pUC18-CpG Is an Effective Adjuvant for a Duck Tembusu Virus Inactivated Vaccine

Duck Tembusu virus (DTMUV) is an emerging pathogenic flavivirus responsible for massive economic losses in the duck industry. However, commercially inactivated DTMUV vaccines have been ineffective at inducing protective immunity in ducks. The widely used adjuvant cytosine-phosphate-guanine oligodeoxynucleotides (CpG ODNs) reportedly improve humoral and cellular immunities in animal models. However, its effectiveness in DTMUV vaccines requires validation. Here, we assessed the protective efficacy of pUC18-CpG as an adjuvant in an inactivated live DTMUV vaccine in ducks. Our results revealed that the serum hemagglutination inhibition (HI) antibody titers, positive rates of anti-DTMUV antibodies, the concentration of serum cytokines, and protection efficacy were significantly increased in ducks immunized with pUC18-CpG compared to that in the control group. Moreover, ducks immunized with a full vaccine dose containing a half dose of antigen supplemented with 40 μg of pUC18-CpG exhibited the most potent responses. This study suggests that pUC18-CpG is a promising adjuvant against DTMUV, which might prove effective in treating other viral diseases in waterfowl.


Introduction
In 2010, a newly emerging infectious disease in ducks broke out in many Chinese coastal provinces. The disease is characterized by a substantial drop in egg-laying and low mortality [1]. The causative agent has been identified as duck Tembusu virus (DTMUV), an enveloped positive sense single-stranded RNA virus that belongs to the genus Flavivirus, family Flaviviridae [2][3][4]. DTMUV, similar to other flaviviruses, is a mosquito-borne Flavivirus, and was first isolated from mosquitoes of the genus Culex in 1970s in Malaysia [5].
Since its emergence, the Tembusu virus infection has led to major economic losses in the Chinese poultry industry [6]. Nearly all duck species have been reported to carry the DTMUV, including Cherry Valley ducks, Pekin ducks, and shelducks [7]. Further, this virus is efficient at infecting other animal species, including chickens [8,9], geese [10,11], mice [12], pigeons [1], and sparrows [13], thus indicating that DTMUV has an extensive host range. A recent report has demonstrated that DTMUV Viruses 2020, 12, 238 3 of 11 procedures were approved by the Animal Care and Use Committee at the Institute of Animal Sciences from the Chinese Academy of Agricultural Sciences, China (IACUC.No.PJ.2011-012-03).

Vaccine Preparation
The vaccines were prepared according to the production standards established by Ringpu Biological Pharmaceutical Co., Ltd. (Baoding, China). The allantoic fluid was inactivated by adding formaldehyde to a final concentration of 0.5% and mixing thoroughly, followed by incubating at 37 • C for 48 h. The vaccines were created by combining allantoic fluid containing inactivated DTMUV-HB with a sterilized mixture containing 94% (v/v) Marcol 52, 6% (v/v) span-80, and 2% (w/v) aluminum stearate. Control vaccines were prepared similarly using allantoic fluid from duck embryos inoculated with PBS. pUC18-CpG was added as the adjuvant to the inactivated oil vaccines. The vaccine formulations are presented in Table 1.

Quantification of Serum Antibody Titers via HI Assays
Serum was isolated from post-vaccination blood samples, and antibody titer quantitation was conducted using a modified HI assay. Briefly, DTMUV infected allantoic fluid was inactivated by beta-propiolactone, and viral antigens were filtrated for HI assays. The unit of HA antigen in each preparation was determined by a hemagglutination (HA) test, which is defined as the reciprocal of the highest dilution of purified antigen that can still cause complete hemagglutination. The antigen was diluted with 0.4% bovine serum albumin borate sodium (BABS) chloride solution, containing 4 HA units per 25 µL. The test serum samples were serially diluted (1:2) in U-bottomed 96-well plates. Four units of HA antigen were added, and the mixture was incubated overnight at 4 • C. Then, 50 µL 0.33% goose red blood cell suspension was added to each well. The plates were incubated for 1 h at 37 • C. DTMUV positive and negative serum controls, as well as a goose erythrocyte control, were included in each plate. The serum HI titers were expressed as the reciprocal of the highest serum dilution that demonstrated complete inhibition of hemagglutination. Positive titers were interpreted as inhibition of hemagglutination at a serum dilution of 1:20 or higher. The positive rates of antibody and the geometric mean titer (GMT) for each group were further determined, and statistical analysis was performed. The GMT is less affected by extreme values than the mean HI titer, which can reduce the influence of individual differences, and is more suitable to reflect the average level of the whole group. Moreover, GMT is a recommended standard statistic for summarizing HI titers [27,28]. The GMT was quantified using the formula, GMT = lg −1 [(N1 × lgX1 + N2 × lgX2 + . . . + Nn lgXn)/(N1 + N2 + . . . + Nn)] (N: number of ducks; X: the value of HI titer; lg −1 (X) = 10 x ).

Determination of Serum Cytokine Levels
To evaluate changes in cytokine expression following vaccination, the concentrations of interferon-alpha (IFN-α), interferon-gamma (IFN-γ), interleukin-2 (IL-2), and interleukin-6 (IL-6) were quantified in serum samples from groups A1, B1, E1, and G1 at different time points using standard, cytokine-capture sandwich ELISA kits (Huabo Deyi, Beijing, China), according to manufacturer's instructions. Briefly, diluted standards and serum samples (1:4) were added to microtiter plates coated with corresponding monoclonal antibodies. HRP-conjugated goat anti-duck IgG was used to detect bound antibodies, and the optical density (OD) was measured at 450 nm. The concentrations of cytokines were calculated using standard curves.

Virus Challenge
On D42, after boost immunization, six ducks randomly selected from each group were intramuscularly challenged with 100 DID 50 (50% duck infection dose) of DTMUV-HB in a volume of 0.5 mL. Previous studies have observed that the Tembusu virus can be propagated in both chicken and duck embryos, and causes death typically 60 h after virus inoculation [17]. Blood was collected from the ducks on D2 post-challenge, and the resulting serum was inoculated into five SPF chicken embryos that were 6-days-old for virus isolation. Between 24-72 h after viral inoculation, embryo death was monitored and recorded, and the PD 50 (50% protection dose) for each vaccine batch was determined using the Spearman-Kärber method [29].

Virus Isolation and RT-PCR
Virus isolation was further confirmed by a flavivirus-specific reverse transcriptase PCR (RT-PCR). Briefly, RNA was extracted from the deceased embryos using MiniBEST Universal RNA Extraction Kit (Takara, Beijing, China), according to manufacturer's protocols. Two micrograms of total RNA from each sample were reverse transcribed into cDNA using PrimeScript™ RT Master Mix (Takara, Beijing, China), according to the manufacturer's instructions. The detection primers were designed according to conserved DTMUV NS5 gene sequences [30]: forward: 5 -TCAAGGAACTCCACATGA-3 ; reverse: Viruses 2020, 12, 238 5 of 11 5 -GTGTCCCATCCTGCTGTGTCATCAGCATACA-3 . The expected length of the amplified fragment was 998 bp, and the specific gene products were visualized on a 1% agarose gel.

Statistical Analysis
Statistical analysis was performed using GraphPad Prism 6 software (GraphPad Software, Inc., San Diego, CA, USA). Differences within each treatment at various time points were analyzed using two-way ANOVA. Data is shown graphically as the geometric mean of the fold change plus the standard error of the mean (SEM). A p-value < 0.05 was considered significant.

pUC18-CpG Enhanced Serum Antibody Responses
Serum from vaccinated ducks was subjected to quantify the HI titers of the test groups. The details of results are summarized in Supplementary Material Table S1. No detectable antibody responses were obtained in any of the control ducks during the whole immunization. Note that the titer of HI antibody in groups A and B improved significantly during 14-24 dpi (day post-immunization) (except for HI titer in group A3, which improved during 14-35 dpi), and was significantly higher than that in group E and group G at 24 dpi and 35 dpi at any dose ( Figure 1). The HI antibody titer increased with increasing dose but the difference was not statistically significant except for the comparison of 0.5-0.125 pair at 35 dpi (Supplementary Material Figure S1).
San Diego, CA, USA). Differences within each treatment at various time points were analyzed using two-way ANOVA. Data is shown graphically as the geometric mean of the fold change plus the standard error of the mean (SEM). A p-value < 0.05 was considered significant.

pUC18-CpG Enhanced Serum Antibody Responses
Serum from vaccinated ducks was subjected to quantify the HI titers of the test groups. The details of results are summarized in Supplementary Material Table S1. No detectable antibody responses were obtained in any of the control ducks during the whole immunization. Note that the titer of HI antibody in groups A and B improved significantly during 14-24 dpi (day postimmunization) (except for HI titer in group A3, which improved during 14-35 dpi), and was significantly higher than that in group E and group G at 24 dpi and 35 dpi at any dose ( Figure 1). The HI antibody titer increased with increasing dose but the difference was not statistically significant except for the comparison of 0.5-0.125 pair at 35 dpi (Supplementary Material Figure S1).
We further implemented the statistical analysis of the positive antibody rates and GMT, as shown in Table S1 and Figure 2. The positive rates of group A and group B under different doses showed a similar trend during the experiment, reaching 100% and remaining at the same level for subsequent periods (except for positive rates of group B3, which declined after reaching 100% at 35dpi). The positive rates of group A and group B at each time point were much higher than that of group E and group G. The positive rates of group E and group G were basically the same, and the positive rates decreased with the decrease of dose. The GMT trend of group A was inverted V-shaped with a peak at 24 dpi at full and 1/2 doses/at 35 dpi at a 1/4 dose. GMT in group B showed a plateau (24 dpi-35 dpi) at full and 1/2 doses and then decreased, whereas at a 1/4 dose, the trend of group B was similar to that of group A3. The GMT relationship among the four groups was A > B > E > G, and the GMT values of group A and B were much higher than those of group E and G.  We further implemented the statistical analysis of the positive antibody rates and GMT, as shown in Table S1 and Figure 2. The positive rates of group A and group B under different doses showed a similar trend during the experiment, reaching 100% and remaining at the same level for subsequent periods (except for positive rates of group B3, which declined after reaching 100% at 35 dpi). The positive rates of group A and group B at each time point were much higher than that of group E and group G. The positive rates of group E and group G were basically the same, and the positive rates decreased with the decrease of dose. The GMT trend of group A was inverted V-shaped with a peak at 24 dpi at full and 1/2 doses/at 35 dpi at a 1/4 dose. GMT in group B showed a plateau (24 dpi-35 dpi) at full and 1/2 doses and then decreased, whereas at a 1/4 dose, the trend of group B was similar to that of group A3. The GMT relationship among the four groups was A > B > E > G, and the GMT values of group A and B were much higher than those of group E and G. HB. The HI titer is expressed as the reciprocal form. Data are expressed as means ± SEM. (* p < 0.05, ** p < 0.01, *** p < 0.001).

pUC18-CpG Enhanced both Th1-and Th2-Type Cytokine Production
To identify potential immunological correlates with protection, we evaluated changes in the expression of specific proteins, including IFN-α ( Figure 3A), IFN-γ ( Figure 3B), IL-2 ( Figure 3C), and IL-6 ( Figure 3D), in response to vaccination, via ELISA. Ducks vaccinated with a full vaccine dose (A1, B1, E1, G1) were selected for this assay. As shown in Figure 3, the expressions of 3 of the 4 proteins were found to be upregulated in the serum of the groups immunized with pUC18-CpG compared to those immunized without the adjuvant. Specifically, the protein expression of IFN-γ (p < 0.001) and IL-6 (p < 0.05) in group B1 were significantly higher than those in group G1 at 14 dpi, whereas significant difference was observed in pair-comparison of group A1 and E1 for IFN-γ expression (p < 0.05) at that time point. Moreover, IL-2 expression level in group B1 was significantly higher (p < 0.01; p < 0.05) than that in group G1 at 24 dpi and 35 dpi. However, no significant differences were observed in IFN-α production among time points.

pUC18-CpG Enhanced both Th1-and Th2-Type Cytokine Production
To identify potential immunological correlates with protection, we evaluated changes in the expression of specific proteins, including IFN-α ( Figure 3A), IFN-γ ( Figure 3B), IL-2 ( Figure 3C), and IL-6 ( Figure 3D), in response to vaccination, via ELISA. Ducks vaccinated with a full vaccine dose (A1, B1, E1, G1) were selected for this assay. As shown in Figure 3, the expressions of 3 of the 4 proteins were found to be upregulated in the serum of the groups immunized with pUC18-CpG compared to those immunized without the adjuvant. Specifically, the protein expression of IFN-γ (p < 0.001) and IL-6 (p < 0.05) in group B1 were significantly higher than those in group G1 at 14 dpi, whereas significant difference was observed in pair-comparison of group A1 and E1 for IFN-γ expression (p < 0.05) at that time point. Moreover, IL-2 expression level in group B1 was significantly higher (p < 0.01; p < 0.05) than that in group G1 at 24 dpi and 35 dpi. However, no significant differences were observed in IFN-α production among time points.

pUC18-CpG Enhanced Protection Efficacy
To evaluate the protective efficacy of the vaccines, the ducks were challenged with DTMUV-HB, and serum samples were collected after challenge. The control ducks exhibited clinical signs, such as green watery diarrhea, reduced feed intake, and depression. The RT-PCR data are consistent with the virus isolation results.
The detailed results of virus isolation in each duck, along with the respective protective rates and PD 50 for each group are presented in Table S1 and Table 3. One-hundred percent of ducks in group A and group B receiving full and 1/2 doses of the vaccines with pUC18-CpG were protected against the virulent virus challenge. However, only 50% and 66.7% of ducks in the A3 and B3 groups, which received a 1/4 dose, were found to be protected, respectively. Furthermore, the PD 50 values for group A and group B were determined to be 4.0 and 4.5, respectively. The protective rates of group E1 and group G1 were both 66.7% at a full dose, and decreased with the dose reduction. The PD 50 value of group E was found to be only 1.4, which was lower than the 1.8 in group G. Viruses 2020, 12, x FOR PEER REVIEW 7 of 11

pUC18-CpG Enhanced Protection Efficacy
To evaluate the protective efficacy of the vaccines, the ducks were challenged with DTMUV-HB, and serum samples were collected after challenge. The control ducks exhibited clinical signs, such as green watery diarrhea, reduced feed intake, and depression. The RT-PCR data are consistent with the virus isolation results.
The detailed results of virus isolation in each duck, along with the respective protective rates and PD50 for each group are presented in Table S1 and Table 3. One-hundred percent of ducks in group A and group B receiving full and 1/2 doses of the vaccines with pUC18-CpG were protected against the virulent virus challenge. However, only 50% and 66.7% of ducks in the A3 and B3 groups, which received a 1/4 dose, were found to be protected, respectively. Furthermore, the PD50 values for group A and group B were determined to be 4.0 and 4.5, respectively. The protective rates of group E1 and group G1 were both 66.7% at a full dose, and decreased with the dose reduction. The PD50 value of group E was found to be only 1.4, which was lower than the 1.8 in group G.

Discussion
Duck Tembusu virus disease, an acute infectious disease caused by DTMUV [31], has not only caused significant economic losses in the poultry industry, but has also posed potential threats to public health [14,32]. Vaccination has proven effective in protecting ducks against DTMUV infection and is being used in specific areas to control the disease [17]. However, the ducks receiving an inactivated vaccine gradually developed antibodies yet exhibited low and non-protective antibody titers, highlighting the urgency of an effective vaccine.
CpG ODN is a standard adjuvant, employed in previous studies, that effectively enhances the protective efficacy of vaccines [33,34]. However, due to the high cost of synthetic CpG ODNs, it is not conducive to large-scale production and application. Within the field of DNA vaccines, the inclusion of antibiotic resistance genes in vector plasmids has been controversial. Several groups have opted to use different antibiotic-free selection systems to circumvent the potential antimicrobial resistance [35][36][37], Viruses 2020, 12, 238 8 of 11 but the technology was not yet mature and major deficiencies in these strategies cannot be addressed could not be addressed (such as the negative effect of increasing the number of plasmid copies on the bacteria growth during fermentation, the harm to humans caused by plasmids carrying redundant bacterial genes, high costs, unavailable large-scale production, etc.) [38][39][40], and no plasmid system has yet been approved for industrial mass production. In addition, antibiotics are added to feed during current breeding processes, making it difficult to evaluate in practice. Moreover, insertion of CpG motifs into the pUC-18 plasmid vector has also been proven to enhance the immune effect of vaccines [22,23]. In view of these, we finally circumvented the antibiotic-free selection system and chose the pUC18 plasmid vector. The large-scale fermentation technique in E. coli in our laboratory is mature, with the most production and lowest cost.
We utilized the pUC-18 vector containing CpG ODN-2006 as an adjuvant to generate a batch of vaccines for our study. Following vaccination, in pair-comparison of core vaccine groups, DTMUV vaccine (group E) generated higher values of antibody-related parameters and cytokine expression than 1/2 DTMUV vaccine (group G) at almost all time points and immunization doses. There are two possible reasons: one is that the inactivated DTMUV vaccines were not able to elicit an adequate immune response, another is that the generation of anti-DTMUV antibody may be antigen dose-dependent within a certain range.
When pUC18-CpG was added to vaccines, not only did more ducks show early immune responses at 14 dpi, but also a significantly improved HI titer (p < 0.001) during 14 dpi-24 dpi and a longer-term enhanced antibody response were observed in group A and B. In addition, the HI titer, antibody-positive rates, and GMT were higher than core vaccine groups (group E and group G) at all time points, and the difference was significant in HI titer between group A/B and group E/G (Table S1, Figures 1 and 2). During the experiment, the average antibody level of group A was higher than that of group B at any dose while the protein expression level of cytokines in group B was found to be higher than that of group A when a 0.5 mL dose was administered, but the differences were not statistically significant (p > 0.05). The possible reason for the similar effect between group A and group B may be that in the case of pUC18-CpG addition, both the efficiency of antigen utilization and the level of immune response are all enhanced, and excessive antigen means more non-structural proteins, which may cause side effect.
It has become increasingly apparent that the IFN-γ-dependent Th1 cellular responses are required to elicit protection against Flavivirus infection [41,42]. Similarly, we found that ducks immunized with pUC18-CpG exhibited significantly higher levels of IFN-γ compared to those vaccinated without the adjuvant at 14 dpi (Figure 3), promoting Pekin ducks' resistance to the virus at early stages. Meanwhile, interleukins have been described to mediate T and B lymphocyte activation, proliferation, and differentiation during immune activation and regulation [43][44][45]. Our results further demonstrate that the administration of pUC18-CpG with vaccines elicits higher IL-2 and IL-6 expression levels in ducks at 24 dpi, 35 dpi, and 14 dpi, respectively ( Figure 3). IFN-γ and IL-2 are cytokines secreted by Th1 cells that mainly mediate cellular immune response, whereas IL-6 is a cytokine secreted by Th2 cells which mainly regulate humoral immune response, and the upregulation of these cytokines induces an enhanced Th1-and Th2-type response. These findings suggest that pUC18-CpG may stimulate the body continuously after immunization.
To further validate the immunostimulatory effects of pUC18-CpG, we also determined the efficacy of the different vaccines by assessing the proportion of ducks in each experimental group that had the live virus present in their serum following vaccination. We verified the presence of the virus in serum through RT-PCR methods. Although the traditional inactivated vaccine (group E and group G) reduced the incidence of morbidity and mortality in infected ducks, it was not capable of ultimately preventing infection. Alternatively, ducks vaccinated with CpG (groups A1, A2, B1, and B2) were found to be fully protected post-challenge, and the PD 50 value of the pUC18-CpG adjuvant vaccines (A: 4.0; B: 4.5) was more than twice that of the traditional vaccine (E: 1.4; G: 1.8) (Table 3). Surprisingly, the PD 50 in group G is higher than that in group E, which may be due to insufficient sample size and Viruses 2020, 12, 238 9 of 11 inevitable difference between individuals. Importantly, the control ducks exhibited clinical signs after receiving a challenge with virulent viruses, and tested positive by virus isolation from the serum of selected control ducks, confirming the validity of the protection criteria that we applied to our study.
Two limitations of this study need to be considered. First, groups vaccinated with a 1/2 or 1/4 dose should be also guaranteed 11 Pekin ducks for experiments. Second, the study just divided three vaccine doses for the calculation of PD 50 , whereas the dose of pUC18-CpG used was not optimized, but only referring previous publication. However, due to insufficient funds and isolators to support more grouping, we had to temporarily ignore these two issues.
Cumulatively, these results suggest that employing pUC18-CpG as an adjuvant induces a positive effect on the protective efficacy against DTMUV, and administration of a full dose of 1/2 DTMUV + 40 µg pUC18-CpG was the most commercial and effective.