd dose. Statistical test for multiple comparison for all groups except PBS were analyzed using one-way ANOVA with Tukey's or Holm-Sidak's multiple comparison tests. (*, p < 0.05 and **, p < 0.01). Error bars are plotted as standard deviation from the mean value. + LTIIb-B5, and FliC-ZDIII, groups ( Figure 3A). However, the FliC-ZDIII and FliC-ZDIII + LTIIb-B5 immunized groups showed an increased but still relatively low level of IL-4 production, and no detectable levels were found for the ZDIII and ZDIII + LTIIb-B5 groups ( Figure 3B). For IL-17A and IL-22 production, we found that FliC-ZDIII and FliC-ZDIII + LTIIb-B5 had significant titers in the spleen compared to all other immunized groups ( Figure 3C,D). Therefore, intranasal immunization with FliC-ZDIII fusion protein induced potent Th1 and Th17 responses.

Protective Immunity in Immunized Mice by Subcutaneous and Intravaginal ZIKV Challenges
To further assess the protective immunity, the same intranasal immunization regimens were conducted in groups of BALB/c mice intranasally immunized with three doses

Protective Immunity in Immunized Mice by Subcutaneous and Intravaginal ZIKV Challenges
To further assess the protective immunity, the same intranasal immunization regimens were conducted in groups of BALB/c mice intranasally immunized with three doses of the FliC-ZDIII antigen at 40 µg per dose for 6 weeks, and the immunized mice were challenged at week 9 with 10 8 PFU of ZIKV (PRVABC59 strain) via the subcutaneous or intravaginal route. All of the immunized BALB/c mice received 2 mg MAR1-5A3 (IFNAR1-blocking mAb) by intraperitoneal injection one day before the virus challenge at week 9 ( Figure 4A). For intravaginal challenges, an additional injection of 2 mg of medroxyprogesterone acetate (DMPA) was administered via the subcutaneous route four days before the virus challenge ( Figure 4A). The antisera, BALFs, and VFs after the third dose but before the virus challenge indicated that both the FliC-ZDIII and FliC-ZDIII + LTIIb-B5 groups elicited increased titers of anti-ZDIII IgG, IgA and neutralizing antibodies in sera and IgA antibodies in BALFs and VFs as compared to the PBS-immunized group ( Figure 4B,C). For the subcutaneous virus challenge, the survival rate recorded at 14 days post-challenge was 60% (FliC-ZDIII), 33% (FliC-ZDIII + LTIIb-B5), and 20% (PBS) ( Figure 4D). The number of the immunized mice in the FliC-ZDIII + LTIIb-B5 group was n = 3 due to the loss of two mice during the second and the third dose immunization periods before the subcutaneous virus challenge, thus resulting in a 33% survival rate ( Figure 4D). However, the survival rate has no statistically significant difference among these three groups ( Figure 4D). The clinical scores recorded for 14 days after the subcutaneous virus challenge were relatively lower for both the FliC-ZDIII and FliC-ZDIII + LTIIb-B5 groups than in the PBS group ( Figure 4D). For the intravaginal challenge, the survival rate was recorded at 14 days after the virus challenge, and VFs were collected on one, three, and five days after virus inoculation and virus titers were measured. The survival rates for all groups were 100% at 14 days after the virus challenge (data not shown). However, the virus titers in VFs were significantly lower for the FliC-ZDIII + LTIIb-B5 group, followed by the FliC-ZDIII group, than the PBS group, one, three, and five days after virus inoculation ( Figure 4E).

Immune Responses Elicited by the Second-Generation FliCΔD3-2ZDIII and FliCΔD2ΔD3-3ZDIII Fusion Proteins to Reduce FliC-Specific Adaptive Response
We further constructed the second-generation FliC-ZDIII fusion proteins by replacing FliC D3 with another ZDIII (FliCΔD3-2ZDIII) or replacing FliC D2 and D3 with two

Figure 5. Expression and characterization of second-generation FliCΔD
FliCΔD2ΔD3-3ZDIII antigens. (A) FliCΔD3-2ZDIII and FliCΔD2ΔD3-3ZDIII were pET-22b (+) vector for second-generation antigen expression; (B) FliCΔD3-2ZDIII an 3ZDIII were produced by E. coli and purified using nickel-chelated affinity chroma fied proteins were verified by SDS-PAGE staining with Coomassie blue. The color lecular weight marker; (C) TLR5 functionality in recombinant FliCΔD3-2ZDIII and 3ZDIII proteins were confirmed using the TLR5 assay. Error bars are plotted as stan from the mean value.
The same immunization regimen was conducted for these two second-g tigens, 40 μg FliCΔD3-2ZDIII and FliCΔD2ΔD3-3ZDIII without or with 3 μg juvant. The results indicated that the anti-ZDIII IgG titers in the sera for 2ZDIII and FliCΔD2ΔD3-3ZDIII groups without or with LTIIb-B5 adjuvant (not statistically significant) compared the FliC-ZDIII immunized groups (F anti-ZDIII IgA titers in sera had almost the same titers for all immunized for FliCΔD2ΔD3-3ZDIII + LTIIb-B5 ( Figure 6A). For serum neutralizing a found that the PRNT50 titer of the FliCΔD2ΔD3-3ZDIII + LTIIb-B5 group compared to the FliC-ZDIII and FliCΔD3-2ZDIII groups with LTIIb-B5 adj The same immunization regimen was conducted for these two second-generation antigens, 40 µg FliC∆D3-2ZDIII and FliC∆D2∆D3-3ZDIII without or with 3 µg LTIIb-B5 adjuvant. The results indicated that the anti-ZDIII IgG titers in the sera for the FliC∆D3-2ZDIII and FliC∆D2∆D3-3ZDIII groups without or with LTIIb-B5 adjuvant were reduced (not statistically significant) compared the FliC-ZDIII immunized groups ( Figure 6A). The anti-ZDIII IgA titers in sera had almost the same titers for all immunized groups except for FliC∆D2∆D3-3ZDIII + LTIIb-B5 ( Figure 6A). For serum neutralizing antibodies, we found that the PRNT50 titer of the FliC∆D2∆D3-3ZDIII + LTIIb-B5 group was reduced compared to the FliC-ZDIII and FliC∆D3-2ZDIII groups with LTIIb-B5 adjuvant ( Figure 6A). Even without the use of LTIIb-B5 adjuvant, the PRNT50 titer of the FliC∆D2∆D3-3ZDIII group was lower than that of the FliC-ZDIII + LTIIb-B5 and FliC∆D3-2ZDIII + LTIIb-B5 groups ( Figure 6A). We also determined the anti-FliC antibodies in the sera after three-dose intranasal immunizations, and all the results followed the order of FliC-ZDIII > FliC∆D3-2ZDIII > FliC∆D2∆D3-3ZDIII with or without the use of LTIIb-B5 adjuvant ( Figure 6B). For mucosal IgA production, we measured the anti-ZDIII IgA titers in BALFs and VFs. Results from all FliC-ZDIII, FliC∆D3-2ZDIII and FliC∆D2∆D3-3ZDIII immunized groups without or with LTIIb-B5 adjuvant showed no significant differences in IgA titers in VFs ( Figure 6C). In BALFs, the anti-ZDIII IgA titers of the three antigens with the use of LTIIb-B5 adjuvant were slightly higher than the titers without LTIIb-B5 adjuvant but no significant differences were observed among FliC-ZDIII, FliC∆D3-2ZDIII and FliC∆D2∆D3-3ZDIII antigens ( Figure 6C). The immunized mice were further intravaginal challenged with ZIKV, and the results showed that the virus titers in VFs on days 1, 3 and 5 were significantly lower for the FliC-ZDIII, FliC∆D3-2ZDIII, and FliC∆D2∆D3-3ZDIII groups than the PBS control group ( Figure 6D). Therefore, the second-generation FliC∆D3-2ZDIII and FliC∆D2∆D3-3ZDIII antigens resulted in a reduced titer of anti-FliC IgG antibodies in sera but still retained the same titers of serum IgG, IgA, and neutralizing and mucosal IgA antibodies and protective immunity Pharmaceutics 2022, 14, x FOR PEER REVIEW 12 of 19 but no significant differences were observed among FliC-ZDIII, FliCΔD3-2ZDIII and FliCΔD2ΔD3-3ZDIII antigens ( Figure 6C). The immunized mice were further intravaginal challenged with ZIKV, and the results showed that the virus titers in VFs on days 1, 3 and 5 were significantly lower for the FliC-ZDIII, FliCΔD3-2ZDIII, and FliCΔD2ΔD3-3ZDIII groups than the PBS control group ( Figure 6D). Therefore, the second-generation FliCΔD3-2ZDIII and FliCΔD2ΔD3-3ZDIII antigens resulted in a reduced titer of anti-FliC IgG antibodies in sera but still retained the same titers of serum IgG, IgA, and neutralizing and mucosal IgA antibodies and protective immunity

Intranasal Immunization with the Use of LTIIb-B5 Adjuvant for the Second-Generation FliC∆D3-2ZDIII Fusion Proteins
We further used LTIIb-B5 adjuvant formulated with the second-generation FliC∆D3-2ZDIII fusion protein to enhance mucosal immunity. Three immunized groups (PBS, FliC∆D3-2ZDIII, and FliC∆D3-2ZDIII + LTIIb-B5) were administered three doses of intranasal immunization at weeks 0, 3, and 6, and all of the immunized mice were challenged with ZIKV virus by the subcutaneous or intravaginal route at week 9. Intranasal FliC∆D3-2ZDIII immunization with the use of LTIIb-B5 adjuvant increased the anti-ZDIII IgG, IgA, and neutralizing antibodies in sera ( Figure 7A). Moreover, the FliC∆D3-2ZDIII + LTIIb-B5 group elicited a significantly higher titer of anti-ZDIII IgA in VFs ( Figure 7B). Therefore, the use of LTIIb-B5 adjuvant with the FliC∆D3-2ZDIII fusion protein for intranasal immunization improved systemic and mucosal immune responses. The protection was further assessed in these immunized mice following the subcutaneous and intravaginal challenges. For the subcutaneous virus challenge, the survival rates for the FliC∆D3-2ZDIII + LTIIb-B5 group was 100% compared to 60% for the FliC∆D3-2ZDIII group and 40% for the PBS group ( Figure 7C). The virus titers in brain tissue of the FliC∆D3-2ZDIII + LTIIb-B5 group at six days post-infection were significantly lower than those in the FliC∆D3-2ZDIII and PBS groups ( Figure 7D). Notably, the numbers of immunized mice (n = 5 per group) six days after subcutaneous virus challenge dropped to n = 2 for the PBS group and dropped to n = 3 for the FliC∆D3-3ZDIII group. For the intravaginal virus challenge (the survival rates for all immunized groups were 100%), the virus titers in VFs at five days post challenge were lower for the FliC∆D3-2ZDIII and FliC∆D3-2ZDIII + LTIIb-B5 groups compared to the PBS group ( Figure 7E). Therefore, intranasal immunization with three doses of FliC∆D3-2ZDIII + LTIIb-B5 was found to elicit systemic and mucosal anti-ZIKV immunity and protection against subcutaneous and intravaginal virus challenges.

Discussion
Intranasal administration is an effective mucosal vaccine delivery route to eli gen-specific systemic and mucosal immune responses [38][39][40]. Antigens can be nized in nasopharynx-associated lymphoid tissues that contain M cells, antigen-p ing cells, T cells, and B cells to trigger mucosal immune responses such as throu secreting B cells (or plasma cells) [38][39][40]. In this study, we constructed FliC-ZDII

Discussion
Intranasal administration is an effective mucosal vaccine delivery route to elicit antigen-specific systemic and mucosal immune responses [38][39][40]. Antigens can be recognized in nasopharynx-associated lymphoid tissues that contain M cells, antigen-presenting cells, T cells, and B cells to trigger mucosal immune responses such as through IgA-secreting B cells (or plasma cells) [38][39][40]. In this study, we constructed FliC-ZDIII fusion proteins for intranasal immunization and the results demonstrated the elicitation of both systemic and mucosal responses and protection against subcutaneous and intravaginal ZIKV challenges. We previously reported the success for expression of the FliC and four-serotype dengue virus (DENV1-4) DIII fusion proteins in E. coli and showed that the FliC-DENV2 DIII fusion protein enhanced the neutralizing antibody titers up to 100 folds than the titer obtained from the mixture of FliC and DENV2 DIII alone [32]. Direct fusion of FliC with ZDIII provides the advantage for a simultaneous delivery of antigen and adjuvant together to the same antigen-presenting cells, eliminates the FliC protease sensitivity and tendency to aggregate, and induces a strong affinity for TLR5 receptors to trigger more localized immune stimulations [41]. The FliC molecule consists of four globular domains (D0, D1, D2, and D3): the D0 domain contains the N'-terminal and C-terminal elements to trigger the NLRC4 inflammasome [42,43]; the D1 domain contains a critical amino acid residue from three helices to interact with TLR5 [44][45][46]; the hypervariable D2 and D3 domains are immunodominant and can be deleted to abrogate intrinsic antigenicity of flagellin without affecting adjuvant stimulatory activity [47]. In this study, we constructed the secondgeneration antigens of FliC∆D3-2ZDIII (by deleting FliC D3 and replacing with 1 × ZDIII) and FliC∆D2∆D3-3ZDIII (by deleting FliC D2 and FliC D3 and replacing with 2 × ZDIII) ( Figure 5). Our results indicated that immunization with only the second-generation FliC∆D3-2ZDIII antigen resulted in a reduced titer of anti-FliC IgG antibodies in the sera ( Figure 6B) and still retained the same levels of serum IgG, IgA, and neutralizing antibodies and mucosal IgA antibodies ( Figure 6A,C). As the bacterial flagellin is an antigen commonly expressed by commensal and pathogenic bacteria in the guts, the flagellin-specific adaptive immunity may link to the development of chronic relapsing intestinal inflammation of the inflammatory bowel diseases such as Crohn's disease and ulcerative colitis [48][49][50]. Therefore, the second-generation FliC∆D3-2ZDIII antigen can reduce FliC-specific adaptive response without compromising the vaccine antigenicity.
Groups of BALB/c mice were immunized intranasally with FliC-ZDIII fusion proteins for a three-dose regimen in a three-week interval. The titers of anti-ZDIII IgG and IgA antibodies were detected in the first-dose sera, and gradually increased to higher values in the second and third dose sera from the groups immunized with FliC-ZDIII and FliC-ZDIII + LTIIb-B5 as compared to the relatively low titers detected in the groups immunized with ZDIII and ZDIII + LTIIb-B5 ( Figure 2B,C). The results also demonstrated titers of the ZDIII-specific IgG and IgA antibodies in BALFs and VFs only after the thirddose immunization for the groups immunized with FliC-ZDIII and FliC-ZDIII + LTIIb-B5 ( Figure 2D,E). Since mucosal vaccinations generally require multiple booster doses to induce effective immune responses, three-dose intranasal regimens were reported using influenza hemagglutinin proteins incorporated into chitosan nanoparticles [51,52] or formulated with recombinant flagellin in oil-in-water emulsions [53]. Our studies showed that three-dose intranasal immunization was required to elicit potent anti-ZIKV antibody responses in the sera, BALFs, and VFs. We conducted the three-dose regimen for intranasal immunizations with 20 µg FliC-ZDIII without or with 1 µg LT-IIbB5 (Figure 2). We further increased the antigen dose to 40 µg FliC-ZDIII and the amount of 3 µg LTIIb-B5 adjuvant to assess the protection (Figure 4). The resulted showed an approximately six-fold difference in neutralizing antibody titers (PRNT-50 values) in the third dose sera in between groups of mice immunized with FliC-ZDIII or FliC-ZDIII + LTIIb-B5. However, the IgG and IgA titers in the third dose sera were similar in between groups of mice immunized with FliC-ZDIII or FliC-ZDIII + LTIIb-B5. Compared to the results that we previously reported using FliC-ZDIII for two-dose intramuscular immunization or the first dose adenovirus vector-priming and the second dose FliC-ZDIII booster in BALB/c mice (33), the three-dose intranasal immunization with FliC-ZDIII and LTIIb-B5 adjuvant elicited higher titers of neutralizing antibodies in sera ( Figures 2B, 4B, 6A, and 7A). Therefore, intranasal immunization with FliC-ZDIII fusion protein for a three-dose regimen can elicit a comparable systematic immune response similarly to intramuscular immunization. To our knowledge, this is the first report using mucosal delivery of ZIKV recombinant vaccines. Intranasal immunizations with the FliC-ZDIII fusion proteins induced potent Th1 and Th17 responses as compared to the ZDIII proteins alone (Figure 3). The results indicated that the FliC-ZDIII fusion proteins even without the use of LTIIb-B5 adjuvant elicited approximately the same Th17 cellular responses for IL-17A and IL-22 production in the spleen ( Figure 3C,D). Mucosal delivery of antigens with flagellin adjuvant has been reported to stimulate TLR5 signaling in dendritic cells to activate processing, presentation, co-stimulatory function and secretion of IL-12, IL-23, or IL-6, thereby promoting the activation of Th1 and Th17 cellular and antibody-mediated immunity as well as innate lymphoid cells [54][55][56]. Flagellin was found to activate TLR-5 signaling in airway epithelial cells, and to trigger proinflammatory responses to recruit neutrophils and dendritic cells to mucosal sites [57] or recapitulate the transcriptional signature of lung responses [58]. The adjuvant potential of flagellin by intranasal delivery was dependent on TLR-5 but not on NLRC4 in radioresistant lung cells [57]. However, a recent finding using MVA vaccinia vector encoding flagellin demonstrated that flagellin can induce NLRC4 inflammasome response to enhance secretory IgA production in the lung and intestinal mucosa [59]. Additional investigations are required to understand the modes of action for the FliC-ZDIII fusion proteins to trigger mucosal immune response. Moreover, we did not measure the CTL response elicited by intranasal immunizations with FliC-ZDIII plus LTIIb-B5 adjuvant in the present study. Further examination is needed to determine the cell-mediated immunity may also contribute to an effective protective immune response.
We assessed the protective immunity using the immunocompetent BALB/c mice administered IFNAR-1-blocking mAb before subcutaneous or intravaginal challenge with ZIKV ( Figure 4A). This IFNAR1-blocking model can retain a competent immune system in the immunized mice to investigate the elicited immunity for protection, rather than the use of the immune-incompetent mice such as Ifnar1−/− mice or AG129 mice (interferon-α/β and -γ receptor-knock-out mice) [60][61][62]. Our results indicated that intranasal immunization with FliC-ZDIII and FliC∆D3-2ZDIII fusion proteins protected from subcutaneous live virus challenges, resulting in a 60% survival rate and relatively lower clinical scores ( Figures 4D and 7C). We also investigated the protection of the FliC-ZDIII and FliC∆D3-2ZDIII immunized mice through intravaginal challenge, by treating mice with DMPA to synchronize the mouse estrus cycle at a prolonged diestrus phase, rendering mice susceptible to vaginal virus infection [63]. Our results showed that immunization with FliC-ZDIII and FliC∆D3-2ZDIII fusion proteins lowered the viral load of ZIKV titer in vaginal washes on day 5, as compared to the control group ( Figures 4E and 7D).

Conclusions
In this study, we constructed FliC-ZDIII fusion protein as a mucosal vaccine candidate and characterized for its elicitation of ZDIII-specific IgG, IgA, and virus neutralizing antibodies in sera, BALFs, and VFs. We further constructed the second-generation FliC∆D3-2ZDIII fusion protein to reduce the elicitation of anti-FliC IgG antibodies in sera without compromising the vaccine antigenicity. Intranasal immunization with the secondgeneration FliC∆D3-2ZDIII fusion proteins formulated with LTIIb-B5 elicited the greatest protective immunity against subcutaneous and intravaginal ZIKV challenges. Our findings indicated that the combination of FliC∆D3-2ZDIII fusion proteins and LTIIb-B5 adjuvant for intranasal immunization can be a viable candidate for developing ZIKV mucosal vaccines.

Institutional Review Board Statement:
All experiments were conducted in accordance with the guidelines of the Laboratory Animal Center of the National Tsing Hua University (NTHU). Animal use protocols were reviewed and approved by the NTHU Institutional Animal Care and Use Committee (approval no. 10533).
Informed Consent Statement: Not applicable.