- Email: [email protected]
S1173 Evidence for the Autophagy Gene ULK1 As Susceptibility Gene in Crohn's Disease
AGA Abstracts
S1171
prioritized. For these genes, a total of 76 haplotype tagging single nucleotide polymorphisms (SNPs) were selected using Haploview 4.0 software, including the region 10kB upstream to 5kB downstream of each gene. These haplotype tagging SNPs were genotyped in an exploratory cohort of CD patients (CD1, n=480), a second independent CD cohort (CD2, n=374) and healthy controls (n=392), using the Sequenom ® platform. Results. The frequency of the T-allele of the intronic haplotype tagging SNP rs12303764 in ULK1 (human homologue of yeast Atg1) was significantly higher in CD1 patients (63%) and CD2 patients (63%) vs. healthy controls (57%, both p-values 0.02), with a p-value of 0.006 for the combination of CD1 and CD2 (Haploview permutation p-value 0.03). Also the T-allele at rs3923716 in ULK1 was seen at a higher frequency in CD1 and CD2 patients (both 10%) vs. healthy controls (7%, respective p=0.05 and p=0.04), with a p-value of 0.03 for the combination of CD1 and CD2. (permutation p-value 0.1). The ULK1 risk alleles were not associated with disease location or behavior. None of the other tested polymorphisms were associated with susceptibility to CD in this cohort. Conclusions. Two haplotype tagging polymorphisms in ULK1 were found to be associated with susceptibility to CD. ULK1 (ubiquitously expressed in mammalian tissue) is an interesting candidate gene for IBD, as it is required for autophagy, while overexpression may also inhibit autophagy, probably due to domains with different regulatory or localization functions. Moreover, the gene has an essential role in membrane trafficking and recruitment of regulatory molecules. In order to further clarify the role of ULK1 in CD pathogenesis, saturation of the involved haplotype blocks will be necessary as a first step, and is ongoing.
Lack of Association of Previously Identified Disease Loci 10q12, 11p15, 20q13 and 1q21.2 (ECM1) in IBD Orazio Palmieri, Anna Latiano, Elisabetta Colombo, Maria Rosa Valvano, Tiziana Latiano, Giuseppe Corritore, Renata D'Inca, Maurizio Vecchi, Renzo Caprilli, Sandro Ardizzone, Stefano Nobile, Domenica De Venuto, Concetta Sferlazzas, Annamaria Staiano, Veronica Lodde, Giuseppe Iacono, Cinzia Bascietto, Brunislava Papadatou, Marseglia Antonio, Massimo Castro, Salvatore Cucchiara, Vito Annese Inflammatory bowel disease (IBD), a chronic disorder affecting the intestinal mucosa, normally presents as either of two subphenotypes, Crohn's disease (CD) or ulcerative colitis (UC). The complex genetic basis of IBD has been highlighted by recent genome-wide association studies (GWAs) and although several susceptibility loci for CD (including NOD2, 5q31, Il23R, ATG16L1, MST1, TNFSF15, ECM1) and recently also for UC, have been replicated in more than one study, some of the reported associations have been found only in single study to date. Aims: To investigate variants of four previously reported susceptibility loci (NELL1, C10ORSF67, TNFRSF6B and ECM1) in a Italian samples with CD or UC and healthy controls (HC) and search for potential correlation with clinical phenotypes in patients with adult and pediatric onset of disease. Results: The polymorphisms rs1793004 in 11p15.1 (NELL1), rs1398024 in 10q12.2 (C10ORSF67), rs4809330 in 20q13 (TNFRSF6B), rs3737240 (T130M) and rs13294 (G290S) in ECM1 gene were tested in 762 CD (439 male) with a mean age at diagnosis of 31 yrs (range 1-79), 1071 UC (627 male) with a mean age at diagnosis of 31 yrs (range 1-83) and 552 HC, by using the TaqMan allelic discrimination method. No significant difference for either allele and genotype frequencies of the selected SNPs was found in both CD and UC patients compared with controls (MAF: rs1793004 21% in CD and UC vs 24% in HC; rs1398024 27% in UC vs 26% in HC; rs4809330 24% in CD, 22% in UC vs 25% in HC; rs3737240 42% in UC vs 40% in HC; rs13294 43% in UC vs 40% in HC). The frequencies remained still not significant different for all variants after stratifying the cohort in adult (CD: 618 patients > 18 yrs at diagnosis; UC: 794 patients >18 yrs at diagnosis) and pediatric-onset patients, although for T130M a week association in UC pediatric-onset patients (164 pts) in both allele (P=0.043; OR 1.29, CI 1.01-1.66) and genotype (P=0.045; OR 1.47, CI 1.01-2.15) frequencies, was found. No correlation with any clinical sub-phenotypes for all loci was observed. Conclusions: The investigated SNPs on the identified loci 11p15.1, 10q12.2 and 20q13 and ECM1 gene, have no significant role in disease susceptibility in IBD Italian population. However, a weak association with variation 130M and UC pediatric-onset patients, was found.
S1174 Global DNA Hypomethylation and DNA Methyltransferase-1 Downregulation in Nonneoplastic Colonic Mucosae of Ulcerative Colitis Patients Alexandru Olaru, Yulan Cheng, Bogdan C. Paun, Takatsugu Kan, Florin M. Selaru, James P. Hamilton, Zhe Jin, Stefan David, Rachana Agarwal, Jian Yang, John M. Abraham, Stephen J. Meltzer, Yuriko Mori Ulcerative colitis (UC) is a risk factor for colon cancers. The cancer risk is especially high for UC with widespread, severe, or long-standing disease, which indicates the central role of inflammation in the UC-associated carcinogenesis. Global DNA hypomethylation is a common early-stage abnormality in multiple cancers that promotes chromosomal instability and transcriptional activation of transposons. The cause of global DNA hypomethylation in cancer has not been fully understood. Putative mechanisms include abnormalities in methyl base supply and in DNA methyltransferases (DNMTs). Notably, global DNA hypomethylation has also been observed in several chronic inflammatory disorders. In the current study, we analyzed global DNA methylation status and the expression of DNMT1, the enzyme responsible to maintenance DNA methylation, in UC colonic mucosae and their association with the presence of persistent local inflammation. Macroscopically dissected snap-frozen nonneoplastic mucosae from affected and unaffected colonic regions of nine patients with localized UC were analyzed. We also analyzed colonic mucosal biopsies from 15 cases with no previous or current colonic adenoma/carcinoma or inflammatory bowel disease as the control. The global DNA methylation status was measured by utilizing bisulfite pyrosequencing analysis of long interspersed nuclear element-1 (LINE-1) loci. The DNMT1 expression status was assessed by utilizing a real-time quantitative RT-PCR method. The Lack of association between LINE-1 methylation level (i.e., %methylated DNA) and age or anatomical sites was verified in control specimens. LINE-1 methylation level was significantly lower in UC colonic mucosae from affected region than in control mucosae (63+/-3% versus 70 +/- 3%, p=3x105), but not in UC colonic mucosae from unaffected region (73+/-4%). LINE-1 methylation level was also significantly lower in affected colonic regions than in their corresponding unaffected colonic regions (p=3x10-6). Similarly, the DNMT1 mRNA level was significantly decreased in affected colonic regions relative to their corresponding unaffected colonic regions (average reduction by 43%, p=0.007). Furthermore, levels of DNMT1 mRNA and LINE-1 methylation demonstrated a significant linear correlation (R=0.5, p=0.03). In conclusion, we observed global DNA hypomethylation and DNMT1 downregulation unique to inflamed nonneoplastic UC colonic mucosa. Levels of DNA hypomethylation and DNMT1 downregulation were correlated linearly. These data suggest that chronic inflammation causes local downregulation of DNMT1, which ultimately contributes to the early stage carcinogenesis in UC patients.
S1172 The Differential Association of the FcgR2a and FcgR3a Genes in Inflammatory Bowel Diseases Across Three Caucasian Populations Rinse K. Weersma, Bart Crusius, Bobby P. Koeleman, Rebecca L. Roberts, Rogelio J. Palomino-Morales, Simone C. Wolfkamp, Jade E. Hollis-Moffatt, Eleonora A. Festen, Sander Meisner, Roel Heijmans, Richard B. Gearry, Murray L. Barclay, Maria GomezGarcia, Miguel A. Lopez-Nevot, Antonio Nieto, Luis Rodrigo, Timothy R. Radstake, Ad A. van Bodegraven, Cisca Wijmenga, Tony R. Merriman, Pieter Stokkers, Amado S. Pena, Javier Martin, Behrooz Z. Alizadeh Background: The Fc receptors II, and III (FcgR2a, and FcgR3a) have a crucial role in the activation of immune cells in response to immune complexes. FcgR2a*519GG and FcgR3a*559CC genotypes have been associated with several autoimmune diseases globally. The genomic region is poorly covered in recently performed genome wide association studies. In a relatively large study of 6,351 people, we tested whether the FcgR2a, and FcgR3a genes are also involved in inflammatory bowel diseases (IBD) including Crohn's disease (CrD), and ulcerative colitis (UC). Methods: In this multi-center two-staged study, we genotyped the common FcgR2a*A519G and FcgR3a*A559C functional variants in 4,205 IBD patients in six well-phenotyped Caucasian IBD cohorts and 2,365 ethnically matched controls recruited from the Netherlands, Spain and New Zealand. Results: In the initial Dutch study, we found a significant association of FcgR2a genotypes with IBD (p-genotype=0.02); while the FcgR2a*519GG was more common among controls (23%) than patients with IBD (18%; odds ratio(OR)=0.75; 95%CI 0.61-0.92; P=0.005). This association was corroborated when a pooled analysis was performed across all study populations (Mantel-Haenszel populations weighted OR=0.84; 0.74-0.95; P=0.005) in replication stage. The Fcgr2a*GG genotype was associated with both CrD (MH-OR=0.84; 0.73-0.97; P=0.01), or to UC (MH-OR=0.84; 0.720.97; P=0.02), suggesting a protective role for this genotype in IBD. There was no association of FcgR3a*A559C genotypes with IBD, CrD, or UC in any of the populations or in the overall analysis. Conclusion: The FcgR2a*519G functional variant is associated with IBD and in contrary to other autoimmune diseases, it reduces susceptibility to IBD, CrD, and UC. There is no association between FcgR3a*5A559C and IBD, CrD or UC. This finding may have clinical implications for our understanding of the pathogenesis of IBD.
S1175 Association Between Non-Synonymous Variant in the MST1 Gene and IBD in Italian Population Anna Latiano, Orazio Palmieri, Maria Rosa Valvano, Elisabetta Colombo, Maurizio Vecchi, Renata D'Inca, Tiziana Latiano, Giuseppe Corritore, Gabriele Riegler, Alessandro Ferraris, Gian L. de Angelis, Claudio Romano, Osvaldo Borrelli, Paolo Lionetti, Arrigo Barabino, Maria Pastore, Giuliano Lombardi, Massimo Castro, Salvatore Cucchiara, Vito Annese IBD locus on chromosome 3p21 has been identified and confirmed in several genome-wide association studies. A single nonsynonymous coding variant rs3197999 (R689C) in the macrophage-stimulating 1 gene (MST1) has been associated to both Crohn's disease (CD) and ulcerative colitis (UC). Aim: To investigate rs3197999 MST1 polymorphism and searched for likely correlation with clinical phenotypes in CD and UC patients with adult and pediatric onset of disease. Methods: 677 CD patients (392 male) with a mean age at diagnosis of 26 yrs (range 1-79), 953 UC patients (539 male) with a mean age at diagnosis of 30 yrs (range 1-83), and 651 healthy controls (HC) were genotyped for rs3197999 polymorphism, by using the TaqMan allelic discrimination method. Genotype frequencies were compared with respect to age at diagnosis, disease localization and behaviour (Montreal classification), need of resective surgery, and response to medical therapy (steroids, immunosuppressive drugs). Results: The investigated SNP was in Hardy-Weinberg equilibrium in controls. Highly significant allele association with CD (P = < 0.0001; OR 1.5; CI 1.3-1.8), and UC (P = 0.0008; OR 1.3; CI 1.1-1.5) was shown. By combining homo- and heterozygous carriers of the risk allele, the difference was significant for both diseases (CD: P = < 0.0001; OR 1.7; CI 1.3-2.1; and UC: P = 0.00225; OR 1.4; CI 1.1-1.7). The difference of genotype frequency remained still significant for both diseases after stratifying the cohort in adult (CD: 412 patients >18 yrs at diagnosis; P = 5 x 10-4; UC: 697 patients >18 yrs at diagnosis;
S1173 Evidence for the Autophagy Gene ULK1 As Susceptibility Gene in Crohn's Disease Liesbet Henckaerts, Isabelle Cleynen, Wouter Van Moerkercke, Marko Brinar, Paul J. Rutgeerts, Severine Vermeire Evidence for the autophagy gene ULK1 as susceptibility gene in Crohn's disease. Liesbet Henckaerts, Isabelle Cleynen, Wouter Van Moerkercke, Marko Brinar, Paul Rutgeerts, Séverine Vermeire Background and aims. Variants in ATG16L1 and IRGM genes have recently been associated with susceptibility to Crohn's disease (CD). This discovery has highlighted the role for autophagy in CD pathogenesis. So far, none of the other genes in the extensive autophagy pathway have been studied in inflammatory bowel disease (IBD). Our aim was to study polymorphisms in selected autophagy genes for their association with susceptibility to CD. Methods. We screened all known human functional homologues of yeast autophagy (Atg) genes for their location in a known IBD locus or in a genomic region detected in a genome wide association study (GWAS) or GWAS meta-analysis. Thus, 12 genes were
AGA Abstracts
A-206
S1171
prioritized. For these genes, a total of 76 haplotype tagging single nucleotide polymorphisms (SNPs) were selected using Haploview 4.0 software, including the region 10kB upstream to 5kB downstream of each gene. These haplotype tagging SNPs were genotyped in an exploratory cohort of CD patients (CD1, n=480), a second independent CD cohort (CD2, n=374) and healthy controls (n=392), using the Sequenom ® platform. Results. The frequency of the T-allele of the intronic haplotype tagging SNP rs12303764 in ULK1 (human homologue of yeast Atg1) was significantly higher in CD1 patients (63%) and CD2 patients (63%) vs. healthy controls (57%, both p-values 0.02), with a p-value of 0.006 for the combination of CD1 and CD2 (Haploview permutation p-value 0.03). Also the T-allele at rs3923716 in ULK1 was seen at a higher frequency in CD1 and CD2 patients (both 10%) vs. healthy controls (7%, respective p=0.05 and p=0.04), with a p-value of 0.03 for the combination of CD1 and CD2. (permutation p-value 0.1). The ULK1 risk alleles were not associated with disease location or behavior. None of the other tested polymorphisms were associated with susceptibility to CD in this cohort. Conclusions. Two haplotype tagging polymorphisms in ULK1 were found to be associated with susceptibility to CD. ULK1 (ubiquitously expressed in mammalian tissue) is an interesting candidate gene for IBD, as it is required for autophagy, while overexpression may also inhibit autophagy, probably due to domains with different regulatory or localization functions. Moreover, the gene has an essential role in membrane trafficking and recruitment of regulatory molecules. In order to further clarify the role of ULK1 in CD pathogenesis, saturation of the involved haplotype blocks will be necessary as a first step, and is ongoing.
Lack of Association of Previously Identified Disease Loci 10q12, 11p15, 20q13 and 1q21.2 (ECM1) in IBD Orazio Palmieri, Anna Latiano, Elisabetta Colombo, Maria Rosa Valvano, Tiziana Latiano, Giuseppe Corritore, Renata D'Inca, Maurizio Vecchi, Renzo Caprilli, Sandro Ardizzone, Stefano Nobile, Domenica De Venuto, Concetta Sferlazzas, Annamaria Staiano, Veronica Lodde, Giuseppe Iacono, Cinzia Bascietto, Brunislava Papadatou, Marseglia Antonio, Massimo Castro, Salvatore Cucchiara, Vito Annese Inflammatory bowel disease (IBD), a chronic disorder affecting the intestinal mucosa, normally presents as either of two subphenotypes, Crohn's disease (CD) or ulcerative colitis (UC). The complex genetic basis of IBD has been highlighted by recent genome-wide association studies (GWAs) and although several susceptibility loci for CD (including NOD2, 5q31, Il23R, ATG16L1, MST1, TNFSF15, ECM1) and recently also for UC, have been replicated in more than one study, some of the reported associations have been found only in single study to date. Aims: To investigate variants of four previously reported susceptibility loci (NELL1, C10ORSF67, TNFRSF6B and ECM1) in a Italian samples with CD or UC and healthy controls (HC) and search for potential correlation with clinical phenotypes in patients with adult and pediatric onset of disease. Results: The polymorphisms rs1793004 in 11p15.1 (NELL1), rs1398024 in 10q12.2 (C10ORSF67), rs4809330 in 20q13 (TNFRSF6B), rs3737240 (T130M) and rs13294 (G290S) in ECM1 gene were tested in 762 CD (439 male) with a mean age at diagnosis of 31 yrs (range 1-79), 1071 UC (627 male) with a mean age at diagnosis of 31 yrs (range 1-83) and 552 HC, by using the TaqMan allelic discrimination method. No significant difference for either allele and genotype frequencies of the selected SNPs was found in both CD and UC patients compared with controls (MAF: rs1793004 21% in CD and UC vs 24% in HC; rs1398024 27% in UC vs 26% in HC; rs4809330 24% in CD, 22% in UC vs 25% in HC; rs3737240 42% in UC vs 40% in HC; rs13294 43% in UC vs 40% in HC). The frequencies remained still not significant different for all variants after stratifying the cohort in adult (CD: 618 patients > 18 yrs at diagnosis; UC: 794 patients >18 yrs at diagnosis) and pediatric-onset patients, although for T130M a week association in UC pediatric-onset patients (164 pts) in both allele (P=0.043; OR 1.29, CI 1.01-1.66) and genotype (P=0.045; OR 1.47, CI 1.01-2.15) frequencies, was found. No correlation with any clinical sub-phenotypes for all loci was observed. Conclusions: The investigated SNPs on the identified loci 11p15.1, 10q12.2 and 20q13 and ECM1 gene, have no significant role in disease susceptibility in IBD Italian population. However, a weak association with variation 130M and UC pediatric-onset patients, was found.
S1174 Global DNA Hypomethylation and DNA Methyltransferase-1 Downregulation in Nonneoplastic Colonic Mucosae of Ulcerative Colitis Patients Alexandru Olaru, Yulan Cheng, Bogdan C. Paun, Takatsugu Kan, Florin M. Selaru, James P. Hamilton, Zhe Jin, Stefan David, Rachana Agarwal, Jian Yang, John M. Abraham, Stephen J. Meltzer, Yuriko Mori Ulcerative colitis (UC) is a risk factor for colon cancers. The cancer risk is especially high for UC with widespread, severe, or long-standing disease, which indicates the central role of inflammation in the UC-associated carcinogenesis. Global DNA hypomethylation is a common early-stage abnormality in multiple cancers that promotes chromosomal instability and transcriptional activation of transposons. The cause of global DNA hypomethylation in cancer has not been fully understood. Putative mechanisms include abnormalities in methyl base supply and in DNA methyltransferases (DNMTs). Notably, global DNA hypomethylation has also been observed in several chronic inflammatory disorders. In the current study, we analyzed global DNA methylation status and the expression of DNMT1, the enzyme responsible to maintenance DNA methylation, in UC colonic mucosae and their association with the presence of persistent local inflammation. Macroscopically dissected snap-frozen nonneoplastic mucosae from affected and unaffected colonic regions of nine patients with localized UC were analyzed. We also analyzed colonic mucosal biopsies from 15 cases with no previous or current colonic adenoma/carcinoma or inflammatory bowel disease as the control. The global DNA methylation status was measured by utilizing bisulfite pyrosequencing analysis of long interspersed nuclear element-1 (LINE-1) loci. The DNMT1 expression status was assessed by utilizing a real-time quantitative RT-PCR method. The Lack of association between LINE-1 methylation level (i.e., %methylated DNA) and age or anatomical sites was verified in control specimens. LINE-1 methylation level was significantly lower in UC colonic mucosae from affected region than in control mucosae (63+/-3% versus 70 +/- 3%, p=3x105), but not in UC colonic mucosae from unaffected region (73+/-4%). LINE-1 methylation level was also significantly lower in affected colonic regions than in their corresponding unaffected colonic regions (p=3x10-6). Similarly, the DNMT1 mRNA level was significantly decreased in affected colonic regions relative to their corresponding unaffected colonic regions (average reduction by 43%, p=0.007). Furthermore, levels of DNMT1 mRNA and LINE-1 methylation demonstrated a significant linear correlation (R=0.5, p=0.03). In conclusion, we observed global DNA hypomethylation and DNMT1 downregulation unique to inflamed nonneoplastic UC colonic mucosa. Levels of DNA hypomethylation and DNMT1 downregulation were correlated linearly. These data suggest that chronic inflammation causes local downregulation of DNMT1, which ultimately contributes to the early stage carcinogenesis in UC patients.
S1172 The Differential Association of the FcgR2a and FcgR3a Genes in Inflammatory Bowel Diseases Across Three Caucasian Populations Rinse K. Weersma, Bart Crusius, Bobby P. Koeleman, Rebecca L. Roberts, Rogelio J. Palomino-Morales, Simone C. Wolfkamp, Jade E. Hollis-Moffatt, Eleonora A. Festen, Sander Meisner, Roel Heijmans, Richard B. Gearry, Murray L. Barclay, Maria GomezGarcia, Miguel A. Lopez-Nevot, Antonio Nieto, Luis Rodrigo, Timothy R. Radstake, Ad A. van Bodegraven, Cisca Wijmenga, Tony R. Merriman, Pieter Stokkers, Amado S. Pena, Javier Martin, Behrooz Z. Alizadeh Background: The Fc receptors II, and III (FcgR2a, and FcgR3a) have a crucial role in the activation of immune cells in response to immune complexes. FcgR2a*519GG and FcgR3a*559CC genotypes have been associated with several autoimmune diseases globally. The genomic region is poorly covered in recently performed genome wide association studies. In a relatively large study of 6,351 people, we tested whether the FcgR2a, and FcgR3a genes are also involved in inflammatory bowel diseases (IBD) including Crohn's disease (CrD), and ulcerative colitis (UC). Methods: In this multi-center two-staged study, we genotyped the common FcgR2a*A519G and FcgR3a*A559C functional variants in 4,205 IBD patients in six well-phenotyped Caucasian IBD cohorts and 2,365 ethnically matched controls recruited from the Netherlands, Spain and New Zealand. Results: In the initial Dutch study, we found a significant association of FcgR2a genotypes with IBD (p-genotype=0.02); while the FcgR2a*519GG was more common among controls (23%) than patients with IBD (18%; odds ratio(OR)=0.75; 95%CI 0.61-0.92; P=0.005). This association was corroborated when a pooled analysis was performed across all study populations (Mantel-Haenszel populations weighted OR=0.84; 0.74-0.95; P=0.005) in replication stage. The Fcgr2a*GG genotype was associated with both CrD (MH-OR=0.84; 0.73-0.97; P=0.01), or to UC (MH-OR=0.84; 0.720.97; P=0.02), suggesting a protective role for this genotype in IBD. There was no association of FcgR3a*A559C genotypes with IBD, CrD, or UC in any of the populations or in the overall analysis. Conclusion: The FcgR2a*519G functional variant is associated with IBD and in contrary to other autoimmune diseases, it reduces susceptibility to IBD, CrD, and UC. There is no association between FcgR3a*5A559C and IBD, CrD or UC. This finding may have clinical implications for our understanding of the pathogenesis of IBD.
S1175 Association Between Non-Synonymous Variant in the MST1 Gene and IBD in Italian Population Anna Latiano, Orazio Palmieri, Maria Rosa Valvano, Elisabetta Colombo, Maurizio Vecchi, Renata D'Inca, Tiziana Latiano, Giuseppe Corritore, Gabriele Riegler, Alessandro Ferraris, Gian L. de Angelis, Claudio Romano, Osvaldo Borrelli, Paolo Lionetti, Arrigo Barabino, Maria Pastore, Giuliano Lombardi, Massimo Castro, Salvatore Cucchiara, Vito Annese IBD locus on chromosome 3p21 has been identified and confirmed in several genome-wide association studies. A single nonsynonymous coding variant rs3197999 (R689C) in the macrophage-stimulating 1 gene (MST1) has been associated to both Crohn's disease (CD) and ulcerative colitis (UC). Aim: To investigate rs3197999 MST1 polymorphism and searched for likely correlation with clinical phenotypes in CD and UC patients with adult and pediatric onset of disease. Methods: 677 CD patients (392 male) with a mean age at diagnosis of 26 yrs (range 1-79), 953 UC patients (539 male) with a mean age at diagnosis of 30 yrs (range 1-83), and 651 healthy controls (HC) were genotyped for rs3197999 polymorphism, by using the TaqMan allelic discrimination method. Genotype frequencies were compared with respect to age at diagnosis, disease localization and behaviour (Montreal classification), need of resective surgery, and response to medical therapy (steroids, immunosuppressive drugs). Results: The investigated SNP was in Hardy-Weinberg equilibrium in controls. Highly significant allele association with CD (P = < 0.0001; OR 1.5; CI 1.3-1.8), and UC (P = 0.0008; OR 1.3; CI 1.1-1.5) was shown. By combining homo- and heterozygous carriers of the risk allele, the difference was significant for both diseases (CD: P = < 0.0001; OR 1.7; CI 1.3-2.1; and UC: P = 0.00225; OR 1.4; CI 1.1-1.7). The difference of genotype frequency remained still significant for both diseases after stratifying the cohort in adult (CD: 412 patients >18 yrs at diagnosis; P = 5 x 10-4; UC: 697 patients >18 yrs at diagnosis;
S1173 Evidence for the Autophagy Gene ULK1 As Susceptibility Gene in Crohn's Disease Liesbet Henckaerts, Isabelle Cleynen, Wouter Van Moerkercke, Marko Brinar, Paul J. Rutgeerts, Severine Vermeire Evidence for the autophagy gene ULK1 as susceptibility gene in Crohn's disease. Liesbet Henckaerts, Isabelle Cleynen, Wouter Van Moerkercke, Marko Brinar, Paul Rutgeerts, Séverine Vermeire Background and aims. Variants in ATG16L1 and IRGM genes have recently been associated with susceptibility to Crohn's disease (CD). This discovery has highlighted the role for autophagy in CD pathogenesis. So far, none of the other genes in the extensive autophagy pathway have been studied in inflammatory bowel disease (IBD). Our aim was to study polymorphisms in selected autophagy genes for their association with susceptibility to CD. Methods. We screened all known human functional homologues of yeast autophagy (Atg) genes for their location in a known IBD locus or in a genomic region detected in a genome wide association study (GWAS) or GWAS meta-analysis. Thus, 12 genes were
AGA Abstracts
A-206