Co-Administration of a Plasmid Encoding CD40 or CD63 Enhances the Immune Responses to a DNA Vaccine Against Bovine Viral Diarrhea Virus in a Mouse Model


 Bovine viral diarrhea virus (BVDV) causes substantial economic losses in the livestock industry worldwide. Plasmids encoding the BVDV E2 protein are potential DNA vaccines against BVDV, but their immunogenicity has been insufficient. Here, we investigated the adjuvant effect of CD40 and CD63 on the immune responses to a BVDV E2 DNA vaccine in a mouse model. We constructed pUMVC4a-based plasmids encoding the BVDV E2 protein (pE2), mouse CD40 (pCD40), or mouse CD63 (pCD63). Protein expression by each plasmid was confirmed through Western blot analysis and immunofluorescence staining of cultured cell lines. BALB/c mice were immunized intradermally twice with pE2 in combination with, or without, pCD40 or pCD63, with 3 weeks between the two doses. pE2 with pCD40 induced significantly higher neutralizing antibody titers against BVDV than pE2 alone. Furthermore, pE2 with pCD40 or pCD63 induced significantly increased lymphocyte proliferation and IFN-γ production in response to BVDV ex vivo, compared with E2 alone. These results suggest that a plasmid encoding CD40 or CD63 can be used as an adjuvant to enhance immune responses to DNA vaccines against BVDV.


Introduction
Bovine viral diarrhea virus (BVDV), which belongs to the Flaviviridae family and genus Pestivirus, is an enveloped single-stranded positive-sensed RNA virus [1,2]. BVDV is mainly a pathogen of cattle, but it also infects buffaloes, deer, pigs, sheep, goats, yaks, and wild ruminants [3]. The clinical manifestations of BVDV are acute infection, fetal infection, and mucosal disease. Fetal infections are particularly associated with congenital malformations and abortion. BVDV can be transmitted to newborn calves through the placenta, potentially resulting in persistent infection in calves that can transmit the virus throughout their lives [4][5][6]. BVDV infection has caused huge economic losses to the global animal husbandry industry through reduced milk production, abortions, and the decreased lifespan of infected animals [7]. Fatality losses due to BVDV have been estimated at US$5.26 per infected animal, with performance losses at US$88.26 per infected animal [8].
Vaccination is used to control BVDV infection [9]. However, several problems exist in the current BVDV vaccines. Although modi ed live virus (MLV) vaccines induce greater immune responses and longer duration of protection than killed virus (KV) vaccines, MLV vaccines are at risk of generating a pathogenic revertant and vertical transmission in pregnant cattle. KV vaccines do not carry such a risk, but they requires multiple immunizations to achieve protection against infection [4]. Furthermore, unlike MLV vaccines, KV vaccines do not induce cell-mediated immune responses e ciently [10][11][12].
DNA vaccines are plasmids that encode protective antigens. Compared with conventional MLV and KV vaccines, DNA vaccines are more stable, easier to manufacture, and safer during handling [13]. DNA vaccines induce not only humoral immune responses but also cell-mediated immune responses by directly expressing antigens on antigen-presenting cells [14]. DNA vaccines have been investigated for various applications, including therapy of cancer [15], allergies [16], autoimmunity, [17] and infectious diseases [18]. Some DNA vaccines have been approved for veterinary use, including those against West Nile virus in horses, canine melanoma, growth hormone releasing hormone in pigs, and infectious hematopoietic necrosis virus in sh [19][20][21][22][23]. DNA vaccines encoding BVDV E2 protein have been shown to induce virus neutralizing antibodies [24][25][26]. However, their immunogenicity remains insu cient.
The co-stimulatory molecule CD40 is a member of the tumor necrosis factor (TNF) receptor superfamily and is constitutively expressed on the surfaces of diverse immune and non-immune cell types, including B cells, dendritic cells, macrophages, and endothelial cells [27]. CD40 directly binds TRAF2, TRAF3, TRAF5, and TRAF6 and indirectly associates with TRAF1 [28]. These interactions result in transcription factor activation, activation of mitogen and stress-activated protein kinase cascades, cytokine secretion, proliferation, differentiation of B cells into Ig-secreting plasma cells, and formation of humoral memory. Anti-CD40 antibodies have enhanced immune responses to DNA or synthetic peptide vaccines in mouse models [29][30][31][32], and the co-administration of plasmids encoding CD40 has enhanced humoral and cellular immune responses to DNA vaccines [33][34][35].
CD63 is a member of the tetraspan family and is expressed on various cellular membranes, including extracellular vesicles. Also known as lysosomal-associated membrane protein 3, CD63 is involved in antigen presentation [36][37][38]. CD63 represents an activation-induced reinforcing element, the triggering of which promotes sustained and e cient T-cell activation and expansion [39]. A plasmid encoding an ovalbumin (OVA)-CD63 fusion protein enhanced cellular immune responses to OVA, whereas coadministration of a plasmid encoding CD63 did not enhance cellular immune responses to the OVA encoded by another plasmid [40].
In this study, we used a mouse model to investigate whether co-administration of a plasmid encoding CD40 or CD63 enhanced immune responses to a BVDV E2 DNA vaccine.

Cell lines and virus
MDBK and HEK-293T cells were cultured in Dulbecco's Modi ed Eagle's Medium (Nissui, Tokyo, Japan). Bovine fetal muscle cells within 20 passages were cultured in Eagle's Minimal Essential Medium (Nissui) supplemented with 10% tryptose phosphate broth. CHO-K1 and P3U1 cells were cultured in RPMI 1640 medium (Nacalai Tesque, Kyoto, Japan). These three media were supplemented with 5% heat-inactivated fetal bovine serum (FBS; Thermo Fisher Scienti c, Waltham, MA, USA), 100 units/mL penicillin (Nacalai Tesque), 100 μg/mL streptomycin (Nacalai Tesque), and 2 mM l-glutamine (Thermo Fisher Scienti c). Cells were cultured at 37°C in a humidi ed atmosphere containing 5% CO 2 and 95% air. The FBS used was con rmed to be free of anti-BVDV antibodies by using a virus neutralizing assay (described below). BVDV 1 (Nose strain; GenBank accession no. AB078951) was propagated in MDBK cells. At 72 to 96 h after infection at an MOI of 0.1 to 0.01, the cells underwent three freeze-thaw cycles; cell debris were removed by centrifugation at 1710 × g for 15 min at 4°C. The virus titer was determined according to the 50% tissue culture infective dose (TCID 50 ) for bovine fetal muscle cells in 96-well at culture plates. The virus stock was stored at -70°C until use.

Construction of DNA plasmids
For the construction of plasmids encoding mouse CD40 (pCD40) or CD63 (pCD63), total RNA was extracted from kidney or splenocytes, respectively, of BALB/c mice by using TRIzol reagent (Thermo Fisher Scienti c). For construction of a plasmid encoding BVDV E2 protein (pE2), TRIzol-LS reagent (Thermo Fisher Scienti c) was used to extract total RNA from the culture supernatant of MDBK cells infected with BVDV 1 Nose strain. A PrimeScript RT Reagent Kit (Takara Bio, Shiga, Japan) was used for reverse transcriptions of mRNA or the viral RNA genome. cDNA was synthesized by incubating at 37°C for 15 min and 85°C for 5 s in a T100 Thermal Cycler (BioRad, Hercules, CA, USA).
The BVDV E2 gene was PCR-ampli ed by using a speci c primer set that was designed to amplify a partial E2 fragment that lacked the transmembrane region located at the 3′ end (33 amino acids). Furthermore, the forward primer was located in the C-terminal hydrophobic region of glycoprotein E1 so that the construct contained 16 amino-acid residues as a leader peptide. The primer sequences were (restriction enzyme sites are underlined): E1 forward primer (ApaI + E1), 5′-GTGGGCCCACCATGGTACAGGGCATCCTATGGCTA-3′ and E2 reverse primer (KpnI + E2), 5′-ACGGTACCTCAAGCGAAGTAATCCCGGT-3′. PCR ampli cations were performed by using KOD One PCR Master Mix (Toyobo). Reaction mixtures consisted of 40 ng/1 μL of cDNA, 1 μL each of 10 μM forward and reverse primers, 5 μL of KOD One (Toyobo), and 2 μL of sterilized ultrapure water. Ampli cations were performed under the following conditions: initial denaturation at 94°C for 2 min, followed by 30 cycles of denaturation at 98°C for 10 s, annealing at 63°C for 5 s, and extension at 68°C for 25 s.
Each ampli ed DNA fragment was enzyme-digested and inserted into pUMVC4a expression vector (Aldevron, South Fargo, ND, USA) to yield pE2, pCD40, or pCD63. The plasmid constructs were veri ed by DNA sequencing (3500 Genetic Analyzer, Applied Biosystems, Carlsbad, CA, USA). The plasmids were ampli ed in E. coli DH5α cells and puri ed by using GenElute HP Endotoxin-Free Plasmid Maxiprep Kit (Sigma-Aldrich, St. Louis, MO, USA) according to the manufacturer's instructions. Plasmids were eluted by using nuclease-free ultrapure water and diluted to a concentration of 3 to 10 μg/μl.

Expression of recombinant BVDV E2 protein in E. coli
The partial E2 gene (492 bp corresponding to the N-terminal 164 amino acids) of BVDV 1 Nose strain with N-terminal and C-terminal 6× histidine tags was synthesized by using codon optimization (Euro ns Genomics, Tokyo, Japan). This partial E2 fragment was restriction digested and cloned into pET-41a expression vector (Merck, Darmstadt, Germany). The resulting plasmid pET-41a/E2 was used to transform E. coli BL21 (DE3) cells (Merck). To induce the expression of recombinant BVDV E2 protein (rE2), isopropyl-β-d-thiogalactopyranoside ( nal concentration, 1.0 mM) was added to the culture, which was then incubated for 6 h at 30°C. Inclusion bodies containing rE2 were sonicated and solubilized in phosphate-buffered saline (PBS) containing protein denaturants (8 M urea and 10 mM bmercaptoethanol). rE2 was a nity puri ed on Co 2+ -charged TALON resin (Takara Bio) under denaturing conditions, according to the manufacturer's instructions. This was followed by buffer exchange to remove denaturants. The resulting rE2 protein was used as the antigen for ELISA and the production of monoclonal antibodies (mAbs).
Hybridoma production BALB/c mice were injected intraperitoneally with rE2 (100 μg/dose); Imject Alum (Thermo Fisher Scienti c) was used as an adjuvant. After three booster immunizations at 1-week intervals, splenocytes were harvested and then fused with P3U1 cells by using PEG1500 (Roche Diagnostics, Indianapolis, IN, USA). The resulting hybridomas were grown in RPMI 1640 medium supplemented with 10% FBS and hypoxanthine-aminopterin-thymidine selection medium supplement (Thermo Fisher Scienti c). Hybridomas were chosen on the basis of the reactivity of their culture supernatants against rE2 in ELISAs. We established two hybridoma clones: E2-1a-1 and E2-1a-6. The reactivity of these mAbs was con rmed by Western blot analysis and immuno uorescence staining. Protein G Sepharose 4 Fast Flow (Cytiva, Marlborough, MA, USA) was used to purify the mAbs from the supernatants of hybridomas cultured in Hybridoma-SFM medium (Thermo Fisher Scienti c).

DNA vaccination
Speci c-pathogen-free, age (8 weeks)-matched female BALB/c mice were purchased from Japan SLC (Shizuoka, Japan) and kept under standard conditions. All procedures and animals used in this study were approved by the Iwate University Animal Care and Use Committee (no. A201909).
The mice were anesthetized through intraperitoneal injection of a combination of anesthetics (0.3 mg/kg medetomidine, 4.0 mg/kg midazolam, and 5.0 mg/kg butorphanol). BALB/c mice (n = 50) were randomly divided into 10 groups; details of the vaccinated groups are shown in Table 1. pCD40 or pCD63 (0.25, 0.5, 1.0, or 2.0 μg/μL) was mixed with pE2 (2.0 μg/μL), and 20 μL of the resulting plasmid solution was administered intradermally at two different sites. The mice received a booster immunization at 3 weeks after the initial immunization; 2 weeks after the booster immunization, blood samples were collected, and serum was analyzed for humoral immune responses. Splenocytes were harvested at this same time, and BVDV E2 antigen-speci c cell proliferation and IFN-g production were evaluated, as described below.

Humoral immune responses
Virus neutralization titers were determined by using a 96-well microplate assay. Sera were inactivated for 30 min at 56°C. Two-fold serially diluted serum samples were mixed with an equal volume of a BVDV 1 (Nose strain) suspension containing 200 TCID 50 /25 µL; the resulting solutions were placed in microplate wells, and the plate was incubated at 37°C for 60 min. After this incubation, bovine fetal muscle cells (10 5 cells/well) were added to each mixture, which was incubated at 37°C in a humidi ed atmosphere of 5% CO 2 in air for 5 days. Each serum dilution was titered twice. The neutralizing antibody titer was expressed as the reciprocal of the highest dilution of the test serum that completely prevented cytopathic effects.
The enzymatic reaction was conducted by using 1-Step Ultra TMB-ELISA (Thermo Fisher Scienti c). Optical density was measured at a wavelength of 450 nm on an In nite 200 PRO multimode plate reader (Tecan Group, Männedorf, Switzerland). These reactions were performed at 37°C, using a total sample volume of 50 µL.
To determine BVDV-speci c IFN-γ production, splenocytes were cultured for 72 h in the presence of BVDV 1 as described for the proliferation assay; the culture supernatants were then collected and stored at -30°C. The concentration of IFN-γ in the supernatant was measured by using a Mouse IFN-γ Uncoated ELISA Kit (Thermo Fisher Scienti c) according to the manufacturer's protocol. Optical density was measured at a wavelength of 450 nm on an In nite 200 PRO multimode plate reader (Tecan Group).

Statistical analysis
Serum neutralization antibody titers are presented as geometric means. E2-speci c ELISA antibody titers are presented as means. All data were analyzed by using GraphPad Prism 9 (GraphPad software, San Diego, CA, USA) to perform one-way analysis of variance followed by Bonferroni's multiple comparison test or the Kruskal-Wallis test followed by Dunn's multiple comparison test. Statistically signi cant differences were de ned as P < 0.05.

Construction of expression plasmids
The coding sequences of a partial E2 (1023 bp) with the C-terminal 16 amino acids of E1, CD40 (870 bp), and CD63 (717 bp) were PCR-ampli ed and cloned into the pUMVC4a expression vector. The recombinant protein expressed by pE2, pCD40, or pCD63 was detected by Western blot analysis by using a BVDV E2-, CD40-, or CD63-speci c mAb in the cell lysate of HEK-293T cells transfected with each expression vector (Fig. 1a). Because the partial E2 gene lacked the hydrophobic transmembrane region, recombinant E2 protein was detected as a secreted protein in the culture supernatant of pE2-transfected 293T cells (Fig. 1b). In addition, the recombinant protein expressed by pE2, pCD40, or pCD63 was detected in the cytoplasm of transfected CHO-K1 cells through immuno uorescence staining with a BVDV E2-, CD40-, or CD63-speci c mAb (Fig. 2). These results indicated that pE2, pCD40, and pCD63 expressed recombinant BVDV E2 protein, CD40, and CD63, respectively, in eukaryote cells.

Humoral immune responses
To investigate the effect of CD40 and CD63 on the humoral immune responses to a BVDV E2 DNA vaccine, virus neutralization titers were compared between the mice that received pE2 only and those that received pE2 with either pCD40 or pCD63. Neutralizing antibodies were detected in all groups except the mock-immunized group (Fig. 3a), and neutralizing antibody titers were signi cantly higher in the groups that received pE2 with 10 or 20 μg of pCD40 than in those that received pE2 only (P < 0.01). In contrast, neutralizing antibody titers did not differ between the group that received pE2 alone and those that received pE2 with pCD63.
BVDV E2-speci c IgG 1 and IgG 2a antibody titers were measured through ELISA and compared between mice that received pE2 only and those that received pE2 with pCD40 or pCD63. E2-speci c IgG 1 antibodies were detected in all groups except the mock-immunized group (Fig. 3b), and E2-speci c IgG 1 titers did not differ between the group that received pE2 only and those that received pE2 with pCD40 or pCD63 (Fig. 3b). However, E2-speci c IgG 2a titers were signi cantly higher in the group that received pE2 with 5 mg of pCD63 than in those that received pE2 only (P < 0.05; Fig. 3c). In addition, the IgG 2a :IgG 1 ratio did not differ between the group that received pE2 only and those that received pE2 with either pCD40 or pCD63 (Fig. 3d).

Cellular immune responses
To investigate the effect of CD40 and CD63 on the cellular immune responses to a BVDV E2 DNA vaccine, BVDV-speci c lymphocyte proliferation was compared between mice that received pE2 only and those that received pE2 with p40 or p63. BVDV-speci c lymphocyte proliferation occurred in all other groups apart from the mock-immunized group (Fig. 4a). BVDV-speci c proliferation was signi cantly greater in the group that received pE2 with 5 μg or 10 μg of pCD40 than in those that received pE2 only (P < 0.01 and P < 0.0001, respectively). In addition, BVDV-speci c lymphocyte proliferation was greater in the group that received pE2 with 10 μg of pCD63 than in those that received pE2 only (P < 0.01).
We then compared BVDV-speci c IFN-g production between mice that received pE2 only and those that received pE2 with either pCD40 or pCD63. IFN-γ production did not differ between the group that received pE2 only and the mock-immunized group (Fig. 4b), but it was signi cantly higher in the group that received pE2 with 5 μg of pCD40 (P < 0.05). In addition, IFN-γ production was signi cantly higher in mice that received both pE2 and 10 μg of pCD63 than in those that received pE2 only (P < 0.01).

Discussion
In this study, we investigated the effects of CD40 and CD63 on the immune responses to a DNA vaccine against BVDV. First, we constructed plasmids encoding BVDV E2 protein (pE2), mouse CD40 (pCD40), or mouse CD63 (pCD63) and used Western blot analysis and immuno uorescence staining to con rm that these plasmids expressed these proteins in cultured cell lines. We then immunized BALB/c mice by using pE2 in combination with, or without, pCD40 or pCD63. The combination of pE2 with pCD40 induced signi cantly higher neutralizing antibody titers against BVDV than did pE2 only. In addition, pE2 combined with either pCD40 or pCD63 induced signi cantly greater lymphocyte proliferation and IFN-γ production in response to BVDV than did pE2 only.
Our results showed that co-administration of pCD40 enhanced humoral and cellular immune responses to pE2. These results are consistent with those of previous studies. For example, a plasmid encoding CD40 enhanced humoral and T-cell immune responses to a DNA vaccine against foot-and-mouth disease [35]. In another study, a plasmid encoding CD40 enhanced the production of type 2 helper T (Th2) cytokines and antibody responses to a DNA vaccine against H5N1 highly pathogenic avian in uenza [33]. These results suggest that co-administration of a plasmid encoding CD40 is a promising means of enhancing humoral and cellular immune responses to DNA vaccines.
In addition, our current study demonstrates that co-administration of pCD63 enhances cellular immune responses to pE2. These results are inconsistent with those of a previous study, in which coadministration of a plasmid encoding CD63 failed to increase cellular immune responses to an OVAencoding plasmid [40]. The reason for this apparent discrepancy is unclear, but it may re ect differences in antigen (BVDV E2 or OVA), plasmid (pUMVC4a or pCl, although both use the same cytomegalovirus promoter), administration route (intradermal or intramuscular), mouse strain (BALB/c or C57BL/Cj), or other experimental conditions. Further studies are needed to clarify whether co-administration of a plasmid encoding CD63 enhances cellular immune responses to DNA vaccines.
In the current study, we did not observe dose-dependency in the adjuvant effect due to pCD40 or pCD63. For example, the highest dose (40 µg) of neither pCD40 or pCD63 enhanced the immune responses to pE2, whereas 10 or 20 µg of pCD40 induced the greatest neutralizing antibody responses and 10 µg of pCD63 induced the greatest lymphocyte proliferation and IFN-γ responses. These results are in harmony with those of previous reports, in which excess doses of plasmids decreased expression levels of foreign genes [41,42]. In this study, the excess dose (40 µg) of pCD40 or pCD63 may have hampered the expression of either CD40 or CD63, respectively, or it may have competitively diminished E2 expression from pE2. The optimal dose of a plasmid encoding CD40 or CD63 for use as a DNA vaccine adjuvant will need to be determined in the target animal species.
Administration of pE2 induced IgG 1 antibody responses, and co-administration of pCD40 or pCD63 did not signi cantly change the IgG 2a :IgG 1 ratio. This result is consistent with those of previous studies, in which the administration of DNA vaccines encoding secreted antigen induced predominantly IgG1 antibody production [43,44]. In contrast, a plasmid encoding BVDV E2 dominantly induced IgG2a antibody responses [45,46]. The reason for this apparent discrepancy is unknown. It has been reported that the route of administration of DNA vaccines affects Th1/Th2 bias: intradermal administration of DNA vaccines induces Th2 responses, whereas intramuscular administration induces Th1 responses [47,48]. However, our current and preliminary studies showed that both intradermal or intramuscular administration of pE2 predominantly induced IgG1 antibody responses (data not shown), suggesting that a secreted form of E2 antigen induces an IgG1 response regardless of the route of administration.
In conclusion, the results that we have presented indicate that co-administration of a plasmid encoding CD40 or CD63 enhances humoral and cellular immune responses to a DNA vaccine encoding BVDV E2 protein. Our current results further suggest that these adjuvant plasmids have the potential to compensate for the weak immunogenicity of DNA vaccines-not only for BVDV but also for various other applications, including other infectious diseases, therapy of cancer, allergies, and autoimmunity.   between groups were assessed by one-way analysis of variance followed by Bonferroni's multiple comparison test, P < 0.05 (*) and P < 0.01 (**). The presented data are representative of two independent experiments. , pUMVC4a; , pE2; ○, pE2 and pCD40; △, pE2 and pCD63.

Figure 4
Adjuvant effects of CD40 and CD63 on cellular immune responses to a BVDV E2 vaccine. BALB/c mice were immunized intradermally twice with pE2 in combination with, or without, pCD40 or pCD63 (3 weeks between doses). (a) BVDV-speci c lymphocyte proliferation. Splenocytes were stimulated with BVDV for 48 h, after which WST-8 was added and the absorbance at a wavelength of 450 nm was measured.
Horizontal bars indicate mean values. Differences between groups were assessed by one-way analysis of variance followed by Bonferroni's multiple comparison test. (b) BVDV-speci c interferon (IFN)-γ production. Splenocytes were stimulated with BVDV for 72 h. IFN-γ in the cell culture supernatants were measured through ELISA. Horizontal bars indicate mean values. Differences between groups were assessed by using the Kruskal-Wallis test followed by Dunn's multiple comparison test, P < 0.05 (*) and P < 0.01 (**). The data presented are representative of two independent experiments. , pUMVC4a; , pE2; ○, pE2 and pCD40; △, pE2 and pCD63.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Table1Leng.xlsx