A DNA vaccine targeting TcdA and TcdB induces protective immunity against Clostridium difficile

Background Clostridium difficile-associated disease (CDAD) constitutes a great majority of hospital diarrhea cases in industrialized countries and is induced by two types of large toxin molecules: toxin A (TcdA) and toxin B (TcdB). Development of immunotherapeutic approaches, either active or passive, has seen a resurgence in recent years. Studies have described vaccine plasmids that express either TcdA and/or TcdB receptor binding domain (RBD). However, the effectiveness of one vector encoding both toxin RBDs against CDAD has not been evaluated. Methods In the study, we constructed highly optimized plasmids to express the receptor binding domains of both TcdA and TcdB from a single vector. The DNA vaccine was evaluated in two animal models for its immunogenicity and protective effects. Results The DNA vaccine induced high levels of serum antibodies to toxin A and/or B and demonstrated neutralizing activity in both in vitro and in vivo systems. In a C. difficile hamster infection model, immunization with the DNA vaccine reduced infection severity and conferred significant protection against a lethal C. difficile strain. Conclusions This study has demonstrated a single plasmid encoding the RBD domains of C. difficile TcdA and TcdB as a DNA vaccine that could provide protection from C. difficile disease.


Background
Clostridium difficile (C. difficile) is one of the most predominant pathogens causing nosocomial intestinal infections in industrialized countries. This bacterial species causes about 10-20 % of the cases of antibiotic-associated diarrhea, up to 70 % of the cases of antibiotic-associated colitis, and the vast majority of cases of pseudomembranous colitis. Clostridium difficile-associated disease (CDAD) causes economic loss of billions of US dollars in many industrialized countries [1][2][3][4]. There is an increasing incidence of CDAD in China caused by rapid economic development and the frequent use of antibiotics. CDADs are mainly caused by antibiotic-induced alteration of the normal flora of the intestine, particularly the long-term use of broad-spectrum antibiotics, thereby allowing C. difficile to proliferate. Cancer chemotherapeutics, hospitalization and immune deficiency are also risk factors, especially in the immunocompromised and the elderly [5]. The clinical manifestation of CDAD is complicated, ranging from being a symptomless carrier to contracting life-threatening pseudomembranous colitis. The prognosis of severe cases indicate that the chance of mortality is 40 %. Metronidazole and Vancomycin are the major treatment drugs for CDAD [6]. However, the C. difficile genome has been found to contain multiple-antibiotic resistant genes and C. difficile clinical isolates resistant to both Metronidazole and Vancomycin have been reported [7], which increase the difficulty for treatment of C. difficile in the future. For all these reasons, the design of vaccines against CDAD is very important.
Over the past two decades, great progress has been achieved in the vaccine development against CDAD [15,[17][18][19][24][25][26]. However, most vaccine research for C. difficile target a single antigen, either TcdA or TcdB or a surface-layer protein (SLP) [24]. Furthermore, the incidence of A -B + C. difficile strains appears to be increasing worldwide over the past decade. These strain types now represent a substantial number of C. difficile isolates. New therapeutic approaches for CDAD treatment such as toxin binders, passive immunotherapy or active immunization through vaccination will now need to target both TcdA and TcdB.
DNA vaccination is an effective platform to generate antigen-specific antibodies and cell-mediated immunity. The most prominent advantage of developing multivalent DNA vaccines is that a plasmid vector with multiple antigen epitopes can be cloned. Several other groups have described vaccine plasmids that express either TcdA and/or TcdB RBD against CDAD [15,19]. In this study, we created a DNA vaccine of C. difficile, which encodes both toxin RBDs of C. difficile. The DNA vaccine was evaluated in two animal models for its immunogenicity, the ability to induce toxin neutralizing antibodies and in vivo anti-toxin protective immunity.

Protein expression
COS7 cells were plated in a 6-well dish at a density of 2 × 10 6 cells per well in Dulbecco's Modified Eagle's Medium (DMEM) with 10 % Fetal Calf Serum (FCS) (v/v) and 2 % penicillin-streptomycin (v/v). 24 h postplating, COS7 cells were transfected with 10 μg of DNA vaccine vectors (pTA, pTAB and pTB). At 48 h post-transfection, the cell lysates and supernatant were collected and stored at -80°C. The supernatant was centrifuged at 16,000 × g for 45 min prior to Western blotting.

Murine immunogenicity study
Six-week-old female BALB/c mice (6 mice per group) were obtained from the Laboratory Animal Unit ( Table 1. Each mouse experiment was repeated in two independent experiments. Blood samples were drawn by tail vein bleeding on days 0, 21, and 35 for immunologic evaluation. Mice were challenged with C. difficile TcdA or TcdB. Toxin a TcdA-specific IgG antibody responses in mouse sera collected at 7 days after each vaccination. b: TcdA-specific IgG1 and IgG2a antibody responses in mouse sera obtained 7 days after the last boost. c TcdB-specific IgG antibody responses in mouse sera collected at 7 days after each vaccination. d TcdB-specific IgG1 and IgG2a antibody responses in mouse sera obtained 7 days after the last boost challenge was performed by inoculating mice intraperitoneally (i.p) with 100 % of the minimal lethal dose (MLD) of the toxin. Mice were monitored for 14 days and survival was recorded for each vaccination group. The MLD of both toxins were confirmed by titration on age match control BALB/c mice. The MLD of toxins A and B were identified to be 50 ng and 25 ng, respectively.

Hamster immunogenicity study
Golden Syrian adult female hamsters (6-week-old, weighing ∼ 100 g) were purchased from LAU of HKU and were housed individually in micro-isolator cages of Department of Microbiology. Hamster experiments were also approved by CULATR of HKU (Approve No. 2903-12). Hamsters were vaccinated by i.m. injection for three times in the thigh, on day 1, 14 and 28, with 100 μg pTAB, pTA or pTB, respectively. Controls were vaccinated with empty plasmid (pIRES, p). Serum samples were collected on days 21, and 35 for immunologic evaluation. On three consecutive days (day 36, 37 and 38), each hamster was treated with 10 mg/kg of clindamycin p.o. On day 39, each hamster received an intragastric challenge of 1x10 8 CFU vegetative bacteria of C. difficile BI/NAP1/027. Hamsters were monitored at 12-h intervals. Each experiment was repeated in two independent experiments.

ELISA
The TcdA and TcdB antibody titers were determined by enzyme-linked immunosorbent assay (ELISA). Briefly, TcdA and TcdB (0.5 μg/ml in 0.05 M carbonate/bicarbonate buffer, pH9.6, and 200 μL/well) were coated on ELISA plates (Nunc, Roskilde, Denmark) by incubation overnight at 4°C. Plates were then blocked with PBS-5 % (w/v) non-fat milk for 3 h at 37°C and washed for 4 times with 0.05 % Tween in PBS. Two-fold serially diluted mice sera were then added into the wells and incubated for 1 h at 37°C. Plates were then washed 6 times with PBS-0.05 % Tween and incubated with HRP-conjugated goat anti-mouse IgG/IgG1/IgG2a for 1 h at 37°C. Color was developed by using Trimethyl Borane (TMB) solution (Sigma) and absorbance was measured using an ELISA reader at 450 nm. The end-point serum antibody titers represent the reciprocal dilution of the last dilution providing an O.D. 2.1-fold higher than the O.D. of negative controls at the lowest performed dilution. A sample of pre-immune serum obtained from mice and hamsters were used as a negative control.

Neutralizing antibody test
Toxin neutralizing titers of the antiserum were determined by using Vero cells and both toxins. Vero cells were grown in Eagle's Minimum Essential Medium (EMEM) containing 10 % fetal calf serum. For neutralizing antibody test, 0.5 ng (100 μl) TcdA or 0.1 ng (100 μl) TcdB was incubated with 100 μl serial dilutions of serum obtained from immunized mice or hamsters. After mixing the antiserum and toxin at 37°C for 90 min, the mixtures were added to 96 well plates containing 1x10 5 Vero cells, and the plates were incubated in 5 % CO 2 at 37°C for 24 h. Incubation of Vero cells with toxin resulted in a loss of cell adherence and a change in cell morphology, which was detected by methyl thiazolyl tetrazolium staining of toxin treated Vero cells after discarding the non-adherent cells. The plates were read on a microtiter plate reader at a wavelength of 490 nm. The neutralization titer of an antiserum was recorded as the serum dilution which gives a 50 % reduction in toxin activity (ED50).

Statistical analysis
Log-rank (Mantel-Cox) analysis was used to analyze the statistical significance of the data from the lethal challenge experiment. Analyses were performed using GraphPad Prism 5 (GraphPad Software, United States) and a p-value of < 0.05 was determined to indicate statistical significance.

Protein expression
Supernatants and cell lysates were harvested at 48 h post-transfection and detected for protein expression via Western blotting with anti-His antibody. The target proteins were highly expressed in the supernatants of the cell lysates (Fig. 1c).

Immunogenicity of the DNA vaccine in mice
To investigate the antibody titers of TcdA-specific and TcdB-specific antibodies in the sera of immunized mice, the levels of IgG, IgG1, IgG2a antibodies were detected by ELISA 7 days after the third immunization. As shown in Fig. 2a, toxin A-specific IgG antibodies were detected in mice immunized with pTA and pTAB, with mean Sera were obtained 7 days following the third immunization. The data are expressed as geometric mean toxin neutralizing titer ± Standard Error of the Mean (SEM) of 10 mice per group c Balb/C mice (10 mice/group) were challenged i.p with 50 ng TcdA or/and 25 ng TcdB 10 days following the second boost of DNA immunization titers of 1.0 × 10 3 and 1.6 × 10 3 respectively. High levels of TcdB-specific IgG antibodies were also detected in mice immunized with pTB and pTAB, both reaching 2.56 × 10 4 (Fig. 2c).
To test the functional activity of the DNA vaccine induced antibodies to neutralize native toxin proteins, toxin neutralization tests were performed (Table 1). Serum samples from the pTAB group were found to have TcdA and TcdB neutralizing activity. Additionally, in mice immunized with pTA (pTB), the serum antibodies can also neutralize TcdA (TcdB) activity. Toxin challenge mouse model was used to evaluate the protective efficacy of DNA vaccine. Immunized Balb/C mice were injected i.p. with 50 ng TcdA or/and 25 ng TcdB 10 days following the second boost of DNA immunization. Following TcdA challenge, 100 % (10/10) of the mice in the p group and 90 % (9/10) in the pTB group died (Fig. 4a). But for the pTAB and pTA groups, 100 and 90 % of the mice survived the lethal TcdA challenge, respectively (Fig. 3a). In the p and pTA group, all mice died following TcdB challenge (Fig. 4b). In contrast, 100 % pTAB and pTB immunized mice survived the lethal TcdA challenge (Fig. 3b). The result (Fig. 3c) shows that the pTAB had significantly improved the survival of the mice following TcdA plus TcdB challenge. Survival rate of 80 % was observed in mice immunized with pTAB. However, in the p, pTA and pTB group, all mice died within 2 days after TcdA plus TcdB challenge.

Protective efficacy of pTAB in a hamster model
Hamster model is the gold standard for evaluation of vaccine against CDAD which can be induced by clindamycin following a challenge with C. difficile. To evaluate the protective efficacy of the DNA vaccine, hamsters (n = 6) were vaccinated for three times. Serum samples obtained 7 days following each immunization were checked by ELISA to detect antibodies. Significant levels of anti-TcdA and TcdB antibodies were detected after the third immunization by pTAB. Only background (titers < 10) were observed in empty vector immunized controls. As observed in the mice immunogenicity study, hamster anti-TcdB titers (pTAB: GMT = 1.0 × 10 5 , pTB: GMT = 6.8 × 10 4 ) were consistently higher than those observed for anti-TcdA (pTAB: GMT = 7.1 × 10 3 , pTA: GMT = 4.9 × 10 3 ) (Fig. 4).
To evaluate in vivo protective efficacy, 8 days following the third immunization, hamsters were treated with clindamycin and challenged with 1 × 10 8 CFU C. difficile BI/ NAP1/027. Similar to the BALB/c mice model, the DNA vaccine group has significantly improved survival following 1 × 10 8 CFU C. difficile BI/NAP1/027 challenge. Survival rate of 100, 50 and 66.7 % were observed in hamster immunized with pTAB, pTA and pTB respectively (Fig. 5). Specifically, CDAD was detected in 30 % of the empty vector immunized hamsters within 24 h. At 48 h, almost all hamsters in the group showed moderate to severe disease and had all died by day 5. However, hamsters immunized with pTAB did not have signs of CDAD until 60 h. The disease noted in pTAB group was less severe and 70 % of the hamsters recovered to normal health by day 6 (data not shown). Of most concern, 100 % survival was observed in the pTAB group at day 14. Additionally, all hamsters in pTAB group exhibited mild to moderate CDAD in the early stages of the experiments and remained symptom free at the end of the study.

Discussion
C. difficile secretes two toxins: TcdA (both an enterotoxin and cytotoxin) and TcdB (a potent cytotoxin). These two toxins can mediate the pathogenesis of CDAD. Since 2002, researches have isolated a novel epidemic typed BI/ NAP1/027 strain [28], which produces 16-fold higher level of TcdA and 24-fold higher level of TcdB than the nonhypervirulent strain VPI 10463 [8]. The role of TcdA and TcdB in CDAD has been confirmed in numerous studies.
According to an early study in which two purified toxins were administered by intragastrically, CDAD was only detected after the administration of purified TcdA, and TcdB cannot induce severe disease unless it is coadministered with TcdA, suggesting that TcdA is the primary pathogenic factor and the toxins act synergistically [29]. After outbreaks of A -B + pathogenic C. difficile variants, an important role for TcdB in C. difficile pathogenicity was established [30][31][32]. Since both toxins have the same importance to the pathogenesis of C. difficile, any immunotherapeutic drugs must target both TcdA and TcdB.
In this study we describe a DNA vaccine -pTAB, consisting of the TcdA RBD (15 of the 31 repetitive oligopeptide sequences) and the TcdB RBD (23 of the 24 repetitive oligopeptide sequences) joined by IRES sequence. This DNA vaccine candidate focuses on the TcdA and TcdB RBD portions. This choice is supported by previous research demonstrating that: (i) antibodies targeting the RBD of both TcdA and TcdB have toxin neutralizing activity [19,33]. (ii) the passive transfer of anti-TcdA-RBD or TcdB-RBD antibodies are protective in animal CDAD models and [23] (iii) hamsters immunized with RBD of TcdA and/or TcdB are protected against CDAD [17]. Several other groups have described vaccine plasmids that express either TcdA and/or TcdB RBD [15,16,19]. In this report, we describe a new vaccine plasmid pTAB, that expresses both toxin RBD sequences. The pTAB was constructed from a commercial mammalian expression vector, pIRES, that allows high level expression of two genes of interest from the same bicistronic mRNA transcript. The vector contains the encephalomyocarditis virus (ECMV) internal ribosome entry site (IRES) flanked by two multiple cloning sites (MCS A and B), an arrangement that allows cap-independent translation of the gene cloned into MCS B [25]. In the DNA vaccine, TcdB-RBD and TcdA-RBD was cloned into the MCS A and MCS B, respectively. The pTAB vaccine plasmid is much more cost-effective than creating two separate plasmids.
Since antibody responses to both RBDs are important for control of CDAD. When DNA vaccine plasmids expressing either A-RBD or B-RBD are co-delivered, it seems that A-RBD dominates the immune response suggesting antigen interference [15]. The pTAB was designed in a way that the A-RBD was placed downstream of a partially disabled IRES sequence [25] to reduce the rate at which the TcdA-RBD is translated relative to that of TcdB-RBD, thus provoking the generation of a higher titer of B-RBD antibody compared to A-RBD. Immunization of mice and hamsters elicited the generation of both TcdA and TcdB antibodies (Figs. 2, 3 and 4) which were capable of neutralizing toxin in vivo assays (Table 1). Immunization with pTAB   TcdA or TcdB  MLD challenge, and also produced 80 % protection  against TcdA MLD plus TcdB MLD challenge. pTAB also demonstrated protective efficacy in the hamster CDAD model, reducing the severity and time till onset of CDAD and significantly protecting the hamster from mortality induced by challenge with BI/ NAP1/027. The significance of this hamster model is characterized by a very rapid progression of CDAD and high mortality. In the empty vector group, hamsters challenged with 10 8 CFU BI/NAP1/027 had all died by day 5, but 100 % survival was observed in the pTAB group at day 14. While hamsters in the pTAB group that exhibited CDAD were characterized as mild to moderate, all recovered and were symptom free by the end of study (14 days). Since BI/NAP1/027 is a double positive (A + B + ) strain, immunized pTA and pTB cannot exhibit a 100 % survival.
Although recently one research has shown that two DNA plasmids encoding the TcdA and TcdB RBDs respectively can induce protective antibody responses if used together in mouse model, the study did not investigate whether a single-plasmid (pARBD or pBRBD) can protect mice from TcdA plus TcdB challenge [15]. Our experiments showed that neither pTA nor pTB single immunization was sufficient to protect mice from a TcdA plus TcdB MLD double challenge. In contrast, compared to the prior study, our research utilized one plasmid (pTAB) instead of two separate plasmids for immunization, achieving even better protection after challenged with double lethal dose toxins in mouse model.

Conclusions
This study has demonstrated a single plasmid encoding the RBD domains of C. difficile TcdA and TcdB as a DNA vaccine that could provide protection from C. difficile disease.