Abstract
Salmonella Enteritidis (SE) and, to a lesser extent, Salmonella Typhimurium (ST) are associated with egg-related outbreaks in people. Recently, monophasic strains of ST (mST; lacking one phase of the flagellar antigen) have been described, and they have officially been classified as variants of ST and thus may contribute to human exposure to contaminated eggs. Currently used vaccination programmes are licensed for use against biphasic variants of ST, and their efficacy against mST has not yet been investigated. In this study, the effectiveness of four vaccination programmes currently in use in the UK poultry industry was evaluated against challenge with one SE strain, one ST strain and two mST strains. A Bayesian model was used to estimate the impact of vaccination on the rate of faecal shedding and on egg contamination. For the majority of vaccine/challenge strain combinations, there was little or no effect of vaccination on the proportion of birds shedding Salmonella for either biphasic or monophasic strains. However, vaccination was effective at reducing egg contamination. A significantly lower proportion of eggshells were positive for the vaccinated birds compared with non-vaccinated birds, including the mST strains (vaccination resulted in a 55% and 21% reduction for the two mST strains). Calculated across all strains, the estimated rate of positive egg contents was lower in vaccinated birds (Bayesian median was 0.13% for vaccinated birds versus 0.27% for non-vaccinated birds). For both vaccinated and unvaccinated birds, there was also an apparent difference in the proportion of positive organs between strains, with the SE strain resulting in a lower proportion of positive organs at post-mortem examination compared with the other strains.
Introduction
Salmonella is a major cause of diarrhoea and systemic infections in people, most commonly as a result of contamination of foodstuffs of animal origin. In the latest report produced by the European Food Safety Authority (EFSA), Salmonella remained the most frequently detected causative agent in the food-borne outbreaks reported (26.6% of outbreaks). Eggs and egg products are frequently associated with Salmonella outbreaks. Salmonella Enteritidis (SE) and, to a lesser extent, Salmonella Typhimurium (ST) are associated with egg-related outbreaks. Recently, monophasic strains of ST (mST; lacking one phase of the flagellar antigen) have been described, and they have officially been classified as variants of ST (EFSA, 2010). Two mST variants are emerging worldwide, respectively bearing (1,4,5,12:i:-) and lacking (1,4,12:i:-) the O5 flagellar antigen (Switt et al., 2009). Interestingly, mST strains resitant to amoxicillin, streptomycin, sulphonamides and tetracycline are widely spread in Europe in people and animals, particularly in pigs and cattle, and they have also been isolated from poultry (EFSA, 2010).
To reduce the number of egg-related Salmonella outbreaks in people, the European Commission and the Member States have agreed to implement programmes for control of SE and ST in laying hens. Since May 2011, the isolation of mST from a flock of laying hens triggers the same restrictions as the isolation of SE and ST (EC 517/2011).
Vaccination of laying hens has been shown to confer protection against Salmonella infection and to decrease the level of on-farm contamination (Van Immerseel et al., 2005). In some European countries (Austria, Belgium, the Czech Republic, Germany, and Hungary) vaccination of laying flocks is compulsory, in others it is allowed and recommended (Bulgaria, Belgium, Cyprus, Estonia, France, Greece, Italy, Latvia, Lithuania, the Netherlands, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, and the UK), and in others it is banned (Denmark, Finland, Sweden and Ireland) (Galis et al., 2013). In the UK, most commercial-scale egg producers subscribe to the British Egg Industry Council Quality Assurance Scheme that provides a code of practice (Lion Code) on farm hygiene and welfare standards. Vaccination against Salmonella began in laying flocks in 1998 for farms that subscribe to the British Egg Industry Council Lion Code Scheme (Ward et al., 2000; Cogan & Humphrey, 2003; O'Brien, 2013).
Both live and killed vaccines are available to vaccinate laying flocks (EFSA, 2004). Live vaccines generally confer better protection than the killed vaccines, as they are able to induce both cell-mediated and humoral immune response (Gantois et al., 2006; Galis et al., 2013). As SE and ST are considered the most important serovars for public health in Europe, existing commercially available live and inactivated Salmonella vaccines for poultry are intended for use against one or both of these serovars. In the UK, three live vaccines and two killed vaccines are currently available (Clifton-Hadley et al., 2002; Gantois et al., 2006; Springer et al., 2011). These vaccines are used alone or in combination. To maximize protection, vaccination programmes that combine live and killed vaccines are sometimes used. Within these programmes, oral vaccines are administered in two or three doses during the rearing period of the pullets and are complemented by one or two injections of killed vaccine (normally close to point of lay) (EFSA, 2004). Currently used vaccination programmes are licensed for use against biphasic variants of ST, and their efficacy against mST has not yet been investigated. ST vaccines probably have a similar protective effect for mST as for ST, but there are no data available concerning the efficacy of current vaccination programmes. In the UK, the limited number of poultry flocks that tested positive for mST within the National Control Programme for Salmonella were either unvaccinated against Salmonella (broiler and turkey flocks) or had received vaccination against SE but not ST (laying flocks) (Davies & Mueller-Doblies, 2013).
In this study, the effectiveness of four vaccination programmes currently in use in the UK poultry industry was evaluated against challenge with one SE PT14b strain, one ST DT8 strain, one 4,5,12:i:- strain and one 4,12:i:- strain.
The objective of this part of the study was to apply a modelling approach to estimating the effect of vaccination on the proportion of birds positive for cloacal swabs, and the rate of eggshell and egg content contamination. In particular, the modelling approach is able to estimate the infection prevalence from the cloacal swab data, so that the rate of eggshell and egg content contamination can be estimated relative per infected bird, which is not possible using a classical statistics approach. Where relevant, data were amalgamated between studies in order to ensure there was sufficient statistical power to detect differences in rates of contamination between strains and vaccines.
Materials and Methods
Data
Vaccination protocols
There were four different vaccination programmes tested in three separate studies:
Vaccination programme 1 (Study 1): two doses of an inactivated SE PT4/ST DT104 vaccine, administered at 12 and 16 weeks of age by injection in the breast muscle.
Vaccination programme 2 (Study 2): two doses of a double-attenuated (adenine–histidine auxotrophic) SE mutant vaccine (administered by oral gavage at 1 and 6 weeks of age) plus one dose of an inactivated SE PT 4/ST PT 104 vaccine (at 18 weeks of age, via injection in the breast muscle).
Vaccination programme 3 (Study 2): three doses of a double-attenuated (adenine–histidine auxotrophic) SE mutant vaccine, administered by oral gavage at 1, 6 and 13 weeks of age.
Vaccination programme 4 (Study 3): three doses of one live attenuated metabolic drift SE vaccine strain combined with one live attenuated metabolic drift ST vaccine strain (administered by oral gavage at 1,7 and 13 weeks of age).
Experimental design
Two hundred and forty (300 hundred in Study 2) commercial Hy-line layer chickens were sourced at 1 day old to be used in each of the three trials. One hundred and twenty birds remained unvaccinated against Salmonella for the duration of each study. The remaining 120 birds (180 birds in Study 2) were vaccinated, according to the vaccination programme for that study. At point of lay (19 to 20 weeks of age), the birds were split into eight groups of 30, four vaccinated (six vaccinated in Study 2) and four unvaccinated. In each study, the same four challenge isolates were used to infect the birds (details of the challenge isolates are given below), each applied to one vaccinated group and one unvaccinated group. Study 2 differed from the others in that a second vaccine was also tested against two of the challenge isolates, leading to two extra groups of 30 vaccinated birds.
All birds were challenged by oral gavage with one of four strains, and cloacal swabs, eggs and tissue samples were collected at regular intervals during the experiment, as described below. The project was approved by the Animal Health and Veterinary Laboratories Agency ethics committee and all procedures were conducted in accordance with the UK Animals Scientific Procedures) Act 1986 under the jurisdiction of HO project licence PPL 70_7319.
All birds were observed twice daily for the duration of the experiment.
Salmonella strains and challenge dose
Four Salmonella strains were used to challenge birds in this study. Strain 1 (denoted by “ST”) is a fully typeable ST DT8, originally isolated from a duck table egg laying flock in the UK (Noble et al., 2012). Strains 2 and 3 (denoted “mST1” and “mST2” respectively) are tetra-resistant (amoxicillin, streptomycin, sulphonamides and tetracycline) mST 4,12:i:- and 4,5,12:i:- strains of UK origin, respectively. Strain 4 (denoted by “SE”) is an egg-invasive SE PT14b strain that spread to the UK through imported contaminated eggs, included in the study for comparison (Janmohamed et al., 2011). Strains ST and SE were additionally tested against the second vaccine used in Study 2. To prepare the challenge dose, a single colony of each of the four strains was incubated overnight (16 to 18 h) aerobically at 37°C in 20 ml Luria–Bertani broth, shaking at ∼170 rpm.
Before the challenge, to enhance the susceptibility of the hens to Salmonella (Suzuki, 1994), the feed was withdrawn for 8 h (overnight) and 2 ml of 10% sodium bicarbonate was administered to each bird via oral gavage before the Salmonella challenge dose (∼5 × 108 colony forming units of strains ST, mST1, mST2 and SE) was administered.
Cloacal swabs and tissue samples
Cloacal swabbing was performed for each bird at various time points during the study although the days post challenge (d.p.c.) on which samples were taken varied between vaccines. For Study 1, swabs were taken at 2, 6, 10, 13, 17 and 21 d.p.c. For Study 2, swabs were taken at 1, 2, 6, 9, 13, 16 and 20 d.p.c. For Study 3, swabs were taken at 2, 9, 16 and 23 d.p.c. For all vaccines, the liver, spleen, ovary and caeca sampled at post-mortem examination were tested by direct culture and enrichment at 3 d.p.c. (five birds), 7 d.p.c. (five birds), 24 d.p.c. (10 birds) and 27 d.p.c. (10 birds). The experiment was concluded at 27 d.p.c. in all three trials.
To enumerate Salmonella from tissues, 1 g sub-samples were homogenized, and decimal dilutions of homogenates were made in phosphate-buffered saline (pH 7.4), plated onto Rambach agar (Clifton-Hadley et al., 2002) and incubated at 37 ± 1°C for 24 ± 3 h. Cloacal swabs were plated directly on Rambach agar for enumeration.
Any negative cloacal swab or tissue was enriched in 9 ml buffered peptone water (BPW; Merck, Nottingham, UK) and incubated at 37 ± 1°C for 16 to 20 h and subsequently 0.1 ml was plated onto modified semi-solid Rappaport–Vassiliadis (MSRV; Mast Ltd, Bootle, UK) enrichment agar at 41.5 ± 1°C for 24 ± 3 h. Spreading growth on MSRV was sub-cultured onto Rambach agar at 37 ± 1°C for 24 ± 3 h. Cloacal swabs and tissue samples were plated directly onto Rambach agar, and incubated as described above. In order to enumerate Salmonella from tissues, 1 g sub-samples were homogenized, and decimal dilutions of homogenates were made in phosphate-buffered saline (pH 7.4) and plated onto Rambach agar (Clifton-Hadley et al., 2002). Cloacal swabs and tissues were also enriched in BPW and any negative tissues from the necropsy were enriched in MSRV media and re-plated onto Rambach agar. Slide agglutination tests on isolates from samples were also carried out to confirm positive results. A selection of the positive samples was subjected to serotyping and phage typing for confirmation (Anderson et al., 1977; Wattiau et al., 2011; OIE, 2012)
Eggs
The eggs were collected daily from 1 to 27 d.p.c. from each room and tested individually, shells and contents separately.
The eggshell was swabbed and the egg was then immersed into 70% ethanol twice (allowing the surface to dry between washes), before being cracked open to test the contents. Eggshell swabs were incubated in 9 ml BPW (Merck) at 37 ± 1°C for 16 to 20 h and subsequently 0.1 ml was plated onto MSRV (Mast Ltd) enrichment agar at 41.5 ± 1°C for 24 ± 3 h. Spreading growth on MSRV was sub-cultured onto Rambach agar at 37 ± 1°C for 24 ± 3 h and cultured as described above. The egg contents were introduced directly into 225 ml BPW and cultured as described above.
Statistical modelling
The count of positive cloacal swab samples for direct culture and after enrichment at each of the sampling times was fitted using a Bayesian model of diagnostic test evaluation in the absence of a gold standard for two conditionally independent tests (Branscum et al., 2005); a model assuming conditional dependence between the result of the direct culture and that after enrichment showed that there was no significant correlation between the two results (results not shown). The output of this Bayesian model was an estimate of the infection prevalence for each time point at which cloacal swabs were taken and the sensitivity of cloacal swab sampling for direct culture and after enrichment.
While cloacal sample data were available for six time points (Study 1), eight time points (Study 2) and four time points (Study 3), eggs were sampled daily. The prevalence of infection at each time point in each vaccination/strain group was therefore inferred on those days on which no cloacal sampling was performed by interpolation; that is, fitting a line through the observed data to provide an estimate for the infection prevalence on those days for which cloacal sampling was not performed. This was done by assuming that the prevalence of infection over the study followed a logistic regression curve; that is, the prevalence of infection at time t was given by: 
(1)
The proportion of positive eggshells and contents was assumed to be linearly dependent on the number of infected birds on the day of sampling, and for each the estimated rate of contamination for vaccinated and non-vaccinated birds was estimated.
The rate of egg contamination and the proportion of birds infected were compared between the following groups:
Vaccinated and unvaccinated birds for each strain/vaccine combination separately.
Between the vaccinated and non-vaccinated birds for each strain (i.e. combining data for the different vaccines for each strain)
Between vaccinated and non-vaccinated birds for each vaccine (i.e. combining the data for different strains for each vaccine)
Results
The challenge dose was selected according to published data and our experience to minimize the onset of any Salmonella-related signs. None of the birds showed clinical signs related to the Salmonella challenge dose during this study.
Cloacal swabs
There was a very high prevalence of positive cloacal swabs at 1 d.p.c., with all or nearly all birds positive regardless of vaccination—except for strain ST, which had a lower proportion of positive cloacal swabs than the other strains, particularly for Studies 1 and 2 (Figures 1 and 2). As the d.p.c. increased, then the number of positive cloacal swabs decreased for all strains, although there was variability between strains in the rate that the cloacal swab prevalence declined (Figures 1 to 3).
Published online:
24 March 2014![]({"type":"image","src":"/na101/home/literatum/publisher/tandf/journals/content/cavp20/2014/cavp20.v043.i02/03079457.2014.896990/20140408/images/medium/cavp_a_896990_f0001_b.jpg"})
Published online:
24 March 2014![]({"type":"image","src":"/na101/home/literatum/publisher/tandf/journals/content/cavp20/2014/cavp20.v043.i02/03079457.2014.896990/20140408/images/medium/cavp_a_896990_f0002_b.jpg"})
Published online:
24 March 2014![]({"type":"image","src":"/na101/home/literatum/publisher/tandf/journals/content/cavp20/2014/cavp20.v043.i02/03079457.2014.896990/20140408/images/medium/cavp_a_896990_f0003_b.jpg"})
For the majority of vaccine/strain combinations, there was little or no effect of the vaccine on the proportion of birds testing positive (Tables 1 and 2). The only statistically significant reductions in the proportion of birds being positive for cloacal swabs was for vaccines 2, 3 and 4 for strain ST, and for vaccines 2 and 4 for strain SE.
Table 1. Number of positive cloacal, organ and egg samples for chickens experimentally infected with Salmonella, according to the vaccine used and the Salmonella strain.
Table 2. Estimated proportion of positive cloacal swabs for each group of birds experimentally infected with Salmonella and the estimated rate of contamination of eggshells per infected bird.
The Bayesian model of diagnostic test evaluation suggested a small rate of under-detection of excreting birds by the cloacal swabs after enrichment, with an estimated sensitivity of 93.5% (95% credible interval [CrI]: 91.6 to 95.3%) and a relatively low sensitivity for direct culture (25.3%, 95% CrI: 23.4 to 27.6%).
Eggs
There was a very low proportion of positive egg contents. In Study 1 there were 0/1345 and 1/1204 positive egg contents in vaccinated and non-vaccinated birds respectively, in Study 2 a total of 3/1439 and 3/823 positive egg contents for vaccinated and non-vaccinated birds respectively, and for Study 3 1/1401 and 1/1589 positive contents for vaccinated and unvaccinated birds respectively. While the estimated rate of positive contents was lower in vaccinated birds (Bayesian median was 0.13% for vaccinated versus 0.27% for non-vaccinated birds), the very low prevalence and small number of positives meant that this difference was not statistically significant so a common rate of contaminated egg contents was fitted to both the vaccinated and non-vaccinated birds. The Bayesian estimate of egg content contamination per infected bird was equal to 0.23% (95% CrI: 0.12 to 0.41%).
A lower proportion of eggshells were positive for the vaccinated birds compared with non-vaccinated birds in all three studies: Study 1, relative rate for vaccinated compared with non-vaccinated birds equal to75.2% (95% CrI 66.9 to 83.9%); Study 2, relative rate equal to 64.0% (95% CrI 50.5 to 81.4%); and Study 3, relative rate equal to 40.8% (95% CrI 32.9 to 51.0%). The relative reduction in eggshell contamination per infected bird varied between strains, with a significant reduction of vaccination on strain mST2 (relative rate of egg contamination of vaccinated to non-vaccinated birds of 44.9% [95% CrI: 37.2 to 53.9%]), strain SE (72.3% [95% CrI: 62.1 to 83.9%]) and strain mST1 (78.9% [95% CrI: 67.4 to 91.6%]) but not for strain ST (58.4% [95% CrI: 33.3 to 100.3%]).
The rate of shell contamination per infected bird differed between the studies for Study 1 unvaccinated birds had a rate of eggshell contamination per infected bird of 59.1% (95% CrI: 55.4 to 63.2%), in Study 2 it was 40.6% (95% CrI: 33.9 to 48.0%), and in Study 3 it was 41.3% (95% CrI: 36.9 to 46.2%); that is, the lack of overlap of the CrI suggests a significantly higher rate of shell contamination in Study 1 compared with Studies 2 and 3. The rate of eggshell contamination for vaccinated birds also differed between Study 1 and Study 2, and was equal to 45.8% (95% CrI: 41.3 to 49.9%) in Study 1, 27.1% (95% CrI: 22.7 to 32.2%) in Study 2 and 30.9% (95% CrI: 26.2 to 36.4%) in Study 3.
Organs/tissues
There were differences in the proportion of each organ testing positive at post-mortem, with ovaries having the lowest proportion positive, followed by the liver, and with the caeca having the highest proportion positive. This was consistent between the different studies (Table 1). There was also an apparent difference in the proportion of positive organs between strains, with strain ST resulting in a lower proportion of positive organs at post-mortem examination compared with the other strains, most probably as a result of the lower proportion of birds infected with this strain compared with the other strains.
For the majority of strain/vaccine combinations, the estimated proportion of positive organs at post-mortem was lower for the vaccinated birds compared with the unvaccinated birds (Table 3), but the difference was small and there was no statistically significant impact of vaccination on the proportion of positive organs—except for Study 3, where considered over all four strains the vaccine achieved a significant reduction in the proportion of organs positive (estimated reduction = 0.45 [95% CrI: 0.25 to 0.75]). At strain level, the greatest impact of vaccination was observed for strain SE, with a reduction of approximately 20% in the proportion of organs positive, across all four organs tested (Table 4). There was a significant effect of vaccination on the proportion of ovaries positive for strain mST2 (Table 4), and the average efficacy over all strains combined (Table 4).
Table 3. Estimated proportion of positive organs for chickens challenged with Salmonella according to the strain and whether the birds were vaccinated.
Table 4. Estimated Salmonella vaccine efficacy in terms of percentage of positive organs for vaccinated birds relative to unvaccinated birds, averaged over three different vaccines (strains mST1 and mST2) and four vaccines (strains ST and SE).
Discussion
By estimating the prevalence of infection at each time using the logistic regression model and the cloacal swabbing data, the present study has been able to relate egg contamination rates to the number of infected birds at each time point.
This relation enables the estimation of vaccine efficacy in terms of its impact on the proportion of eggshells and egg contents from infected birds that are contaminated, which is a parameter of direct biological interest. The Bayesian approach also integrates both the direct culture and enrichment data, so maximizing the power of the analysis. The drawback in relating contamination rates to the proportion of birds infected at each time point is that it relies on correctly inferring the proportion of birds infected at time point where cloacal swabbing was not performed; that is, it relies on the fitting of the logistic regression model to the time points at which cloacal swabbing was performed. However, the model fit appears satisfactory, and a comparison of the estimation of shell contamination rates between the model fitted through all time points and one fitted only at the time points where cloacal swabbing was performed show very little difference between the Bayesian median estimates (data not shown).
Both egg contents and eggshell contamination rates from infected vaccinated birds have been estimated in a previous study (Arnold et al., 2013), where eggs were tested from flocks from England and Wales, detected through the National Control Programme for Salmonella in the UK. In that field study, the egg contamination rate was similar to the value found in the present study (0.24% of eggs contaminated from infected birds in the previous study compared with 0.23% in the present study, when averaged over vaccinated and unvaccinated birds). However, the rate of shell contamination was much higher in the present study than that estimated in Arnold et al. (2013). For example, in Arnold et al. (2013) the highest rate of shell contamination found in a flock (estimated to be 18% of eggshells contaminated from infected birds) was much lower than the maximum rate observed in the present study (67% of eggshells contaminated from infected birds) The proportion of contaminated eggshells was higher than that observed in the field (Davies & Breslin, 2004), and the level of eggshell contamination detected in this study is probably partly attributable to the high challenge dose and to the fact that eggs were laid in contaminated litter, as the birds refused to use the nest boxes provided (Olsen et al., 2013). Floor-raised birds and birds that can have contact with all other birds in the flock have been demonstrated to have a higher transmission rate than those in cages. This is attributed to the much greater level of contact with faeces and contaminated litter (De Vylder et al., 2011).
The vaccines were effective at reducing shell contamination, with significant reductions for strains mST1, mST2, and SE. The lack of statistical significance for strain ST may be a reflection of the low rate of infection for the experimentally infected birds; since there were fewer birds infected for strain ST, there was less power in the data to detect differences between the vaccinated and non-vaccinated birds in terms of eggshell contamination. In particular, the Bayesian median estimate of the effect of vaccination for strain ST was lower than that of strains mST1 and SE, and was very close to being statistically significant (upper CrI was 100.3%). Strain ST was chosen because of the occurrence of DT8 infection in some laying hen flocks where there were also ducks on site and the known egg invasiveness of this strain on the farm of origin, but it is possible that the particular strain may have been less well adapted to adult chickens than to ducks. There is a considerable degree of variability in the level of invasiveness between ST strains (Rabsch et al., 2002).
Vaccination was found to have greatest effect on strain mST2 (reducing shell contamination for vaccinated birds to 44% of that of the unvaccinated birds) and least effect on strain mST1.
The loss of O5 somatic antigen expression has been suggested to represent an evolutionary mechanism, through which Salmonella improves its chances of evading the host immune response (Hauser et al., 2011), and this might justify why strain mST1 (O5−) was more successful at evading the vaccinal protection than strain mST2 (O5+).
While no significant difference was found in the rate of content contamination between vaccinated and non-vaccinated birds, there was a difference in the Bayesian median estimate (0.27% for non-vaccinated birds and 0.13% for vaccinated birds). The very low rate of content contamination from Salmonella-infected birds makes it problematic to collect sufficient samples to detect a significant difference, should a difference exist. The results from the present study do provide a useful starting point for estimating the sample size required to detect significant differences, since they give estimates of the probable difference required to be detected.
The majority of the cloacal swabs were positive immediately after challenge, but, as the number of d.p.c. increased, the number of birds positive in cloacal swabs decreased for all strains. This is not unexpected, and has been showed in previous studies (Gast & Beard, 1990; Gast, 1994; Bichler et al., 1996).
There was little or no effect of the vaccine on the proportion of individual birds positive in cloacal swabs. The estimated proportion of positive organs at post-mortem was lower for the vaccinated birds compared with the unvaccinated birds, but this difference was small and often not statistically significant. Vaccination is not always effective in significantly reducing faecal shedding or tissue invasion, especially in the presence of high challenge doses (EFSA, 2004). The greatest impact of vaccination on the proportion of positive organs was for strain SE—a consistent 20% reduction across organs (Table 4). This suggests that vaccination for S. Enteritidis may appear to be more successful than for S. Typhimurium because of the tendency of the latter to clear spontaneously as a result of a greater natural inflammatory and immune response after infection (Withanage et al., 2005; Carrique-Mas et al., 2009). This could reduce the difference between vaccinated and control groups.
Differences were found in the rate of egg contamination and the proportion of birds that were positive via cloacal swabs between the three studies, where Study 1 had higher contamination rates than the other studies. This could be due to the fact that live vaccines are more effective at inducing a cell-mediated immune response, which is believed to be more effective against Salmonella (Gantois et al., 2006). The colonization of organs with salmonellas is reduced primarily by a cell-mediated immune mechanism (Beal et al., 2006; Lehmann et al., 2006; Carvajal et al., 2008). Live vaccines may evoke both cellular and humoral immune responses, which may generate a more effective immune response against challenge (EFSA, 2004).
In summary, this study confirms the egg content contamination rate for artificially infected birds (0.24%) to be similar to that observed within field studies (0.23%). The proportion of contaminated eggshells was higher in this study compared with that observed in field situations, and this was probably due to the high challenge dose administered to the birds and to the design of the bird accommodation. Vaccination was effective in reducing the number of contaminated eggs produced by chickens challenged with mST strains. The 4,12:i:- strain appeared to be more effective at evading vaccinal protection than the 4,5,12:i:- strain, which could be due to the fact that the loss of the O5 somatic antigen represents an evolutionary advantage for this strain.
The effect of vaccination was less evident in the reduction of number of positive cloacal swabs (and therefore number of individual birds shedding Salmonella), and a significant reduction of the number of positive tissues was only observed in the birds challenged with SE.
In conclusion, this study confirms that vaccination is a valuable control measure for Salmonella in laying flocks, particularly in relation to egg contamination. This conclusion also applies to birds infected with mST strains.
| Organs positive (out of 30 tested) | Eggs positive/eggs tested | |||||||
|---|---|---|---|---|---|---|---|---|
| Strain/study | Vaccine | Cloaca positive/tested | Liver | Spleen | Ova | Caeca | Shells | Contents |
| ST/1 | None | 39/135 | 7 | 11 | 5 | 8 | 11/246 | 0/246 |
| ST/1 | Vaccine 1 | 35/135 | 9 | 11 | 6 | 8 | 8/309 | 0/309 |
| ST/2 | None | 36/155 | 7 | 9 | 5 | 8 | 0/159 | 0/159 |
| ST/2 | Vaccine 2 | 13/155 | 7 | 6 | 1 | 3 | 0/246 | 0/246 |
| ST/2 | Vaccine 3 | 9/155 | 4 | 5 | 2 | 1 | 4/217 | 0/217 |
| ST/3 | None | 48/90 | 13 | 16 | 9 | 23 | 74/332 | 0/332 |
| ST/3 | Vaccine 4 | 19/90 | 8 | 7 | 1 | 8 | 11/410 | 1/410 |
| mST1/1 | None | 126/135 | 11 | 19 | 6 | 29 | 187/352 | 1/352 |
| mST1/1 | Vaccine 1 | 126/135 | 11 | 18 | 9 | 28 | 142/352 | 0/352 |
| mST1/2 | None | 114/155 | 10 | 11 | 6 | 11 | 38/213 | 0/213 |
| mST1/2 | Vaccine 2 | 113/155 | 12 | 10 | 5 | 20 | 48/278 | 0/278 |
| mST1/3 | None | 65/90 | 8 | 7 | 1 | 8 | 104/331 | 0/331 |
| mST1/3 | Vaccine 4 | 56/90 | 9 | 6 | 2 | 25 | 56/375 | 0/375 |
| mST2/1 | None | 115/135 | 9 | 14 | 10 | 28 | 174/301 | 0/301 |
| mST2/1 | Vaccine 1 | 106/135 | 11 | 14 | 8 | 29 | 88/369 | 0/369 |
| mST2/2 | None | 112/155 | 10 | 11 | 8 | 16 | 21/256 | 3/256 |
| mST2/2 | Vaccine 2 | 118/155 | 11 | 12 | 4 | 22 | 15/250 | 0/249 |
| mST2/3 | None | 74/90 | 9 | 6 | 2 | 25 | 108/347 | 0/347 |
| mST2/3 | Vaccine 4 | 64/90 | 8 | 10 | 4 | 24 | 45/392 | 0/392 |
| SE/1 | None | 120/135 | 13 | 17 | 11 | 30 | 172/305 | 0/305 |
| SE/1 | Vaccine 1 | 112/135 | 10 | 18 | 10 | 30 | 150/315 | 0/315 |
| SE/2 | None | 119/155 | 13 | 16 | 9 | 23 | 58/195 | 0/195 |
| SE/2 | Vaccine 2 | 131/155 | 11 | 15 | 10 | 23 | 52/207 | 2/207 |
| SE/2 | Vaccine 3 | 92/155 | 14 | 15 | 8 | 19 | 22/242 | 1/242 |
| SE/3 | None | 57/90 | 8 | 10 | 4 | 24 | 118/391 | 1/391 |
| SE/3 | Vaccine 4 | 32/90 | 14 | 18 | 7 | 16 | 15/412 | 0/412 |
| Bayesian estimate (95% credible interval) | |||
|---|---|---|---|
| Strain/study | Vaccine | Percentage of cloacal swabs positive | Rate of eggshell contamination per infected bird (%) |
| ST/1 | None | 29.6 (22.3, 37.9) | 25.9 (13.3, 44.6) |
| ST/1 | Vaccine 1 | 26.3 (19.5, 34.2) | 17.8 (8.2, 34.6) |
| ST/2 | None | 19.7 (14.3, 25.9) | 6.9 (0.2, 36.1) |
| ST/2 | Vaccine 2 | 7.3 (4.2, 11.7) | 11.5 (0.4, 65.6) |
| ST/2 | Vaccine 3 | 5.3 (2.6, 9) | 62.0 (21.4, 97.5) |
| ST/3 | None | 53.3 (43.4, 63.1) | 38.7 (29.8, 51) |
| ST/3 | Vaccine 4 | 22.1 (14.3, 31.1) | 17.3 (8.5, 34) |
| mST1/1 | None | 93.6 (88.8, 97.1) | 54.6 (48.5, 62) |
| mST1/1 | Vaccine 1 | 93.5 (88.6, 97) | 48.1 (41.6, 54.9) |
| mST1/2 | None | 61.6 (54.2, 68.4) | 45.2 (33.3, 58.6) |
| mST1/2 | Vaccine 2 | 61.0 (54.1, 67.9) | 33.3 (24.6, 45.3) |
| mST1/3 | None | 71.9 (62.3, 80.6) | 45.7 (38.3, 53.7) |
| mST1/3 | Vaccine 4 | 62.2 (52.1, 71.7) | 21.5 (13.4, 29.4) |
| mST2/1 | None | 85.6 (79, 90.7) | 66.1 (59.1, 73.8) |
| mST2/1 | Vaccine 1 | 78.9 (71.4, 85.2) | 31.3 (25.2, 38.1) |
| mST2/2 | None | 60.5 (53.4, 67.2) | 27.3 (18.3, 38.8) |
| mST2/2 | Vaccine 2 | 63.8 (56.7, 70.5) | 11.9 (6.9, 19.3) |
| mST2/3 | None | 81.8 (73.1, 89.1) | 38.6 (29.4, 46.7) |
| mST2/3 | Vaccine 4 | 70.8 (61.2, 79.2) | 12.4 (8.9, 21.8) |
| SE/1 | None | 89.2 (83.3, 93.7) | 67.0 (59.5, 74.7) |
| SE/1 | Vaccine 1 | 83.3 (76.3, 88.8) | 56.6 (49.4, 64.9) |
| SE/2 | None | 64.2 (57.3, 71.1) | 51.7 (40.3, 65.5) |
| SE/2 | Vaccine 2 | 70.6 (63.9, 77) | 37.1 (28.1, 48.4) |
| SE/2 | Vaccine 3 | 49.6 (42.7, 56.5) | 27.6 (17.7, 40.4) |
| SE/3 | None | 63.2 (52.9, 72.9) | 45.8 (38.3, 54.1) |
| SE/3 | Vaccine 4 | 35.9 (26.2, 46) | 10.1 (3.2, 18.5) |
| Strain | |||||
|---|---|---|---|---|---|
| Organ | Vaccinated | ST | mST1 | mST2 | SE |
| Liver | Yes | 0.24 (0.17, 0.31) | 0.36 (0.26, 0.46) | 0.34 (0.25, 0.44) | 0.41 (0.32, 0.5) |
| Liver | No | 0.27 (0.19, 0.37) | 0.37 (0.28, 0.47) | 0.36 (0.26, 0.46) | 0.41 (0.32, 0.52) |
| Spleen | Yes | 0.24 (0.17, 0.33) | 0.38 (0.29, 0.48) | 0.4 (0.3, 0.5) | 0.55 (0.46, 0.64) |
| Spleen | No | 0.36 (0.27, 0.46) | 0.46 (0.36, 0.56) | 0.5 (0.4, 0.6) | 0.5 (0.4, 0.6) |
| Ovary | Yes | 0.09 (0.05, 0.15) | 0.18 (0.11, 0.27) | 0.18 (0.11, 0.27) | 0.29 (0.22, 0.38) |
| Ovary | No | 0.17 (0.11, 0.26) | 0.22 (0.14, 0.31) | 0.31 (0.22, 0.41) | 0.33 (0.25, 0.43) |
| Caeca | Yes | 0.17 (0.11, 0.24) | 0.81 (0.72, 0.88) | 0.83 (0.74, 0.9) | 0.73 (0.65, 0.8) |
| Caeca | No | 0.27 (0.18, 0.36) | 0.63 (0.53, 0.73) | 0.72 (0.62, 0.81) | 0.8 (0.71, 0.87) |
| Strain | |||||
|---|---|---|---|---|---|
| Organ | ST | mST1 | mST2 | SE | All strains |
| Liver | 98.8 (69.8, 131.2) | 97.7 (72.5, 130.4) | 84.2 (59.3, 115.9) | 79.9 (23.2, 137) | 87.6 (73.1, 105.3) |
| Spleen | 107.6 (97.7, 126.2) | 108.1 (88.4, 132.4) | 114.2 (94.6, 135) | 80 (23, 136.7) | 104.2 (93.1, 117) |
| Ovary | 96.6 (65.3, 133.5) | 70.3 (45.8, 107.1) | 44.6 (25.1, 74.6) | 81.1 (22.8, 136.4) | 69.4 (53.5, 88.3) |
| Caeca | 98.4 (78.2, 123.2) | 88.9 (67.2, 118.4) | 75.6 (55.5, 102.1) | 81.7 (23, 136.9) | 91.1 (77.2, 107.8) |
Acknowledgements
This work was supported by the Department for Environment, Food and Rural Affairs (UK Government) under grant OZ0341.
References
- Anderson, E.S., Ward, L.R., Saxe, M.J. & de Sa, J.D. (1977). Bacteriophage-typing designations of Salmonella typhimurium. Journal of Hygiene (London), 78, 297–300. 10.1017/S0022172400056187 [Crossref], [PubMed], [Google Scholar]
- Arnold, M.E., Martelli, F., McLaren, I. & Davies, R.H. (2013). Estimation of the rate of egg contamination from Salmonella-infected chickens. Zoonoses and Public Health, 61, 18–27. doi:10.1111/zph.12038 [Crossref], [PubMed], [Google Scholar]
- Beal, R.K., Powers, C., Davison, T.F., Barrow, P.A. & Smith, A.L. (2006). Clearance of enteric Salmonella enterica serovar Typhimurium in chickens is independent of B-cell function. Infection and Immunity, 74, 1442–1444. 10.1128/IAI.74.2.1442-1444.2006 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Bichler, L.A., Nagaraja, K.V. & Halvorson, D.A. (1996). Salmonella enteritidis in eggs, cloacal swab specimens, and internal organs of experimentally infected White Leghorn chickens. The American Journal of Veterinary Research, 57, 489–495. [PubMed], [Web of Science ®], [Google Scholar]
- Branscum, A.J., Gardner, I.A. & Johnson, W.O. (2005). Estimation of diagnostic-test sensitivity and specificity through Bayesian modeling. Preventive Veterinary Medicine, 68, 145–163. 10.1016/j.prevetmed.2004.12.005 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Carrique-Mas, J.J., Breslin, M., Snow, L., McLaren, I., Sayers, A.R. & Davies, R.H. (2009). Persistence and clearance of different Salmonella serovars in buildings housing laying hens. Epidemiology and Infection, 137, 837–846. 10.1017/S0950268808001568 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Carvajal, B.G., Methner, U., Pieper, J. & Berndt, A. (2008). Effects of Salmonella enterica serovar Enteritidis on cellular recruitment and cytokine gene expression in caecum of vaccinated chickens. Vaccine, 26, 5423–5433. 10.1016/j.vaccine.2008.07.088 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Clifton-Hadley, F.A., Breslin, M., Venables, L.M., Sprigings, K.A., Cooles, S.W., Houghton, S. & Woodward, M.J. (2002). A laboratory study of an inactivated bivalent iron restricted Salmonella enterica serovars Enteritidis and Typhimurium dual vaccine against Typhimurium challenge in chickens. Veterinary Microbiology, 89, 167–179. 10.1016/S0378-1135(02)00169-4 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Cogan, T.A. & Humphrey, T.J. (2003). The rise and fall of Salmonella Enteritidis in the UK. Journal of Applied Microbiology, 94 Suppl, 114S–119S. 10.1046/j.1365-2672.94.s1.13.x [Crossref], [Google Scholar]
- Davies, R. & Breslin, M. (2004). Observations on Salmonella contamination of eggs from infected commercial laying flocks where vaccination for Salmonella enterica serovar Enteritidis had been used. Avian Pathology, 33, 133–144. 10.1080/03079450310001652040 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
- Davies, R. & Mueller-Doblies, D. (2013). Observations on the occurence and persistence of monophasic Salmonella Typhimurium (S. 4, 5, 12:i:-) in poultry farms. In I3th International Symposium Salmonella and Salmonellosis (pp. 281–282). St Malo, France. [Google Scholar]
- De Vylder, J., Dewulf, J., Van Hoorebeke, S., Pasmans, F., Haesebrouck, F., Ducatelle, R. & Van Immerseel, F. (2011). Horizontal transmission of Salmonella enteritidis in groups of experimentally infected laying hens housed in different housing systems. Poultry Science, 90, 1391–1396. 10.3382/ps.2010-00944 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- EFSA. (2004). The use of vaccines for the control of Salmonella in poultry. EFSA Journal, 114, 1–74. [Google Scholar]
- EFSA. (2010). Scientific opinion on monitoring and assessment of the public health risk of “Salmonella Typhimurium-like” strains. EFSA Journal, 8, 236–282. [Google Scholar]
- Galis, A., Marcq, C., Marlier, D., Portetelle, D., Van, I., Beckers, Y. & Thewis, A. (2013). Control of Salmonella contamination of shell eggs—preharvest and postharvest methods: a review. Comprehensive Reviews in Food Science and Food Safety, 12, 155–182. 10.1111/1541-4337.12007 [Crossref], [Web of Science ®], [Google Scholar]
- Gantois, I., Ducatelle, R., Timbermont, L., Boyen, F., Bohez, L., Haesebrouck, F., Pasmans, F. & van Immerseel, F.2006. Oral immunisation of laying hens with the live vaccine strains of TAD Salmonella vac E and TAD Salmonella vac T reduces internal egg contamination with Salmonella Enteritidis. Vaccine, 24, 6250–6255. 10.1016/j.vaccine.2006.05.070 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Gast, R.K. (1994). Understanding Salmonella enteritidis in laying chickens: the contributions of experimental infections. International Journal of Food Microbiology, 21, 107–116. 10.1016/0168-1605(94)90204-6 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Gast, R.K. & Beard, C.W. (1990). Production of Salmonella enteritidis-contaminated eggs by experimentally infected hens. Avian Diseases, 34, 438–446. 10.2307/1591433 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Hauser, E., Junker, E., Helmuth, R. & Malorny, B. (2011). Different mutations in the oafA gene lead to loss of O5-antigen expression in Salmonella enterica serovar Typhimurium. Journal of Applied Microbiology, 110, 248–253. 10.1111/j.1365-2672.2010.04877.x [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Janmohamed, K., Zenner, D., Little, C., Lane, C., Wain, J., Charlett, A. & Adak, B., Morgan, D. (2011). National outbreak of Salmonella Enteritidis phage type 14b in England, September to December 2009: case-control study. Euro Surveillance, 16. [PubMed], [Google Scholar]
- Lehmann, J., Springer, S., Werner, C.E., Lindner, T., Bellmann, S., Straubinger, R.K., Selbitz, H.J. & Alber, G. (2006). Immunity induced with a Salmonella enterica serovar Enteritidis live vaccine is regulated by Th1-cell-dependent cellular and humoral effector mechanisms in susceptible BALB/c mice. Vaccine, 24, 4779–4793. 10.1016/j.vaccine.2006.03.024 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Noble, D.J., Lane, C., Little, C.L., Davies, R., De Pinna, E., Larkin, L. & Morgan, D. (2012). Revival of an old problem: an increase in Salmonella enterica serovar Typhimurium definitive phage type 8 infections in 2010 in England and Northern Ireland linked to duck eggs. Epidemiology and Infection, 140, 146–149. 10.1017/S0950268811000586 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- O'Brien, S.J. (2013). The “Decline and Fall” of nontyphoidal Salmonella in the United Kingdom. Clinical Infectious Diseases, 56, 705–710. 10.1093/cid/cis967 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- OIE. (2012). Manual of Diagnostic Tests and Vaccines for Terrestrial Animals. p. 1268. Paris: World Organisation for Animal Health (OIE). [Google Scholar]
- Olsen, J.E., Hoegh-Andersen, K.H., Rosenkrantz, J.T., Schroll, C., Casadesus, J., Aabo, S. & Christensen, J.P. (2013). Intestinal invasion of Salmonella enterica serovar Typhimurium in the avian host is dose dependent and does not depend on motility and chemotaxis. Veterinary Microbiology, 165, 373–377. 10.1016/j.vetmic.2013.04.008 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Rabsch, W., Andrews, H.L., Kingsley, R.A., Prager, R., Tschape, H., Adams, L.G. & Baumler, A.J. (2002). Salmonella enterica serotype Typhimurium and its host-adapted variants. Infection and Immunity, 70, 2249–2255. 10.1128/IAI.70.5.2249-2255.2002 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Springer, S., Lindner, T., Ahrens, M., Woitow, G., Prandini, F. & Selbitz, H.J. (2011). Duration of immunity induced in chickens by an attenuated live Salmonella enteritidis vaccine and an inactivated Salmonella enteritidis/typhimurium vaccine. Berliner und Munchener tierarztliche Wochenschrift, 124, 89–93. [PubMed], [Web of Science ®], [Google Scholar]
- Suzuki, S. (1994). Pathogenicity of Salmonella enteritidis in poultry. International Journal of Food Microbiology 21, 89–105. 10.1016/0168-1605(94)90203-8 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Switt, A.I., Soyer, Y., Warnick, L.D. & Wiedmann, M. (2009). Emergence, distribution, and molecular and phenotypic characteristics of Salmonella enterica serotype 4,5,12:i. Foodborne Pathogens and Disease, 6, 407–415. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Van Immerseel, F., Methner, U., Rychlik, I., Nagy, B., Velge, P., Martin, G., Foster, N., Ducatelle, R. & Barrow, P.A. (2005). Vaccination and early protection against non-host-specific Salmonella serotypes in poultry: exploitation of innate immunity and microbial activity. Epidemiology and Infection, 133, 959–978. 10.1017/S0950268805004711 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Ward, L.R., Threlfall, J., Smith, H.R. & O'Brien, S.J. (2000). Salmonella enteritidis epidemic. Science, 287, 1753–1754; author reply 1755–1756. 10.1126/science.287.5459.1753c [Crossref], [PubMed], [Google Scholar]
- Wattiau, P., Boland, C. & Bertrand, S. (2011). Methodologies for Salmonella enterica subsp. enterica subtyping: gold standards and alternatives. Applied Environmental Microbiology, 77, 7877–7885. 10.1128/AEM.05527-11 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Withanage, G.S.K., Wigley, P., Kaiser, P., Mastroeni, P., Brooks, H., Powers, C., Beal R.K., Barrow, P.A., Mascall, D. & McConnell, I. (2005). Cytokine and chemokine responses associated with clearance of a primary Salmonella enterica serovar Typhimurium infection in the chicken and in protective immunity to rechallenge. Infection and Immunity, 73, 5173–5182. 10.1128/IAI.73.8.5173-5182.2005 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]