Evaluation of a polysaccharide conjugate vaccine to reduce colonization by Campylobacter jejuni in broiler chickens

Vaccine

Capsular polysaccharide of C. jejuni strain 81-176 was purified and conjugated to diphtheria toxoid CRM₁₉₇ to form capsular polysaccharide conjugate (CPSconj) as described previously 17].

Immunogenicity in the absence of maternal antibodies

Specific pathogen free chicks (SPF, N2 strain, Cornell University Center for Animal
Resources and Education, Cornell, New York, USA, 5 chicks per experimental group),
lacking detectable maternal antibodies at the time of vaccination, were raised in
the isolation facilities of the Ontario Veterinary College and vaccinated subcutaneously
at 3 and 4 weeks of age with 10 ?g CPSconj with either 20 ?g Quil A (Brenntag Biosector,
Frederickssund, Denmark) or 10 ?g CpG oligodeoxynucleotide (CpG ODN) 2007 (Eurofins
Operon, Huntsville, AL, USA) as adjuvants, or received 20 ?g CPSconj with 20 ?g Quil
A at 3, 4 and 5 weeks of age. Vaccination of these chicks thus occurred under favorable
conditions (absence of maternal antibodies) beginning at an age when immune function
is considered to be mature 27], 28]. Control (placebo) birds received phosphate buffered saline (PBS) in place of vaccine.

Vaccination and challenge of commercial broiler chicks

One-day-old female commercial broiler chicks (Ross 308 chicks from Stratford Chick
Hatchery, Stratford, ON, Canada) were housed in isolation facilities at the Ontario
Veterinary College. Chicks were allocated (8 chicks per group) to receive 25 ?g CPSconj
with 10 ?g CpG or 100 ?l Addavax™ (squalene based adjuvant, InvivoGen, San Diego,
CA, USA) as adjuvant, or 25 ?g CPSconj without adjuvant, or PBS as a placebo by subcutaneous
injection in a 200 ?l volume at 7 days post-hatch. Birds received a booster dose of
10 ?g of the same antigen preparation as previously, or PBS at 21 days post-hatch.
In contrast to the chicks used in the immunogenicity experiment above, these broiler
chicks were obtained from a commercial source and had detectable serum (maternal)
antibodies at the time of primary vaccination. At 29 days post-hatch birds were challenged
orally with 2 × 10⁷ CFU of C. jejuni strain 81-176 prepared by the method of Davis and DiRita 30] (see below). Birds were necropsied at 9 days post-challenge (38 days post-hatch)
and dilutions of the cecal contents were plated onto Mueller–Hinton agar containing
Preston Campylobacter Selective Supplement (Oxoid, Basingstoke, Hampshire, UK) and
incubated at 42°C for 48 h in a microaerobic environment to quantitate CFU of C. jejuni. CFU per gram of cecal contents were calculated based on plate counts, adjusting
for dilutions. An additional eight birds (serving as negative controls for challenge)
were housed in a separate isolation room, were not vaccinated and were not challenged,
but were necropsied at the same time as challenged birds. Blood was collected at 7,
21, 29, 34 and 38 days post-hatch for serological testing. The experiment was repeated
a second time using chicks from the same source, following the identical protocol
in the same facilities, and the results were pooled for statistical analysis.

Animal ethics

All experiments were approved by the Animal Care Committee of the University of Guelph
(Animal Utilization Protocol number 10R086-1836) and followed the guidelines of the
Canadian Council on Animal Care.

Enzyme-linked immunosorbent assay (ELISA) for serum IgG antibodies

Purified capsular polysaccharide of C. jejuni diluted to 40 ?g/ml in PBS buffer was coated onto Nunc 96 well (MicroWell™ untreated
polystyrene, Thermo Fisher Scientific, Rochester, New York, USA) plates, 100 ?l/well,
at 37°C for 3 h. Plates were washed four times with wash buffer consisting of PBS
with 0.5% fish skin gelatin (Sigma, St. Louis, MO, USA) and 0.05% Tween 20 (Sigma).
Sera were diluted 1/20 in wash buffer and 100 ?l was added to wells in duplicate,
followed by a 2 h incubation at 37°C. After washing four times, bound antibodies were
detected using rabbit anti-chicken IgG (Fc specific) horse radish peroxidase conjugate
(Jackson ImmunoResearch Laboratories, West Grove, PA, USA) diluted 1/350, incubating
at 37°C for 1 h. After washing four times, ABTS (2,2?-azino-di (3-ethyl-benzthiazoline-6-sulfonate))
substrate (Kirkegaard and Perry Laboratories, Gaithersburg, Maryland, USA) was added,
100 ?l/well; 1% sodium dodecylsulfate (Bio-Rad, Hercules, CA, USA) was added as stop
solution after 1 h and the optical densities were evaluated at 405 nm. A dilution
series of a positive control serum consisting of pooled high titre sera from mature
hens was run on each plate. Titres were estimated using the method of Sacks et al.
31]. The limit of detection of the assay was 3.32 log₂ units.

Bacterial culture

Campylobacter jejuni strain 81-176 was cultured according to the method of Davis and DiRita 30] to provide bacteria for experimental oral challenge. Briefly, a 10-?L loop of frozen
C. jejuni culture maintained at ?80°C, was inoculated onto Mueller–Hinton agar (Oxoid) and
incubated for 18 h in a sealed jar using gas packs (CampyGen, Oxoid) to maintain microaerobic
conditions. Subsequently, several colonies of C. jejuni were inoculated into 100 mL fresh Mueller–Hinton broth and incubated at 41°C in microaerobic
conditions for 40 h. The broth culture was centrifuged at 3,500×g for 10 min and the bacteria were diluted with PBS (pH 7.4) to attain an optical density
corresponding to approximately 4.0 × 10⁷ CFU/ml, based on previous analysis of growth curves. The number of viable C. jejuni received by the chickens at the time of challenge was established retrospectively
by plating dilutions of the inocula onto Mueller–Hinton agar plates.

Statistical analysis

Data from the two challenge trials were pooled. CFU/gm of C. jejuni in cecal contents were log₁₀ transformed before analysis using a mixed model (Proc Mixed) including trial as a
random variable, in SAS version 9.2. Vaccine groups were subsequently compared using
Tukey’s test. Antibody titres were expressed on a log₂ scale. Titres were not normally distributed for the majority of time points; a nonparametric
test (Proc Npar1way [Kruskal–Wallis test]) was used to compare antibody titres among
groups. Seroconversion (increase in titre by ?2 log₂ units) rates were compared using Fisher’s exact test. Correlation between serum antibody
titres on the day of necropsy and CFU of C. jejuni in cecal contents was evaluated using Pearson’s correlation coefficient using Proc
Corr in SAS.