Gene Gun Bombardment with DNA-Coated Golden Particles Enhanced the Protective Effect of a DNA Vaccine Based on Thioredoxin Glutathione Reductase of Schistosoma japonicum

Schistosomiasis, caused by infection with Schistosoma species, remains an important parasitic zoonosis. Thioredoxin glutathione reductase of Schistosoma japonicum (SjTGR) plays an important role in the development of the parasite and for its survival. Here we present a recombinant plasmid DNA vaccine, pVAX1/SjTGR, to estimate its protection against S. japonicum in BALB/c mice. The DNA vaccine administrated by particle bombardment induced higher protection than by intramuscular injection. All animals vaccinated with pVAX1/SjTGR developed significant specific anti-SjTGR antibodies than control groups. Moreover, animals immunized by gene gun exhibited a splenocyte proliferative response, with an increase in IFN-γ and IL-4. The recombinant plasmid administrated by gene gun achieved a medium protective efficacy of 27.83–38.83% (P < 0.01) of worm reduction and 40.38–44.51% (P < 0.01) of liver egg count reduction. It suggests that different modes of administering a DNA vaccine can influence the protective efficacy induced by the vaccine. Interestingly, from the enzymatic activity results, we found that worms obtained from pVAX1/SjTGR-vaccinated animals expressed lower enzymatic activity than the control group and the antibodies weakened the enzymatic activity of SjTGR in vitro, too. It implies that the high-level antibodies may contribute to the protective effects.


Introduction
Schistosomiasis is an important disease distributed in many parts of the world, most of which are the places with poor sanitation or irrigation areas, and it is estimated that 779 million people are at risk of schistosomiasis [1]. S. japonicum is the most difficult form of schistosomiasis to control among the 5 Schistosoma species which infect humans [2][3][4]. Schistosomiasis is a chronic and debilitating disease [5,6] which is always accompanied by emaciation and anemia, and even death.
Over the past decades of years, the Chinese government has implemented several control programs, including community-based praziquantel chemotherapy [7], health education, improved sanitation, environmental modi�cation, and snail control. However, schistosomiasis remains an important public health concern in China. As snail control [8] is always difficult to achieve, and praziquantel has no effect on reinfection [9,10], the disease is difficult to control. erefore, a complementary approach to integrate chemotherapy, vaccination for example, is needed.
Since the 20th century, scientists have been trying to develop an effect vaccine against S. japonicum for �eld use [11][12][13], mainly for yellow cattle and water buffalos. rough decades of efforts, several kinds of vaccines have been developed, including cercariae-attenuated vaccines [14], natural or recombinant protein vaccines, nucleic acid vaccines, and multivalent affiliate vaccines. Currently, DNA vaccines have received increased attention and are considered advantageous compared to other vaccine preparations [15,16], for low cost and easy preparation. However, the mode of delivering a DNA vaccine can in�uence the effect induced by the vaccine [17].
e schistosome tegument is a single syncytium that covers the surface of the parasite body [18]. Although there remains many unresolved questions in relation to the structure and function of the tegument, the dynamic host-interactive layer tegument is believed to involve in nutrient uptake, immune evasion and modulation, sensory reception, and signal transduction, and is important from a vaccine perspective [19][20][21][22]. A number of described vaccine candidates are membrane proteins [23,24], muscle proteins [25], and enzymes [26][27][28]. ioredoxin glutathione reductase of S. japonicum (SjTGR) is also a tegument antigen mainly distributed in the tegument of adult worms [29]. Adult schistosome worms, which reside in the hepatic portal system, are exposed to reactive oxygen compounds from metabolism and the host immune response. In eukaryocyte, two major systems, the thioredoxin (Trx) system and the glutathione (GSH) system, exist to detoxify reactive oxygen species (ROS). However, it has been proved that there are no separate Trx reductase and GSH reductase enzymes in S. japonicum, instead of the linked thioredoxin-glutathione system (TGR) [30]. As such, this union enzyme, thioredoxin glutathione reductase, is thought to be an attractive vaccine antigen candidate.
In this paper, a recombinant DNA plasmid was constructed containing a complete open reading fragment of SjTGR and immunized with two different modes, particle bombardment, and needle inoculation to evaluate the ability to protect BALB/c mice against S. japonicum challenge and explore the conceivable immune protective mechanism.

Experimental Mice and Parasites.
Male BALB/c mice, 6-8 weeks old, were purchased from Slac Animal Laboratory (Shanghai, China). e freshwater snail, Oncomelania hupensis, was maintained in the Shanghai Veterinary Research Institute, Chinese Academy of Agricultural Sciences. Cercariae were collected by exposing infected snails to light and the number and viability of cercariae were determined under a light microscope before challenge. Animal care and all procedures involving animals were conducted according to the principles of Shanghai Veterinary Research Institute for the Care and Use of Laboratory Animals.
Speci�c anti-SjTGR serum was collected from BALB/c mice thrice immunized with recombinant protein SjTGR.

Construction of Recombinant
Plasmid DNA. e eukaryotic expression plasmid, pVAX1, which contains the strong cytomegalovirus (CMV) promoter and bovine growth hormone (BGH) polyadenylation signal, was used as the vector. e entire SjTGR open reading fragment was ampli-�ed from the S. japonicum adult worm cDNA library, with primers: 5 � -CGCGGATCCATGCCTCCGATTGAT-3 � and 5 � -GCCTCGAGTCAGCAACCGGTTACC-3 � and subcloned into cleaved pVAX1 to construct the recombinant expression plasmid, pVAX1/SjTGR. e recombinant plasmid was sequenced to ensure the insert sequence was cloned correctly. en, the expression plasmid was transferred into DH5 , a type of competent cell, for large-scale preparation and puri�cation using a Qiagen Plasmid Maxi Kit (Qiagen), followed by the manufacturer's protocol.

Expression in 293T
Cells. Transfections of plasmid (pVAX1/SjTGR, pVAX1) were done by Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol to detect gene expression in 293T cells. One day before transfection, cells were plated into a 6-well plate in 2 mL/well of growth medium without antibiotics so that the cells will be 90%-95% con�uent when transfection was performed. For each well, 5 L Lipofectamine 2000 and 10 g DNA were mixed gently and incubated for 20 min at room temperature. en, the complex was volume to 2 mL with Opti-MEM and added to the monolayer of cells in each well.
Aer incubating cells at 37 ∘ C in a 5%CO 2 incubator for 48 h, the monolayer cells were �xed with 80% ethanol, and thrice washed with PBS-0.05%/Tween-20 (PBST). Speci�c anti-SjTGR serum was added to each well and incubated at 37 ∘ C for 2 h, then thrice washed with PBST. Aer that, Cy3labeled goat anti-mouse antibodies (Beyotime) were used as secondary antibodies at a dilution of 1 : 5000 and added to each well. Aer 1 h incubation at 37 ∘ C keeping in dark, the plates were thrice washed again. Finally, the potential protein in the cells was detected using a converted �uorescence microscope. e tests were assayed in triplicate.

Immunization with the Helios Gene Gun
System. Cartridges were prepared prior to the day of the experiments, followed by general methods [31]. First, the amount of DNA and gold required for each transformation was calculated. e DNA loading ratio (DLR) used was 5 g DNA/mg gold, and the microcarrier loading quantity (MLQ) was 0.5 mg/cartridge. For the duration of producing bullets, polyvinylpyrrolidone (PVP) (Sigma) served as an adhesive. At higher discharge pressures from the nitrogen source, DNA mixed with nanolevel gold particle was coated in the walls of the tubes (BioRad) and is referred to as bullet. Similarly, at higher discharge pressures from the helium source, murine abdominal epidermis was bombarded with the Helios gene gun system (BioRad). e optimum pressure for mice was determined to occur at 600 psi.

Immunization Schedule and Challenge Infection.
Two mice vaccinations were carried out in this study. In trail 1, �y-male BALB/c mice were randomly divided into 5 groups (10 each group, pVAX1/i.m., pVAX1/SjTGR/i.m., pVAX1/g.g., pVAX1/SjTGR/g.g., PBS). All mice were given 2 intramuscular immunizations 3 weeks apart called primeboost inoculation by two different modes, particle bombardment and needle inoculation. In trail 2, forty-�ve male BALB/c mice were randomly divided into 3 groups (15 each group, pVAX1/g.g., pVAX1/SjTGR/g.g., and PBS) and immunized with a gene gun with the same schedule. Ten days aer each immunization in the trail 2, blood from each animal was collected. Serums were separated and stored at −20 ∘ C for antibody assays and cytokine detections. Two weeks later, mice in each group of the two trails were infected with 40 ± 2 S. japonicum cercaria and sacri�ced 6 weeks aer challenge and blood was collected. e total worm and liver egg burden was determined ( Figure 1).

2.�. Detection of Speci�c Antibodies in Serum by Enzyme�
Linked Immunosorbent Assay (ELISA). In trail 2, the levels of speci�c IgG antibodies against SjTGR were detected by ELISA following standard methods [32]. A 96-well �at-bottomed plate was coated with recombinant protein SjTGR at 4 ∘ C overnight (1 g/well), thrice washed with PBS-0.05%/Tween-20 (PBST), blocked with 150 L/well of PBST-1.5% (m/v) normal bovine serum albumin (BSA) for 2 h at room temperature (25 ∘ C), then thrice washed with PBST. e serum samples collected in the previous section were diluted with PBST in 1 : 100, added to the plate (100 L/well), incubated at 37 ∘ C for 2 h, and thrice washed again with PBST. Horseradish peroxidase labeled goat antimouse IgG, IgG1, and IgG2a (BD Pharmingen) were used as secondary antibodies at a dilution of 1 : 5000 and added at 100 L/well. Aer a 1 h incubation at 37 ∘ C, the plates were thrice washed and the substrate, 3,3 � 5,5 � -tetramethyl benzidine dihydrochloride (TMB), was added (100 L/well). e plates were incubated for 10 min at room temperature in the dark and the reaction was stopped using 2 M H 2 SO 4 (50 L/well). All of the samples were assayed in triplicate. e results were detected in a microplate reader (BioTek) and the absorbance was measured at 450 nm.

Calculation of the Percentage of CD4+ and CD8+ Cells and Cytokine Determination by Flow Cytometry.
Five mice in each group in trail 2 were sacri�ced 2 weeks aer the booster immunization and splenocytes were collected. 1 mL RPMI 1640 medium (Gbico) with 10% FBS was added to each spleen and grinded. Aer grinding, 1.5 mL FACS lysing solution (BD Pharmingen) was added to the cells, and thrice washed by centrifuging at 3000 g for 5 min at 4 ∘ C. e cells were adjusted to 10 7 /mL and cultured overnight and the next day stimulated with 2.5 L (1 L/mL) PMA (Sigma) 2 L (1 g/mL) Ionomycin (Sigma) and 3.4 L Monensin (eBioscience), for 6 h at 37 ∘ C in a 5% CO 2 incubator. en, 0.25 g PE-labeled Cy5 CD3 and 0.25 g PE-labeled Cy7 CD8 (BD Pharmingen) were added to each sample, and incubated at 25 ∘ C for 20 min at dark, thrice washed as usual. en 1 mL dyeing buffer was used to wash the cells for three times as usual. Aer that, 0.5 mL cell-�xed liquid was added to each sample for 20 min and washed thrice as usual. Cells were resuspended with 100 L permeabilization for another 20 min. en for each sample, 0.25 g PE-labeled anti-IL-4 antibodies and 0.25 g FITC-labeled anti-IFN-antibodies (BioLegend) were added and thrice washed as usual. Finally, 0.6 mL cell staining buffer was added to resuspend the cells, and a �ow cytometry system (Beckman) was used to detect the interferon-gamma (IFN-) and interleukin (IL)-4 levels. e criteria for this study were set according to the blank measurement. e ratio of CD4+ and CD8+ T cells in total cells was examined and the rates of T cells producing IFNor IL-4 were reported.

Count of Worm and Liver Egg
Burden. Forty-two days aer challenge, all mice (10 in each group) were euthanized in the two independent trails, and worms were collected by perfusion from the hepatic portal system then counted.
To determine the liver egg burden, each mouse liver were weighed, homogenized, and digested for approximately 1 h at 56 ∘ C with 10 mL 10% NaOH. e suspensions were agitated, and 1-mL aliquots, collected from the middle of each tube, were transferred into 1.5-mL Eppendorf tubes and spun at a low speed to sediment the particles. Pellets were then resuspended in 200 L of PBS and the egg counts were determined under a microscopy.
Reductions in the parasite burden were calculated as follows: worm reduction rate (%) = ((average number of recovered worms of control group-average number of recovered worms of experimental group)/average number of recovered worms of control group) × 100; and egg reduction rate (%) = ((average number of eggs/g liver in control group-average number of eggs/g liver in experimental group)/average number of eggs/g liver of control group) × 100.

Enzyme Activity Analysis by ioredoxin Reductase Assay
Kit. e enzyme activity analysis was referenced to Han [29]. Six-week-old worms were collected in trail 2 and stored at −80 ∘ C with Dulbecco's phosphate buffered saline and protease inhibitor cocktail (Sigma) at a ratio of 1 : 1000. en, all of the worms (50 worms in each group) were grinded with Ready Prep Mini Grinders (BioRad) on ice and all the samples were cracked thoroughly by freezing and thawing three times. e thawed lysates were centrifuged at 10000 ×g for 20 min at 4 ∘ C and the supernatants containing thioredoxin glutathione reductase were used to detect the enzymatic activity in each individual group.
e enzyme activity was assessed using a ioredoxin Reductase Assay Kit (Sigma) for an easy and simple colorimetric assay [33]. It is based on the reduction of 5,5 �dithiobis(2-nitrobenzoic) acid (DTNB) with NADPH to 5thio-2-nitrobenzoic acid (TNB), which produces a strong yellow color that is measured at 412 nm. Components were added to a cuvette with a �nal volume of 1 mL, and 30 L of DTNB in DMSO (100 mM) was added immediately before detection with a ermo NanoDrop ND-2000C (ermo). e enzymatic kinetic program was set as follows using the spectrophotometer: delay = 120 sec, interval = 10 sec, and number of readings = 6. During the test, an inhibitor solution for speci�c inhibition of mammalian thioredoxin reductase contained in the kit was used, to determine the reduction of DTNB due only to thioredoxin reductase activity of SjTGR. All of the samples were detected in triplicate independently. And the data was calculated as the following computational formula: where, ΔA 412 /min (thioredoxine reductase) = [ΔA 412 /min (sample) −ΔA 412 /min (sample + inhibitors)], dil = sample dilution factor, vol = volume of reaction in mL, and enzol = volume of enzyme in mL. An in vitro test was carried out to evaluate the weakened e�ect of speci�c anti-SjTGR serum to thioredoxin reductase. Forty-two-day-old worms were carefully collected and soluble adult worm antigen preparations (SWAPs) were extracted using the above methods. Anti-SjTGR and normal mouse sera were added to SWAPs (1 mg/mL) at a ratio of 1 : 1, and both of the mixtures were incubated at room temperature for 2 h. e enzymatic activity of SjTGR was then detected with the methods described above.

Transient Expression of Recombinant Plasmid in 293T
Cells. e SjTGR entire open reading frame ampli�ed by PCR with special primers was subcloned into the plasmid pVAX1 with T4 DNA ligase and was con�rmed correct by restriction enzyme digestion and sequencing. en, the recombinant plasmids were transiently transfected into 293T cells.
Forty-eight hours aer transfection of recombinant plasmid pVAX1/SjTGR, the cells were �xed to detect the expression of the protein of interest. Red �uorescence was observed on the cells transferred with pVAX1/SjTGR (Figure 2(a)), but not on those transferred with pVAX1 vector DNA alone (Figure 2(b)).

Evaluation of Protective Efficacy Induced by pVAX1/
SjTGR. e total worm burden and eggs per gram (EPG) in each group, as well as the percent reduction in the worm burden and EPG in the vaccinated group compared with PBS control group are summarized in Table 1. In trail 1, animals administrated with particle bombardment induced better protective efficacy. Mice vaccinated with pVAX1/SjTGR by bombarding murine epidermis with the Helios gene gun system resulted in a signi�cant worm burden reduction of 27.83% ( < 1) and 38.83% ( < 1) compared to the PBS control group in two independent trails, respectively. And no protection was observed in pVAX1/SjTGR/i.m. group as well as in two pVAX1 vaccinated groups. Mice in the vaccinated group by gene gun resulted in a signi�cant liver egg burden reduction of 40.38% ( < 1) and 44.51% ( < 1) to the blank control group, in trail 1 and 2, respectively. Signi�cant egg reduction also observed in pVAX1/SjTGR/i.m. group, but not in pVAX1 vaccinated groups.

Antibody Assay.
Total IgG antibodies and its subtypes of IgG1 and IgG2a were detected by ELISA in trail 2, as described. SjTGR-speci�c IgG antibody was detected 10 days aer immunization, signi�cantly increased aer boost vaccination, in the pVAX1/SjTGR immunized mice, and the Differences were signi�cant at . 5 ( * ); data was presented in 95% con�dence interval. a Percent reduction was determined using total worms in immunized group compared to PBS group. b Percent reduction was calculated using liver egg burden in immunized group compared to PBS group. c e number of animals in each group when perfusion. speci�c antibody had no obvious change in mice that received pVAX1 vector DNA or PBS only (Figure 3(a)).
Both SjTGR-speci�c IgG1 (Figure 3(b)) and IgG2a antibodies (Figure 3(c)) increased signi�cantly aer the booster immunization with pVAX1/SjTGR, and the level of IgG1 was higher than IgG2a; the IgG1/IgG2a ratio was signi�cantly increased aer boost immunization (Figure 3(d)). No sig-ni�cant changes were noted in the two control groups in speci�c IgG1 and IgG2a antibody levels. We also found that different inoculation modes can induce different levels of antibodies, and the level of speci�c IgG antibody induced by recombinant plasmids pVAX1/SjTGR is signi�cantly higher when delivered by gene gun than that by i.m. (Figure 4).

T Cell Subsets and Cytokine Determination.
Aer the last immunization, the splenic lymphocytes of animals from each group in trail 2 were collected. Grinded cells were cultured, stimulated, stained by �uorescent-labeled antibodies, and detected with a �ow cytometry system (FMC). In this paper, particle bombardment immunization pushs T cells forward to CD4+ T cells (Table 2). And the percent of cells producing IL-4 ( Figure 5) or IFN- (Figure 6) in the pVAX1/SjTGRvaccinated group were signi�cantly increased compared to those in pVAX1 or PBS group.

Activation Changes of Enzyme.
SWAPs were extracted from worms in each individual group and enzymatic activity was detected. e detectable products of substrate were increased with incubation time. With same amount of SWAPs, the thioredoxin reductase (TR) enzymatic activity in the worms from pVAX1/SjTGR-immunized group expressed much lower enzymatic activity than that of the pVAX1 or PBS control groups (Figure 7). Furthermore, SWAPs treated with anti-SjTGR antibodies expressed a lower TR enzymatic activity than untreated SWAPs, and the enzyme activity was not affected aer incubating with normal mouse serum, indicating that anti-SjTGR antibodies in�uenced the thioredoxin reductase activity of SjTGR to catalyze DNTB into NTB, which can be measured at 412 nm ( Figure 8).

Discussion
ioredoxin glutathione reductase of Schistosome japonicum is a membrane protein of about 65 kDa, which is considered as a promising vaccine candidate antigen based on its immunogenicity [29] and its important role in parasite metabolism as a vital enzyme to balance redox equilibrium [34]. In the current study, we focused on evaluation of a recombinant plasmid pVAX1/SjTGR as a DNA vaccine based on SjTGR against schistosomiasis japonicum with two different modes. And pVAX1 was chosen as the vaccination regimen because it is mentioned by FDA before (http://www .fda.gov/ohrms/dockets/ac/00/transcripts/3664t1_a.pdf).
e main methods of delivery plasmid DNA into animals include intramuscular injection and intradermal delivery into skin by a gene gun system [35]. And the administration mode of delivering a DNA vaccine can in�uence the type of immune response [17] and somehow in�uence the result of immunoprotective efficacy. In this study, mice vaccinated with pVAX1/SjTGR by bombarding murine epidermis with the Helios gene gun system (BioRad) elicited a much stronger IgG antibody response speci�c for SjTGR by ELISA ( Figure  3(a)), and achieved a prominent reduction of worm (27.83%, 38.83%, . , Table 1) and liver EPG (40.38%, 44.51%, . , Table 1), which was considered as a high-efficiency method for less cost (2.5 g/mouse), compared with needle injection (20 g/mouse). Yoshida et al. compared the two methods in reproducible induction of speci�c immune responses and found that gene gun DNA delivery appeared to bring about highly reproducible and reliable results while the results obtained by intramuscular inoculation vary significantly [36]. ey also thought that intramuscular injection appears to favor 1 responses, while gene gun prefers to promote 2 responses. And our results are like theirs to some extent. e DNA vaccine transferred into cells by gene gun bombardment with golden particles under a special-high pressure (600 psi) will be assimilated by professional antigenpresenting cells or some other cells, which can induce humoral and cellular immunity [37]. DNA-coated golden particles delivery by gene gun predominantly produces IgG1 and induces 2-type responses [17]. In our experiment, the dose of DNA plasmid that used in particle bombardment is 8fold less than needle injection. But the former received much better protection. And it is considered that the CpG motif provided by the vector pVAX1 can in�uence the immune response, too [38].
In this paper, mice immunized pVAX1/SjTGR by particle delivery induced both anti-IgG1 and IgG2a antibodies increasing. e antibodies induced by SjTGR DNA constructs were dominantly IgG1 type which is proved to be e�cient in complement �xation and with cytophilic property in antibody-dependent cell-mediated cytotoxicity [39], and the result is the same with Da' dara et al. [40], who stated that gene gun immunization resulted in signi�cantly higher named 2 response [37,41]. On the other hand,  cells can activate B cells and regulate their differentiation and antibody producing. IL-4 can activate B lymphocytes to produce IgG1 subtype antibodies, which is bene�cial to transform between antibody subtypes and enhances 2-type immune responses [41,42]. And mice immunized with the recombinant plasmid by gene gun elicited both 1-type cytokine IFN-and 2type cytokine IL-4 secreting augment, suggesting that miceinoculated pVAX1/SjTGR plasmid induced mixed 1/2 immune responses. Sawant [43] suggested that IL-4 may be more appropriate as a genetic adjuvant than IFN-for ND (Newcastle) DNA vaccine. One strategy to improve DNA vaccine-induced immune responses is the utility of cytokine cDNA as a molecular adjuvant [44]. Coimmunization of these cytokine molecular cassettes is an effective method to modulate the direction of the immune responses (humoral or cellular immune response). In addition, SWAPs were prepared to determine SjTGR thioredoxin reductase activity in vitro. SWAPs of worms from immunized animals expressed a much lower enzymatic activity in catalyzing DTNB into TNB (Figure 7) and SWAPs treated with speci�c anti-SjTGR serum displayed lower enzyme activity compared to untreated SWAP (Figure 8). is may be an important evidence to explain reduction of worms and eggs. We speculate that the complex of speci�c anti-SjTGR antibodies and the interest enzyme lead to the changes of spatial structure of SjTGR, which may be a good competitive inhibitor of interaction between the enzyme and substrate. Consequently, the development progress of worms may be blocked partially, which is responsible for the protective effect against schistosomiasis. On the other hand, the DNA vaccine inoculated by golden-particle bombardment induced much higher antibody level (Figure 4), which is probably bene�cial to the protection.
In this study we investigated the protective ability of pVAX1/SjTGR as a DNA vaccine by inoculating with a gene gun system. We found that animals immunized with pVAX1/SjTGR induced both humoral and cellular immunity.
And the enzymatic activity of SjTGR was weakened to some extent by high-titer antibodies. Based on our work, SjTGR may be considered as a prospective vaccine candidate antigen against schistosomiasis in a BALB/c mouse model.