Development of oral-DNA-based vaccine against Clostridium perfringens using Lactobacillus expressing genetically modified beta toxin

1. Presents a novel oral-DNA vaccine targeting Clostridium perfringens, a major cause of enterotoxaemia in animals.
2. Genetic Modification: Utilizes Lactobacillus species engineered to express a modified beta toxin, enhancing immunogenicity and safety.
3. Oral Administration: Highlights the benefits of an oral vaccine, including ease of administration and potential for widespread application in livestock.
4. Immunological Response: Describes the elicitation of robust immune responses, providing protection against C. perfringens infections.
5. Safety Profile: Emphasizes the safety of the genetically modified Lactobacillus strain, with minimal risk of pathogenicity.

Introduction

One of the common reasons of sickness in cattle especially calves is diarrhea. Diarrhea in animals can cause huge economic losses around the world yearly (1-4). There are various causes of diarrhea in farm animals as pathogenic bacteria, viruses and intestinal parasites. Escherichia coli and Salmonella spp. are considered as the most known bacterial causes of diarrhea (5). In addition, C. perfringens type B, C and E is considered as a dangerous cause of haemorrhagic necrotizing enteritis in calves, sheep, chicken and piglets (6,7). C. perfringens is positive to Gram staining, sporogenic and anaerobic pathogen may cause many diseases in animals and human including diarrhea (8,9). C. perfringens subdivided into five types; type A, type B, type C, type D, and type E depending on type of toxins that produced; alpha (CPA), beta (CPB), C. perfringens enterotoxin (CPE), epsilon (ETX), Iota (ITX), and necrotic enteritis B-like toxin (NetB) which can lead to gas gangrene, necrotic enteritis and enterotoxaemia (4,10). Importantly, β toxin (CPB) is a crucial virulence factor in C. perfringens pathogenesis in addition to the other toxins causes intestinal lesions (11,12). CPB toxin produced by types B, C and E of C. perfringens and it is encoded by cpb gene, it is beta-pore-forming toxins (BPFT) (13,14). Beta toxin is vulnerable to trypsin, posing a threat of CPB-mediated disease in newborn animals due to their limited production of gastric proteases, compounded by the presence of trypsin inhibitors in colostrum (15,16). Recent studies indicate that vaccinating calves with a blend of native toxins from C. perfringens generates antibodies that confer protection against C. perfringens challenges in bovine intestinal models with enteritis. Furthermore, immunization solely against alpha toxin offers significant protection against experimental gas gangrene, although both alpha toxin and perfringolysin O play roles in the disease's pathogenesis (17-19). Consequently, developing vaccines against C. perfringens is crucial in animal husbandry, as it can decrease treatment costs and improve animal welfare (20,21).

This study aims to evaluate the capacity of genetically modified Clostridium perfringens beta toxin (CPB) to induce an immune response in an animal model and immunize against C. perfringens safely.

Materials and methods

Ethical approval

Based on the data for the submitted research project detailed below and approved by the Scientific Research Ethics Committee (established by administrative decision number 4914 on 14/11/2024), which is based on relevant regulations and systems in the United States of America concerning the care and use of animals and humans in scientific research, the Dean of the College of Veterinary Medicine at Al-Qadisiyah University, have decided to approve and accept the research plan.

Bacterial strains preparation

The bacteria that used as a live vector was isolated from new born calf stool. Lactobacillus was selectively isolated from the stool samples using MRS selective agar. Then Gram stain was performed to recognize the Lactobacillus. Also, laboratory E. coli pet 21 strain was used to transform the constructed plasmid pUC19 and also used for oral immunization of the experimental animals. Five feces samples were collected from new born calves in AL-Qadisiyah governorate using sterile borosilicate glass bottles. The samples were placed in a nutrient medium (nutrient broth). Then it is placed in the incubator for a full day after that, the bacteria were grown in a special culture medium (selective media). MRS selective medium was used to isolate Lactobacillus. Also, LB agar was used for E. coli growth. The media were prepared according to the manufacturer's instruction and pH was adjusted to 7.5 and then autoclaved at 121ºC, at 15 psi steam for 20 min.

Experimental design

Beta toxin protein (CPB) was selected among different toxins secreted by C. perfringens to design vaccine candidate against this pathogen. The protein was modified by deleting signal peptides following (18). CPB-C (143-311) was synthesized and cloned into pUC expression plasmid by biological company (SYNBIO, USA). Mice were used as an experimental animal, three groups of mice (V1, V2 and control,) were obtained as 5 mice in each. Oral administration of bacterial suspension, 5 × 10¹⁰ cfu/mL was given at day 0, 4 and 7 to the animals. V1 was immunized with 200 μL of E. coli carrying the constructed plasmid, V2 with Lactobacillus carrying the constructed plasmid, and control group with PBS. Blood samples were collected from the animals at days 4,7 and 14. Complete blood count (CBC) test and total IgG measuring by ELIZA were performed.

Sequence retrieval of protein

We used NCBI to get the beta toxin sequence of C. perfringens strain FDAARGOS-903. Beta toxin is a main virulence factor responsible for intestinal lesions of hemorrhagic necrotizing enteritis in animals affected by C. perfringens (22, 23). Consequently, we utilized the expression of the C-terminal domain (CPB-C 143-311) for the development of an oral DNA-based vaccine.

Antigenicity prediction

The antigenicity of the modified protein was evaluated using the IEDB analysis. The target organism was designated as bacteria, with a prediction threshold set at 0.4.

Profile of beta toxin properties

The modified protein's features were analysed using the ProtParam web server. The vaccine's antigenicity was confirmed with the VaxiJen v2.0 web server. The modified protein's secondary structure elements, including beta-turns, random coils, and beta-helices, were predicted using the PSIPRED v3.3 web server.

Forecasting the three-dimensional structure (Tertiary structure)

The three-dimensional structure of the modified protein was created using the I-TASSER web server. A high-quality model was chosen based on a confidence score that varied from -5 to 2.

Model vaccine validation and refinement

The optimization of the altered protein was performed utilizing the Galaxy web server. The Ramachandran plot was analysed through the RAMPAGE web server to validate the refined structure.

Docking of the modified protein with the immune receptor TLR-3

The ClusPro v2.0 web server was employed to explore the interaction affinity between the modified toxin and the TLR-3 receptor, which was sourced from the PDB database. Comprehensive evaluation of the binding affinity between the modelled vaccine and the TLR-3 receptor was performed utilizing the HDOCK web server.

Plasmid construction

Beta-toxin gene of C. perfringens (GenBank: CP065680.1) was synthesized by a biological company (SYNBIO, USA). After deleting the toxic amino acids, the resulted fragment of beta-toxin was CPB-C (143-311). Then, the fragment was cloned in to pUC expression vector and transformed into E. coli pet 21 strain by the same company. Conjugation between the E. coli that carrying the recombinant plasmid and the isolated Lactobacillus strain was performed in our lab following Samperio et al. (2021) (19). ApaLI enzyme was used to cconfirm presence of the constructed gene.

Oral immunization

Five-week-old female mice, each weighing between 45 and 50 grams, were randomly assigned to three groups (n = 5 per group): two treated groups and a PBS control group. The recombinant Lactobacillus and E. coli were cultured and centrifuged, washed with PBS, and re-suspended in PBS at a concentration of 5 × 10¹⁰ cfu/mL. The treated groups mice received 200 μL of the recombinant bacteria and the PBS for the control group. The immune and sampling procedures for the mice were conducted following (20) with some modification. Briefly, all mice were immunized orally at day 0, 4 and 7.

Detection of total IgG by ELISA

On day 4, 7 and 14 after immunization, serum samples were collected from 15 mice. Serum total IgG was detected by ELISA. In brief, to determine total Immunoglobulin G (IgG) levels in mice, a standard protocol involves several steps. First, collect blood samples from the mice, usually via tail vein. Allow the blood samples to clot at room temperature for approximately 30 minutes, then centrifuge at 3000 rpm for 10 minutes to separate the serum. Next, use an enzyme-linked immunosorbent assay (ELISA), which is a common method for quantifying total IgG. Prepare the ELISA plate by coating them with an IgG capture antibody specific to mouse IgG and incubate overnight at 4°C. After washing the plates, add the serum samples diluted in assay buffer, followed by incubation for 1 h to allow the antigen-antibody binding. A secondary antibody conjugated to an enzyme is then added, followed by a substrate solution that produces a colorimetric change proportional to the amount of IgG present. Followed by two times washing using wash buffer supplied with the kit. Measure the optical density of the wells with a microplate reader at the 450 nm wavelength, p ≤ 0.05. Finally, calculate the total IgG concentration by comparing the results to a standard curve derived from known concentrations of mouse IgG.

Results

Sequence of beta toxin

The selected beta toxin (accession no. CP065680.1) of C. perfringens strain FDAARGOS_903 has 336 amino acids was modified manually. The modifying was by deleting the signal amino acids of the protein, C-terminal domain (CPB-C(143-311)) with 170 amino acids was selected (Figure 1). The antigenicity was assessed using the VaxiJen v2.0 webserver. The results indicated that the toxin, following deletion, is likely to be an antigen, exhibiting an antigenic score of 0.7483, which exceeds the established threshold.

Figure 1: Amino acids sequence of CPB toxin (336 A.A.). The green highlighted part is the selected part.

Analysis of physiochemical and immunological properties of the modified toxin

The prediction results indicate that the protein is non-allergenic, antigenic, water-soluble, and non-toxic (Table 1). The physicochemical characteristics of the modified toxin were analysed using the ProtParam tool. The protein has a molecular weight of 19,178.39 Da, an estimated in vitro half-life of 20 hours in mammalian cells, and an isoelectric point (pI) of 9.34.

Table 1: Immunogenic and physicochemical characteristics of the modified vaccine

Forecasting the secondary structure

The secondary structure of the modified protein was predicted using PSI-PRED v3.3 and the NPS@ server. The analysis indicated that the structure comprises 37% helix, 48% coil, and 13% strand. During infection, antibodies recognize the unfolded regions of the protein within the host. As illustrated in Figure 2, these unfolded areas are represented by the random coil, depicted as grey amino acids in the figure 2.

Figure 2: Prediction of the secondary structure elements of the engineered modified beta toxin (CPB) utilizing PSIPRED 4.0. The colour coding is as follows: grey represents coils, pink indicates helices, and yellow denotes strands.

Construction, verification, and enhancement of the three-dimensional model

The structural predictions relied on locating similar protein structures within the PDB database. Five leading models were produced, each accompanied by confidence scores of -3.19, -2.09, -3.65, -4.15, and -3.91. The C-score for these models ranges from -5 to 2, with structures exhibiting a C-score below -1.5 typically classified as reliable. Of the five models, Model 1 was selected as the recommended 3D representation of the vaccine construct, boasting a C-score of -3.19, an estimated TM-score of 0.36 ± 0.12, and an estimated RMSD of 14.2 ± 3.8 Å (see Figure 3A). This model underwent additional refinement through the Galaxy web server, with the Ramachandran plot analysis illustrated in Figure 3B. In summary, the structural analysis of the modified protein reveals a significant level of flexibility.

Figure 3: Refinement and validation of the 3D model. A: Development of the vaccine construct model using I-TASSER. B: Ramachandran plot analysis of the model.

Molecular docking of the constructed model with the TLR-3 receptor

The server analysis identified ten promising docking models between the designed vaccine and the TLR-3 receptor. The selected model included 91 components and demonstrated the lowest binding energy of -694. Additional assessment was performed using the HDOCK web server to examine the interaction between TLR-3 and the modelled vaccine in greater detail, yielding a minimum binding energy of -299 (Figure 4). The schematic diagram of vector constructed in this study was shown in figure 5 and their structures were confirmed by restriction enzyme digestion analysis.

Figure 4: Molecular docking of the modelled CPB toxin with TLR2 receptor Plasmid construction

Figure 5: pUC vector constructed with CPB-C 143-311.

Lactobacillus isolation

Lactobacillus strain that used in the study as a life bacterial vector was isolated from clinical feces samples were collected from healthy calves. The bacteria were diagnosed by MRS selective media and Gram staining.

Figure 6: Lactobacillus on MRS selective media, and under microscope after Gram staining.

Using of live bacterial vector

The constructed plasmid pTuqa was prepared and transformed in to E. coli pet 21 strain by biological company (SYNBIO, USA). The recombinant E. coli was grown on LB plate with ampicillin 50 μl/ml and incubated overnight. Broth culture was performed with the same antibiotic and plasmid was extracted then digested to confirm presence of the inserted protein. Conjugation between the E. coli pet 21 and Lactobacillus was performed and the resulted Lactobacillus was tested by plasmid extraction to confirm presence of the plasmid. In addition to its growth on the plates with ampicillin.

Figure 7: E. coli pet 21 carrying the constructed plasmid p Tuqa (pUC with CPB) on gel electroporation shows closed and linear plasmid.

Figure 8: Lactobacillus carrying the constructed plasmid p Tuqa (pUC with CPB) on LB agar plate with ampicillin after conjugation.

Evaluation of vaccine safety and efficiency

There are no any adverse clinical signs observed in any of the groups after oral administration of the vaccine. There is no post-mortem changes were observed in the intestine or other internal organs after 7 days of immunization. Utilizing Complete Blood Count (CBC) as a biomarker for evaluating cellular immune response (On the 14th day of oral immunization). Lymphocytes percentage was significantly different between the groups. It was significantly increase in the lymphocyte’s percentage of treated groups (V1 and V2) compared to the control. While other blood elements' values were similar between all animals’ groups.

Figure 9: Experimental mice look normal and active after immunization (day 7).

Figure 10: Anatomical section of experimental mice, normal colour and appearance after 7 days of immunization.

Figure 11: Lymphocyte percentage (%) of control, V1: mice immunized with E. coli carrying the constructed plasmid and V2: groups mice immunized with Lactobacillus carrying the constructed plasmid.

Detection of IgG in serum samples after immunization

The bar chart illustrates the progression of IgG values over time (Day 0 to Day 14) for three groups: control, v1, and v2. Both v1 and v2 show a steady and substantial increase in IgG values across all time points, while the control group exhibits only a slight rise. On Day 7 and Day 14, the values for v1 and v2 are significantly higher than those of the control group, as indicated by the red asterisks, suggesting a statistically meaningful difference likely due to treatment effects. By Day 14, v1 reaches the highest value (1200 pg/ml), followed closely by v2 (1100 pg/ml), while the control group remains much lower at (560 pg/ml). This pattern indicates that the interventions represented by v1 and v2 are not only effective but increasingly impactful over time.

Furthermore, the early divergence seen from Day 4 onward suggests that both treatments begin to exert measurable effects shortly after initiation, with the effects becoming more pronounced as time progresses. Overall, the results support the efficacy of v1 and v2 treatments in enhancing the measured outcome compared to the control. 

Figure 12: Fourteen days following immunization. Mice exhibit humeral immunological response 1: 200 serum dilution of total IgG from various groups of mice.

Discussion

Clostridium perfringens is a zoonotic pathogen commonly found in various environments and as part of the normal gut flora in humans and animals. It can lead to severe conditions such as muscle necrosis, necrotizing enteritis, and foodborne illness in humans. Consequently, its impact can result in significant economic losses in livestock sectors (24-28). The primary pathogenic mechanism of this bacterium involves the release of toxins, which have recently been categorized into seven distinct types A through G (29-31). Unfortunately, there are no effective treatments for severe C. perfringens infections, and while antibiotics have been used both preventively and therapeutically, their overuse has raised significant global public health concerns (32,33). Therefore, discovering new strategies that could replace antibiotics is becoming increasingly crucial.

Vaccination is a proven strategy for disease prevention. Several toxoid and subunit vaccines aimed at C. perfringens are available and have shown varying degrees of efficacy. Structural vaccinology, a field of study focusing on the epitopes that induce immunity, allows for the engineering of chimeric vaccines that incorporate only protective epitopes from different toxins while omitting irrelevant domains. This streamlines the production process, enabling the creation of a single vaccine capable of providing immunity against multiple toxins rather than requiring multiple vaccine formulations. Research efforts have concentrated on elucidating the mechanism of toxicity associated with the CPB toxin, yielding valuable insights for vaccine development (34,35).

In the present study, the CPB toxin was selected for immunization against C. perfringens. In addition to the immune response that is induced by the modified toxin, using Lactobacillus as a vector may improve the microflora when the bacterial suspension has given orally. In a previous study, an oral vaccine was able to enhance the immunity in human and mice. S. cerevisiae and yeast-cell microcapsule (YCM) increased the Firmicutes and Bacteroides in the gut when they used in oral vaccination in rats (36,37). We found a high level of IgG and a significant increase in the lymphocyte percentage of the treated groups compared with the control, which indicates the role of the constructed vaccine to induce the mucosal immunity of the experimental rats. Furthermore, oral administration of the vaccine did not negatively impact the mice's behaviour or lead to any clinical signs or anatomical abnormalities. Overall, these findings suggest that the designed vaccine is non-lethal and capable of inducing an immune response in the host. Nonetheless, further tests and parameters must be evaluated to ensure its safety and efficacy.

Conclusion

In the current study, oral vaccine based on DNA toxin using lactobacillus expressing modified beta toxin was developed. At first step, sequence of beta toxin was obtained, then modified was performed of this sequence. The modified toxin was evaluated in term antigenicity, sensitivity and Allergenicity using immunoinformatic tool. The results show that it was immunogenic and it can be binding to the selected immune cell receptor. Furthermore, the modified toxin was non-toxic and non-allergic. At the second step, the modified toxin sequence was synthesised, cloned in plasmid and transformed into E coli then transferred to Lactobacillus via conjugation. Finally, the candidate toxin was evaluated using animal model (mice). Taken together, the candidate vaccine induced immune response in the animal model with no evidence of toxicity or Allergenicity on the animal model.

Acknowledgments

The authors would like to thank the Deanship of college of veterinary medicine al-Qadisiyah university for supporting this work.

Conflict of interest

The authors declare that there is no conflict of interest.