Veterinary Immunology and Immunopathology 161 (2014) 32–41

Contents lists available at ScienceDirect

Veterinary Immunology and Immunopathology journal homepage: www.elsevier.com/locate/vetimm

Research paper

Association between nucleotide oligomerisation domain two (Nod2) gene polymorphisms and canine inflammatory bowel disease A. Kathrani a , H. Lee a , C. White a , B. Catchpole b , A. Murphy c , A. German d , D. Werling b , K. Allenspach a,∗ a b c d

Department of Veterinary Clinical Sciences and Services, Royal Veterinary College, University of London, Hatfield, Hertfordshire, UK Department of Pathology and Infectious Diseases, Royal Veterinary College, University of London, Hatfield, Hertfordshire, UK Department of Veterinary Basic Sciences, Royal Veterinary College, University of London, Hatfield, Hertfordshire, UK Department of Clinical Sciences, School of Veterinary Science, University of Liverpool, Liverpool, UK

a r t i c l e

i n f o

Article history: Received 19 April 2014 Received in revised form 6 June 2014 Accepted 19 June 2014 Keywords: Canine Inflammatory bowel disease Nucleotide oligomerisation domain Polymorphisms

a b s t r a c t The most important genetic associations that have been implicated to play a role in the etiology of Crohn’s disease (CD) in humans are single nucleotide polymorphisms (SNPs) in nucleotide oligomerisation domain 2 (NOD2). The aim of this study was to investigate whether SNPs in the canine NOD2 gene are associated with inflammatory bowel disease (IBD) in German shepherd dogs (GSDs) and other canine breeds. A mutational analysis of the NOD2 gene was carried out in 10 randomly selected GSDs with IBD. The mutational analysis identified five non-synonymous SNPS, of which four in exon 3 of the NOD2 gene were evaluated in a case-control study using sequence based typing. Sequencing information from 55 GSDs with IBD were compared to a control group consisting of 61 GSDs. In addition, 85 dogs of other breeds with IBD and a breed-matched control group consisting of 162 dogs were also genotyped. All four SNPs were in complete linkage and, in the GSD population, were found to be in Hardy–Weinberg equilibrium. When the GSD case population was compared to the GSD control group, the heterozygote genotype for all four SNPs was more frequently found in the IBD population (p = 0.03, OR = 2.30, CI = 1.07–4.94). However, these results were not mirrored in other canine breeds. Our study suggests that the four SNPs in exon 3 of NOD2 are significantly associated with IBD in GSDs when analyzed in an over-dominant model. However, these results were not mirrored in other canine breeds with IBD. This suggests that the etiology of this disease is complex and may involve the interaction of SNPs present in several genes or pathways to bring about the inflammatory changes seen in the intestine. © 2014 Elsevier B.V. All rights reserved.

1. Introduction

∗ Corresponding author. E-mail address: [email protected] (K. Allenspach). http://dx.doi.org/10.1016/j.vetimm.2014.06.003 0165-2427/© 2014 Elsevier B.V. All rights reserved.

The healthy intestine is able to regulate inflammatory responses to commensal bacteria and food antigens whilst maintaining the capacity to respond to pathogens (Abreu et al., 2005; Magalhaes et al., 2007; RakoffNahoum et al., 2004). This distinction relies on pattern

A. Kathrani et al. / Veterinary Immunology and Immunopathology 161 (2014) 32–41

recognition receptors (PRRs) such as Toll-like receptors (TLRs) and Nucleotide oligomerisation domain (NOD) receptors; which are ideally situated in intestinal epithelia cells and recognize microbe associated molecular patterns (MAMPs) (Akira et al., 2006). NOD2 recognizes muramyl dipeptide (MDP) from peptidoglycan present in bacteria. A breakdown in this homeostatic response could lead to inflammatory bowel disease (IBD) in people as well as in dogs (Abreu et al., 2005). Although the exact etiology of IBD is unknown it is considered a multifactorial disease with the mucosal immune system, microbiota, environment and genetics all playing a role (Blumberg et al., 1999; Fritz et al., 2008; Hendrickson and Gokhale, 2002; Xavier and Podolsky, 2007). In people a number of genetic polymorphisms have been reported to play a role in the pathogenesis of IBD. Polymorphisms in NOD2 are considered to be the most important genetic susceptibility factor for Crohn’s disease (CD), a form of human IBD. Thirty non-conservative single nucleotide polymorphisms (SNPs) have been identified in this gene, although 3 SNPs; Arg702Trp, Gly908Arg and Leu1007fsinsC account for approximately 82% of the mutated alleles (Hugot et al., 2001; Ogura et al., 2001). Heterozygous carriage of the risk alleles confers a 2–4 fold increased risk and homozygotes or compound heterozygotes have a 20–40 fold increased risk (Cuthbert et al., 2002) of developing CD. NOD2 is upregulated in a canine colonic epithelial cell line following various infectious stimuli (Swerdlow et al., 2006), suggesting that it could be involved in canine IBD; however, to date, no genetic studies of NOD2 in canine IBD have been reported. We have demonstrated a significant association between SNPs in TLR4 and TLR5 and IBD in the German shepherd dog (GSD) breed (Kathrani et al., 2010). In addition, SNPs in TLR5 were also significantly associated with IBD in other canine breed (Kathrani et al., 2011). Similarly, studies have highlighted the association of SNPs in TLR2, 4 and 5 with various phenotypes of human IBD (Browning et al., 2007; Franchimont et al., 2004; Gewirtz et al., 2006; Pierik et al., 2006; Torok et al., 2004). As there is evidence of overlap between susceptibility genes in both human and canine IBD, it would be valuable to investigate the role of NOD2 in canine IBD given its major role in the human disease. The aim of this study was to carry out a mutational analysis of NOD2 in ten GSDs with IBD to identify non-synonymous SNPs present. The identified non-synonymous SNPs were tested further in a casecontrol population consisting of healthy GSDs and those diagnosed with IBD. This was replicated in an IBD casecontrol population consisting of 38 canine breeds other than GSDs.

2. Materials and methods 2.1. Ethics statement Recruitment of healthy German shepherd dogs was approved from the Ethics Committee at the Royal Veterinary College (URN 2010 1035).

33

2.2. Sample population The two case populations included 55 GSDs; (38 were diagnosed with inflammatory bowel disease (IBD) at the Royal Veterinary College (RVC), UK and 17 at the Small Animal Teaching Hospital, University of Liverpool, UK) and 85 dogs of 38 other breeds (diagnosed with IBD at the RVC). For all cases, the diagnostic work-up was reviewed. All cases had complete medical records consisting of a detailed history at first examination and at subsequent re-examinations, hematology and biochemistry results, detailed abdominal ultrasonography report and detailed histopathology reports. All known causes of gastrointestinal inflammation including enteropathogenic bacteria were ruled out by routine hematology and biochemistry, fecal analysis, ultrasound and measurement of serum trypsin-like immunoreactivity concentration. Definitive diagnosis of IBD was based on histological evidence of an inflammatory infiltrate within the lamina propria of endoscopically-obtained intestinal biopsies. The two control populations consisted of 47 unrelated GSDs and 162 dogs of other breeds. These dogs had visited the hospital for non-inflammatory, non-immune-mediated or non-infectious illnesses. They were retrospectively phenotyped, either by owner or veterinary surgeon telephone contact, to ensure they had not developed any inflammatory, immune-mediated or infectious disease. Fourteen additional GSDs were also included in the GSD control group; six of these were enrolled from the blood donor program at the RVC and eight were recruited exclusively for the study. Ethics approval was granted from the Royal Veterinary College, UK for the recruitment of these eight dogs. Blood stored in ethylenediaminetetraacetic acid (EDTA) was collected from the DNA archive for those cases that were seen at the RVC. For the cases diagnosed at Liverpool University, paraffin sections of intestinal tissue were available. Residual blood from the blood donor dogs and from the eight dogs following collection of blood for folate and cobalamin concentrations as a screening test for intestinal health were collected into EDTA tubes. 2.3. Mutational analysis of NOD2 exons A mutational analysis of the 12 exons of NOD2 was carried out in 10 GSDs diagnosed with IBD, to determine the presence and frequency of non-synonymous SNPs in the coding regions of the gene, so that these could be investigated further in a case-control association study to determine their significance in IBD in the GSD population and other canine breeds. Ten cases of GSDs with IBD were selected from the 38 diagnosed at the RVC and genomic DNA was extracted from blood stored in EDTA using the Qiagen DNeasy blood and tissue kit (Qiagen, Crawley, UK) according to the manufacturer’s protocol. Forward and reverse primers were designed for each of the 12 exons using the Integrated DNA technology oligoanalyzer 3.1 program (http://eu.idtdna.com/analyzer/ Applications/OligoAnalyzer) (Table 1) based on the full sequences for canine NOD2.

34

A. Kathrani et al. / Veterinary Immunology and Immunopathology 161 (2014) 32–41

Table 1 Primers used in PCR for full-length amplification of NOD2 gene for mutational analysis in IBD. Primer

Exon

NOD2

NOD2 Exon 1 NOD2 Exon 2 NOD2 Exon 3

NOD2 Exon 4 and 5 NOD2 Exon 6 and 7 NOD2 Exon 8 and 9 NOD2 Exon 10 NOD2 Exon 11 NOD2 Exon 12

Primer pos. 67,529,505 67,530,078 67,537,327 67,537,671 67,539,104 67,540,934 67,540,911 67,541,726 67,541,251 67,541,856 67,545,884 67,546,466 67,548,567 67,548,986 67,550,512 67,551,507 67,552,863 67,555,219 67,556,486 67,556,828 67,558,017 67,558,505

Direction

Primer sequence

Forward Reverse Forward Reverse Frag 1

5 -TGC ACT CAT CTA CCT CCC-3 5 -GTC TTG GTG ACT AAG CAC-3 5 -AGG TAG GTG TGT ACC ACG-3 5 -CTG GAG TTT TGC TCT TCA C-3 5 -TGC TCA CAT CCT ACC TC-3 5 -CAA CAA CAG CTC CTG GTG-5 5 -TCT CAT GGA TGG TGT CC AAA-3 5 -GCA GAA CGT CAG CTT CAG GT-3 5 -CAG ATC ACG GCA GCC TT-3 5 -GGA CAC TCA ATA AAC CTG GC-3 5 -GTC CTT GAG GTG TTT TCC-3 5 -AGT CCA GTA GAG GCC AA-3 5 -TCT GGG ACA CAC TTT CC-3 5 -CCC AAA TAG TTC CGA GG-3 5 -GGG CAC TGT TAA CAC ATG-3 5 -CAA AGG CCA ACC GCT CA-3 5 -TTC TAA CTG GGG GTG CAA G-3 5 -CTG ACA CTA CTG CCA AAG T-3 5 -CTT AAG GCC TTG AGC CTT-3 5 -GAA TGC TAC AGA ATG TCA GG-3 5 -TTT TCT CTG CAG AGT CAC GG-3 5 -CAA AGA TGA CCC ACT GTG C-3

Frag 2 Frag 3 Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse

For Rev For Rev For Rev

Primers used in PCR for full-length amplification of NOD2 gene for mutational analysis of ten GSDs with IBD to determine number and nature of nonsynonymous SNPs. Primer pos. Primer position number within chromosome 2. NOD2 mRNA NCBI Reference Sequence XM 544412.

Easy A 2× master mix (Stratagene, California, USA) was used for polymerase chain reaction (PCR) with reactions heated to 95 ◦ C for 10 min, followed by 30 cycles of 95 ◦ C for 1 min, 52 ◦ C for 1 min and 72 ◦ C for 3 min then a further 72 ◦ C for 7 min. PCR products were electrophoresed with 6X orange loading dye (Fermentas, UK) on a 1% agarose/safeview nucleic stain (NBS Biologicals, Huntington, UK) gel in 1X Tris–Borate EDTA (TBE) buffer using the power pack FB 300 (Fischer Scientific, UK). The gels were visualized under UV light (Ultraviolet Transilluminator, UVP, California, USA), and the bands cut out and placed into separate eppendorf tubes. DNA was extracted using the Qiagen mini-elute extraction kit (Qiagen, Crawley, UK), according to manufacturer’s instructions. Gel purified PCR products were sent to Geneservice, Cambridge, UK for sequencing using sequencing primers specifically designed to cover the entire sequence of the exons (Table 1). Sequence data was compared to the canine genome (www.ensembl.org/Canis familiaris) and CLC DNA Workbench version 5.1 (CLC Bio) used to identify nonsynonymous SNPs present in the genes. 2.4. Determining the allele frequency of NOD2 SNPs in exon 3 in healthy GSDs and those affected with IBD using Sanger sequencing Non-synonymous SNPs identified by mutational analysis in the NOD2 exon gene were analyzed further in a case population consisting of 55 GSDs with IBD and a control population consisting of 61 GSDs; 47 with noninflammatory disease and 14 healthy dogs. Genomic DNA was extracted from blood stored in EDTA from the remainder of the 28 IBD cases seen at the RVC and

from the 61 control GSDs using the Qiagen DNeasy blood and tissue kit (Qiagen, Crawley, UK) according to the manufacturer’s protocol. This kit along with the manufacturer’s recommendations was also used to extract genomic DNA from the paraffin sections from the 17 cases diagnosed with IBD at the University of Liverpool. PCR was carried out on all samples using primers to amplify NOD2 exon 3 Fragment 2 (Table 1). Immolase DNA polymerase (Bioline) was used for PCR with reactions heated to 95 ◦ C for 10 min, followed by 35 cycles of 95 ◦ C for 1 min, 55 ◦ C for 1 min and 72 ◦ C for 2 min, then a further 72 ◦ C for 7 min. Five micro-liters of PCR product from each sample was added to 2 ␮l of ExoSAP-IT (USB), and mixed thoroughly and incubated for an hour at 37 ◦ C and then at 75 ◦ C for 15 min. Five microliters of diluted PCR product was used as the template for sequencing reactions, using Terminator ready reaction mix (2 ␮l), buffer (2 ␮l) and 1 ␮l of 3.2 pmol sequencing primer. Sequencing reactions were performed using an M J Research Peltier Thermal cycler-2000 (Bio-Rad Laboratories) in a 96-well plate format. Samples were incubated at 96 ◦ C for 10 s, 50 ◦ C for 4 s and 60 ◦ C for 4 min for 25 cycles. Following the sequencing reaction, DNA was purified by ethanol/sodium acetate precipitation as follows; 5 ␮l of DNA, 1 ␮l of sodium acetate and 25 ␮l of 100% ethanol were mixed by pippetting and incubated at room temperature for 15 min. Following incubation, plates were centrifuged at 2500 × g for 30 min. Plates were tapped dry and an additional 125 ␮l of 70% ethanol added and centrifuged at 2500 × g for 10 min. To remove all residual ethanol, plates were tapped dry and incubated at 37 ◦ C for 15 min. DNA was resuspended in Hi–Di Formamide (Applied Biosystems) then denatured by heating to 95 ◦ C for 3 min following by rapidly cooling on ice. Nucleotide sequences were determined on an ABI PRISM 3100 Genetic Analyser (Applied Biosystems).

A. Kathrani et al. / Veterinary Immunology and Immunopathology 161 (2014) 32–41

35

Table 2 Signalment of cases and controls. Controls Counts Sex Age

Male Female 1–4 years 5–9 years >10 years

34 27 15 36 10

Cases Proportions 0.56 0.44 0.25 0.59 0.16

Counts 30 25 24 21 10

p-Value Proportions 0.55 0.45 0.44 0.38 0.18

0.54 0.0017

Signalment of 55 German shepherd dog (GSD) cases and 61 GSD controls included in the NOD2 association study of inflammatory bowel disease.

2.5. Determining the allele frequency of NOD2 SNPs in exon 3 in healthy dogs of other breeds and those affected with IBD using Sanger sequencing Non-synonymous SNPs identified by mutational analysis in the NOD2 exon gene were analyzed further in a case population consisting of 85 dogs of 38 different breeds with IBD and a control population consisting of 162 breedmatched controls. Genomic DNA was extracted and used in PCR for sequencing using the same methods as described above.

2.6. Statistical analysis Allele frequency, association and linkage disequilibrium data for all the SNPs were statistically assessed using Haploview software package (version 4.1; http://www. broad.mit.edu/personal/jcbarret/haplo/). The calculation of Hardy–Weinberg equilibrium and further genotype and haplotype associations for all the SNPs were carried out using SNPSTATS program (http://bioinfo.iconcologia.net/ index.php?module=Snpstats). Statistical significance was set at p < 0·05.

3.2. Case-control association study of NOD2 SNPs in German shepherd dogs with IBD An association study was carried out to assess the significance of the non-synonymous SNPs identified in exon three of NOD2 (A1537G, T1578C, C1693G and G1885A) in GSDs with IBD. The case group consisted of 55 unrelated GSDs; 38 diagnosed with IBD at the Royal Veterinary College (RVC), UK and 17 diagnosed with IBD at the Small Animal Teaching Hospital, University of Liverpool (Table 2). The control population consisted of 61 GSDs, of which 47 dogs were diagnosed with non-inflammatory disease at the RVC, UK and 14 healthy GSDs. The most common diagnosis for the group of GSDs diagnosed with non-inflammatory disease was neoplasia. Nineteen dogs fell into this latter category, 18 had neurological disease, three had musculoskeletal disease, two had gastrointestinal disease and there was one dog in the following etiological categories; cardiovascular, dermatological and urogenital. Two dogs had other diseases that did not fit the above etiological categories: one had suffered a road traffic accident, whilst the other had post-ovariohysterectomy hemorrhage (Table 3).

Table 3 Disease phenotype of control dogs. Cases

3. Results

Neurological

3.1. Mutational analysis of NOD2 A mutational analysis of the NOD2 gene in 10 GSDs with IBD was carried out to determine the number and nature of non-synonymous single nucleotide polymorphisms (SNPs) present. This analysis revealed four non-synonymous SNPs; A1537G (ENSCAFT00000015589), T1578C (ENSCAFT00000015589), C1693G (ENSCAFT 00000015589), G1885A (ENSCAFT00000015589) in exon three and one non-synonymous SNP; G2569A (ENSCAFT00000015589) in exon six of the NOD2 gene. Six out of 10 GSDs with IBD carried SNPs in the heterozygous genotype at A1537G, T1578C, C1693G, G1885A and none carried them in the homozygous genotype. Nine out of 10 GSDs carried the SNP G2569A in the homozygous genotype and one case carried it in the heterozygous genotype. Only one of the non-synonymous SNPs identified in NOD2, T1578C resulted in a change in the class of amino acid coded from tryptophan (non-polar) to arginine (basic).

Musculo-skeletal Gastro-intestinal Neoplasia

Cardiovascular Dermatological Uro-genital Other Healthy

Degenerative myelopathy Idiopathic epilepsy Intervertebral disc disease Fibro-cartilaginous embolism Hip degenerative joint disease Idiopathic generalized osteopenia Pharyngeal cyst Idiopathic megaoesophagus Sarcoma Hemangiosarcoma Leukemia Lymphoma Insulinoma Nerve sheath tumor Glial cell tumor Idiopathic pericardial effusion Peri-vascular dermatitis Protein losing nephropathy Road traffic accident Bleeding post-spay Blood donor Recruited for study

6 4 4 4 2 1 1 1 5 4 3 3 3 1 1 1 1 1 1 1 6 8

Disease phenotype of 61 German shepherd dogs (GSDs) included in the control group in the NOD2 association study of IBD using Sanger sequencing method (case = 55 unrelated GSDs, control = 61 unrelated GSDs; 47 GSDs with non-inflammatory disease and 14 healthy GSDs).

36

A. Kathrani et al. / Veterinary Immunology and Immunopathology 161 (2014) 32–41

Table 4 NOD2 single nucleotide polymorphism allele association with inflammatory bowel disease in German shepherd dogs. SNP

Associated allele

Minor allele freq.

p-Value

NOD2 A1537G NOD2 T1578 NOD2 C1693G NOD2 G1885A

A T C G

0.265 0.265 0.265 0.265

0.585 0.585 0.585 0.585

Association of NOD2 single nucleotide polymorphism (SNP) alleles in a case-control association study of inflammatory bowel disease in German shepherd dogs (GSDs) carried out using Sanger sequencing method (case = 55 unrelated GSDs, control = 61 GSDs; 47 GSDs with noninflammatory disease and 14 healthy GSDs).

The minor allele frequency for the SNPs was >10% (Table 4). The presence of Hardy–Weinberg equilibrium was tested using the exact Hardy–Weinberg test (Guo and Thompson, 1992) in the SNPSTATS program (Sole et al., 2006) (http://bioinfo.iconcologia.net/index.php?module= Snpstats). All NOD2 SNP alleles were in Hardy–Weinberg equilibrium (p > 0.05). The significance of the SNP alleles was determined using the Haploview software package (version 4.1; http://www.broad.mit.edu/personal/jcbarret/haplo/). None of the SNP alleles were found to be significantly associated with IBD in GSDs (Table 4). All four of the NOD2 SNPs were in complete linkage disequilibrium (Fig. 1) and were found to be significant when the genotypes were analyzed in an overdominant model (p = 0.03, OR = 2.30, CI = 1.07–4.94) (Tables 5–7). 3.3. Case-control association study of NOD2 SNPs in other canine breeds with IBD An association study was performed to assess the significance of the non-synonymous SNPs present in exon

three of NOD2 (A1537G, T1578C, C1693G and G1885A) in other canine breeds with IBD. The case group consisted of 85 unrelated non-GSDs diagnosed with IBD at the RVC, UK (Table 5). The control population consisted of 162 breed-matched controls diagnosed with non-inflammatory disease at the RVC, the phenotypes of which are presented in Table 8. The minor allele frequency for all the SNPs were >10% (Table 9) The presence of Hardy–Weinberg equilibrium was tested using the exact Hardy–Weinberg test (Guo and Thompson, 1992) using the SNPSTATS programme (Sole et al., 2006) (http://bioinfo.iconcologia.net/ index.php?module=Snpstats). None of the SNPs were found to be in Hardy Weinberg equilibrium (p < 0.001). The significance of the SNP alleles was determined using the Haploview software package (version 4.1; http://www. broad.mit.edu/personal/jcbarret/haplo/). None of the SNP alleles were found to be significantly associated with IBD in the breeds included in this study (p > 0.05) (Table 9). In addition, none of the genotypes were found to be significantly associated with IBD in any of the models tested (co-dominant, dominant, recessive, over-dominant and additive) (p > 0.05). 4. Discussion Single nucleotide polymorphisms in NOD2 have been shown to be the most influential susceptibility gene in CDs (Cho, 2001; Hugot et al., 2001). No genetic studies of NOD2 have been undertaken in canine IBD. In the current study, we identified four non-synonymous SNPs in exon 3 of the NOD2 gene to be significantly associated with IBD when their genotype forms were analyzed in an overdominant model in the GSD breed. However, in other canine breeds, the NOD2 SNPs were not significantly associated with IBD when their genotypes

Fig. 1. Linkage disequilibrium of single nucleotide polymorphisms in canine NOD2. Linkage disequilibrium plot of the four non-synonymous NOD2 SNPs in the GSD and other breed populations obtained from a case-control association study of IBD.

A. Kathrani et al. / Veterinary Immunology and Immunopathology 161 (2014) 32–41

37

Table 5 NOD2 A1537G, T1578C, C1693G and G1885A genotype association with inflammatory bowel disease in German shepherd dogs. NOD2 A1537G SNP association with IBD Model

Genotype

Control

Case

OR (95% CI)

p-Value

AIC

BIC

Codominant

37 (60.7%) 18 (29.5%) 6 (9.8%) 37 (60.7%) 24 (39.3%) 55 (90.2%) 6 (9.8%) 43 (70.5%) 18 (29.5%) –

26 (47.3%) 27(49.1%) 2 (3.6%) 26 (47.3%) 29 (52.7%) 53 (96.4%) 2 (3.6%) 28 (50.9%) 27 (49.1%) –

1.00 2.13 (0.98–4.65) 0.47 (0.09–2.54) 1.00 1.72 (0.82–3.60) 1.00 0.35 (0.07–1.79) 1.00 2.30 (1.07–4.94) 1.20 (0.67–2.17)

0.063

161

169.2

0.15

162.4

167.9

0.18

162.7

168.2

0.03

159.8

165.3

Log-additive

G/A A/G A/A G/G A/G–A/A G/G–A/G A/A G/G–A/A A/G –

0.53

164.1

169.6

Model

Genotype

Control

Case

OR (95% CI)

p-Value

AIC

BIC

Codominant

37 (60.7%) 18 (29.5%) 6 (9.8%) 37 (60.7%) 24 (39.3%) 55 (90.2%) 6 (9.8%) 43 (70.5%) 18 (29.5%) –

26 (47.3%) 27 (49.1%) 2 (3.6%) 26 (47.3%) 29 (52.7%) 53(96.4%) 2 (3.6%) 28 (50.9%) 27 (49.1%) –

1.00 2.13 (0.98–4.65) 0.47 (0.09–2.54) 1.00 1.72 (0.82–3.60) 1.00 0.35 (0.07–1.79) 1.00 2.30 (1.07–4.94) 1.20 (0.67–2.17)

0.063

161

169.2

0.15

162.4

167.9

0.18

162.7

168.2

0.03

159.8

165.3

Log-additive

C/C C/T T/T C/C C/T–T/T C/C–C/T T/T C/C–T/T C/T –

0.53

164.1

169.6

Model

Genotype

Control

Case

OR (95% CI)

p-Value

AIC

BIC

Codominant

G/G C/G C/C G/G C/G–C/C G/G–C/G C/C G/G–C/C C/G –

37 (60.7%) 18 (29.5%) 6 (9.8%) 37 (60.7%) 24 (39.3%) 55 (90.2%) 6 (9.8%) 43 (70.5%) 18 (29.5%) –

26 (47.3%) 27 (49.1%) 2 (3.6%) 26 (47.3%) 29 (52.7%) 53 (96.4%) 2 (3.6%) 28 (50.9%) 27 (49.1%) –

1.00 2.13 (0.98–4.65) 0.47 (0.09–2.54) 1.00 1.72 (0.82–3.60) 1.00 0.35 (0.07–1.79) 1.00 2.30 (1.07–4.94) 1.20 (0.67–2.17)

0.063

161

169.2

0.15

162.4

167.9

0.18

162.7

168.2

0.03

159.8

165.3

0.53

164.1

169.6

OR (95% CI)

p-Value

AIC

BIC

1.00 2.13 (0.98–4.65) 0.47 (0.09–2.54) 1.00 1.72 (0.82–3.60) 1.00 0.35 (0.07–1.79) 1.00 2.30 (1.07–4.94) 1.20 (0.67–2.17)

0.063

161

169.2

0.15

162.4

167.9

0.18

162.7

168.2

0.03

159.8

165.3

0.53

164.1

169.6

Dominant Recessive Overdominant

NOD2 T1578C SNP association with IBD

Dominant Recessive Overdominant

NOD2 C1693G SNP association with IBD

Dominant Recessive Overdominant Log-additive

NOD2 G1885A SNP association with IBD Model

Genotype

Control

Codominant

A/A A/G G/G A/A A/G–A/A A/A–A/G G/G A/A–G/G A/G –

37 (60.7%) 18 (29.5%) 6 (9.8%) 37 (60.7%) 24 (39.3%) 55 (90.2%) 6 (9.8%) 43 (70.5%) 18 (29.5%) –

Dominant Recessive Overdominant Log-additive

Case 26 (47.3%) 27 (49.1%) 2 (3.6%) 26 (47.3%) 29 (52.7%) 53(96.4%) 2 (3.6%) 28 (50.9%) 27 (49.1%) –

Frequency and significance of the NOD2 A1537G, T1578C, C1693G and G1885A single nucleotide polymorphism genotypes in a case-control association study in German shepherd dogs with inflammatory bowel disease using Sanger sequencing method (case = 55 unrelated, control = 61 GSDs; 47 GSDs with non-inflammatory disease and 14 healthy GSDs (OR-Odds ratio, CI-Confidence interval, AIC—Akaike information criteria).

where analyzed in any genetic model. Dog breeds have developed into closed breeding populations representing isolated genetic pools as a result of differential selection for traits important in behavior and appearance (Ostrander and Wayne, 2005; Parker et al., 2004). This may explain why the four NOD2 SNP alleles were not in Hardy–Weinberg equilibrium in other breeds included in this study (Short et al., 2007). In addition, this may also explain why no significant association was found between

the NOD2 polymorphisms and IBD in breeds other than the GSD. NOD2 polymorphisms in people are absent in some ethnicities such as Asians (Yamazaki et al., 2002; Li et al., 2008) and the same may also be true in some canine breeds as these SNPs may have developed after the formation of dog breeds. Therefore, the potential absence of NOD2 SNPs in some canine breeds may have diluted the true association of these SNPs with IBD. It may be beneficial to investigate these SNPs separately in the different canine

38

A. Kathrani et al. / Veterinary Immunology and Immunopathology 161 (2014) 32–41

Table 6 Breed distribution of IBD cases and controls. Breed

Number of IBD cases

Cavalier King Charles spaniel Giant Schnauzer Boxer English bull terrier Staffordshire bull terrier Golden retriever Rottweiler Hungarian Vizla Weinmaraner Cocker spaniel Yorkshire terrier Old English sheepdog Standard poodle Bearded collie Bichon Frise Border Collie Miniature poodle West Highland white terrier Pomeranian Jack Russell terrier English springer spaniel Great Dane Labrador Rhodesian ridgeback Italian spinone Border terrier Cairn terrier Bassett hound Corgi German short haired pointer Irish setter Welsh springer spaniel Scottish West Highland terrier Doberman Greyhound English setter Miniature schnauzer Kerry Blue terrier Total

Table 8 Disease phenotype of controls. Number of controls

2 1 6 6 3 6 4 4 2 2 2 1 2 2 1 3 1 3 1 2 3 1 9 1 1 1 1 2 1 3 1 1 1 2 2 1 1 1

3 2 6 6 4 9 6 6 4 4 6 1 5 5 4 7 3 7 3 4 5 3 18 2 3 3 2 4 2 6 2 2 3 3 4 2 2 2

85

162

Cases Neurological

Musculo-skeletal

Gastro-intestinal

Neoplasia

Cardiovascular

Respiratory

Uro-genital

Number of IBD cases and controls in each breed class in case-control association study of NOD2 SNPs in canine IBD.

breeds rather than in a collective population. In addition, only a relatively small number of cases with IBD from each of the different 38 breeds was included in this study. As the effect of the SNPs in NOD2 were modest in the GSD population and could have a similar penetrance in other canine breeds, it is possible that a larger sample size would be able to detect a significant association in the other breeds. The potential complex polygenic nature of canine IBD may also explain the absence of a significant association

Other

3 Brain tumor 16 Idiopathic epilepsy 11 Intervertebral disc disease 4 Fibro-cartilaginous embolism 3 Vestibular disease Spinal fracture 1 Hip dysplasia 1 Degenerative joint disease 3 Elbow dysplasia 3 Pelvic fracture 2 Gastric foreign body 1 Idiopathic megaoesophagus 1 Esophageal foreign body 3 Biliary mucocele 1 1 Hemangiopericytoma Osteosarcoma 3 2 Plasma cell tumor 2 Pulmonary adenocarcinoma 1 Multiple myeloma 8 Lymphoma Nasal adenocarcinoma 4 Disseminated histiocytic sarcoma 2 1 Chemodectoma 3 Mast cell tumor 1 Oral melanoma 3 Hepatocellular carcinoma 1 Salivary adenocarcinoma Prostatic carcinoma 2 Hemangiosarcoma 7 Phaeochromocytoma 1 1 Insulinoma 2 Idiopathic pericardial effusion Congenital patent ductus arteriosus 2 Mitral valve dysplasia 4 Pulmonic stenosis 2 Cardiomyopathy 10 7 Tracheal collapse Laryngeal paralysis 2 Nasal foreign body 1 Brachycephalic airway obstruction 1 2 Idiopathic renal haemorrhage Urethral sphincter mechanism incompetence2 1 Congenital ectopic ureter 1 Road traffic accident Bleeding post-spay 2 2 Behavioral problems 1 Diabetes insipidus 1 Hyperlipidaemia 1 Metaldehyde intoxication 22 Diagnosis open

Disease phenotype of 162 breed controls included in the control group in the NOD2 association study of IBD using the Sanger sequencing method.

Table 7 Signalment of cases and controls. Controls

Sex Age

Male Female 1–4 years 5–9 years >10 years

Cases

p-Value

Counts

Proportions

Counts

Proportions

102 60 36 71 55

0.63 0.37 0.22 0.44 0.34

52 33 30 39 16

0.61 0.39 0.35 0.46 0.19

Signalment of non-GSD breeds with IBD and controls included in the NOD2 association study of inflammatory bowel disease.

0.78 0.018

A. Kathrani et al. / Veterinary Immunology and Immunopathology 161 (2014) 32–41 Table 9 NOD2 single nucleotide polymorphism allele association with inflammatory bowel disease in other canine breeds. SNP

Associated allele

Minor allele freq.

p-Value

NOD2 A1537G NOD2 T1578C NOD2 C1693G NOD2 G1885A

A T C G

0.328 0.328 0.328 0.328

0.6004 0.6004 0.6004 0.6004

Association of NOD2 single nucleotide polymorphism (SNP) alleles in a case-control association study of inflammatory bowel disease in non-German shepherd dogs (GSDs) carried out using Sanger sequencing method (case = 85 unrelated non-GSD breeds, control = 162 breed matched controls).

between NOD2 polymorphisms and IBD in other breeds. The polygenic etiology of human IBD is demonstrated by the discovery of several loci associated with susceptibility in this disease (Bonen and Cho, 2003). Several studies have proven that interactions exist between genetic loci (Malmberg et al., 2005; Moore, 2005; Segre et al., 2005). Interestingly, gene–gene interactions have been shown, not only to enhance individual gene effects, but also to weaken them (Culverhouse et al., 2002; Moore, 2003). Studies have demonstrated the importance of gene–gene interactions in human IBD:, such as the additive effect reported between ATG16L1 and NOD2 (Roberts et al., 2007). These complex gene–gene interactions may be considered more important than the independent effects of single susceptibility loci and may explain the lack of replication of single-locus results seen in many human IBD genetic studies (Csongei et al., 2010). Similar gene–gene interactions may exist in canine IBD and may therefore explain the lack of replication of NOD2 SNPs seen in other breeds with IBD. Genome wide association studies in canine IBD will help to identify additional susceptibility loci implicated in the disease and further studies may help to highlight the complex interactions involved between genetic loci. Linkage disequilibrium (LD) in dogs is up to 100 times more extensive than in human beings (Sutter et al., 2004). It is, therefore, possible that the associations between SNPs in exon 3 of NOD2 and IBD are not the cause of the aberrant inflammation but are in LD with the causative gene. If this is the case, this may explain why no association was seen in other canine breeds with IBD. Functional analysis of these SNPs will help to confirm if NOD2 plays a significant role in the pathogenesis of IBD in GSDs. However, the role of NOD2 in the pathogenesis of human IBD and murine colitis has been reported in the literature. An association between SNPs in NOD2 and CD in people was the major breakthrough in helping to determine the complex pathogenesis of IBD in people. NOD2 is the major susceptibility gene found in CD and approximately 30-50% of patients carry variants in this gene (Hugot et al., 2001). In addition, the relative risk of developing CD is 3-fold for heterozygotes, 38-fold for homozygotes and 40-fold for compound heterozygotes (Hugot et al., 2001). Functional studies have confirmed that NOD2 plays a key role in the pathogenesis of human IBD. Human embryonic kidney cells transfected with mutated NOD2 containing the

39

1007fsCins failed to up-regulate NF-kB compared to wild type, confirming a loss-of function mutation (Hisamatsu et al., 2003). In contrast, knock-in mice models where wild-type NOD2 was replaced by the most common NOD2 polymorphism; 1007fsCins showed an increased NF-kB activation and increased secretion of IL-1␤ (Maeda et al., 2005), suggesting that NOD2 immuno-regulatory function is lost in-vivo. It has also been shown that NOD2 is required for the expression of cryptidins, which are intestinal antimicrobial peptides secreted by paneth cells in the crypts. (Kobayashi et al., 2005) Polymorphisms in NOD2 may, thus, directly affect the permeability of the intestinal barrier and thus predispose to CD. Recent studies have also shown that chronic NOD2 stimulation in human macrophages down-regulates proinflammatory cytokines on NOD2- or TLR-re-stimulation (Hedl and Abraham, 2010). This suggests that functional NOD2 receptors mediate cytokine down-regulation and is therefore required for intestinal homeostasis. In addition, one study demonstrated that the leucine-rich repeat domain of NOD2 is a direct antimicrobial agent and this activity is generally deficient in patients carrying polymorphisms in this gene (Perez et al., 2010). Although, in this study, the SNP alleles in NOD2 were not associated with IBD in GSDs, when analyzed in their genotype forms in an over-dominant model they were found to be significant. However, as this significance was modest (p = 0.03), repeating the methods in an independent sample would be needed to confirm these findings. In addition, the significant increase in heterozygotes in the case population may represent sampling of more dogs with this genotype rather than being directly associated with IBD. However, in cattle it has been shown that over-dominant SNPs in NOD2 are associated with Johne’s disease, a chronic intestinal disease caused by Mycobacterium avium subspecies paraturberculosis (Pinedo et al., 2009a, 2009b) and therefore our results may be a true relevant finding in GSDs with IBD. The SNPs identified in exon 3 of the NOD2 gene result in a change in the amino acid coded. Although when their location was mapped using the SMART web server none of them were found to be located within the leucine rich repeat domains. However, although studies in people with IBD have confirmed that the three most common polymorphisms in NOD2 are found in or around the leucine rich repeat region (Hugot et al., 2001), one study in pigs showed that SNPs outside this region are also functionally significant (Jozaki et al., 2009). Therefore the association of the 4 SNPs in NOD2 in IBD in the GSD breed may also have functional significance. Mutational analysis revealed a non-synonymous SNP in exon 6 of NOD2 in GSDs with IBD that when mapped using the SMART web server was found to be in the leucine rich repeat domain. Remarkably, this SNP was found in all 10 GSDs with IBD (9 were homozygous for the SNP allele and one was a heterozygote) used in the mutational analysis and, therefore, may play a significant role in canine IBD. However, this SNP would need to be assessed in a case-control population to determine if this is the case. Unfortunately, this could not be performed in the current study as the SNP was present in a different exon and therefore resources were unavailable to allow further

40

A. Kathrani et al. / Veterinary Immunology and Immunopathology 161 (2014) 32–41

sequencing of this region in the case and control populations. A possible limitation of our present study may be the inclusion of control dogs with non-inflammatory disease. Although retrospective phenotyping was carried out to ascertain that none of these dogs had inflammatory, infectious or immune-mediated disease, this type of inquiry can be imprecise and, therefore, some dogs may have been unsuitable as controls as they could subsequently have developed such diseases. However, to control for this, we aimed only to recruit dogs older than 4 years and indeed, 75% and 65% of dogs in our GSD and other breed control groups were 5 years or older respectively. This resulted in a significant difference in the age groups between the case and control population. In conclusion, this study has confirmed the significant association of four non-synonymous SNPs in NOD2 with IBD in the GSD breed. Although this association was not apparent in other breeds with IBD, a larger sample size may be needed to detect a significant difference in other breeds. Similarly to previous findings in human IBD, our results highlight the potential importance of NOD2 in the pathogenesis of canine IBD in the GSD breed. However, as the significance in the GSD population was modest, replication in an independent sample would be needed to confirm these findings. References Abreu, M.T., Fukata, M., Arditi, M., 2005. TLR signaling in the gut in health and disease. J. Immunol. 174, 4453–4460. Akira, S., Uematsu, S., Takeuchi, O., 2006. Pathogen recognition and innate immunity. Cell 124, 783–801. Blumberg, R.S., Saubermann, L.J., Strober, W., 1999. Animal models of mucosal inflammation and their relation to human inflammatory bowel disease. Curr. Opin. Immunol. 11, 648–656. Bonen, D.K., Cho, J.H., 2003. The genetics of inflammatory bowel disease. Gastroenterology 124, 521–536. Browning, B.L., Huebner, C., Petermann, I., Gearry, R.B., Barclay, M.L., Shelling, A.N., Ferguson, L.R., 2007. Has toll-like receptor 4 been prematurely dismissed as an inflammatory bowel disease gene? Association study combined with meta-analysis shows strong evidence for association. Am. J. Gastroenterol. 102, 2504–2512. Cho, J.H., 2001. The Nod2 gene in Crohn’s disease: implications for future research into the genetics and immunology of Crohn’s disease. Inflamm. Bowel Dis. 7, 271–275. Csongei, V., Jaromi, L., Safrany, E., Sipeky, C., Magyari, L., Farago, B., Bene, J., Polgar, N., Lakner, L., Sarlos, P., Varga, M., Melegh, B., 2010. Interaction of the major inflammatory bowel disease susceptibility alleles in Crohn’s disease patients. World J. Gastroenterol. 16, 176–183. Culverhouse, R., Suarez, B.K., Lin, J., Reich, T., 2002. A perspective on epistasis: limits of models displaying no main effect. Am. J. Hum. Genet. 70, 461–471. Cuthbert, A.P., Fisher, S.A., Mirza, M.M., King, K., Hampe, J., Croucher, P.J., Mascheretti, S., Sanderson, J., Forbes, A., Mansfield, J., Schreiber, S., Lewis, C.M., Mathew, C.G., 2002. The contribution of NOD2 gene mutations to the risk and site of disease in inflammatory bowel disease. Gastroenterology 122, 867–874. Franchimont, D., Vermeire, S., El Housni, H., Pierik, M., Van Steen, K., Gustot, T., Quertinmont, E., Abramowicz, M., Van Gossum, A., Deviere, J., Rutgeerts, P., 2004. Deficient host–bacteria interactions in inflammatory bowel disease? The toll-like receptor (TLR)-4 Asp299gly polymorphism is associated with Crohn’s disease and ulcerative colitis. Gut 53, 987–992. Fritz, J.H., Le Bourhis, L., Magalhaes, J.G., Philpott, D.J., 2008. Innate immune recognition at the epithelial barrier drives adaptive immunity: APCs take the back seat. Trends Immunol. 29, 41–49. Gewirtz, A.T., Vijay-Kumar, M., Brant, S.R., Duerr, R.H., Nicolae, D.L., Cho, J.H., 2006. Dominant-negative TLR5 polymorphism reduces adaptive immune response to flagellin and negatively associates with Crohn’s disease. Am. J. Physiol. Gastrointest. Liver Physiol. 290, G1157–G1163.

Guo, S.W., Thompson, E.A., 1992. Performing the exact test of Hardy–Weinberg proportion for multiple alleles. Biometrics 48, 361–372. Hedl, M., Abraham, C., 2010. Secretory mediators as a critical mechanism for Nod2-mediated tolerance in human macrophages. Gastroenterology. Hendrickson, B.A., Gokhale, R., Cho, J.H., 2002. Clinical aspects and pathophysiology of inflammatory bowel disease. Clin. Microbiol. Rev. 15, 79–94. Hisamatsu, T., Suzuki, M., Reinecker, H.C., Nadeau, W.J., McCormick, B.A., Podolsky, D.K., 2003. CARD15/NOD2 functions as an antibacterial factor in human intestinal epithelial cells. Gastroenterology 124, 993–1000. Hugot, J.P., Chamaillard, M., Zouali, H., Lesage, S., Cezard, J.P., Belaiche, J., Almer, S., Tysk, C., O’Morain, C.A., Gassull, M., Binder, V., Finkel, Y., Cortot, A., Modigliani, R., Laurent-Puig, P., Gower-Rousseau, C., Macry, J., Colombel, J.F., Sahbatou, M., Thomas, G., 2001. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn’s disease. Nature 411, 599–603. Jozaki, K., Shinkai, H., Tanaka-Matsuda, M., Morozumi, T., Matsumoto, T., Toki, D., Okumura, N., Eguchi-Ogawa, T., Kojima-Shibata, C., Kadowaki, H., Suzuki, E., Wada, Y., Uenishi, H., 2009. Influence of polymorphisms in porcine NOD2 on ligand recognition. Mol. Immunol. 47, 247–252. Kathrani, A., House, A., Catchpole, B., Murphy, A., German, A., Werling, D., Allenspach, K., 2010. Polymorphisms in the TLR4 and TLR5 gene are significantly associated with inflammatory bowel disease in German shepherd dogs. PLoS One 5, 15740. Kathrani, A., House, A., Catchpole, B., Murphy, A., Werling, D., Allenspach, K., 2011. Breed-independent toll-like receptor 5 polymorphisms show association with canine inflammatory bowel disease. Tissue Antigens 78, 94–101. Kobayashi, K.S., Chamaillard, M., Ogura, Y., Henegariu, O., Inohara, N., Nunez, G., Flavell, R.A., 2005. Nod2-dependent regulation of innate and adaptive immunity in the intestinal tract. Science 307, 731–734. Li, M., Gao, X., Guo, C.C., Wu, K.C., Zhang, X., Hu, P.J., 2008. OCTN and CARD15 gene polymorphism in Chinese patients with inflammatory bowel disease. World J. Gastroenterol. 14, 4923–4927. Maeda, S., Hsu, L.C., Liu, H., Bankston, L.A., Iimura, M., Kagnoff, M.F., Eckmann, L., Karin, M., 2005. Nod2 mutation in Crohn’s disease potentiates NF-kappaB activity and IL-1beta processing. Science 307, 734–738. Magalhaes, J.G., Tattoli, I., Girardin, S.E., 2007. The intestinal epithelial barrier: how to distinguish between the microbial flora and pathogens. Semin. Immunol. 19, 106–115. Malmberg, R.L., Held, S., Waits, A., Mauricio, R., 2005. Epistasis for fitnessrelated quantitative traits in Arabidopsis thaliana grown in the field and in the greenhouse. Genetics 171, 2013–2027. Moore, J.H., 2003. The ubiquitous nature of epistasis in determining susceptibility to common human diseases. Hum. Hered. 56, 73–82. Moore, J.H., 2005. A global view of epistasis. Nat. Genet. 37, 13–14. Ogura, Y., Bonen, D.K., Inohara, N., Nicolae, D.L., Chen, F.F., Ramos, R., Britton, H., Moran, T., Karaliuskas, R., Duerr, R.H., Achkar, J.P., Brant, S.R., Bayless, T.M., Kirschner, B.S., Hanauer, S.B., Nunez, G., Cho, J.H., 2001. A frameshift mutation in NOD2 associated with susceptibility to Crohn’s disease. Nature 411, 603–606. Ostrander, E.A., Wayne, R.K., 2005. The canine genome. Genome Res. 15, 1706–1716. Parker, H.G., Kim, L.V., Sutter, N.B., Carlson, S., Lorentzen, T.D., Malek, T.B., Johnson, G.S., Defrance, H.B., Ostrander, E.A., Kruglyak, L., 2004. Genetic structure of the purebred domestic dog. Science 304, 1160–1164. Perez, L.H., Butler, M., Creasey, T., Dzink-Fox, J., Gounarides, J., Petit, S., Ropenga, A., Ryder, N., Smith, K., Smith, P., Parkinson, S.J., 2010. Direct bacterial killing in vitro by recombinant Nod2 is compromised by Crohn’s disease-associated mutations. PLoS One 5, e10915. Pierik, M., Joossens, S., Van Steen, K., Van Schuerbeek, N., Vlietinck, R., Rutgeerts, P., Vermeire, S., 2006. Toll-like receptor-1, -2, and -6 polymorphisms influence disease extension in inflammatory bowel diseases. Inflamm. Bowel Dis. 12, 1–8. Pinedo, P.J., Buergelt, C.D., Donovan, G.A., Melendez, P., Morel, L., Wu, R., Langaee, T.Y., Rae, D.O., 2009a. Association between CARD15/NOD2 gene polymorphisms and paratuberculosis infection in cattle. Vet. Microbiol. 134, 346–352. Pinedo, P.J., Wang, C., Li, Y., Rae, D.O., Wu, R., 2009b. Risk haplotype analysis for bovine paratuberculosis. Mamm. Genome 20, 124–129. Rakoff-Nahoum, S., Paglino, J., Eslami-Varzaneh, F., Edberg, S., Medzhitov, R., 2004. Recognition of commensal microflora by toll-like receptors is required for intestinal homeostasis. Cell 118, 229–241.

A. Kathrani et al. / Veterinary Immunology and Immunopathology 161 (2014) 32–41 Roberts, R.L., Gearry, R.B., Hollis-Moffatt, J.E., Miller, A.L., Reid, J., Abkevich, V., Timms, K.M., Gutin, A., Lanchbury, J.S., Merriman, T.R., Barclay, M.L., Kennedy, M.A., 2007. IL23R R381Q and ATG16L1 T300A are strongly associated with Crohn’s disease in a study of New Zealand Caucasians with inflammatory bowel disease. Am. J. Gastroenterol. 102, 2754–2761. Segre, D., Deluna, A., Church, G.M., Kishony, R., 2005. Modular epistasis in yeast metabolism. Nat. Genet. 37, 77–83. Short, A.D., Kennedy, L.J., Barnes, A., Fretwell, N., Jones, C., Thomson, W., Ollier, W.E., 2007. Hardy Weinberg expectations in canine breeds: implications for genetic studies. J. Hered. 98, 445–451. Sole, X., Guino, E., Valls, J., Iniesta, R., Moreno, V., 2006. SNPStats: a web tool for the analysis of association studies. Bioinformatics 22, 1928–1929. Sutter, N.B., Eberle, M.A., Parker, H.G., Pullar, B.J., Kirkness, E.F., Kruglyak, L., Ostrander, E.A., 2004. Extensive and breed-specific linkage disequilibrium in Canis familiaris. Genome Res. 14, 2388–2396.

41

Swerdlow, M.P., Kennedy, D.R., Kennedy, J.S., Washabau, R.J., Henthorn, P.S., Moore, P.F., Carding, S.R., Felsburg, P.J., 2006. Expression and function of TLR2, TLR4, and Nod2 in primary canine colonic epithelial cells. Vet. Immunol. Immunopathol. 114, 313–319. Torok, H.P., Glas, J., Tonenchi, L., Mussack, T., Folwaczny, C., 2004. Polymorphisms of the lipopolysaccharide-signaling complex in inflammatory bowel disease: association of a mutation in the Toll-like receptor 4 gene with ulcerative colitis. Clin. Immunol. 112, 85–91. Xavier, R.J., Podolsky, D.K., 2007. Unravelling the pathogenesis of inflammatory bowel disease. Nature 448, 427–434. Yamazaki, K., Takazoe, M., Tanaka, T., Kazumori, T., Nakamura, Y., 2002. Absence of mutation in the NOD2/CARD15 gene among 483 Japanese patients with Crohn’s disease. J. Hum. Genet. 47, 469–472.