The in vivo efficacy of two administration routes of a phage cocktail to reduce numbers of Campylobacter coli and Campylobacter jejuni in chickens

Abstract

Background: Poultry meat is one of the most important sources of human campylobacteriosis, an acute bacterial enteritis which is a major problem worldwide. Campylobacter coli and Campylobacter jejuni are the most common Campylobacter species associated with this disease. These pathogens live in the intestinal tract of most avian species and under commercial conditions they spread rapidly to infect a high proportion of the flock, which makes their treatment and prevention very difficult. Bacteriophages (phages) are naturally occurring predators of bacteria with high specificity and also the capacity to evolve to overcome bacterial resistance. Therefore phage therapy is a promising alternative to antibiotics in animal production. This study tested the efficacy of a phage cocktail composed of three phages for the control of poultry infected with C. coli and C. jejuni. Moreover, it evaluated the effectiveness of two routes of phage administration (by oral gavage and in feed) in order to provide additional information regarding their future use in a poultry unit.

Results: The results indicate that experimental colonisation of chicks was successful and that the birds showed no signs of disease even at the highest dose of Campylobacter administered. The phage cocktail was able to reduce the titre of both C. coli and C. jejuni in faeces by approximately 2 log10 cfu/g when administered by oral gavage and in feed. This reduction persisted throughout the experimental period and neither pathogen regained their former numbers. The reduction in Campylobacter titre was achieved earlier (2 days post-phage administration) when the phage cocktail was incorporated in the birds' feed. Campylobacter strains resistant to phage infection were recovered from phage-treated chickens at a frequency of 13%. These resistant phenotypes did not exhibit a reduced ability to colonize the chicken guts and did not revert to sensitive types.

Conclusions: Our findings provide further evidence of the efficacy of phage therapy for the control of Campylobacter in poultry. The broad host range of the novel phage cocktail enabled it to target both C. jejuni and C. coli strains. Moreover the reduction of Campylobacter by approximately 2 log10cfu/g, as occurred in our study, could lead to a 30-fold reduction in the incidence of campylobacteriosis associated with consumption of chicken meals (according to mathematical models). To our knowledge this is the first report of phage being administered in feed to Campylobacter-infected chicks and our results show that it lead to an earlier and more sustainable reduction of Campylobacter than administration by oral gavage. Therefore the present study is of extreme importance as it has shown that administering phages to poultry via the food could be successful on a commercial scale.

Figure 1

Electron micrographs of the Campylobacter phages that composed the cocktail: (a) Phage phiCcoIBB12; (b) Phage phiCcoIBB35; (c) Phage phiCcoIBB37. Phages were stained with 1% uranyl acetate and observed with a transmission electron microscopy. There was no difference in morphology between the three phages. They have an icosahedral head of approximately 100 nm in diameter and a contractile tail with 140 × 17 nm average length. This morphology is typical of the members of the Myoviridae family.

Figure 2

Single-step growth curve of the Campylobacter phages that composed the cocktail: (a) Phage phiCcoIBB35; (b) Phage phiCcoIBB37; (c) Phage phiCcoIBB12. Single-step growth experiments were performed in order to assess the latent period and burst size of a single round of phage replication: phage phiCcoIBB35 has a burst size of 24 pfu and a latent period of 52.5 min; phage phiCcoIBB37 has a burst size of 9 pfu and a latent period of 67.5 min; phage phiCcoIBB12 has a burst size of 22 pfu and a latent period of 82.5 min. Samples were taken every 15 min for 4 h. The data was fitted to a four-parameter symmetric sigmoid model. Non-linear regression was performed to calculate the latent period and burst size. Error bars represent the standard deviation.

Figure 3

Colonization of chicks by Campylobacter jejuni 2140CD1 after challenge with a range of dose levels. Eighteen, one day-old chicks were randomly assigned to one of three groups receiving by oral gavage different concentrations of 0.1 ml of PBS C. jejuni 2140CD1:low dose (7.5 × 10⁴cfu); medium dose (1.0 × 10⁶cfu) and high dose (5.5 × 10⁷cfu). Faecal samples were collected from all birds at intervals and Campylobacter and phages enumerated. Error bars represent the standard error of the mean.

Figure 4

Numbers of Campylobacter jejuni 2140CD1 (a) and phages (b) in faeces from broilers orally administered a phage cocktail by gavage. Thirty day-old chicks were inoculated with Campylobacter jejuni 2140CD1. One week later the birds were randomly assigned to a treated group or an untreated group and were inoculated by oral gavage with antacid containing 1 × 10⁶pfu of a phage cocktail, or antacid only respectively. Faecal samples were collected from all birds at intervals and Campylobacter and phages enumerated. Error bars represent the standard error of the mean. At 2 dpa, 4 dpa and 7 dpa there is a significant difference between control and infected group at P < 0.05.

Figure 5

Numbers of Campylobacter coli A11 (a) and phages (b) in faeces from broilers orally administered phage by food or by oral gavage. Forty-five, day-old chicks were inoculated with Campylobacter coli A11. One week later the birds were randomly assigned to one of three groups, a non-treated group and two treated groups: a group receiving the phage cocktail by oral gavage; and a group receiving the phage cocktail in feed. Birds were inoculated with antacid only, antacid containing 1 × 10⁶pfu phage cocktail or antacid followed by feeding with the phage cocktail laced with 1.5 × 10⁷pfu, respectively. Faecal samples were collected from all birds at intervals and Campylobacter and phages enumerated. Error bars represent the standard error of the mean. At 1 dpa, 2 dpa, 4 dpa and 7 dpa there is a significant difference between control and infected groups at P < 0.05.