Effect of the housing system on shedding and colonization of gut and internal organs of laying hens with Salmonella Enteritidis

Effect of the housing system on shedding and colonization of gut and internal organs of laying hens with Salmonella Enteritidis J. De Vylder,* 1 S. Van Hoorebeke, R. Ducatelle,* F. Pasmans,* F. Haesebrouck,* J. Dewulf, and F. Van Immerseel * * Department of Pathology, Bacteriology and Avian Diseases, and Department of Reproduction, Obstetrics and Herd Health, Research Group Veterinary Public Health and Zoonoses, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, B-9820 Merelbeke, Belgium ABSTRACT As a result of welfare considerations, conventional cage systems will be banned in the European Union from 2012 onward. Currently, there is limited information on the level of contamination with zoonotic pathogens related to the laying hen housing system. Therefore, 2 studies were designed to investigate the effect of the housing system on colonization of layers with Salmonella. In both studies, layers were housed in 3 different housing systems: a conventional cage system, a furnished cage, and an aviary. At 18 wk of age, all birds were orally inoculated with Salmonella Enteritidis. Shedding and colonization were measured at regular time points. The results did not show an increased risk for alternative housing systems compared with the conventional battery cage system. In contrast, in one study, a faster decline in shedding was noted for layers housed in the alternative cage systems in comparison with the conventional cage system. This study does not give indications that housing layers in alternative systems will yield a risk for increased Salmonella contamination. Key words: Salmonella Enteritidis, conventional cage, furnished cage, aviary, laying hen 2009 Poultry Science 88 :2491 2495 doi: 10.3382/ps.2009-00203 INTRODUCTION 2009 Poultry Science Association Inc. Received April 21, 2009. Accepted August 28, 2009. 1 Corresponding author: jantina.devylder@ugent.be Salmonella Enteritidis is still a major cause of foodborne disease worldwide, mainly due to the consumption of contaminated hen eggs. A retrospective study, analyzing data from 1985 to 1999, reported that 80% of the outbreaks of Salmonella Enteritidis in the United States, which had a known foodborne origin, were eggassociated (Patrick et al., 2004). Data of the European Food Safety Authority (EFSA, 2007a) showed that Salmonella caused the majority of the foodborne disease outbreaks reported in 2006 in the European Union. Eggs and egg products were the most frequently identified food vehicles in these outbreaks. This illustrates the importance of hens eggs, derived from flocks infected with Salmonella Enteritidis, as a source of human salmonellosis. Although conventional battery cages are the predominant housing type in many countries worldwide, they have been criticized because of their failure in satisfying hen s welfare needs (Baxter, 1994). As a result of laying hens welfare considerations, the European Union has issued the Council Directive 1999/74/EC prohibiting the housing in conventional battery cages from January 1, 2012 onward (Anonymous, 1999). Deep litter systems and alternative cage systems, such as the furnished cage and the aviary, are proposed to replace the conventional cages (Tauson, 2005). The effect of alternative cage systems on animal welfare and productivity has been investigated extensively. These new systems allow a broader range of natural behavior to be expressed, although more aggressive behavior is observed in alternative cage systems, mainly in large-scale units (Tauson, 2002; Rodenburg et al., 2005; Shinmura et al., 2006). Concerns also arose regarding air quality and the potential increase of contamination of eggs by infectious and zoonotic agents when using the enriched cages and alternative housing systems (De Reu et al., 2005). A recent field study, using fecal sampling, reported that 18.3% of the laying hen holdings in the European Union were positive for Salmonella Enteritidis in 2006 (EFSA, 2007b). Cage housing, compared with barn, free-range, and organic housing, was found to be associated with a higher risk of Salmonella positivity, but this finding may have been confounded by farm and flock size, age of the housing system, and geographical spread of the housing systems (EFSA, 2007b). There 2491

2492 is thus urgent need for qualitative and quantitative research on the microbiology of eggs produced in different housing systems under controlled conditions. A study was therefore initiated to analyze shedding and colonization of the gut and internal organs after Salmonella infection of chickens housed in different systems but under standardized conditions. These include environmental parameters, the inoculated Salmonella strain and inoculum size, and sampling and analysis methods. Birds MATERIALS AND METHODS The ISA Brown laying hens, not vaccinated against Salmonella, were bought from the same commercial farm. They were housed in the animal units described below at least 2 wk before the start of the experiments. Strain The nalidixic acid-resistant Salmonella Enteritidis phage type 4 strain 76Sa88 was used in the experimental trials. This strain was isolated on a poultry farm and is well characterized (Van Immerseel et al., 2004; Bohez et al., 2008). The strain was cultured in Luria-Bertani broth (Sigma-Aldrich, St. Louis, MO) supplemented with 20 µg/ml of nalidixic acid (Sigma-Aldrich) for 20 h at 37 C while shaking. The suspension was centrifuged for 10 min at at approximately 1,000 g at 4 C and the pellet was resuspended in PBS. Serial dilutions of the suspension were made on brilliant green agar (BGA, LabM, Lancashire, UK) supplemented with 20 µg/ml of nalidixic acid to determine the number of colony-forming units per milliliter. The suspension was stored overnight at 4 C and was then further diluted with PBS to a concentration of 1.5 10 8 cfu/ml (experiment 1) and 2.25 10 9 cfu/ml (experiment 2). Housing Systems Three different housing systems, a battery cage, a furnished cage, and an aviary, were used. All cages were supplied by Big Dutchman (Vechta, Germany). The housing systems were installed in separate rooms with the same environmental conditions (feed, water, temperature, air humidity, light scheme). All rooms were provided with a high-efficiency particulate air filter. Conventional Cage. The conventional cages had a width of 44 cm and a depth of 50 cm. Height varied between 38 and 42 cm. In each cage, 4 hens were housed with 550 cm 2 area per hen. The conventional cage system consisted of 3 columns of 3-tier cages at both sides. Furnished Cage. The furnished cage [Eurovent EV625a-EU60 (Kleinvoliere), Big Dutchman] measured 125 cm 361.8 cm in total. The height varied between De Vylder et al. 45 and 52.5 cm. The cage was designed for 60 hens and provided an area of 750 cm 2 per hen. Perches were available at one level, resulting in 15 cm of perch per hen. A litter mat (0.57m 2 ) was available at the front of the cage. A nest (0.57m 2 ), obscured by curtains, was installed at the back of the cage. Aviary. The aviary (Natura-Nova, Big Dutchman) had a width of 120 cm and a length of 190 cm. The height varied between 148 and 156 cm. A laying nest, with a tilting nest floor and obscured by curtains, was installed at the back of the cage. The nest measured 46.8 cm 120 cm and was closed at night. A littered floor area, measuring 185 cm 255 cm, was provided at the front of the cage. Perches of a total length of 9.6 m were available on 3 levels. The cage provided an area of 1,166 cm 2 per hen. Experimental Setup Experiment 1. Two hundred non-salmonella-vaccinated commercial laying hens were housed in 3 different housing systems from the age of 16 wk: (1) 80 layers in a conventional battery cage, (2) 60 layers in a furnished cage, and (3) 60 layers in an aviary. Before the start of the study, the Salmonella status of the hens was evaluated by bacteriological analysis of cloacal swabs of all the hens and by serological testing for Salmonella antibodies using a previously described ELISA test based on detection of anti-lipopolysaccharide antibodies (Desmidt et al., 1996). At the age of 18 wk, all laying hens were orally inoculated with 1.5 10 8 cfu of a nalidixic acid-resistant Salmonella Enteritidis strain, 76Sa88, phage type 4. Cloacal swabs were taken at d 1, 4, 7, 14, 21, and 28 postinfection to monitor shedding. At these time points, also a number of chickens of each group were killed and samples of the ceca were taken for bacteriological analysis (Table 1). Experiment 2. One hundred ninety-two 16-wk-old non-salmonella-vaccinated commercial laying hens were housed in 3 different housing systems: (1) 72 layers in a conventional battery cage, (2) 60 layers in a furnished cage, and (3) 60 layers in an aviary. The Salmonella status of the hens was evaluated before the start of the study by bacteriological analysis of cloacal swabs of all the hens and by serological testing for Salmonella antibodies using a previously described ELISA test based on detection of anti-lps antibodies (Desmidt et al., 1996). All hens were orally inoculated at the age of 18 wk with 2.25 10 9 cfu of Salmonella Enteritidis 76Sa88. Cloacal swabs were taken at d 1, 4, 7, 14, and 21 postinfection. To exclude possible influences on colonization from removing birds from the groups during the trial, all layers were killed at the end of the trial. At necropsy, samples of the ceca and spleen were taken for bacteriological analysis. In both experiments, all layers had free access to feed and drinking water. A 16L/8D lighting scheme was applied. For each animal unit, 2 animal technicians (Federation of European Laboratory Animal Science Asso-

EFFECT OF HOUSING ON SALMONELLA COLONIZATION OF HENS 2493 Table 1. Number of Salmonella-positive samples after selective enrichment to the total number of tested samples for the different housing systems on the different sampling days in experiment 1 1 Day p.i. 2 Housing system Swab Ceca Spleen Liver Ovary Oviduct 3 Infundibulum Magnum Isthmus Uterus Vagina 1 Conventional cage 9/76 12/12 0/12 0/12 0/12 0/12 0/12 0/12 0/12 0/12 Furnished cage 14/56 8/8 0/8 0/8 0/8 0/8 0/8 0/8 1/8 2/8 Aviary 8/56 8/8 0/8 0/8 0/8 0/8 0/8 0/8 0/8 0/8 4 Conventional cage 28/64 14/16 13/16 13/16 6/16 0/16 2/16 0/16 5/16 7/16 Furnished cage 20/48 12/12 11/12 11/12 4/12 0/12 1/12 1/12 2/12 4/12 Aviary 12/48 10/12 9/12 9/12 4/12 1/12 2/12 0/12 1/12 2/12 7 Conventional cage 5/48 12/16 12/16 10/16 6/16 1/16 2/16 3/16 3/16 1/16 Furnished cage 6/36 12/12 10/12 10/12 4/12 0/12 0/12 0/12 0/12 0/12 Aviary 1/36 8/12 7/12 7/12 4/12 1/12 2/12 2/12 1/12 0/12 14 Conventional cage 1/32 11/12 12/12 11/12 5/12 2/12 2/12 2/12 2/12 3/12 Furnished cage 1/24 4/8 5/8 6/8 4/8 0/8 0/8 0/8 0/8 0/8 Aviary 0/24 4/8 3/8 3/8 2/8 0/8 1/8 0/8 0/8 0/8 21 Conventional cage 0/20 9/12 9/12 8/12 7/12 1/12 0/12 1/12 1/12 1/12 Furnished cage 0/16 7/8 4/8 4/8 4/8 1/8 2/8 0/8 0/8 0/8 Aviary 0/16 6/8 7/8 5/8 4/8 0/8 0/8 1/8 1/8 1/8 28 Conventional cage 0/8 3/8 3/8 3/8 3/8 1/8 2/8 1/8 1/8 2/8 Furnished cage 0/8 3/8 5/8 4/8 4/8 1/8 1/8 1/8 2/8 0/8 Aviary 0/8 2/8 6/8 3/8 1/8 1/8 2/8 1/8 1/8 1/8 2 Isolation at days postinoculation. 3 No significant differences (P < 0.05) were detected between the number of positive samples of laying hens housed in different housing systems at any time point. ciations category B) were responsible for the daily care and the collection of the samples during both experiments. All experiments were approved by the Ethical Committee of the Faculty of Veterinary Medicine, Ghent University. Bacteriological Analysis Cloacal swabs were directly plated on BGA supplemented with 20 µg/ml of nalidixic acid, and incubation was done for 20 h at 37 C. Swabs that were negative after direct plating were preenriched in 2 ml of buffered peptone water (Oxoid, Basingstoke, UK) for 20 h at 37 C and enriched by addition of 1 ml of the preenrichment suspension to 9 ml of tetrathionate brilliant green broth (Merck, Darmstadt, Germany) for 20 h at 37 C. A loopful of this suspension was plated on BGA. Samples of ceca, spleen, liver, ovary, and oviduct were homogenized and 10-fold dilutions were made in PBS. Each dilution was plated on BGA plates containing 20 µg/ml of nalidixic acid. For samples that were negative after direct plating, preenrichment and enrichment were performed as described above. Statistical Analysis Data of experiment 1 were analyzed by means of multivariate logistic regression taking into account both the housing type and the day of sampling to enable the evaluation of the effect of the housing system for the whole observation period in 1 analysis. For each outcome variable (cloacal swabs, ceca, spleen, liver, ovary, and oviduct), a new model was fitted. Data of experiment 2 were first analyzed by means of a univariate logistic regression model for the cloacal swabs at each time point separately and for the different organs at the end of the observation period. Second, a multivariate logistic regression model was developed to evaluate the effect of the housing system on the results of bacteriological analysis of the cloacal swabs for the whole observation period. All data analysis was done in SPSS 17.0 (SPSS Inc., Chicago, IL). Statistical results were considered to be significant when P-values were lower than 0.05. Experiment 1 RESULTS Before the experiment, excretion of Salmonella was not detected and all birds were found to be serologically negative for antibodies against Salmonella Enteritidis. Table 1 shows the ratio of the total number of positive samples after enrichment to the total number of samples, for cloacal swabs and samples of ceca, liver, spleen, ovary, and the different segments of the oviduct, at d 1, 4, 7, 14, 21, and 28 postinfection. Although some differences between the housing systems were observed in the number of positive cloacal swab samples at some time points, no overall significant differences were observed over time. The percentage of positive swabs decreased for all groups over time, and at 3 and 4 wk postinfection, all cloacal swabs were negative. All cecal samples were positive for Salmonella Enteritidis at d 1 postinfection, although only a low percentage (number) of shedding birds, ranging from 11.84% (9/76) to 25% (14/56), depending on the housing sys-

2494 De Vylder et al. Table 2. Number of Salmonella-positive cloacal swabs after selective enrichment to the total number of tested samples for the different housing systems on the different sampling days in experiment 2 1 Housing system Day 1 Day 4 Day 7 Day 14 Day 21 Conventional cage 41/72 a 52/72 a 36/72 a 19/71 a 22/71 a Furnished cage 52/60 b 55/60 b 50/60 b 13/60 a 8/60 b Aviary 50/60 b 46/60 a 48/60 b 18/60 a 6/59 b a,b Different superscripts indicate a significant difference (P < 0.05) between the number of positive swabs of birds in the different housing systems at that time point. tem, was detected by cloacal swabbing at that time point. Over time, a decline in the number of Salmonella-positive cecal samples was noted. Although shedding was not recorded at 3 and 4 wk postinfection, positive cecal samples were detected. No overall statistically significant differences were detected in the number of positive cecal samples between groups. In addition, no overall significant differences were observed in the number of positive spleen, liver, ovary, and oviduct samples between the different treatment groups (Table 1). Experiment 2 Before the experiment, excretion of Salmonella was not detected and all birds were found to be serologically negative for antibodies against Salmonella Enteritidis. Shedding was significantly lower at d 1 and 7 postinfection in the group of layers housed in the conventional cage system compared with the birds housed in the aviary and the furnished cage. Over time, the percentage of positive cloacal swabs decreased for all the groups, but this decrease was most prominent for birds housed in the aviary and the furnished cage system (Table 2). When analyzing the full observation period, a significantly (P < 0.05) lower Salmonella excretion was observed in the battery cage in comparison to the furnished cage and aviary. At the end of the trial (i.e., at 21 d postinfection), ceca were Salmonella-positive in 46.5, 45, and 42% of the birds housed in the conventional cage, furnished cage, and aviary, respectively. No significant differences were detected in the number of positive ceca. The spleen was less frequently colonized (borderline significant) in birds housed in the furnished cage (P < 0.06) and the aviary (P < 0.07), compared with the conventional cage. Liver was significantly (P < 0.05) less colonized in the furnished cage and the aviary, compared with the conventional cage (Table 3). No significant differences were found in the number of positive ovary and oviduct samples for the different housing systems at the end of the trial (Table 3). DISCUSSION Based on epidemiological data from Pieskus et al. (2008) and data from the current study, no increases in Salmonella shedding and colonization of the gut and internal organs are to be expected when moving laying hens from conventional battery cage systems to alternative cage systems. Until now, epidemiological data have mainly focused on comparing conventional cage systems with noncage systems. A study of Mollenhorst et al. (2005) concluded that the housing system with the lowest risk of a Salmonella Enteritidis infection was the cage system with wet manure. The study considered only cage systems with wet or dry manure and deep litter systems with or without outdoor run. In contrast, a study of Methner et al. (2006) showed a higher prevalence of Salmonella for layer flocks housed in conventional cages (46.3%) in comparison with flocks housed in alternative systems (33%), such as organic farming and floor management. A study of EFSA (2007b), sampling feces from 5,310 holdings, showed that cage systems were more frequently Salmonella-positive compared with organic flocks, barn, and free-range systems. A field study of Pieskus et al. (2008), in which 47 flocks of 8 laying hen farms in Lithuania were sampled, showed no significant differences in the prevalence of Salmonella between laying hens reared in conventional cages, furnished cages, and aviaries. All of the above-mentioned epidemiological data, however, have the risk of being confounded to a certain degree by factors such as flock size, age Table 3. Number of Salmonella-positive tissue samples after selective enrichment to the total number of tested samples for the different housing systems at d 21 postinfection in experiment 2 1 Housing system Ceca Spleen Liver Ovary Oviduct Conventional cage 33/71 a 29/71 a 22/71 a 11/71 a 8/71 a Furnished cage 27/60 a 15/60 a 8/60 b 7/60 a 12/60 a Aviary 25/59 a 15/59 a 6/59 b 8/59 a 6/59 a a,b Different superscripts indicate a significant difference (P < 0.05) between the number of positive tissue samples of birds housed in different housing systems.

EFFECT OF HOUSING ON SALMONELLA COLONIZATION OF HENS 2495 of the holdings, and sampling method. In general, it is difficult to determine any causal relationships between housing type itself and Salmonella contamination level based on the results of these field studies. Therefore, in the present study, a direct comparison of 3 different housing systems was carried out using experimental infections with the same strain, dosage, and environmental factors (temperature, lighting). These trials did not show an increased risk for alternative housing systems compared with the conventional battery cage system, which is similar to the results of the field study of Pieskus et al. (2008). Even more, in our experiments, a faster decline in the shedding level of birds housed in the alternative cage systems in comparison with the conventional cage system was noticed, as well as a slightly decreased internal organ colonization level in one trial. Direct bird-to-bird contact, mechanical vectors, and airborne transmission of Salmonella Enteritidis in dust or aerosols can play a role in the dissemination of infection within flocks (Gast et al., 1998). All of these factors can differ depending on the housing system but were not investigated separately in these experiments. Differences in cage area, animal density, and cage environment could all have contributed to the results. In conclusion, in addition to existing epidemiological data, the experimental infection trials carried out in the current study indicate that moving laying hens to more welfare-friendly housing systems, other than the conventional battery cage, will not yield an increased risk of layers being colonized with Salmonella. ACKNOWLEDGMENTS This research was funded by the European Union, Framework Program 6, under the contract 065547 (Safehouse Project). REFERENCES Anonymous. 1999. Council Directive 1999/74/EC of 19 July 1999: Minimum standards for the protection of laying hens. Off. J. Eur. Communities L203:53 57. Baxter, M. R. 1994. The welfare problems of laying hens in battery cages. Vet. Rec. 134:614 619. Bohez, L., J. Dewulf, R. Ducatelle, F. Pasmans, F. Haesebrouck, and F. Van Immerseel. 2008. The effect of oral administration of a homologous hila mutant strain on the long-term colonization and transmission of Salmonella Enteritidis in broiler chickens. Vaccine 26:372 378. De Reu, K., K. Grijspeerdt, M. Heyndrickx, J. Zoons, K. De Baere, M. Uyttendaele, J. Debevere, and L. Herman. 2005. Bacterial eggshell contamination in conventional cages, furnished cages and aviary housing systems for laying hens. Br. Poult. Sci. 46:149 155. Desmidt, M., R. Ducatelle, F. Haesebrouck, P. A. De Groot, M. Verlinden, R. Wijffels, M. Hinton, J. A. Bale, and V. M. Allen. 1996. Detection of antibodies to Salmonella Enteritidis in sera and yolks from experimentally and naturally infected chickens. Vet. Rec. 138:223 226. EFSA. 2007a. The community summary report on trends and sources of zoonoses, zoonotic agents, antimicrobial resistance and foodborne outbreaks in the European Union in 2006. EFSA J. 130:34 117. EFSA. 2007b. Report of task force on zoonoses data collection on the analysis of the baseline study on the prevalence of Salmonella in holdings of laying hen flocks of Gallus gallus. EFSA J. 97:1 84. Gast, R. K., B. W. Mitchell, and P. Holt. 1998. Airborne transmission of Salmonella Enteritidis infection between groups of chicks in controlled environment isolation cabinets. Avian Dis. 42:315 320. Methner, U., R. Diller, R. Reiche, and K. Böhland. 2006. Occurrence of salmonellae in laying hens in different housing systems and conclusion for the control. Berl. Munch. Tierarztl. Wochenschr. 119:467 473. Mollenhorst, H., C. J. van Woudenbergh, E. G. M. Bokkers, and I. J. M. de Boer. 2005. Risk factors for Salmonella Enteritidis in laying hens. Poult. Sci. 84:1308 1313. Patrick, M. E., P. M. Adcock, T. M. Gomez, S. F. Altekruse, B. H. Holland, R. V. Tauxe, and D. L. Swerdlow. 2004. Salmonella Enteritidis infections, United States, 1985 1999. Emerg. Infect. Dis. 10:1 7. Pieskus, J., E. Kazeniauskas, C. Butrimaite-Ambrozeviciene, Z. Stanevicius, and M. Mauricas. 2008. Salmonella incidence in broiler and laying hens with the different housing systems. Jpn. Poult. Sci. 45:227 231. Rodenburg, T. B., F. A. Tuyttens, B. Sonck, K. De Reu, L. Hermans, and J. Zoons. 2005. Welfare, health and hygiene of laying hens housed in furnished cages and alternative housing systems. J. Appl. Anim. Welf. Sci. 8:211 226. Shinmura, T., Y. Eguchi, K. Uetake, and T. Tanaka. 2006. Behavioral changes in laying hens after introduction to battery cages, furnished cages and an aviary. Anim. Sci. J. 77:242 249. Tauson, R. 2002. Furnished cages and aviaries: Production and health. World s Poult. Sci. J. 58:49 63. Tauson, R. 2005. Management and housing systems for layers Effects on welfare and production. World s Poult. Sci. J. 61:467 490. Van Immerseel, F., J. De Buck, F. Boyen, L. Bohez, F. Pasmans, J. Volf, M. Sevcik, I. Rychlik, F. Haesebrouck, and R. Ducatelle. 2004. Medium-chain fatty acids decrease colonization and invasion through hila suppression shortly after infection of chickens with Salmonella enterica serovar Enteritidis. Appl. Environ. Microbiol. 70:3582 3587.