Evaluation of the efficacy of a Clostridium perfringens type A toxoid vaccine for pigs using a toxin challenge model and under field conditions

The toxovar C of C. perfringens is of major importance for pigs. It forms the alpha and beta toxin and causes necrotizing enteritis of suckling piglets (NE), which can lead to high losses of suckling piglets in affected herds (Posthaus et al. 2020). Beta toxin plays a key role in the pathogenesis of NE (Johannsen et al. 1986, Sayeed et al. 2007, Miclard et al. 2009, Richard et al. 2019) .

Toxovar A has also been described as a cause of diarrhea in suckling piglets (Garmory et al. 2000, Bueschel et al. 2003, Jost et al. 2005, Springer et al. 2012, Kiu and Hall 2018). Since C. perfringens type A is part of the physiological gut flora of piglets during the first days of age (Melin et al. 1997, Songer and Uzal 2005) it is often difficult to relate Clostridium perfringens type A to occurrence of diarrhea. Suckling piglets usually develop serious diarrhea within the first 5 days of age, which may be associated with transient toxemic general disturbances (Nabuurs et al. 1983, Johannsen et al. 1993). Morbidity is high but mortality tends to be low. Severe outbreaks have been described in association with the birth of weak piglets as a result of infection with PRRSV and after mixed infections with Rotaviruses, enterotoxic E. coli and Cystoisospora suis as part of neonatal diarrhea complex (Gresham 1997, Holmgren et al. 2004). The pathogenesis of the disease is not fully understood, yet. All C. perfringens strains, including the toxovar A, form the alpha toxin (zinc metallophospholipase C). Alpha toxin leads to cell lysis by degradation of membrane phospholipids, tissue necrosis, hemolysis, platelet aggregation, contraction of blood vessels, superoxide generation, cytokine storm and ultimately death (Uzal et al. 2010, Li et al. 2013, Oda et al. 2015). The alpha toxin also plays a major role as a virulence factor in bovine necro-hemorrhagic enteritis (Goossens et al. 2017).

In addition to alpha toxin, beta2 toxin was described as a virulence factor for swine (Gibert et al. 1997). It has a molecular weight of 28 kDa and thus differs significantly from the beta toxin (molecular weight 35 kDa) (Uzal et al. 2010). Based on genetic studies of the beta2 toxin gene (cpb2) of various C. perfringens strains of different origin, a differentiation is made between the consensus beta2 toxin gene (ccpb2) and the atypical beta2 toxin gene (acpb2). C. perfringens isolates from pigs in particular harbor the ccpb2 variant (Jost et al. 2005). The mode of action of the toxin is not fully understood (van Asten et al. 2010). A pore-forming effect of the toxin leading to a change in cell membrane permeability which finally results in membrane disruption and cell death is suggested (Smedley et al. 2004, Fisher 2006). Benz et al. (2022) were able to show that consensus beta2 toxin has a pore-forming component with an extremely high activity in lipid bilayers, although it does not have the typical structure of a pore-forming protein. Zeng et al. (2016) demonstrated moderate cytotoxicity of a recombinant beta2 toxin on NC460 human epithelial cells. This cytotoxicity was neutralized by monoclonal antibodies. Luo et al. (2020) indicated that beta2 toxin can cause a dose-dependent inhibition of growth of a porcine jejunum epithelial cell line (IPEC-J2) and results in cell inflammation. Beta2 toxin simultaneously reduces a tight junction protein that leads to barrier dysfunction of the intestinal epithelium. Furthermore, beta2 toxin increases the accumulation of Reactive Oxygen Species (ROS), the reduction of mitochondrial membrane potential and the expression of apoptosis-related genes. These processes lead to apoptosis of IPEC-J2 cells. In contrast, attempts to demonstrate cytotoxic activity on porcine IPI-21 and Caco-2 cells in vitro failed (Allaart et al. 2014). Springer et al. (2012) speculated an influence of beta2 toxin on the effect of alpha toxin or its resorption in the gut.

Basically, it can be assumed that other factors such as the histotoxic gas production and the formation of the enzymes microbial collagenase (kappa toxin) and sialidase may also have an influence on the pathogenesis of the disease (Kiu and Hall 2018).

For prophylaxis of Clostridium perfringens type A associated diarrhea, autogenous vaccines are commonly used. Furthermore, a vaccine containing alpha, beta and beta2 toxoids (Enteroporc AC, Ceva Santé Animale, Libourne, France), a vaccine containing alpha toxoid (C. perfringens type A) and TcdA and TcdB toxoids of Clostridioides difficile (Suiseng^® Diff/A, Hipra) and a combined vaccine containing the alpha, beta and beta2 toxoid of C. perfringens types A and C and the fimbrial antigens (F4ab, F4ac, F5, F6) of enterotoxic E. coli (Enteroporc COLI AC, Ceva Santé Animale, Libourne, France) are licensed in the EU.

All of the mentioned maternal vaccines are administered two times during the last third of gestation. This ensures high antibody titers in the sows’ colostrum at farrowing. Some vaccines are also approved for threefold vaccination. In this case, gilts are vaccinated twice before insemination and a third time two weeks before parturition.

The safety and efficacy testing of these vaccines is required by Regulation (EU) 2019/6. In this context, proof of efficacy must be provided under controlled laboratory conditions by a provocation test after administration of the vaccine under the recommended conditions of use. Results from the laboratory trials must be supplemented by data from field trials. Because C. perfringens type A associated diarrhea is a multifactorial disease, no infection model has been described that induces diarrhea in suckling piglets, yet. However, Johannsen et al. (1993) were able to induce transient general symptoms after application of bacteria as well as toxin-containing supernatant of C. perfringens type A. Based on this observation, a toxin challenge model that showed differences between vaccinated and non-vaccinated sows after intra-abdominal toxin challenge of piglets was developed (Springer et al. 2012).

The objective of the presented trials was the combined testing of the C. perfringens type A component of a C. perfringens type A/C toxoid vaccine using the described toxin challenge model and in a field trial in accordance with the requirements of the Regulation (EU) 2019/6.

In the field trial (registered under animal experiment application No. 42502-3-834 IDT) 35 pregnant gilts (Dan breed, approximately 340 days old) of a farrowing group in a sow breeding facility with 2000 sows were used. The sows farrowed in a weekly rhythm and the suckling period was 4 weeks. The farm was selected because of diarrhea in piglets during the first week of age. Prior to the field trial, C. perfringens type A was detected in fecal samples of the first days of age. Enterotoxic or enteropathogenic E. coli, C. perfringens type C, Rota- and Coronaviruses were not detected. The selected gilts showed no antibodies against alpha and beta2 toxin in serum before the first vaccination (B0). The allocation of the animals to the vaccination and control group was randomized and the study performed as blinded study. After farrowing a total of 474 suckling piglets (mother: see above, father: Duroc, weight of the piglets > 800 g) deriving from both treatment groups were included in the study.

Blood samples from the jugular vein from each gilt were collected one week before 1st vaccination (B0), immediately before 2nd vaccination (B1) and 4 days after farrowing (B2). During the 2nd pregnancy blood samples were taken immediately before booster vaccination (B3) and at 4 days after farrowing (B4). Colostrum was collected at both times of farrowing. For this purpose, carbetocin (Depotocin 70 µg/ml, Veyx Pharma GmbH, Germany) was administered to the gilts after the birth of the 3rd to 4th piglet. Approximately 40 ml colostrum per sow was milked from the anterior 4 mammary complexes.

In addition, blood samples were collected from the jugular vein of two randomly selected piglets per litter at 1 (PB1), 2 (PB2), 3 (PB3) and 4 weeks (PB4) post farrowing. Always the same piglets were sampled. The serum and colostrum samples were tested for antibodies against the alpha and beta2 toxins. Colostrum and serum samples of the piglets (PB1-4) were further tested for the presence of neutralizing antibodies against the alpha toxin by the Lecithinase Neutralization Test (LNT). Until testing the samples were stored at –20°C.

For animal welfare reasons a challenge was carried out on piglets from only 10 gilts per group after 1st farrowing. The gilts were randomly selected according to their farrowing date. Two piglets per gilt received a challenge by intra-abdominal administration of the toxin (2.0 ml) on day 2 of age.

On the day of challenge the clinical parameters of general condition, behavior and milk uptake were assessed using a scoring system (Table 1). In case of severe disturbance of the general condition (somnolence, body temperature <35°C, no milk intake), the animals were euthanized. Animals that died or had to be euthanized during the observation period were examined gross-pathologically. When lesions typical for the challenge occurred (accumulation of serosanguineous fluid in the thoracic and abdominal cavities), the animal received the highest score of 6. Piglets were weighed and clinically examined daily before and until day 3 post challenge.

Three to 5 days after farrowing the piglets of both groups were given an iron supplement (Ursoferran 200 mg/ml, Serumwerk Bernburg AG, Germany) and a coccidiostat (Baycox 50mg/l, Elanco, US). Treatment with antibiotics was not carried out. If piglets had to be treated for other reasons (i.e. arthritis), they were removed from the study. Cross-fostering of piglets (litter compensation) took place on the 2nd day of age but only within the treatment groups. The suckling piglets were clinically (behavior, diarrhea, skin turgor) examined daily for the individual first 5 days of age (individual examination). Further clinical examinations were performed at the average age of 7, 10, 12, 14, 21 and 26 days. Piglet losses and the occurrence of diarrhea in both groups were recorded. In case of diarrhea of the litter rectal swabs were taken from two piglets per pen and examined bacteriologically for C. perfringens. If C. perfringens was detected, typing was performed by multiplex PCR (Polymerase Chain Reaction).

Dilution series from serum or colostrum samples were prepared with phosphate buffered saline (PBS) and incubated with sterile filtered pre-diluted alpha toxin from a supernatant of a Clostridium perfringens type A culture. 10 μl of each sample and control preparation were then incubated in stamped-out wells of yolk lactose agar and further incubated for 16 to 24 hours at 35–38.5°C.

The turbidity of the area around the wells was estimated visually. If the turbidity court of the first dilution of a sample was clearly reduced or not visible, the sample contained neutralizing antibodies and was evaluated as positive. The last dilution at which neutralization was detected was subsequently given as the reciprocal titer (e.g., 2 for a titer of 1:2). Samples without neutralizing antibodies do not affect the turbidity and were evaluated as negative (0).

Piglets from priming-immunized (5 and 2 weeks before 1st farrowing) gilts showed significantly (p<0.05) higher antibody titers against alpha and beta2 toxins compared to the control group until four weeks after 1st farrowing (Figures 3). Neutralizing antibodies against the alpha toxin were detected in the majority of piglets from vaccinated gilts until 4 weeks post farrowing (PB1: 90.0%, PB2: 83.3%, PB3: 73.3%, PB4: 56.7%). In contrast, only 2 out of 30 piglets (6.7%) of the control gilts showed low titers of neutralizing antibodies (maximum titer 2) against the alpha toxin up to 2 weeks after birth (PB2).

Before booster vaccination (B3) during 2nd gestation the antibody titers in serum against alpha and beta2 toxins had decreased when compared to the titers determined directly after 1st farrowing (B2) (Figures 1). However, booster vaccination significantly gained both, the serum (B4) and colostrum (colostrum 2) antibody titers against alpha and beta2 toxins compared to the first pregnancy (Figures 1 and 2).

All sows vaccinated three times showed high titers of neutralizing antibodies against the alpha toxin in the colostrum (32 to 256) at 2nd farrowing (Figure 2). Neutralizing antibodies were also detected in the colostrum of control sows (maximum titer 32, 9 out of 11 animals).

In addition, the piglets of the 2nd gestation had significantly higher serum antibody titers against alpha and beta2 toxins compared to the piglets of the 1st gestation of the same sows. Similar to the piglets from the 1st gestation the antibody titers decreased over the period of 4 weeks after farrowing (Figure 3).

All piglets of 3 times vaccinated sows showed neutralizing antibodies against the alpha toxin until 3 weeks (PB3) post 2nd farrowing (maximum titer 16). At 4 weeks post farrowing 21 out of 22 piglets had neutralizing antibodies. In the control group 13 (PB1), 10 (PB2), 7 (PB3) and 4 piglets (PB4) out of 22, respectively had neutralizing antibodies against alpha toxin. The maximum titer was 4.

The results of clinical symptoms (clinical score) and the mortality after intra-abdominal toxin challenge performed after the 1st farrowing are given in Table 2. None of the piglets from vaccinated gilts showed clinical symptoms, died or had to be euthanized as a result of the toxin administration. In contrast, most piglets of the control group showed severe clinical symptoms (complete recumbency, somnolence, no milk intake). The mean of the clinical score in the vaccinated group was significantly lower (Mann-Whitney U test, p<0.001) than in the control group. Following administration of the challenge toxin altogether 11 out of 20 piglets of the control group died or had to be euthanized due to severe clinical symptoms. Overall, the difference between the vaccinated group and the control group was significant (Fisher’s exact test, p<0.001).

In the vaccinated group, 13 out of 217 piglets (6%) died until the age of 26 days. In the control group, 16 out of 257 piglets (6.2%) died during the same period. Significant differences between the groups were not present. In both groups most of the losses of piglets occurred during the 1st week of age. Most of the piglets died during this time because of being crushed by the mother sow, because of being a runt or due to aspiration of amniotic fluid and subsequent pneumonia.

The incidence of diarrhea (1 to 26 days of life) in the piglets of the vaccine and the control group are shown in Table 3. Diarrhea occurred most frequently in piglets up to the age of 5 days. Between the age of 2 to 5 days and at the average age of 7 days these differences between the groups were significant (Fisher’s exact test, p<0.05).

Within the entire suckling period there were in total 38.7% of piglets (84 out of 217 piglets) from vaccinated gilts with diarrhea compared to 62.6% piglets (161 out of 257 piglets) from control gilts. This difference was also significant (Fisher’s exact test, p<0.001).

Bacteriological examination of 69 rectal swabs of diarrheal piglets from the vaccinated and control group revealed C. perfringens type A (cpa, cpb2) in 40 and C. perfringens type A (cpa) in 5 of the cases. In the vaccinated group, C. perfringens type A (cpa, cpb2) was detected 17 times and C. perfringens type A (cpa) three times. In the control group C. perfringens type A (cpa, cpb2) occurred 23 times and C. perfringens type A (cpa) occurred twice. Significant differences were not detectable between the groups. Clostridium perfringens type C or enterotoxin producing C. perfringens type A strains were not detected.

When sows were boostered during their 2nd gestation the mean colostrum antibody titer and the mean piglet serum antibody titer were also significantly higher at 2nd farrowing compared to the 1st farrowing. These results are in line with previous reports, that also showed a significant increase of antibodies against beta toxin (C. perfringens type C) and transmission to the piglets after basic and booster vaccination of gilts with a C. perfringens type A/C toxoid vaccine (Richard et al. 2019). The results also support the current vaccination scheme with one booster vaccination prior to every following farrowing (Hogh 1976). However, the results of the determination of antibodies in colostrum and serum of the piglets also show that significantly higher antibody levels occur after three times vaccination than after two times vaccination. To guarantee an adequate supply of antibodies to large litters via colostrum, vaccination three times, e.g., twice at an interval of 3 weeks before insemination and once before the 1st farrowing, can be recommended.

As already shown by Springer et al. (2012), piglets originating from vaccinated sows were significantly better protected after intra-abdominal administration of an alpha and beta2 toxin containing supernatant of a Clostridium perfringens type A strain compared to piglets from non-vaccinated sows. Restrictively, it should be noted that the intra-abdominal toxin admi­nistration model applied in this study is not capable to reproduce the major clinical symptoms (diarrhea) that occur under natural conditions. However, Johannsen et al. (1993) were also able to induce transient signs of systemic toxin effects after oral administration of vegetative Clostridium perfringens type A bacteria and intragastric application of toxin containing CpA supernatants.

The intra-abdominal toxin administration model clearly shows that severe clinical symptoms occur when the toxins are absorbed and can be considered as the worst-case scenario of the outcome of the clinical infection caused by toxins producing Clostridium perfringens type A strains. This may explain the sudden death of young piglets and increased piglet mortality described after infection with Clostridium perfringens type A (Gresham 1997, Holmgren et al. 2004).

In the field trial presented, Rota- and Coronaviruses (examination during selection of the herd), enterotoxin producing Clostridium perfringens type A strains and C. perfringens type C were ruled out by laboratory evaluation before and during the study. Accompanying vaccination of gilts with a vaccine against E. coli diarrhea and metaphylactic treatment of piglets with a coccidiostat further excluded the involvement of enterotoxic E. coli and coccidia. Thus, the study demonstrated that Clostridium perfringens type A was able to cause severe diarrhea in suckling piglets from non-vaccinated animals and could be effectively controlled by vaccination. The occurrence of Clostridium perfringens type A as a causative agent of diarrhea is in accordance with previous communications (Garmory et al. 2000, Bueschel et al. 2003, Jost et al. 2005, Springer et al. 2012) but are contrary to results from Kongsted et al. (2018), who questioned the involvement of Clostridium perfringens type A in diarrhea. However, the occurrence of clinical signs, positive culture, detection of toxins or toxin genes (alpha and beta2 toxin) and exclusion of other relevant infectious agents are required for a confirmed diagnosis. At the same time, it is important to note that Clostridium perfringens type A strains can differ significantly in their in vitro toxin production capacity (Springer et al. 2012).

In the presented field trial vaccination with a C. perfringens type A/C vaccine had no effect on piglet mortality and detection of C. perfringens type A. Nevertheless, mortality is not a typical outcome of Clostridium perfringens type A infection compared to infection with C. perfringens type C, especially in clinical cases, where other enteric pathogens were ruled out. At the same time a reduction of the detection of C. perfringens type A in fecal samples after basic immunization is not expected because the antibodies are primarily directed against the toxins. However, the incidence of diarrhea was significantly reduced by vaccination compared to the control group in piglets between 2 to 5 days of age as well as throughout the entire study period. Including the results of the intra-abdominal toxin administration model, it can be assumed that vaccination can also prevent systemic clinical courses that may occur after resorption of the toxins during mixed infections with other pathogens. The results of the trials also show evidence of the efficacy of the vaccine under laboratory (toxin challenge model) and field conditions, which is required according to Regulation 2019/6. Because C. perfringens belongs to the spore-forming bacteria and has a shorter generation time compared to other bacteria (Kiu and Hall 2018), vaccination should always be accompanied by a check of hygiene measures. Since colostrum uptake is a prerequisite for the efficacy of a maternal vaccination, it should also be verified whether piglets are able to take up sufficient amounts of colostrum.