In vitro and in vivo activity of Lactobacillus sakei L14 strain against Campylobacter jejuni DC3 strain

Campylobacter is one of the most important enteropathogens that cause diarrhoeal diseases. The diseases with this aetiology, referred to as campylobacteriosis, are commonly manifested as gastroenteritis accompanied by mild to severe diarrhoea. The incidence of human campylobacteriosis mainly caused by C. jejuni and C. coli has been increasing worldwide. In some parts of the world, it exceeds those of salmonellosis and shigellosis. The main mode of transmission is via the ingestion of contaminated food, primarily chicken (3), and this species is considered a natural reservoir of Campylobacter, as are turkeys and ducks. Birds are infected via the faecal–oral route and are primarily colonised in the gut. The prevalence of Campylobacter in poultry has been observed to vary depending on geographical location, season, and production system (3, 21). The host–pathogen interaction between chicken and Campylobacter is predominantly commensal and asymptomatic; however, studies have shown that Campylobacter is capable of invading gut cells and entering blood vessels, causing extraintestinal infection of other organs such as the liver and spleen. Evidence suggests that chickens mount an immune response to Campylobacter and other commensal bacteria (21). In addition to Campylobacter, lactic acid bacteria (LAB) are also present in the gastrointestinal tract of birds and these bacteria constitute a Gram-positive, non-spore-forming group that produce lactic acid as the major end-product of carbohydrate fermentation. As examples of LAB that have been isolated from chickens, Enterococcus, Lactobacillus, Pediococcus, and Streptococcus have been shown to produce antibacterial substances such as organic acids, hydrogen peroxide, carbon dioxide, and bacteriocins (13, 19).

The objective of this study is to determine whether the natural resistance of chickens to Campylobacter is mediated by LAB. Specifically, this work aimed to isolate C. jejuni and LAB from the gastrointestinal tracts of healthy poultry chickens; determine the interaction between the isolated C. jejuni and LAB via cell-to-cell contact; determine if the cell-free supernatant of LAB is capable of inhibiting C. jejuni, and determine the effect of C. jejuni co-infection with LAB in an in vivo chicken model. This study determined the role of LAB in C. jejuni colonisation in chickens and may contribute to the current knowledge as to why chickens remain asymptomatic despite being colonised by Campylobacter.

Sample collection. A 4-week-old broiler chicken was obtained from a wet market in the Philippines. The chicken was humanely euthanised via cervical dislocation following the recommendations of the Handbook of Laboratory Animal Management and Welfare (29). The small intestine, large intestine, and cecum were dissected and stored in an ice-cold airtight container, which was transported to the laboratory within 4 h of collection. The samples were homogenised in 10⁻¹ volumes of 0.9% sterile saline. The resulting mixture was weighed and equally divided for the isolation of Campylobacter and endogenous LAB. For exogenous LAB, five poultry chicken intestinal tracts were procured from wet markets in Metro Manila. The gut samples from all five birds were pooled and homogenised in 10⁻¹ volumes of 0.9% sterile saline.

Isolation and biochemical tests. The enrichment protocol for Campylobacter using Bolton broth as specified in the ISO 10272-1:2017 standard (11) was followed with modifications. Briefly, 25 mL of the homogenised sample was enriched in 100 mL of Bolton selective enrichment broth supplemented with cefoperazone (20 mg/L), vancomycin (20 mg/L), trimethoprim (20 mg/L), cycloheximide (50 mg/L), piperacillin-tazobactam (8 mg/1 mg/L), and 5% lysed horse blood. The culture was pre-enriched at 37°C for 4 h under microaerobic conditions and incubated at 42°C for 48 h under the same conditions. The enrichment was streaked onto chromogenic Mueller–Hinton agar (MHA) plates supplemented with cefoperazone (32 mg/L), amphotericin B (10 mg/L), piperacillin-tazobactam (8 mg/1 mg/L), and 5% lysed horse blood and the plates were incubated at 42°C for 48 h under microaerobic conditions. Colonies that were dark matte purple coloured were selected for purification. Presumptive Campylobacter spp. isolates were subcultured on MHA plates with 5% lysed horse blood and incubated at 42°C for 48 h under microaerobic conditions for DNA extraction and PCR amplification. For LAB isolation, 10 mL of each homogenised sample was serially diluted to 10⁻⁶ in 0.9% sterile saline solution. A 100 μL volume of each of the last three dilutions (10⁻⁴, 10⁻⁵, and 10⁻⁶) was pipetted onto de Man, Rogosa, and Sharpe (MRS) agar plates supplemented with 0.5% CaCO₃. The plates were incubated at 37°C for 48–72 h under microaerobic conditions. Colonies with zones of clearing were purified on MRS agar plates, which were incubated as previously described. Gram staining and a catalase test were performed based on the protocols of Smith and Hussey (25) and Reiner (20), respectively, and a hippurate hydrolysis test was conducted as described by Morris et al. (18).

Screening for antimicrobial activity. Primary screening was performed via a cross-streak assay. Lactic acid bacteria isolates were seeded on MHA plates with 5% lysed horse blood until one-third of the surface was covered. The plates were then incubated at 37°C for 48 h under microaerobic conditions. On the free surface of the same MHA plates, the C. jejuni isolate and the C. jejuni ATCC 33560 quality control strain (Microbiologics, St. Cloud, MN, USA) were aseptically streaked horizontally towards the LAB growth. Streaking of the test pathogens was performed in duplicate. The plates were incubated at 42°C for 48 h under microaerobic conditions, after which the LAB isolates that exhibited inhibitory activity against the C. jejuni isolate and the control organism C. jejuni ATCC 33560 were selected for secondary screening.

Secondary screening was performed using the agar plug diffusion method. Lawns of LAB were prepared on MHA plates with 5% lysed horse blood, using cultures adjusted to 0.5 McFarland. A sterile swab previously dipped once into tubes containing the adjusted inoculum was streaked three times on the MHA plates to cover the entire surface of the medium. The plates were then incubated at 37°C for 48 h under microaerobic conditions. Bacterial lawns of the C. jejuni isolate were prepared as previously described using cultures adjusted to 0.5 McFarland and streaked onto MHA plates with 5% lysed horse blood. Five-millimetre agar plugs were bored from LAB-seeded plates and were aseptically transferred onto the C. jejuni isolate lawns. The plates were incubated at 42°C for 48 h under microaerobic conditions, after which the areas of inhibition were measured using a Vernier caliper.

Molecular identification. Genomic DNA of the presumptive Campylobacter isolate and the priority LAB isolate were extracted using a G-spin Genomic DNA Extraction kit (iNtRON Biotechnology, Seongnam, South Korea), following the manufacturer’s recommendations. A 20 μL PCR reaction mixture composed of 10 μL of GoTaq Master Mix (Promega, Wisconsin, USA) (with 50 units/mL of Taq DNA polymerase in a proprietary reaction buffer (pH 8.5), 400 μM each of deoxyadenosine triphosphate, deoxyguanosine triphosphate, deoxycytidine triphosphate, deoxythymidine triphosphate, and 3 mM of MgCl₂), 1.2 μL of each primer (0.6 μM), 6.6 μL of nuclease-free water, and 1 μL of template DNA was prepared. Amplifications were carried out in a Labnet MultiGene Gradient PCR thermal cycler (Labnet International, Edison, NJ, USA) using Campylobacter spp.- and C. jejuni-specific primers (Table 1) and following the optimised PCR conditions described by Subejano and Penuliar (26). The genomic DNA of the C. jejuni isolate and the priority LAB isolate were sent to Macrogen Inc. (Seoul, South Korea) for DNA sequencing. The consensus sequences of the forward and reverse sequences were obtained using Bioedit Sequence Alignment Editor v. 7.0.5 (6) and used to identify the isolates through cross-referencing with the nucleotide sequences deposited in the National Center for Biotechnology Information Nucleotide BLAST database.

Preparation and collection of cell-free supernatant. The growth of C. jejuni DC3 isolate was determined in broth cultures with different preparations of cell-free supernatant (CFS). Two types of CFS were collected: CFS-MC from a monoculture of L. sakei L14 isolate and CFS-CC from a co-culture of L. sakei L14 and C. jejuni DC3. For CFS-MC, 1 mL of 1.0 McFarland-adjusted LAB culture was inoculated into 149 mL of yeast autolysate–peptone–tryptone–Tween 80–glucose (LAPTg) broth. For CFS-CC, 1 mL each of 1.0 McFarland-adjusted LAB and C. jejuni isolate were inoculated together into 148 mL of LAPTg broth. The preparations were incubated at 37°C under microaerobic conditions. Approximately 30 mL of the cultures were harvested after 24, 48, and 72 h of incubation by centrifugation at 12,000 rpm for 1 min. The supernatants were collected, passed through a 0.45 μm polyether sulfone syringe filter (Membrane Solutions, Shanghai, China), and pooled in 50 mL conical tubes.

Campylobacter jejuni in LAB CFS. Campylobacter jejuni DC3 was cultured in different concentrations of CFS-MC and CFS-CC in separate scintillation vials for final volumes of 10 mL. A single concentration of C. jejuni DC3 (3.0 × 10⁸ CFU/mL) was used in all treatments. One millilitre of the 1.0 McFarland-adjusted C. jejuni suspension was cultured in CFS-MC and CFS-CC, each with five different proportions of CFS (0%, 12.5%, 25%, 50%, and 100%) harvested after three different timepoints (24, 48, and 72 h). Uninoculated vials with only the culture medium were prepared to serve as sterility controls. Each treatment was performed in triplicate. The preparations were incubated at 42°C under microaerobic conditions. To measure the growth of C. jejuni DC3, spectrophotometry was performed after 48 h of incubation with the LAB CFS, and absorbance readings were used to plot growth curves using Microsoft Excel (Microsoft, Redmond, WA, USA).

Campylobacter jejuni in neutralised and heat-treated CFS. CFS-MC and CFS-CC were neutralised using 1 N NaOH. Scharlau pH indicator strips (Scharlab, Barcelona, Spain) were used to monitor the changes in pH. A separate set of ten-millilitre volumes of CFS-MC at the proportions previously stated and an equivalent set with CFS-CC were heat-treated at 121°C and 15 psi for 15 min using a BKQ-B II vertical autoclave (BIOBASE, Shandong, China). Campylobacter jejuni DC3 was cultured in different concentrations of CFS in separate scintillation vials for final volumes of 10 mL. The CFS that were harvested after 72 h were used for each set-up. A single concentration of C. jejuni DC3 (3.0 × 10⁸ CFU/mL) was used in all treatments. One millilitre of the 1.0 McFarland-adjusted C. jejuni suspension was cultured in both the CFS previously described, each with three different proportions of CFS (0%, 50%, and 100%) harvested after 72 h. As previously, uninoculated vials with only the culture medium were prepared to serve as sterility controls. Each treatment was performed in triplicate. In this operation also, the preparations were incubated at 42°C under microaerobic conditions. To measure the growth of C. jejuni DC3, spectrophotometry was performed after 48 h of incubation with the CFS and absorbance readings were used to plot growth curves using Microsoft Excel.

Statistical analysis. Significant differences in the absorbance readings obtained for each type of CFS, concentration of CFS, and incubation times used were determined using analysis of variance (ANOVA) (α = 0.05) and post-hoc Tukey’s Honestly Significant Difference (HSD) test.

In vivo challenge with C. jejuni and LAB. Twenty 1-day-old female Cobb broiler chicks were procured and were housed in an animal facility outside the University of the Philippines. Four groups were established: Group 1 for no treatment, Group 2 for inoculation with C. jejuni DC3 only (control group), Group 3 for inoculation with C. jejuni DC3 and L. sakei L14 concurrently, and Group 4 for inoculation with L. sakei L14 only. Separate cages with five chicks each were prepared for each treatment. The chicks were acclimated for 7 days prior to any treatment. The temperature was maintained at 30–36°C and monitored daily and adequate internal air circulation was provided. Sterile commercial chick starter mash/feeds (GMP 100) and sterile drinking water were given ad libitum using designated containers. The chicks were given antibiotics on days 1–7 to deplete their existing gut microflora, following the protocol by Han et al. (7) with modifications. The antibiotic cocktail was composed of broad-spectrum antibiotics, specifically doxycycline (0.5 mg/mL), tiamulin (0.12 mg/mL), amoxicillin (0.125 mg/mL), tylosin (0.08mg/mL), ciprofloxacin (0.125 mg/mL), sulfadimethoxine (0.2 mg/mL), and trimethoprim (0.04 mg/mL), dissolved in sterile drinking water. Faeces were collected using sterile cotton swabs and transferred into 5 mL of Bolton broth with 5% lysed horse blood and 5 mL of MRS broth on day 8. The samples were transported to the lab where they were incubated at 42°C for C. jejuni and 37°C for LAB for 48 h under microaerobic conditions. The enrichments were processed for DNA extraction and were screened by PCR for Campylobacter spp. andL. sakei. Primer sequences for the LAB group and the target LAB were adapted from the literature (26) and are shown in Table 2, and PCR conditions were optimised in the laboratory.

Upon confirming the absence of the target bacteria, each group was treated with C. jejuni and L. sakei via ingestion of a standardised bacterial suspension every 2 days for 31 days. Approximately 25 mL of C. jejuni and LAB at 9 × 10⁸ CFU/mL concentration were prepared separately in lactated Ringer’s solution. On each second day, 25 mL of the standardised bacterial suspension was administered to each group of chicks. Group 1 was neither infected with C. jejuni nor LAB. Group 2 was infected with C. jejuni only for 31 days between days 8 and 39. Group 3 was simultaneously infected with C. jejuni and LAB for 31 days between the same start and end points, and Group 4 was infected with LAB only with identical duration parameters. Faecal samples were collected and processed as previously described to confirm the presence of C. jejuni and LAB at 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29 and 31 days post infection (dpi). At the end of the experiment, the chickens were humanely euthanised via cervical dislocation in order for their digestive tract tissue to be sampled to detect the presence of C. jejuni via PCR.

Ten presumptive Campylobacter spp. were isolated, but only isolates DC1, DC2, DC3, and DC4 were initially identified as C. jejuni using species-specific primers. These four isolates were all Gram-negative and hippurate-positive. The third one, DC3, was chosen as the priority isolate due to its sustained viability. Sequencing of its 16S rDNA region identified DC3 as C. jejuni with 99.9% identity.

There was successful isolation and purification of 12 endogenous and 30 exogenous LAB. The exogenous LAB isolate L14 was chosen as the priority isolate based on its high activity against C. jejuni DC3 and C. jejuni ATCC 33560, in addition to its sustained viability. It was determined to be Gram-positive and catalase-negative. DNA sequencing of the16S rDNA region of L14 identified it as Lactobacillus sakei. The endogenous LAB isolates did not show any inhibitory activity against C. jejuni DC3 or C. jejuni ATCC 33560 (data not shown). For the exogenous LAB, 12 isolates inhibited the growth of C. jejuni DC3 and C. jejuni ATCC 33560 (Fig. 1). Only 10 isolates were subjected to secondary screening, however, because of loss of viability affecting two isolates. Six LAB isolates showed consistent activity against both C. jejuni DC3 and C. jejuni ATCC 33560, with L. sakei L14 exhibiting the most activity (Fig. 2).

Fig. 1

Fig. 2

Lactobacillus sakei L14 CFS generated from the co-culture with and without C. jejuni DC3 was able to inhibit C. jejuni DC3 growth in a dose-dependent manner (Figs 3 and 4). Co-culture–derived CFS harvested at 72 h was most efficient in inhibiting C. jejuni DC3 growth, starting at a 25% proportion of the suspension being supernatant. Neutralised L. sakei L14 CFS was unable to inhibit C. jejuni DC3 growth (P > 0.05), as shown in Fig. 5. On the other hand, heat-treated L. sakei L14 CFS harvested after 72 h was shown to inhibit C. jejuni DC3 growth to greater extents as its proportion increased, as seen in Fig. 6. Significantly different values between the 0% and 50% and the 0% and 100% proportions were determined using ANOVA (α = 0.05) and post-hoc Tukey’s HSD test.

Fig. 3

Fig. 4

Fig. 5

Fig. 6

PCR detection of C. jejuni DC3 from pooled faecal samples from each preparation shows that colonisation of C. jejuni occurred as early as 5 dpi (Table 3a). Campylobacter jejuni was only detected once in Group 2 (infected only with this bacterium) during the 31-day infection period, while in Group 3 (infected with this bacterium and LAB) it was detected, although not consistently, at several points during the observation period (Table 3a).

Lactobacillus sakei L14 was detected as early as 1 dpi in groups 3 and 4 (the former infected with bacteria of both genera). In Group 3, it was evident at 13, 19, and 21 dpi and in Group 4 from 19 to 25 dpi. However, it was undetectable in Group 3 and Group 4 at 23 dpi and 27, respectively (Table 3b).

Table 3c shows that LAB were consistently detected in all treatment groups before and during treatment. The results of PCR amplification of C. jejuni from individual gut samples were consistent with the results of assays to detect it in the pooled faecal samples, suggesting that C. jejuni DC3 was able to successfully colonise the chickens in the co-infection group (Table 4).

Lactic acid bacteria are a diverse group of microaerophilic organisms that can be isolated from many sources, especially fermented foods and animal intestines. Aside from their major role in the food technology industry because of their antimicrobial activity, probiotic LAB have also been considered as an alternative therapy in alleviating discomfort caused by gastrointestinal infections (22). Their antibacterial property is based on the production of lactic acid and secretion of growth-inhibiting substances such as hydrogen peroxide, diacetyl, and bacteriocins (15). One of these bacteria is Lactobacillus sakei, which is naturally found on raw and fermented meats (15, 30). Despite being widely associated with animal muscle tissue and therefore, with food, it has also been isolated from chicken intestines (1, 24), a finding that was confirmed in this study. While there have been numerous studies on L. sakei demonstrating its use as a protective culture for meat products (14, 30), there have also been reports on its probiotic potential, especially in the poultry industry (22, 24).

Lactobacillus sakei and other exogenous LAB in this study exhibited anti-C. jejuni activity in vitro through direct cell-to-cell contact in screening assays. In a similar study, Jones et al. (13) were also able to report the same activity by two strains of L. sakei from meat products. It is interesting to note that none of the endogenous LAB or those isolated from the same chicken intestine as C. jejuni DC3 inhibited the growth of the Campylobacter isolate via direct cell-to-cell contact. It is possible that these LAB isolates are species other than Lactobacillus that do not produce secretory products capable of inhibiting the growth of Campylobacter. It should also be noted that only a small number of endogenous LAB were isolated (n = 12) and it is likely that other isolates with potential anti-C. jejuni activity were present but were not selected during isolation. It is also possible that these LAB isolates exert their antagonistic effects through indirect mechanisms of action. Studies have demonstrated how C. jejuni inhibition is LAB strain- and species-specific (16, 22).

In this study, C. jejuni DC3 growth was inhibited by CFS-MC and CFS-CC, but significant pathogen growth inhibition was observed only by CFS-CC generated after 72 h of incubation. Co-incubation of LAB with its target pathogen has been shown to be effective in inhibiting growth. Drago et al. (5) reported the same findings in L. paracasei that was able to inhibit the growth of Escherichia coli and Salmonella enteritidis in vitro, either in the simultaneous inoculation setup or when the pathogens were added after L. paracasei was first grown overnight. The inhibitory substances produced by LAB are secreted when the bacteria are grown in an appropriate medium. In testing the inhibitory activity of the CFS in this work, LAPTg broth was the culture medium of choice because it is the conventional medium used in LAB cultivation and has been reported to be the optimal medium for bacteriocin production. It also has lower amounts of potentially interfering peptides compared to MRS (4). There are studies that have attributed the antimicrobial activity of LAB CFS against Gram-negative pathogens to lactic acid accumulation and the concomitant lowering of pH (28) and which have posited that neutralisation will eliminate the antagonistic effect of LAB (16, 28), as observed in this study. The antibacterial activity of lactic acid and other organic acids is said to be due to their ability to penetrate the cytoplasmic membrane in their undissociated forms. This results in a reduction of the intracellular environment’s pH and a subsequent disruption of the proton motive force. In Gram-negative bacteria, lactic acid has the ability to eliminate the lipopolysaccharide layer, which confers the permeability barrier property of the outer membrane. This consequently disrupts the outer membrane, which may enhance the action of other antimicrobial substances (2).

The potential of L. sakei L14 to produce bacteriocins that can inhibit C. jejuni was preliminarily explored with the use of LAPTg broth as growth medium. Bacteriocins exhibit species-specific or broad-spectrum antimicrobial activity and may be heat stable or heat labile. Examination of the bacteriocinogenic potential of L. sakei L14 after heat treatment revealed that heat-treated CFS-MC and CFS-CC retained their anti-C. jejuni activity, but did not show any further pathogen growth inhibition, even with increased CFS concentration.

Despite the evident in vitro activity of L. sakei L14 against C. jejuni DC3, the results of the chicken model experiment demonstrate how the LAB isolate failed to inhibit the pathogen in vivo. Although many LAB have been examined for their anti-Campylobacter activity in vitro, only a few studies have confirmed this effect in vivo. The LAB-associated Bifidobacterium longum PCB 133 actinobacterium showed anti-C. jejuni activity in vitro and in vivo (23). Another study reported that L. crispatus JCM 5810, L. acidophilus NCFM, L. gallinarum ATCC 33199, and L. helveticus CNRZ32 were able to reduce the level of C. jejuni colonisation in broiler chicks (19). There are different mechanisms by which LAB can reduce Campylobacter colonisation in the gut. Aside from the production of organic acids and bacteriocins, LAB may also compete with pathogens for nutrients or strengthen the tight junctions in the intestinal epithelium to inhibit pathogen invasion. Lactic acid bacteria can also colonise the intestinal epithelium so that pathogenic bacteria cannot adhere and invade epithelial cells or may directly bind to the pathogen. Immune system modulation by LAB is also a possible mechanism that may facilitate the host’s mounting an effective immune response against pathogens (22).

The in vivo experiment was more complex and included host factors that could have influenced the activity of L. sakei L14, such as avian gut morphology and health, the microbial communities present, and the avian immune system. Taking into consideration these host factors and the different mechanisms by which LAB can deliver beneficial effects to its host, it is likely that L. sakei L14 confers its protective effect through mechanisms other than the production of organic acids and antimicrobial substances. Sim et al. (24) reported that in the murine model, heat-killed cells of two L. sakei strains were able to induce high levels of immune cell proliferation as well as stimulate the production of interleukin (IL) 6 and tumour necrosis factor alpha (TNF-α). Hong et al. (8) were also able to show the immune-inducing potential of L. sakei in the murine model when it exhibited the ability to stimulate IL-12, interferon gamma, and TNF-α production. Based on these previous findings, it is possible that L. sakei L14 may have also stimulated the avian immune system, in addition to stimulating organic acid production and probably the secretion of antimicrobial peptides.

Although L. sakei strains have been isolated from human faeces, they are most often isolated from food products such as fermented cabbage and meat. Because of its presence in food, L. sakei is mostly considered a diet-associated LAB, which implies that the gastrointestinal tract may not be its natural environment (30). This may help explain why L. sakei L14 did not inhibit the pathogen’s growth in vivo. In addition, Todorov et al. (27) reported that the bacteriocins of L. sakei, generally called sakacins, are more active against Gram-positive foodborne pathogens such as Listeria monocytogenes. Furthermore, LAB strains that show inhibitory activity against Gram-negative bacteria are more common in environments where co-evolution with other Gram-negative competitors has occurred (13).

The results of PCR detection of L. sakei L14 in faecal samples in this study suggest that its occurrence may only be transient in the chicken model. Previous studies confirm that most Lactobacillus strains can only temporarily colonise the chicken gut and some remain undetectable after infection (16, 23). As stated earlier, they are mostly diet-associated and may not be native to the gastrointestinal tract. The genome analysis of an L. sakei strain isolated from a French sausage revealed that it employs several strategies in order to grow on meat products. Its robustness enables it to outgrow competitors and survive the environmental stresses associated with food processing such as high salt concentration, cold stress, and the presence of antimicrobial substances (13). These genomic characteristics make it well adapted to food-associated environments.

The absence of Campylobacter in the control group suggests that C. jejuni DC3 failed to colonise the gut, which could be due to the persistence of LAB and other bacteria there despite the antibiotic treatment prior to the challenge. The presence of Campylobacter in the co-infection group having been inconsistent suggests that it may not have been able to maintain its ability to colonise the gut, which could be attributed to the presence of L. sakei L14 and other LAB. While it was detected by PCR, this does not directly suggest that L. sakei did not have the ability to reduce the Campylobacter load, so further examination is recommended for a quantitative evaluation of the anti-Campylobacter activity exerted by L. sakei in vivo. The persistence of LAB across the groups despite antibiotic treatment suggests that an extended antibiotic treatment period may be necessary, in addition to variegation of the antibiotic cocktail administered, to ensure the elimination of all endogenous LAB.

In conclusion, the chicken intestine-isolated L. sakei L14 exhibited antagonistic activity against C. jejuni DC3 in vitro. The initial characterisation of the LAB isolate’s anti-Campylobacter activity suggested the production of organic acids. Its in vivo activity, however, did not match the in vitro findings in that C. jejuni was still able to colonise the chicks, although inconsistently. This suggests that protective mechanisms other than organic acid production might be in play, given the complexity of the avian gut environment. The bacteriocinogenic potential of L. sakei L14 should be explored, considering that it has been isolated from the gut despite being widely associated with food products and therefore with muscle tissue. Its immunomodulatory ability also warrants further examination, as the isolate has the potential to be used as a chicken probiotic aimed at reducing Campylobacter carriage by poultry.