Effects of rearing environment on the gut antimicrobial responses of two broiler chicken lines

Accepted Manuscript Title: Effects of Rearing Environment on the Gut Antimicrobial Responses of Two Broiler Chicken Lines Author: Vanessa L. Butler Catherine A. Mowbray Kevin Cadwell Sherko S. Niranji Richard Bailey Kellie A Watson John Ralph Judith Hall PII: DOI: Reference:

S0165-2427(16)30113-1 http://dx.doi.org/doi:10.1016/j.vetimm.2016.06.004 VETIMM 9518

To appear in:

VETIMM

Received date: Revised date: Accepted date:

22-9-2015 6-5-2016 15-6-2016

Please cite this article as: Butler, Vanessa L., Mowbray, Catherine A., Cadwell, Kevin, Niranji, Sherko S., Bailey, Richard, Watson, Kellie A, Ralph, John, Hall, Judith, Effects of Rearing Environment on the Gut Antimicrobial Responses of Two Broiler Chicken Lines.Veterinary Immunology and Immunopathology http://dx.doi.org/10.1016/j.vetimm.2016.06.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Effects of Rearing Environment on the Gut Antimicrobial Responses of Two Broiler Chicken Lines

Vanessa L. Butler1a, Catherine A. Mowbray1, Kevin Cadwell1, Sherko S. Niranji1, Richard Bailey2, Kellie A Watson2, John Ralph2 and Judith Hall1*

1

Institute for Cell and Molecular Biosciences, Newcastle University, Newcastle upon Tyne, NE2 4HH, UK and 2Aviagen Ltd, Newbridge, Midlothian, EU28 8SZ, UK.

* To whom correspondence should be addressed Correspondence:

Judith Hall Institute for Cell and Molecular Biosciences Newcastle University Newcastle upon Tyne NE2 4HH UK

Tel

+44 (0) 191 208 8346

Fax

+44 (0) 191 208 7424

Email

[email protected]

Short Title: Innate response of the chicken gut

a

VLB present address: School of Biomedical Sciences, Newcastle University. 1

Highlights:    

Rearing conditions and gut innate defences of genetically distinct birds studied Hatch linked to maximal proximal gut antimicrobial activity High hygiene environment linked to early relaxation of gut antimicrobial defences Impact of rearing environment on specific gut defences unique to specific bird line

Abstract To reduce the risk of enteric disease in poultry, knowledge of how bird gut innate defences mature with age while also responding to different rearing environments is necessary. In this study the gut innate responses of two phylogenetically distinct lines of poultry raised from hatch to 35 days, in conditions mimicing high hygiene (HH) and low hygiene (LH) rearing environments, were compared. Analyses focussed on the proximal gut antimicrobial activities and the duodenal and caecal AvBD1, 4 and 10 defensin profiles. Variability in microbial killing was observed between individual birds in each of the two lines at all ages, but samples from day 0 birds (hatch) of both lines exhibited marked killing properties, Line X: 19±11% (SEM) and Line Y: 8.5±12% (SEM). By day 7 a relaxation in killing was observed with bacterial survival increased from 3 (Line Y (LY)) to 11 (Line X (LX)) fold in birds reared in the HH environment. A less marked response was observed in the LH environment and delayed until day 14. At day 35 the gut antimicrobial properties of the two lines were comparable. The AvBD 1, 4 and 10 data relating to the duodenal and caecal tissues of day 0, 7 and 35 birds LX and LY birds revealed gene expression trends specific to each line and to the different rearing environments although the data were confounded by inter-individual variability. In summary elevated AvBD1 duodenal expression was detected in day 0 and day 7 LX, but not LY birds, maintained in LH environments; Line X and Y duodenal AvBD4 profiles were detected in day 7 birds reared in both environments although duodenal AvBD10 expression was less sensitive to bird age and rearing background. Caecal AvBD1 expression was particularly evident in newly hatched birds. These data suggest that proximal gut antimicrobial activity is related to the bird rearing environments although the roles of the AvBDs in such activities require further investigation.

Key Words: Chicken; Rearing Environment; High Hygiene; Low Hygiene; Gut Antimicrobial Activity; Host peptides; SNPs; Gene Expression; Duodenum; Caecum.

2

1.1 Introduction Newly hatched and young birds rely on their innate defences for protection against infection and with the onset of food intake the gut is particularly susceptible to microbial assault. The inability or failure of the gut defences to protect against pathogens can present as chronic gut inflammation (Ramasundara et al., 2009), which in commercial rearing environments can escalate into disease and bird mortality. Hence an understanding of the innate immune responses of the bird gut to microbial challenges, particularly those relating to the immediate rearing environment, is necessary to help direct future breeding programmes. Microbial colonisation of the chicken gut is naturally associated with inflammation linked potentially to the induction of pro-inflammatory cytokines that help prime the gut immune system and facilitate the maturation of the gut (Crhanova et al., 2011). During this period immediate gut protection is mediated through the collective functioning of the gut innate defences including the epithelial barrier, the mucus layer covering the epithelium and the production of proteins and peptides with antimicrobial activity. These host molecules including lysozyme, sPLA2, and the defensins function as endogenous antibiotics with the avian defensins (AvBDs) providing a significant protective barrier due to their broad-spectrum antimicrobial activity against bacteria and fungi, a function linked to their structure and charge (Cuperus et al., 2013). Potential immunomodulatory functions also support a role for the AvBDs in the recruitment of immune cells thus facilitating the development of the adaptive immune response (Soman et al., 2009). In birds, as well as mammals, microbial colonisation of the avian gut occurs in conjunction with defensin gene expression (Bar-Shira and Friedman, 2006; Crhanova et al., 2011; Salzman and Bevins, 2013), which supports a role for the encoded peptides in controlling the indigenous microbial numbers and composition. Following acute microbial challenges the gut defensin responses appear less predictable with studies in birds reporting both up and down regulation of the genes (Akbari et al., 2008; Hong et al.; Meade et al., 2009a; Milona et al., 2007). In reality a variety of factors in addition to the microbial challenge combine to affect gene expression including the location of the tissues along the anterior-posterior axis of the bird gut as well as the age and breed of the birds studied. In fact the importance of bird genetics in the hierarchical functioning of the gut innate defences is illustrated by the identification of AvBD single-nucleotide polymorphisms (SNPs) as molecular markers for selecting poultry resistant to enteric pathogens including Salmonella enteritidis (Hasenstein and Lamont, 2007).

3

The genetic selection of poultry is often compromised by the rearing environment yet to preserve the elite stocks, environmental challenges that influence bird performance and direct genetic selection are often performed in carefully monitored biosecure conditions that support a high hygiene (HH) environment. For some genetic traits this scenario works well with the selection against foot-pad dermatitis in a HH environment also reducing FPD prevalence in birds reared in commercial or low hygiene (LH) conditions (Kapell et al., 2012). However, in relation to genes associated with growth and/or immunity the outcomes appear less transferable. Examination of 12 immune related genes with performance and mortality traits in elite commercial broiler lines revealed that the TGF-β3-MSp1 SNP was significantly associated with mortality in HH, but not LH environments, while progeny of birds with allele 1 of iNOS-Alu1 had a higher 40 day body weight in HH compared to LH conditions (Ye et al., 2006). These data show that a better understanding of interactions between genetic and environmental factors is necessary to underpin the selection of bird stocks with immune systems more robust to environmental changes. The immune responsiveness of young birds is important in selecting and maintaining healthy birds yet the involvement and roles of the host innate defences, particularly the defensins, in protecting birds against disease in low hygiene (LH) environments more reflective of conditions in commercial situations are limited. To address this, two phylogenetically distinct genetic lines of poultry were raised from hatch to 35 days, in controlled conditions mimicing HH and LH environments and the effects of rearing environment on the gut innate responses of such birds compared. The primary focus was the upper gastrointestinal antimicrobial activities and the duodenal and caecal defensin profiles. 1.2 Materials & Methods 1.2.1 Birds Two phylogenetically distinct lines of poultry used in broiler breeding and designated X and Y were studied (Andreescu et al., 2007). The birds, 100 in total and 50 per line, were housed on farms in two different environments: a high biosecure environment referred to as pedigree or high hygiene (HH) where breeding programme candidates are selected and a non bio-secure environment referred to as sib-test or low hygiene (LH) environment resembling commercial conditions. Water and highquality diet were provided ad libitum throughout the growing period. Litter was in the form of wood shavings. Following hatch male birds were randomly selected and reared for up to 35 days under conditions of HH or LH. In the LH locations birds were raised in barns containing a mix of old (mechanically conditioned) and new litter while in the HH barns a complete disinfection process was

4

adopted between stocking regimes and all pens were supplied with fresh litter. Tissue sampling was performed at day 0 (hatch), 7, 14 and 35 respectively. On the day of bird sampling a representative bedding sample of shavings combined with faecal matter (20g) was taken from each pen. Samples were analysed by Poultry Health Service Ltd, UK for colony forming bacteria (CFU) per gram of sample. 1.2.2 Gut Antimicrobial Assays At each sampling time the duodenal loop of each bird was excised, cut longitudinally, the gut contents removed by washing in 0.1M phosphate buffered saline (PBS) and the mucosal layer collected by scraping. Samples were frozen in liquid nitrogen and stored at -80oC. Cationic gut proteins were extracted in 10% acetic acid and following lyophilisation each sample was reconstituted in 0.1M PBS pH7.4. Antimicrobial assays were performed as described previously (Townes et al., 2004), using Salmonella enterica serovar Typhimurium phoP (Behlau and Miller, 1993), and data normalised to protein concentration. Gut antimicrobial activity was presented as percentage of killed bacteria compared to PBS controls. 1.2.3 LC-MS/MS analysis Four volumes of ice-cold acetone were added to each reconstituted gut sample, the samples gently vortexed and stored overnight at -20oC. Following centrifugation the supernatants were removed, and the reconstituted residues subjected to LC/MS analysis (NEPAF Proteome Facility - now Newcastle University Protein & Proteome Analysis (NUPPA)). Annotation of the proteins was achieved through searching and mapping of the LC/MS data to either UNiProt or NCBI sites. Gene ontology analysis www.geneontology.org/ was performed to facilitate protein localisation. 1.2.4 RNA Tissue samples were taken from groups of up to 10 birds sacrificed at each time point ie day 0 (hatch), 7 and 35, and stored at -80oC in Trizol (Invitrogen) or RNA later (ThermoFisher). RNA was prepared using either Trizol methodology or a Promega SV Total RNA Isolation System with yield and purity determined using a Nanodrop ® (Thermoscientific) spectrophotometer. DNase treatment was carried out on all samples either after (Trizol) or during (Promega) the extraction process. 500ng of RNA was reverse transcribed using random hexamers (Roche), MMLV-reverse transcriptase and RNasin (both Promega) as per manufacturer’s instructions. Semi-quantitative analyses was performed using primers (SQ) and annealing conditions shown in Table 1. Quantitative PCR was performed using Sybr Green mastermix (Roche) with primers (Q) and annealing conditions specific for each gene of interest (Table 1), and where appropriate following manufacturer’s instructions. 5

Primers were designed to amplify over an exon-exon junction to eliminate amplification of any remaining genomic DNA and products were verified by cloning and sequencing. qPCR analysis was carried out using the LightCycler 480 (Roche) with the following program: 95oC 10 mins, 45 cycles of 95oC 10 seconds, 58/60/61oC 10 seconds, 72oC 5 seconds, followed by melt curve analysis to confirm generation of a single product. Negative, without RT, and positive, cloned plasmid, controls were included on each plate. Data were normalised to GAPDH and/or 18S housekeeping gene expression data, and presented as arbitrary units (AU).

1.2.5 Statistical Analyses Data are expressed as means ± standard error mean (SEM). Statistical analyses were performed by one-way analysis of variance, followed by the Dunnett’s multiple comparison tests for unpaired data. In all cases a P-value < 0.05 was considered statistically significant.

1.3 Results 1.3.1 Environmental Rearing Conditions Throughout the trial period the LH rearing environment was associated with higher mean temperatures than the HH environment (week 2: 30.3oC ±0.4 compared to 28.4 oC ±0.4 and week 5: 25.2 oC ±0.1 compared to 20.3 oC ±0.4), but lower humidity readings (week 2: 42.2% ±0.6 compared to 55.1% ±0.9 and week 5: 52.0% ±1.3 compared to 63.7% ±3.0). Bacterial bedding counts reflecting the severity of the microbial challenge during rearing (Fig 1) show that Line X day 7 and 14 LH reared birds were exposed to greater external bacterial burdens than those raised under HH conditions. Similar conditions prevailed for Line Y day 7 LH reared birds although by day 14 the bacterial counts of the LH and HH environments were comparable. At day 35 the bacterial challenge was equivalent for both Lines X and Y in the two rearing environments. 1.3.2 Gut Antimicrobial Activity Duodenal mucosal scrape samples from Line X and Y birds raised in the LH and HH environments were analysed for bacterial killing properties using the S. enterica serovar Typhimurium phoP strain, which exhibits increased susceptibility to the activity of antimicrobial peptides (Townes et al., 2004). Data relating to individual birds are shown in Fig 2A, 2B, 2C & 2E. Variability in microbial killing was observed between individual birds in each of the two lines for all data points, but samples from day 6

0 birds (hatch) exhibited marked bacterial killing properties with no significant differences in the mean bacterial survival data from line X: 19±11% (SEM) and line Y: 8.5±12% (SEM) (Fig 2A/D &B/E). At day 7 the bacterial killing properties of all the bird mucosal scrapes were reduced and, in fact, indicated the gut mucosa at this stage of development to be pro-microbial. However, bacterial survival relating to the gut mucosae of day 7 Line X birds raised in LH conditions (Fig2A) was reduced by up to 11 fold compared to that of birds raised in the HH environment (Fig2D). A similar but less marked response (3 fold reduction in killing) was observed for the Line Y birds (Fig 2 B & E). Significant differences (p<0.001) were observed between the X and Y bird lines raised in both the HH (1058 ±113% v 304±53% (SEM)) and LH environments (571 ±43% v 150±16% (SEM)). Notably, this difference occurred at day 7 in the HH environment, but not until day 14 in the LH environment (Fig2 C & F). By 35 the gut antimicrobial properties of the two lines were comparable.

1.3.3 Proteome Analyses The duodenal scrape data suggested a marked relaxation of the bird gut microbial killing properties during the first seven days of rearing that was particularly evident in Line X birds raised in the HH environment. These data suggested a potential immunological switch in the bird innate gut defences and to explore this further the gut proteins of such birds were analysed by LC-MS/MS. Table 1S lists the proteomic data relating to the gut scrapes of Line X birds at day 0 (hatch) and following 7 days in the HH conditions. Essentially 197 host proteins in the day 0 sample and 143 proteins in the day 7 sample were identified with 71 common proteins. Gene ontology analyses of the day 0 and day 7 gut proteins identified the majority as cytoplasmic 37% and 39% respectively with 6% cell membrane associated and 9 and 12% respectively, secreted (Fig 3). Of the proteins common to both samples 38% were cytoplasmic , 9% cell membrane associated and 6% secreted. The proteome analyses did not identify classic innate host defence molecules such as the defensins and /or cytokines in either the day 0 ie hatch or day 7 duodenal samples although these may have been contained within the 6% of the proteins that remained unidentifed. However, in relation to bacterial killing the day 0 duodenal proteome was characterised by the presence of histone and cathespin proteins known to possess potent antimicrobial activity as well as galectin-3 molecules linked to immune regulation. In contrast the day 7 sample was typified by proteins playing a role in digestion and included enzymes such as amylase, sucrase-isomaltase and dipeptidyl-peptidase. 1.3.4 Molecular Analyses

7

Molecular (end-point PCR) analyses of day 0 and 7 duodenal and caecal bird tissues indicated AvBD1-10 gene expression (Fig 4A), which despite the proteomics data, supported the defensins being involved in the antimicrobial response of the bird guts to the rearing environment. Further analyses of the AvBD 1-10 genes relating to the two different genetic lines identified 15 SNPs with 10 being intronic (AvBD 6, 8, 9 and 10), two being associated with the 5’UTR (AvBD 4, 10: Rs16341536 & Rs14411785), one being in 3’UTR region (AvBD9) and three within the mature peptide sequence (AvBD1) encoding synonymous (NYH) and non-synonymous (N/SSY) mutations (Rs15457749; Rs15457749; unreported VB1). The AvBD1 gene SNP resulted in the synthesis of a NYH AvBD1 peptide variant in Line X birds and a SSY AvBD1 peptide variant in Line Y birds (Fig 4B). Subsequent studies focussed on analysing the AvBD1, 4 and 10 genes due to potential impacts of the SNPs on gene expression and/or peptide function. Expression of the AvBD 1, 4 and 10 genes in the duodenal and caecal tissues of day 0, 7 and 35 birds showed marked variability not only between the groups, but between individual birds within each group (Fig 5). The outliers evident in all studies (shown as *) affected the statistical impact of these data with only the duodenal AvBD1 and caecal AvBD10 data reaching statistical significance (LX 0 and day 7LH duodenal AvBD1 differed significantly (p<0.05) from all other groups; LX 0 and day 7LH caecal AvBD10 differed significantly (p<0.001) from all other groups), but distinct trends were evident. The median duodenal AvBD1 expression was higher in day 0 ie hatch, Line X birds compared to those of Line Y (Fig 5A). Expression remained elevated in the 7 day-old Line X birds maintained in LH conditions, but thereafter was nominal. AvBD1 expression remained detectable, but minimal, in the Line Y birds regardless of age or rearing environment. In contrast duodenal AvBD4 expression was comparable between the two bird lines at hatch and, regardless of rearing environment, was detectable up to 7 days post hatch although by day 35, expression appeared minimal (Fig 5C). Likewise, the AvBD10 duodenal expression pattern was consistent between the two lines, but in contrast to AvBD1 and 4, the expression profiles remained comparable up to 35 days post hatch (Fig 5E). Unlike the pattern observed in the proximal gut tissues of day 0 birds, the median caecal AvBD1 expression, at hatch, was comparable between the X and Y Lines. Although again marked by individual bird variability, expression remained elevated in the day 7 Line Y birds raised in LH, but thereafter was nominal in both lines. In particular the Line Y bird data was characterised by numerous outliers (*) indicating AvBD 1 expression from 10 (Y7LH) to 250 (Y7HH) times the median value for the group (Fig 5B). Despite the variability caecal AvBD4 patterns supported elevated 8

expression in young birds (up to day 7) raised in LH conditions (Fig 5D). Caecal AvBD10 expression was again comparable between the two lines, but marked expression was only observed in birds reared in LH conditions and up to 7 days post hatch (Fig 5F). Immunologically, it was noted that AvBD10 expression appeared atypical in that levels in the kidney, liver and testes, tissues not normally associated with an immediate microbial challenge or indigenous microflora were maintained in all birds regardless of age and the rearing environment (Table 2).

1.4 Discussion

With the aim of developing strategies to maintain and improve poultry welfare, particularly since the E.U ban on anti-microbial growth promoters, attention has turned to understanding how factors, particularly those related to the immediate rearing environment, impact on the bird innate defences and consequent overall health. From hatch the chick gut is exposed to a plethora of environmental microbes and failure of the gut defences under such conditions can result in chronic gut inflammation, which in commercial rearing environments may escalate into disease and bird mortality. In this study two genetically different lines of birds were reared in conditions reflecting those of pedigree (HH) and commercial farms (LH) and their gut antimicrobial characteristics compared. Despite the contrasting rearing environments the proximal intestines of all the newly hatched birds were characterised by their potent antimicrobial activities with gut scrape data supporting up to 90% bacterial killing. This indicated an immediate and efficient innate barrier mechanism functioning to protect the gut epithelium from microbial attachment and invasion. The molecular and proteomics data suggested this response to include AvBDs and other proteins with antimicrobial properties, including histones (Tagai et al., 2011), although additional factors including maternally derived antibodies known to play important, albeit passive, protective roles cannot be excluded (Klipper et al., 2004).

By day 7 a marked relaxation in the gut antimicrobial activities of both bird lines was observed presumably linked to changes in the gut defences and microbial colonisation of the GI tract. That this occurred one week earlier in both bird lines raised in the HH environment suggests that gut development and hence the gut defences were strongly influenced by the immediate rearing environment with microbial numbers, and potentially species, a contibutory factor. The relationship 9

between the gut immune defences and the microbiota is known to be a key feature in maintaining healthy birds, with the synthesis of pro-inflammatory cytokines or physiological inflammation a natural and essential part of the gut colonisation process (Crhanova et al., 2011). That the reduction in bacterial killing was more marked in the Line X birds suggests this line may be more susceptible to gut pathology. Molecular analyses confirmed the expression of AvBDs 1-10 in the GI tissues of the newly hatched birds. Defensins are evolutionary ancient molecules characterised not only by their antimicrobial properties, but by the conservation of gene and protein sequences within species (Choi et al., 2012). As expected SNPS were relatively rare within the defensin locus and those that were identified predominantly intronic. However, three SNPs located to protein coding (AvBD1) and promoter (AvBD4 & 10) regions did differentiate the two chicken lines and directed the study of AvBD 1, 4 and 10 expression. Despite the genetic identity of the birds within each line, individual bird AvBD expression values varied widely and this variability was consistant for all AvBD genes and tissues analysed. The intravariability in defensin expression may have simply reflected the impact of the SNPs although such disparity has been observed previously in different lines of birds exposed to bacterial challenge (Meade et al., 2009b). Another explanation is defensin copy number variation. Large scale gene copy number variation has been reported in chicken breeds (Crooijmans et al., 2013), in humans where defensin copy number variation is well established (Hollox, 2008), and it has been reported to impact on the bacterial microbiota (Jones et al., 2014). Conversely the variation may have represented the actual microbial challenge experienced by each bird. For birds raised under commercial conditions the immediate rearing environment has been shown to have a profound impact on the initial bacterial communities with studies reporting the ileal microbiota of birds raised on fresh litter to be dominated by Lactobacillus spp (Cressman et al., 2010; Lu et al., 2003) compared to Clostridiales in those birds maintained on recycled bedding (Cressman et al., 2010). However, as in this study only the bedding microbial numbers were measured and not the microbial composition, the actual impact, if any, of a particular or combination of species on defensin expression profiles remains unknown.

Elevated duodenal AvBD 1 and 4 gene expression recorded in hatch and day 7 birds raised in the LH conditions, was linked to potent gut antimicrobial activity, while the lower expression observed in the HH raised birds was associated with a reduction in gut killing capacity. The duodenal AvBD10 expression profiles in contrast were maintained across all environments and ages suggesting the encoded peptide has physiological roles in addition to microbial killing. This is further supported by 10

the expression profiles observed in the kidney and testes, which in contrast to the gut, are not consistantly exposed to a microbial challenge. Multiple roles for defensins including functions in cell growth and tissue repair (Cuperus et al., 2013), and sperm function (Dorin and Barratt, 2014) have been described, but apart from duck Apl_AvBD2 a predominantly myeloid defensin, which has been shown to display chemotactic properties, such functions have not been widely reported for the avian defensins.

The caecal data followed similar trends to the duodenum with the tissues of younger birds (0-7 days) raised in the LH environments supporting a trend of increased gene expression. The caecum accommodates up to 1012 bacteria cells/g digesta (Gong et al., 2007), and as the defensins classically protect the gut epithelium from microbial assault it was surprising that at 35 days expression was barely detectable. This suggests that the caecal microbiota were either downregulating defensin expression or that by day 35 the mucus layer protecting the caecal epithelium from bacterial contact negated the need for defensin expression and activity. The mucus layer is an integral part of the immunological strategy used by the host to keep a beneficial symbiosis in the host–microbiota relationship (Hooper, 2009), and interestingly caecal bacteria including the Lactobacteria actively contribute to this epithelial barrier protection through inducing mucin 2 (MUC2) and MUC3 production (Lebeer et al., 2010).

This study comparing the effects of rearing environment on the gut innate responses of genetically distinct bird lines indicated that the proximal gut antimicrobial activity of those reared in LH conditions relaxed later than those in HH conditions. Future studies will address the impact of such effects on the gut microbiota and health of such birds.

1.6 Acknowledgements We acknowledge the support of BBSRC through grants BB/H018603/1 (VB & CM), BB/1532845/1 (KTN studentship KC) and BBS/S/M/13127 (Studentship VB). SSN was supported by the Kurdistan Government.

11

1.5 References Akbari, M.R., Haghighi, H.R., Chambers, J.R., Brisbin, J., Read, L.R., Sharif, S., 2008. Expression of antimicrobial peptides in cecal tonsils of chickens treated with probiotics and infected with Salmonella enterica serovar typhimurium. Clin Vaccine Immunol 15, 1689-1693. Andreescu, C., Avendano, S., Brown, S.R., Hassen, A., Lamont, S.J., Dekkers, J.C., 2007. Linkage disequilibrium in related breeding lines of chickens. Genetics 177, 2161-2169. Bar-Shira, E., Friedman, A., 2006. Development and adaptations of innate immunity in the gastrointestinal tract of the newly hatched chick. Developmental and comparative immunology 30, 930-941. Behlau, I., Miller, S.I., 1993. A PhoP-repressed gene promotes Salmonella typhimurium invasion of epithelial cells. J Bacteriol 175, 4475-4484. Choi, M.K., Le, M.T., Nguyen, D.T., Choi, H., Kim, W., Kim, J.H., Chun, J., Hyeon, J., Seo, K., Park, C., 2012. Genome-level identification, gene expression, and comparative analysis of porcine ss-defensin genes. BMC genetics 13, 98. Cressman, M.D., Yu, Z., Nelson, M.C., Moeller, S.J., Lilburn, M.S., Zerby, H.N., 2010. Interrelations between the microbiotas in the litter and in the intestines of commercial broiler chickens. Appl Environ Microbiol 76, 6572-6582. Crhanova, M., Hradecka, H., Faldynova, M., Matulova, M., Havlickova, H., Sisak, F., Rychlik, I., 2011. Immune response of chicken gut to natural colonization by gut microflora and to Salmonella enterica serovar enteritidis infection. Infect Immun 79, 2755-2763. Crooijmans, R.P., Fife, M.S., Fitzgerald, T.W., Strickland, S., Cheng, H.H., Kaiser, P., Redon, R., Groenen, M.A., 2013. Large scale variation in DNA copy number in chicken breeds. BMC Genomics 14, 398. Cuperus, T., Coorens, M., van Dijk, A., Haagsman, H.P., 2013. Avian host defense peptides. Developmental and comparative immunology 41, 352-369. Dorin, J.R., Barratt, C.L., 2014. Importance of beta-defensins in sperm function. Molecular human reproduction 20, 821-826. Gong, J., Si, W., Forster, R.J., Huang, R., Yu, H., Yin, Y., Yang, C., Han, Y., 2007. 16S rRNA gene-based analysis of mucosa-associated bacterial community and phylogeny in the chicken gastrointestinal tracts: from crops to ceca. FEMS Microbiol Ecol 59, 147-157. Hasenstein, J.R., Lamont, S.J., 2007. Chicken gallinacin gene cluster associated with Salmonella response in advanced intercross line. Avian Dis 51, 561-567. Hollox, E.J., 2008. Copy number variation of beta-defensins and relevance to disease. Cytogenet Genome Res 123, 148-155. Hong, Y.H., Song, W., Lee, S.H., Lillehoj, H.S., Differential gene expression profiles of beta-defensins in the crop, intestine, and spleen using a necrotic enteritis model in 2 commercial broiler chicken lines. Poult Sci 91, 1081-1088. Hooper, L.V., 2009. Do symbiotic bacteria subvert host immunity? Nature reviews. Microbiology 7, 367-374. Jones, E.A., Kananurak, A., Bevins, C.L., Hollox, E.J., Bakaletz, L.O., 2014. Copy number variation of the beta defensin gene cluster on chromosome 8p influences the bacterial microbiota within the nasopharynx of otitis-prone children. PLoS One 9, e98269. Kapell, D.N., Hill, W.G., Neeteson, A.M., McAdam, J., Koerhuis, A.N., Avendano, S., 2012. Genetic parameters of foot-pad dermatitis and body weight in purebred broiler lines in 2 contrasting environments. Poult Sci 91, 565-574. Klipper, E., Gilboa, T., Levy, N., Kisliouk, T., Spanel-Borowski, K., Meidan, R., 2004. Characterization of endothelin-1 and nitric oxide generating systems in corpus luteum-derived endothelial cells. Reproduction 128, 463-473.

12

Lebeer, S., Vanderleyden, J., De Keersmaecker, S.C., 2010. Host interactions of probiotic bacterial surface molecules: comparison with commensals and pathogens. Nature reviews. Microbiology 8, 171-184. Lu, J., Idris, U., Harmon, B., Hofacre, C., Maurer, J.J., Lee, M.D., 2003. Diversity and succession of the intestinal bacterial community of the maturing broiler chicken. Appl Environ Microbiol 69, 68166824. Meade, K.G., Higgs, R., Lloyd, A.T., Giles, S., O'Farrelly, C., 2009a. Differential antimicrobial peptide gene expression patterns during early chicken embryological development. Developmental and comparative immunology 33, 516-524. Meade, K.G., Narciandi, F., Cahalane, S., Reiman, C., Allan, B., O'Farrelly, C., 2009b. Comparative in vivo infection models yield insights on early host immune response to Campylobacter in chickens. Immunogenetics 61, 101-110. Milona, P., Townes, C.L., Bevan, R.M., Hall, J., 2007. The chicken host peptides, gallinacins 4, 7, and 9 have antimicrobial activity against Salmonella serovars. Biochem Biophys Res Commun 356, 169174. Ramasundara, M., Leach, S.T., Lemberg, D.A., Day, A.S., 2009. Defensins and inflammation: the role of defensins in inflammatory bowel disease. Journal of gastroenterology and hepatology 24, 202208. Salzman, N.H., Bevins, C.L., 2013. Dysbiosis--a consequence of Paneth cell dysfunction. Seminars in immunology 25, 334-341. Soman, S.S., Nair, S., Issac, A., Arathy, D.S., Niyas, K.P., Anoop, M., Sreekumar, E., 2009. Immunomodulation by duck defensin, Apl_AvBD2: in vitro dendritic cell immunoreceptor (DCIR) mRNA suppression, and B- and T-lymphocyte chemotaxis. Mol Immunol 46, 3070-3075. Tagai, C., Morita, S., Shiraishi, T., Miyaji, K., Iwamuro, S., 2011. Antimicrobial properties of arginineand lysine-rich histones and involvement of bacterial outer membrane protease T in their differential mode of actions. Peptides 32, 2003-2009. Townes, C.L., Michailidis, G., Nile, C.J., Hall, J., 2004. Induction of cationic chicken liver-expressed antimicrobial peptide 2 in response to Salmonella enterica infection. Infect Immun 72, 6987-6993. Ye, X., Avendano, S., Dekkers, J.C., Lamont, S.J., 2006. Association of twelve immune-related genes with performance of three broiler lines in two different hygiene environments. Poult Sci 85, 15551569.

13

Figure Legends

Figure 1: Bacterial Counts (CFU) /g Bedding: Data represents single bedding sample taken at days 7, 14 and 35 respectively from pens in the two different environments housing the two lines of birds. XLH: Line X, low hygiene; YLH: Line Y, low hygiene; XHH: Line X, high hygiene; YHH: Line Y, high hygiene.

14

Figure 2: Bird Gut Antimicrobial Activity: Bacterial S. enterica serovar Typhimurium phoP survival following a time-kill antimicrobial assay using proximal gut extracts from day 0 (hatch), 7, 14 and 35 birds. A: XLH and B: YLH antimicrobial data for individual birds analysed; mean data presented as horizontal lines and C: collated data (mean±SEM):XLH; :YLH. D: XHH and E: YHH antimicrobial data for individual birds analysed; mean data presented as horizontal lines and F: collated data (mean±SEM) :XHH; :YHH *** p<0.001.

15

Figure 3: Bird Proximal Gut Proteome Analysis: Putative cellular locations of the 143 proteins identified in a proximal gut mucosal scrape of the Line X day 7HH birds (A) and the 197 proteins identified in a comparable gut mucosal scrape of the Line X day 0 birds (B).

16

Figure 4: AvBD Gut Tissue Expression: A: End-point PCR analyses of Avian β-defensin (AvBD)1-10 and GAPDH genes in Line X day 0 (Hatch) and day 7LH bird gut tissues. B: Three single nucleotide polymorphisms (SNPs) of AvBD1 gene nucleotide sequences and the corresponding translated amino acids are shown. The AvBD1 peptides most prevalent for each line are designated ‘NYH’ and ‘SSY’; Table indicates the prevalence of each SNP form in each bird line. Percentage values were calculated using data from the Aviagen Ltd. SNP study performed by Illumina (San Diego, U.S.A) in combination with sequencing of pooled bird DNA (n = 120).

17

Figure 5: RT-QPCR analyses of AvBD1, 4 and 10 gene expression in duodenal (A,C, E) and caecal (B,D, F) tissues of Line X and Y birds raised in either LH or HH environments. X0: Line X hatch birds; Y0: Line Y, hatch birds; X7LH: day 7, Line X birds, low hygiene; Y7HH: day 7, Line Y, high hygiene; X35LH: day 35, Line X, low hygiene; Y35HH: day 35, Line Y, high hygiene. Data (AU) presented as median± inter quartile range. * group outliers. LX 0 and day 7LH duodenal AvBD1 differed significantly (p<0.05) from all other groups; LX 0 and day 7LH caecal AvBD10 differed significantly (p<0.001) from all other groups.

18

Table Legends Table 1: End-point PCR (SQ) and QPCR (Q) primers. AvBD1-10 primers utilised for semi-quantitative Gene Forward primer Reverse primer Annealing Lengt temperatu h re (oC) (bp) AvBD1SQ/Q TACCTCTGCTGCAAAAGAATA GAGAAGCCAGGGTGATGT 60 70 (NM_204993.1) TGG CC AvBD2SQ TGCTGCAAATGGCCTTGGAA CTTCTTGCTGCTGAGGCTTT 63 113 (NM_204992) T G AvBD3SQ CTGCTGTGGAAGAGCATATG CTTCCACTGCCACGGTCAT 61 142 (NM_204650.2) AGGT AC AvBD4SQ/Q TGCTGTAGATGGTTGTAGTG ACCGGTACAATGGTTCCCC (NM_00100161 61 100 TGAA A 0.2) AvBD5SQ GCAAGAAAGGAACCTGCC (NM_00100160 GCAAGAAAGGAACCTGCCCT 64 136 CT 8.2) AvBD6SQ TCTTGCTGTGTGAGGAACAG TTAGAGTGCCAGAGAGGC (NM_00100119 61 95 G CA 3) AvBD7SQ GGAGTGCCAGAGAAGCCA (NM_00100119 CTCTTGCTGTGCAAGGGGAT 59 91 TT 4.1) AvBD8SQ TGCCGGACTGTGTACGACTA TTCAGCCCCAAATTCCAGG (NM_00100178 58 112 A TT 1.1) AvBD9SQ GCAAAGGCTATTCCACAGCA CTTCTTGGCTGTAAGCTGG (NM_00100161 62 103 GA AGCA 1.2) AvBD10SQ/Q CTGTTAAACTGCTGTGCCAA TGTTGCTGGTACAAGGGCA (NM_00100160 58 77 GATTC AT 9.1) (SQ) and quantitative analyses (Q), optimal annealing conditions and product size

19

Table 2: RT-QPR analysis of AvBD 10 gene expression in liver, kidney and testes tissues of Line X and Y birds raised in either LH or HH environments. X0: Line X hatch birds; Y0: Line Y, hatch birds; X7LH: day 7, Line X birds, low hygiene; Y7HH: day 7, Line Y, high hygiene; X35LH: day 35, Line X, low hygiene; Y35HH: day 35, Line Y, high hygiene. Data (AU) presented as mean±SEM.

X0

X7LH 247(78)

X7HH

X35LH

5202 (2882)

902(582)

Liver

205 (147)

Kidney

4157(2526) 6836(3653) 23287(9532) 132193 (41029)

Testes

791(747)

2385(1398) 8754(6322)

X35HH 3276 (2013)

Y0 777(524)

Y7LH 567(168)

Y7HH 10003 (302)

99288(56550) 12227(7571) 15067(6583) 43848 (33273)

1244(566) 440(80)

20

423(389)

1033(539)

Y35LH

Y35HH

591(181)

817(383)

31907 (14765)

33556 (14266)

146935(93551) 2428(1099) 1889(1570)