Polymorphisms in the Inflammatory Pathway Genes TLR2, TLR4, TLR9, LY96, NFKBIA, NFKB1, TNFA, TNFRSF1A, IL6R, IL10, IL23R, PTPN22, and PPARG Are Associated with Susceptibility of Inflammatory Bowel Disease in a Danish Cohort

Background The inflammatory bowel diseases (IBD), Crohn's disease (CD) and ulcerative colitis (UC), result from the combined effects of susceptibility genes and environmental factors. Polymorphisms in genes regulating inflammation may explain part of the genetic heritage. Methods Using a candidate gene approach, 39 mainly functional single nucleotide polymorphisms (SNPs) in 26 genes regulating inflammation were assessed in a clinical homogeneous group of severely diseased patients consisting of 624 patients with CD, 411 patients with UC and 795 controls. The results were analysed using logistic regression. Results Sixteen polymorphisms in 13 genes involved in regulation of inflammation were associated with risk of CD and/or UC (p≤0.05). The polymorphisms TLR2 (rs1816702), NFKB1 (rs28362491), TNFRSF1A (rs4149570), IL6R (rs4537545), IL23R (rs11209026) and PTPN22 (rs2476601) were associated with risk of CD and the polymorphisms TLR2 (rs1816702), TLR4 (rs1554973 and rs12377632), TLR9 (rs352139), LY96 (rs11465996), NFKBIA (rs696), TNFA (rs1800629), TNFRSF1A (rs4149570), IL10 (rs3024505), IL23R (rs11209026), PTPN22 (rs2476601) and PPARG (rs1801282) were associated with risk of UC. When including all patients (IBD) the polymorphisms TLR2 (rs4696480 and rs1816702), TLR4 (rs1554973 and rs12377632), TLR9 (rs187084), TNFRSF1A (rs4149570), IL6R (rs4537545), IL10 (rs3024505), IL23R (rs11209026) and PTPN22 (rs2476601) were associated with risk. After Bonferroni correction for multiple testing, both the homozygous and the heterozygous variant genotypes of IL23R G>A(rs11209026) (ORCD,adj: 0.38, 95% CI: 0.21–0.67, p = 0.03; ORIBD,adj 0.43, 95% CI: 0.28–0.67, p = 0.007) and PTPN22 1858 G>A(rs2476601) (ORCD,unadj 0.54, 95% CI: 0.41–0.72, p = 7*10−4; ORIBD,unadj: 0.61, 95% CI: 0.48–0.77, p = 0.001) were associated with reduced risk of CD. Conclusion The biological effects of the studied polymorphisms suggest that genetically determined high inflammatory response was associated with increased risk of CD. The many SNPs found in TLRs suggest that the host microbial composition or environmental factors in the gut are involved in risk of IBD in genetically susceptible individuals.


Introduction
Chronic inflammatory bowel diseases (IBDs), Crohn's disease (CD) and ulcerative colitis (UC), are complex diseases that result from the interaction of numerous genetic and environmental factors [1].
Genetic association studies have identified innate immunity as a critical component in the development of IBD. Until now, more than 163 IBD susceptibility polymorphisms have been confirmed, most of which are associated with both CD and UC, by candidate and genome wide association studies (GWAS) [2][3][4][5][6][7][8][9][10][11]. However, these polymorphisms have been estimated to only account for 20% of the genetic heritage involved in IBD [12].
Functional polymorphisms in genes in the inflammatory pathways may explain some of the genetic heritage involved in IBD. The transcription factor NFkB is a central regulator of inflammation. NFkB can be activated by Toll like receptors (TLRs). TLRs recognize pathogen-associated molecular patterns (PAMPs) that are broadly shared by pathogens but distinguishable from host molecules such as bacterial or viral DNA, flagellin or lipopolysaccharide (LPS). The TLRs initiates a kinase cascade that ultimately activates the IKK-complex, which phosphorylates and degrades the NFkB inhibitor IkBa. NFkB is shuttled from the cytosol to the nucleus where it initiates expression of pro-and antiinflammatory cytokines including TNF-a, IL-6 and IL-10 [13].
To identify susceptibility loci we assessed 39 mainly functional polymorphisms in genes involved in inflammation, particular in the NFkB pathway, in a homogeneous Danish cohort of 624 patients with severe CD, 411 patients with severe UC and 795 healthy controls. The candidate gene approach using functional polymorphisms allows interpretation of the underlying biological mechanisms based on increased or decreased gene expression or protein activity.

Materials and Methods
Cohort A prior anti-TNF naïve Danish cohort of patients with IBD was established. In short, blood samples retrieved as part of the routine screening for latent Mycobacterium tuberculosis at Statens Serum Institut (SSI, Copenhagen, Denmark) and the Department of Respiratory Diseases B or the Department Clinical Microbiology, Aarhus University Hospital (Aarhus, Denmark) were collected from 01.09.2009 to 30.03.2011 (9217 patients). Patients with intestinal diseases (ICD-10 code K50-K63) were identified by linking the unique personal identification number of Danish citizens (CPR-number) from each blood sample with the National Patient Registry (2659 patients). Patient records from 18 medical departments were examined (1378 patients) and identified 1035 ethnic Danish patients with IBD where blood and clinical data were available. The patients either received or were considered candidates to anti-tumor necrosis factor-a (TNF-a) therapy (infliximab or adalimumab). The control group consisted of 795 healthy blood donors recruited from Viborg, Denmark [4].

Selection of polymorphisms
A candidate gene approach was used with focus on polymorphisms in the TNF-a and NFkB pathways. In addition, polymorphisms in genes which have been shown to be associated with CD and/or UC, polymorphisms in inflammatory cytokines and polymorphisms in TLR2 and TLR4 were included [14].
Functional polymorphisms in relevant genes were found by searching pubmed with ''polymorphism AND gene-name AND (reporter gene OR luciferase OR ELISA OR enzyme-linked immunosorbent assay OR RT-PCR OR reverse transcriptase PCR OR EMSA OR electrophoretic mobility shift assay OR flow cytometry)''.

Genotyping
For patients with IBD the DNA was extracted from cryopreserved blood clots by using the Maxwell 16 Blood purification kit (Promega) according to the manufacturers' instructions with a median yield of 4.90 mg (range 0.8-25 mg) pr 300 ml total blood [15]. For the healthy controls, DNA was extracted from EDTAstabilized peripheral blood by either PureGene (Qiagen, Hilden, Germany) or Wizard Genomic (Promega, Madison, Wisconsin, USA) DNA purification kit according to the manufacturers' instructions [4]. Competitive Allele-Specific Polymerase chain reaction (KASP), an end-point PCR technology, was used by LGC Genomics for genotyping (LGC Genomics , Hoddesdon, United Kingdom) (http://www.lgcgenomics.com/). The SNPs studied were TLR2 (rs4696480, rs1816702, rs11938228, rs3804099), Genotyping of TNFA (TNF-a) 2857 C.T (rs1799724) and 2 863 C.A (rs1800630) failed due to their close proximity to each other. All genotyping of 2857 C.T (rs1799724) either failed or were erroneously genotyped as homozygous wild type when the patients were carriers of the AA genotype of 2863 C.A (rs1800630) due to genotyping bias.
The 39 genotypes were replicated in 94 randomly selected samples and yielded .99% identical genotypes.

Statistical analysis
Logistic regression was used to compare genotype distributions among patients with CD, UC and IBD versus healthy controls (Table S1 and S2). Crude odds ratio and odds ratio adjusted for age, gender and smoking status were assessed. A chi-square test was used to test for deviation from Hardy-Weinberg equilibrium in the healthy controls and for haplotype analysis (Table S3, S4, S5).

Ethics statement
The study was conducted in accordance with the Declaration of Helsinki and was approved by the Regional Ethics Committees of

Study population
Characteristics of the Danish patients with CD, UC and healthy controls are shown in Table 1.

Haplotype analysis
Haplotype analyses of TLR2, TLR4 and TLR9 among patients with CD, UC and IBD versus healthy controls are shown in Table  S3, S4, S5, respectively. Four haplotypes in TLR2, three in TLR4 and two in TLR9 described 88%, 94% and 97% of the observed genotypes, respectively.
No associations were found for TLR4.   The TLR9 haplotype combination 11 (rs187084CC and rs352139GG) was associated with increased risk of UC (OR: 1.53, 95% CI: 1.03-2.28, p = 0.04). The TLR9 haplotype combination 12 was associated with increased risk of IBD (OR: 1.35, 95% CI: 1.04-1.76, p = 0.03). Indirectly, these results support the analysis of the individual SNPs, where the variant allele of rs352139 (included in haplotype combination 22 which was used as reference) was fond to be associated with lowered risk of UC.
The biological interpretation indicates that a genetically determined higher activity of the inflammatory genes TLR2 (rs1816702 C.T), TNFRSF1A (2609 G.T) and IL6R (rs4537545 C.T) was associated with increased risk of CD and lower activity of the inflammatory genes NFKB1 (NFkB) (294ins/del ATTG), IL23R (rs11209026 G.A) and PTPN22 (1858 G.A) was associated with reduced risk of CD ( Table 2). The picture was less clear for UC. A genetically determined higher activity of the inflammatory genes TNFRSF1A (2609 G.T), PPARG (rs1801282 C.G) and TLR2 (rs1816702 C.T) was associated with increased risk of UC and lower activity of the inflammatory genes IL23R (rs11209026 G.A) and PTPN22 (1858 G.A) was associated with reduced risk of UC. In contrast, a genetically determined higher activity of the NFkB inhibitor IkBa (NFKBIA 2758 A.G) and lower activity of TLR9 (haplotype 11 (21486CC and 1174GG)) were associated with increased risk of UC. Furthermore, a genetically determined higher activity of LY96 (21625 C.G) and TNFA (2308 G.A) was associated with reduced risk of UC. However, the risk of CD and UC seem to have shared mechanisms through the TLR2 (rs1816702 C.T), TNFRSF1A (2609 G.T), IL23R (rs11209026 G.A) and PTPN22 (1858 G. A) polymorphisms confirming that the inflammatory pathways are involved in risk of both CD and UC [16,17].
This study was unable to confirm the associations between the variant allele of IL10 (rs3024505 C.T) and increased risk of CD [3,16] or the associations between the variant allele of CD14 (2 159 G.A) and increased risk of CD, UC and IBD [22]. This study may be underpowered to detect an association in IL10 (rs3024505 C.T), as a GWA study with more than six thousands CD cases and fifteen thousands control found that the varaint allele was associated with increased risk of CD but with an odds ratio of 1.12 [16]. However, IL10 has previously been found to be associated with risk of CD and UC in the Danish population [3]. The difference of association in CD14 (2159 G.A) may be due to genetic differences between the Korean and Danish population.
The variant allele in NFKBIA (IkBa) (2758 A.G) [23][24][25] and TNFA (2308 G.A) were associated with higher and lower risk of UC, respectively, and the deletion polymorphism in NFKB1 (NFkB) (294ins/del ATTG) was associated with lower risk of CD in our cohort study. However, there seems to be no consensus regarding these polymorphisms in other cohort studies [24][25][26][27]. TLR5stop (1174 C.T) has been found to be associated with reduced risk of CD in a Jewish cohort [28] but not in a non-Jewish and German cohort [28,29]. No association with TLR5stop was found in this cohort study.
The results in this study should be interpreted with care. TLR2 (rs4696480 A.T), TLR4 (rs1554973 T.C), TLR9 (1174 G.A) and TGFB1 (2509 C.T) were not in Hardy-Weinberg equilibrium among the healthy controls which is probable due to chance because of the number of polymorphisms analyzed. When corrected adequately for multiple testing they did not deviate from Hardy-Weinberg equilibrium. In the light of the obtained Pvalues and the number of statistical tests performed, we cannot exclude that some of our positive findings may be due to chance. If the results were corrected for multiple testing only the well known susceptibility polymorphisms in IL23R (rs11209026 G.A) and PTPN22 (1858 G.A) were associated with reduced risk of both CD and IBD. We successfully tested 37 polymorphisms and, assuming a 5% acceptance level, two would be expected to be associated with susceptibility by pure chance. In this study 16 polymorphisms were found to be associated with susceptibility and the found associations were biologically plausible. A major strength was that this clinically homogeneous and well-characterised cohort was rather large including 1035 patients with IBD and 795 healthy controls. All the patients were considered for anti-TNF treatment and were therefore considered to have a severe disease course. Genetic determinants may be expected to be strong among severely ill cases [30].
In conclusion, 16 functional SNPs in 13 genes involved in regulation of inflammation were found to be associated with susceptibility of severe CD, UC or IBD. Eleven of the SNPs have not previously been reported as susceptibility polymorphisms of CD, UC or IBD (Figure 1) although other polymorphisms in most of these genes have previously been associated with susceptibility of CD, UC or IBD. Our results suggest that genetically determined high inflammatory response was associated with increased risk of CD and the large number of polymorphisms in the TLRs associated with risk of CD or UC support that the host microbial composition, diet or environmental molecules in the gut are important factors driving the inflammatory response in genetically susceptible individuals. Figure 1. Sixteen functional single nucleotide polymorphisms (SNPs) in 13 genes involved in regulation of inflammation were found to be associated with susceptability of severe Crohn's disease (CD), ulcerative colitis (UC) or inflammatory bowel diseases (IBD). Eleven of the SNPs have not previously been reported as susceptability polymorphisms of CD, UC or IBD (TLR2 (rs4696480 and rs1816702), TLR4 (rs1554973 and rs12377632), TLR9 (rs187084 and rs352139), LY96 (rs11465996), NFKBIA (rs696), TNFRSF1A (rs4149570), IL6R (rs4537545) and PTPN22 (rs2476601)). doi:10.1371/journal.pone.0098815.g001

Supporting Information
Table S1 Odds ratios (OR) (unadjusted) for genotypes studied among healthy controls and patients with Crohn's disease (CD), ulcerative colitis (UC) and combined inflammatory bowel disease (IBD).

(DOC)
Table S2 Odds ratios (OR) (adjusted for age, sex and smoking status) for genotypes studied among healthy controls and patients with Crohn's disease (CD), ulcerative colitis (UC) and combined inflammatory bowel disease (IBD).