Comparison of 2-day-old and 14-day-old chicken colonization models for Campylobacter jejuni


  • Editor: George Mendz

Correspondence: Victoria Korolik, Institute for Glycomics, Griffith University, PMB 50, Gold Coast Mail Centre, QLD 9726, Australia. Tel.: +61 7 55528321; fax: +61 7 55528908; e-mail


In this study, we compared two types of chicken infection models for Campylobacter jejuni in terms of infectious dose required to colonize the chickens and the susceptibility of chickens of different ages to persistent colonization by C. jejuni. In one model, chickens at day 2 posthatching were used, and in the other, 14-day-old chickens were used. The minimum C. jejuni cell number required to colonize 14-day-old chickens was 5 × 104 cells, and that for 2-day-old chickens was 5 × 103. The ability of various C. jejuni strains to colonize the chicken gastrointestinal tract was the same in both models.


Campylobacter species are now well recognized as one of the major causes of enteric disease in humans (Blaser, 1997). Campylobacter spp., in particular Campylobacter jejuni, are now the most common causes of foodborne disease in the developed world, and have surpassed Salmonella and Shigella spp. as causes of food-related illness (Altekruse et al., 1999; Frost, 2001). Campylobacter disease is zoonotic, as the bacteria are widespread in animals and birds, where they are commensal organisms (Blaser, 1997). The processing of poultry and other animal food products often leads to contamination of the end-product, and consumption of nonpasteurized milk, contaminated water or undercooked meats leads to the transfer of Campylobacter spp. to humans (Skirrow, 1982; Blaser, 1997; Altekruse et al., 1999; Shane, 2000; Frost, 2001).

As campylobacters form part of the natural microbial community in chickens, several models for assessment of colonization of chickens by Campylobacter spp. are currently being used. These models are also used to examine the behavior of isogenic mutants in vivo and cloned genes with potential pathogenicity characteristics (Nachamkin et al., 1993; Wassenaar et al., 1993; Purdy et al., 1999; Ziprin et al., 1999; Fry et al., 2000). There are essentially two types of chicken colonization model: one using very young chickens, 0–2-days-old, and one using 14–16-day-old chickens (Cawthraw et al., 1996; Rice et al., 1997; Korolik et al., 1998, 2001; Chen & Stern, 2001; Ringoir & Korolik, 2003). In this study, we compare the two types of model in terms of infectious dose of the C. jejuni cells required to colonize the chickens of different ages, and the susceptibility of chickens of different ages to colonization by various C. jejuni strains.

It was decided to use, at one extreme, chickens at day 2 posthatching (as this allows us to determine that the chickens are Campylobacter-free and healthy on the day of inoculation), and, at the other extreme, 14-day-old chickens.

Materials and methods

Bacterial strains and animals

Campylobacter strains were either clinical or originally isolated from commercial chicken flocks in Australia. Strains 331, 413, 415, 016 and 019 were from the RMIT (Royal Melbourne Institute of Technology University), and strain 007 was from Fairfield Infectious Diseases Hospital culture collections. Chickens used in the experiments were of the Ross breed.

Inoculum preparation

Campylobacter bacteria were cultured on Columbia agar supplemented with 5% defibrinated horse blood. All strains were laboratory cultured. The plates were incubated for 48 h at 42°C in a 90% N2/5% CO2/5% O2 atmosphere. Bacteria were harvested in phosphate-buffered saline and quantified by spectrophotometry and viable counts.

Isolation of Campylobacter spp. from cloacal swabs and cecal contents

The collected material was cultured on Columbia agar supplemented with 5% defibrinated horse blood and Skirrow antibiotic supplement (Oxoid). The plates were incubated for 48–72 h at 42°C in a 90% N2/5% CO2/5% O2 atmosphere. Colonies with Campylobacter-like morphology were Gram stained, and their identity was confirmed by PCR analysis as required (Korolik et al., 2001).

Inoculation of chickens with Campylobacter strains

Newly hatched chickens collected from a local hatchery were placed in isolation in groups of five. Chickens were given unlimited food and water and checked for the presence of endogenous Campylobacter spp. by cloacal swabbing on days 7 and 11 posthatching to confirm the absence of Campylobacter spp. for the 14-day-old chicken infection model, and on day 1 for the 2-day-old model. Eight groups of chickens were infected with the highly colonizing C. jejuni strain 331 (strain 331 was chosen because it represents a highly colonizing phenotype with respect to colonization of chickens Korolik et al., 1998) at the following levels of viable cells: 5 × 107, 5 × 106, 5 × 105, 5 × 104, 5 × 103, 5 × 102 and 5 × 101. One group of five chickens was kept as a control. Each chicken for every group was tested for the presence of Campylobacter cells on days 1, 2, 3, 5, 7, 10 and 14 after inoculation by cloacal swabs. The chickens were fed 30 min before swabbing, to ensure that the cloacal samples contained fresh feces. Any chickens that did not excrete detectable levels of Campylobacter cells on day 7 postinoculation were kept for another week and were sacrificed on day 14 postinoculation; the cecal contents were examined by plate culture. All the chickens that still secreted detectable levels of Campylobacter cells on day 14 were sacrificed on day 21 postinoculation, and cecal contents were examined. The experiment was repeated at least twice independently. The same schedule was followed to test colonization of 2-day-old chickens.


Comparison of 14-day-old and 2-day-old chicken models for assessment of colonization potential of Campylobacter strains

To compare the two colonization models, groups of the 2-day-old and 14-day-old chickens were orally inoculated with serially diluted C. jejuni 331 cells as described in ‘Materials and methods’.

Tables 1 and 2 show the presence or absence of colonization with respect to size of the C. jejuni 331 inoculum for 2-day-old and 14-day-old chickens respectively. The lowest inoculum able to colonize 14-day-old chickens repeatedly was 5 × 104C. jejuni cells, and for 2-day-old chickens it was 5 × 103 cells. Once established, the colonization persisted for the duration of the experiment, with a maximal bacterial load in the ceca of 108 CFU g−1 of cecal material. The maximum cecal bacterial load for the Ross chickens has been previously reported to be 108 CFU g−1 of cecal material (16).

Table 1.   Colonization of 2-day-old chickens by Campylobacter jejuni 331with respect to the inoculum
InoculumDay 1Day 2Day 3Day 5Day 7Day 10Day 14Day 23
  1. +, positive for Campylobacter for all the birds; −, negative for Campylobacter for all the birds.

  2. Example: 4/5=4 out of 5 birds positive for Campylobacter.

5 × 101
5 × 102
5 × 1032/52/53/54/55/5
5 × 1044/55/55/55/55/55/55/55/5
5 × 1055/54/55/55/55/55/55/55/5
5 × 106++++++++
5 × 107++++++++
Table 2.   Colonization of 14-day-old chickens by Campylobacter jejuni 331 with respect to the inoculum
InoculumDay 1Day 2Day 3Day 5Day 7Day 10Day 14Day 23
  1. −, negative for Campylobacter for all the birds.

  2. Example: 4/5=4 out of 5 birds positive for Campylobacter.

5 × 101
5 × 102
5 × 1031/52/55/55/55/55/5
5 × 1041/51/52/54/55/55/55/55/5
5 × 1051/54/55/55/55/55/55/55/5
5 × 1063/54/55/55/55/55/55/55/5
5 × 1073/55/55/55/55/55/55/55/5

Assessment of colonization status in a 2-day-old chicken colonization model of strains previously assessed in 14-day-old chickens

To further assess both models, previously tested strains that could colonize and strains that could not colonize 14-day-old chickens were tested using 2-day-old chickens.

Both colonizing and noncolonizing strains maintained their colonization status in the 2-day-old chicken colonization model (Table 3). Some variation was noted for C. jejuni strain 415, which took 3 days longer to clear the intestinal tract of 2-day-old chickens compared to the 14-day-old chickens, and for C. jejuni strain 413, which was not detected at all in the 2-day-old chickens, but was cleared from the 14-day-old chickens after 7 days.

Table 3.   Colonization status of Campylobacter strains in 14-day-old and 2-day-old chicken colonization models 21 days after inoculation
StrainC. jejuni 405*C. jejuni 413*C. jejuni 415*C. jejuni 331*C. jejuni 007C. jejuni 016*C. jejuni 019*
  • *

    Culture collection, RMIT University, Department of Applied Biology, Melbourne, Australia.

  • Culture collection, Fairfield Infectious Diseases Hospital, Fairfield, Australia.

  • +, positive for Campylobacter for the whole group; −, negative for Campylobacter for the whole group.

2-day-old chickens+++++
14-day-old chickens (Korolik et al. 1998)+++++


Both models demonstrate that in order to establish colonization with a naive laboratory (not isolated directly from chicken feces or ceca) C. jejuni strain, the number of viable Campylobacter cells required is quite high, even in very young chickens. In previous studies by Cawthraw et al. (1996) and Ringoir et al. (2003), all-or-nothing colonization was observed (Cawthraw et al., 1996; Ringoir & Korolik, 2003). No detectable levels of Campylobacter bacteria were found with low inocula (variable between strains), but with an increase as low as 0.5 log viable bacteria, a maximum cecal colonization up to × 108 CFU−1 g−1 of cecal content was observed. It can be concluded from these studies that fecal or cecal samples should give an indication of colonization by Campylobacter bacteria. By taking swabs, which is less invasive, a time course of colonization for an individual C. jejuni strain can be established. This might be important in assessing mutant strains in the future for changes in colonization efficiency and phenotypic pattern. The 14-day-old chickens could be persistently colonized with minimum of 5 × 104 viable cells, and for 2-day-old chickens the number of viable cells required was 5 × 103. This difference might be attributed to the presence of a mature normal intestinal microbial community in the gastrointestinal tract of 14-day-old chickens. We have also demonstrated that the colonization status of the strains tested was not dependent on the type of model (the age of the chickens), with the exception of some noncolonizing strains, which required several extra days to completely clear the chicken gastrointestinal tract. Therefore, when using 2-day-old chickens for assessment of colonization status of Campylobacter strains, it is necessary to either lower the inoculum or allow up to 2 weeks to establish true colonization status. Campylobacter jejuni strain 413 was not able to colonize the 2-day-old chickens, but was able to colonize the 14-day-old chickens for a short time. This can be explained by multiple passages on medium. The strain was screened in the 14-day-old chickens 6 months before it was screened in the 2-day-old chickens, and multiple passages on plates occurred between two screenings. Loss of expression of genes involved in colonization may explain this phenomenon, as the strain was initially isolated from a commercial flock. Both strains 413 and 415 have lost the capacity to colonize chickens in a sustainable way. It has been established that strains cultured on laboratory media for a longer time may lose expression of colonization factors, and that this can be reversible (Cawthraw et al., 1996; Ringoir & Korolik, 2003).

Colonization could not be established in 2-day-old chickens infected with fewer than 5 × 103 viable cells of a highly colonizing C. jejuni strain. It has been previously established (Cawthraw et al., 1996; Ringoir & Korolik, 2003) that if a strain has been directly isolated from an infected chicken, the cell number required to establish persistent colonization drops by several orders of magnitude. This is thought to account for the rapid spread of campylobacters in chicken flocks at 2–3 weeks of age. It is still, however, unclear where the infection of chicken flocks with campylobacters originates.

The data presented here show that 2-day-old chickens inoculated with fewer then 103 bacteria did not have detectable Campylobacter organisms in their cecum after 2 weeks following inoculation. In this model, an inoculum of at least 1000 cells of a naive strain are required to infect a chicken.

This investigation has shown that both the 2-day-old and the 14-day-old chicken models can be efficiently used for colonization studies, provided that appropriate inoculum levels, laboratory strains and passaged stains are used.


This work is supported by the Australian Rural Industries Research and Development Corporation.