Application of a liposomal subunit vaccine in chickens for reduction of Campylobacter gut colonisation

Campylobacter, microaerophilic, spiral-shaped, Gram-negative microorganisms belonging to the Epsilonproteobacteria class, are the aetiological agent of campylobacteriosis, the most diagnosed zoonosis among residents of European Union member states since 2005. According to the European Food Safety Authority (EFSA), there were 137,107 confirmed cases of Campylobacter infections in 2022 (9). Recently, the incidence rate has remained relatively stable with a slight downward trend, ranging from 58.3 cases per 100,000 people in 2018 to 43.1 cases per 100,000 people in 2022 (9).

Campylobacter infection usually has a low mortality rate, and most patients do not require specific therapy. Antibiotics (macrolides and fluoroquinolones) are used only in severe systemic infections or in patients with impaired immunity (27). However, epidemiological data have recorded cases of serious autoimmune and neurological complications that develop following Campylobacter infections. One such example is neuropathy of the peripheral nervous system, known as Guillain-Barré syndrome (GBS) (11). Current research also indicates a connection between Campylobacter infection and the development of reactive arthritis, inflammatory bowel disease, irritable bowel syndrome, and colorectal cancer (22). It is estimated that for every reported case of campylobacteriosis, there are 30 more cases that go undiagnosed because of the mildness of the usual symptoms of infection.

Epidemiological studies demonstrate that the majority of human campylobacteriosis cases are primarily caused by the consumption of undercooked infected poultry meat. This aligns with the observation that farmed poultry is the main reservoir of Campylobacter (36). These microorganisms colonise the birds’ digestive systems at exceptionally high levels (up to 10⁹ colonyforming units (CFU)/g of intestinal contents), contributing to meat contamination during the production process. Most of the studies conducted to date indicated that high levels of chicken gut colonisation do not cause clinical symptoms in birds, thereby vitiating the isolation of infected individuals from the flock. However, there are also reports describing adverse health effects of Campylobacter colonisation on chicken intestines (1).

On January 1, 2018, the EU Commission Regulation No. 2017/1495 introducing a hygiene criterion for Campylobacter in broiler carcasses came into force. This regulation stipulated that the allowable number of CFU/g of poultry meat should not exceed 1,000. It was assumed that decreasing the population of Campylobacter in the gastrointestinal tract of chickens would significantly reduce the incidence of human campylobacteriosis, and consequently, would lower healthcare costs. According to the updated model estimation, a reduction of 3 log10 could result in a 58% decrease in the risk of Campylobacter infection associated with the consumption of contaminated poultry meat (10). In 2022, an EFSA report indicated that approximately 38% of examined broiler carcasses were contaminated with Campylobacter (9).

One of the promising approaches to reducing the number of Campylobacter in chicken digestive systems is immunoprotection. Over the past twenty years, many attempts have been made to develop an antiCampylobacter vaccine for chickens with varying degrees of success, including whole-cell, DNA, carrierbased and subunit vaccines. Data on this topic have been collected recently in some review papers (30, 31). Many studies have indicated that vaccine formulations based on a single antigen exhibit inadequate protective potential. In our previous study, a recombinant rCjaAD protein was created. It is composed of Campylobacter jejuni antigen A protein (CjaA) which presented three selected Campylobacter jejuni antigen D (CjaD) epitopes on its surface. Administration of rCjaAD to chickens, either encapsulated in liposomes or using a carrier strain of lactic acid bacteria, resulted in a reduction of Campylobacter in the intestines by approximately 2 log₁₀ (24, 25). Hence, in this study, a subunit vaccine for poultry was developed, composed of a mixture of four Campylobacter antigens in the form of two hybrid proteins. We used two immunogenic Campylobacter proteins, CjaA and CjaC (Campylobacter jejuni antigen C), as the “backbone”, and two others – CjaD and EF-Tu (elongation factor thermo unstable), as sources of epitopes, resulting in rCjaAEF-Tu and rCjaCD.

The “backbone” components are glycosylated extracytoplasmic lipoproteins. They belong to the substrate-binding adenosine triphosphate–binding (ABC) cassette transport system and are conserved among various Campylobacter serotypes. Crystallographic analyses of rCjaA have indicated that this protein may be responsible for cysteine transport; however, growth experiments have suggested that other amino acids may also be involved (40). Conversely, histidine is the most probable ligand for CjaC (12). Amino acids constitute the crucial elements of Campylobacter metabolism as they serve not only as a nitrogen source but also as carbon and energy supplies. Therefore, proteins involved in amino acid transport play a significant role in many important physiological processes in this microorganism. Additionally, the expression of the cjaA gene is upregulated in iron-deficient media, suggesting the involvement of CjaA in colonisation processes in vivo (20). It has also been shown that both the CjaA and CjaC proteins occur in higher quantities in cells of fresh clinical isolates than in cells of repeatedly passaged laboratory strains (8). Moreover, they are recognised by maternal antibodies, which protect chicks from Campylobacter during the first weeks of life (8, 35).

The CjaA and CjaC “backbone” proteins were further modified by adding epitopes derived from EF-Tu and CjaD proteins, respectively. Numerous recent studies have shown that EF-Tu may be implicated in bacterial pathogenesis in addition to its involvement in protein synthesis (17). This protein is recognised by secretory immunoglobulin A (IgA) in one-week-old chicks and is responsible for inducing specific IgY during Campylobacter infection (19). The peptidoglycan-associated lipoprotein CjaD, anchored to the outer membrane of bacterial cells, was proved to be highly immunogenic in many pathogens (15).

A subunit vaccine of an assortment of Campylobacter antigens carried in two types of liposomes was administered to chicks to assess the protective potential of the antigens. The study also expanded the body of knowledge on liposome multi-antigen vaccines.

High levels of colonisation of chicken intestines by Campylobacter do not cause disease symptoms in birds and do not affect the efficiency of breeding or the yield of meat. This has led to a lack of interest from the poultry industry in developing a vaccine for chickens against these pathogens. However, the increasing number of cases of campylobacteriosis in humans, complications after infection, and the growing number of antibiotic-resistant strains of Campylobacter have prompted the search for effective strategies to prevent Campylobacter infections.

One approach is to vaccinate people in high-risk groups, such as children under the age of five. The presence of antibodies in people who have suffered infection and the lack of disease in adults in endemic areas indicate that while developing a vaccine for humans is not an easy task, it is feasible. Difficulties arise from, among other things, an incomplete understanding of the pathogenesis of the disease agent and the great diversity of Campylobacter strains. The ability of Campylobacter to cause GBS also raises serious concerns about the safety of vaccines, especially those based on killed or attenuated strains of the microorganism. So far, only one formulation has reached the clinical trial stage. It is a vaccine using the capsule polysaccharides as antigens (US National Library of Medicine ClinicalTrials.gov identifier NCT02067676).

Another solution that could help reduce human infections is to decrease the level of chicken intestinal colonisation by human-pathogenic strains of Campylobacter spp. Various attempts have been made to control the spread of infections in chickens under mass-rearing conditions. However, measures to improve farm hygiene and biosecurity, such as the use of competitive exclusion or feed or drinking water additives (e.g. organic and fatty acids, plant-derived products, bacteriocins, bacteriophages, biofilm formation inhibitors or quorum-sensing inhibitors) are not sufficient to effectively control Campylobacter or eliminate it from the poultry food chain. It appears that the most effective method of anti-Campylobacter prophylaxis may be immunisation of chickens. However, despite intensive research to develop a prototype of such a vaccine, still no vaccine is commercially available.

To date, the question of what type of immune response would effectively protect chickens from Campylobacter colonisation has not been answered. Sahin et al. (33) demonstrated that maternal Campylobacter-specific antibodies played a role in protecting young chickens from colonisation, but this protection was short-lived and did not cover the entire period of broiler production. This study nevertheless indicates that boosting immunity may be a rational and feasible way to control Campylobacter in livestock farming. The relevance of the humoral immune response in preventing Campylobacter colonisation of chickens was confirmed by Hermans et al. (19). Their study showed that passive immunisation, which involved administering IgY obtained from Campylobacter-infected hens, significantly reduced the number of cells of these microorganisms in the intestines and prevented transmission of the pathogen between birds. Furthermore, recent research suggests that antimicrobial peptides and gut microbiota composition may play a crucial role in this process (13).

Numerous vaccination attempts have been made so far using various methods (31). Among recent trials, the utilisation of live, attenuated Campylobacter strains is noteworthy. Those strains carried mutations in the ahpC or katA oxidative stress defence genes, and when administered to chickens, they notably decreased colonisation by wild Campylobacter strains. Nevertheless, they were unable to permanently colonise the intestines of birds and could not survive in aerobic environments because their sensitivity to oxygen was greater (21).

Attention is also given to subunit vaccines, as they are generally regarded as safer than live, attenuated kinds and are easy to produce or modify. To date, several potentially useful proteins have been tested in animal models in design work on anti-Campylobacter vaccines. These include but are not limited to the Campylobacter adhesion-to-fibronectin protein (CadF), DNA binding protein from starved cells (Dps), C. jejuni surface-exposed lipoprotein A (JlpA), component protein of the multidrug efflux pump (CmeC), haemolysin coregulated protein (Hep), flagellin A (FlaA), fibronectin-like protein A (FlpA) and lipoprotein CjaA. The immune response induced by FlaA heat-labile enterotoxin B subunit (FlaA-LTB), CmeC, Hcp, FlpA, CadF and JlpA provided limited protection, that induced by Dps and FlaA did not protect against infection with heterologous strains and the response to CjaA and FlaA was highly variable depending on the dose of the product, route of administration, and application regimen (31). These outcomes suggest that vaccine preparations based on a single antigen offer insufficient protection. Only a multicomponent vaccine has the potential to efficiently diminish Campylobacter colonisation of chicken intestines. This was demonstrated by the results obtained by Neal-McKinney (28), who vaccinated chickens with individual CadF, FlaA or FlpA proteins to reduce C. jejuni colonisation somewhat, but who achieved the highest level of protection with a trifecta protein containing CadF, FlaA and FlpA. Another example of such an approach is the formulation developed by Vandeputte et al. (39), which encompassed six immunodominant antigens: adenosine triphosphate synthase subunit (AtpA), EF-Tu, heat-shock protein 60 (GroEL), trigger factor chaperone (Tig), chemotaxis protein V (CheV) and leucine/isoleucine/valine-binding protein (LivJ). After immunisation of hens, the researchers observed elevated levels of specific IgY against the administered antigens in the egg yolks. Prophylactic feeding of broilers with yolk from vaccinated hens led to a significant reduction in the number of birds colonised by Campylobacter (39). Similarly, a protective effect in hens was observed when birds were fed an avirulent strain of Salmonella that produced two Campylobacter antigens – CadF and FlaA (4). Nonetheless, it is worth considering that the beneficial effect may also have been gained in part through the method of their application.

In the present study, two hybrid proteins – rCjaAEF-Tu and rCjaCD – were prepared. The CjaA protein was frequently used as an antigen, but the results obtained were inconsistent. Some studies reported a reduction in the level of intestinal colonisation by approximately 6 log₁₀, while others showed no effect (3, 42). The second backbone protein, which is CjaC, is highly immunogenic. Studies support the use of CjaA and CjaC lipoproteins as antigens, as they induce strong toll-like receptor 2–mediated innate immune responses in chickens (44). Recent data suggested that the innate immune response may play an important role in protecting chickens from colonisation by Campylobacter (32). One of the proteins used as a source of epitopes was EF-Tu, which apart from its main role in polypeptide synthesis has recently been implicated in bacterial pathogenesis (17). The second source of epitopes was the highly immunogenic CjaD lipoprotein (15).

After a series of structural analyses of backbone proteins combined with predicted epitopes, the two hybrid proteins rCjaAEF-Tu and rCjaCD were obtained and purified. It was confirmed that the epitopes selected for their construction were recognised by specific sera, and that they also induced the production of antibodies which reacted with the native proteins from which the epitopes were derived. The hybrids were then encapsulated in liposomes and orally administered to the chickens. Liposomes are attractive carriers of vaccine antigens. The advantage of liposomes, which are bilayered vesicles composed of amphipathic phospholipids, is that they extend the time of antigen release from their interior and protect their content from excessive proteolysis. In addition, liposomes facilitate uptake by antigen-presenting cells (APCs), leading to stimulation of both humoral and cellular immune responses (34). Furthermore, liposomes can be produced on a large scale, stored for a relatively long time, and most importantly, are considered safe and well tolerated by animals and humans. Some studies have indicated that liposomes can be used successfully to stimulate the immune system in chickens (6, 43).

From our previous experiments, we know that the administration of CjaA and rCjaAD antigens caused an increase in specific IgY and IgA antibodies (14, 25). However, this did not result in any significant reduction of the colonisation of the chicken digestive tract. Therefore, in the planned study, evaluating the protective effect and its outcome was crucial regardless of the immunological status of animals.

Despite the use of hybrid proteins with carefully selected immunogenic epitopes, we did not observe the expected reduction in colonisation in immunised birds. There may be various reasons for this result and further investigation is required. The most probable issues to be considered are liposome–hybrid interaction or antigen selection. Positively charged liposomes were shown to be much more effective than negatively or neutrally charged molecules because they interacted better with negatively charged mucosal surfaces, thus prolonging antigen exposure time (2). They are also more efficiently taken up by APCs (18). However, the results of studies on liposome interactions with APCs, the influence of liposome size, their cargo and the phase state of the bilayer are often contradictory and should be verified on a case-by-case basis. The hybrid proteins we produced were incorporated into the structure of anionic and neutral liposomes, because packing them into cationic forms had failed. A factor deserving investigation may be the charge characteristics of the liposomes. Another aspect requiring investigation is the exact nature of the interaction of the hybrid proteins with liposomes (e.g. their surface exposure). Some studies indicate that liposomes elicit a stronger immune response when antigens are exposed on their surfaces rather than encapsulated within. It was shown that packaging was very efficient; however, one of our hybrids, rCjaAEF-Tu, always fragmented during the incorporation process, regardless of the type of liposome used. This observation suggests that the surface presentation of antigens may be a factor driving the efficiency of such vaccine formulations.

Antigen selection is crucial. Both backbone proteins used (CjaA and CjaC) were established to have immunogenic properties in chickens (8, 25, 42). The source of epitopes for the rCjaAEF-Tu hybrid, EF-Tu, was shown to be involved in the pathogenesis of various microorganisms, and importantly it was capable of stimulating secretory IgA in chickens (19). The immunostimulatory effects of CjaD protein, the source of epitopes for the rCjaCD hybrid, were also documented before (15). A similar approach to ours combining various immunogenic determinants was presented by Lou et al. (26). A hybrid protein containing epitopes of alkyl hydroperoxide reductase subunit C (AhpC), Omp18 (CjaD) and flagellar L-ring protein precursor (FlgH) was used to immunise specific pathogen-free mice from the BALB/c line, other mice were immunised with only one selected from these three proteins or not immunised at all, and all mice were later challenged with Campylobacter. The disease index and protection rate were then calculated based on the disease symptoms observed in the immunised and non immunised mice. The protection rate was higher for the hybrid protein (80%) than for each of the three proteins separately (30–50%). Although all of our chosen proteins exhibited promising features and both hybrids were positive in Western blots with sera obtained from 7-day-old chickens, it is impossible to predict or guarantee an exact protective effect.

The use of liposomes in multi-antigen vaccines has shown potential, as seen in the vaccines against influenza (Inflexal), hepatitis A (Epaxal) and malaria (Mosquirix). Additionally, incorporating immunostimulatory elements like 5′–C–phosphate–G–3′ (ligands of pattern recognition receptors) into vaccine complexes can modulate the immune response (5).

It is also extremely important to clarify the role of gut microbiota in Campylobacter infection in chickens. To date, only a few studies have addressed this issue (41). For example, Kaakoush et al. (23) linked the presence of C. jejuni in the digestive tract of chickens with a lower abundance of Lactobacillus and Corynebacterium and a higher abundance of both Streptococcus and Ruminococcaceae. Changes in the abundance of Clostridiales in response to Campylobacter colonisation have also been observed (7, 38). The effects of removing Campylobacter from the chicken microbiota are unknown. In the event of its eradication through vaccination, a thorough analysis of the microbiota composition and assessment of the wellbeing of immunised birds will undoubtedly be necessary.

Although the immunisation described in the present study did not produce the desired results, further exploration of this approach is still worthwhile.