Final Report Summary - FERAL (Functional characterization of a novel rheumatoid arthritis susceptibility locus on chromosome 6q)
Rheumatoid arthritis (RA) is a chronic disabling condition affecting up to 1 % of the population worldwide. RA is a complex autoimmune disease with a strong genetic component to susceptibility.
The arthritis research United Kingdom epidemiology unit was involved in the ground-breaking 'Wellcome Trust Case Control Consortium' (WTCCC) study in which RA was one of seven diseases investigated using genome-wide association studies (GWAS). After human leukocyte antigen (HLA) and protein tyrosine phosphatase non-receptor type 22 (PTPN22), the next strongest association with RA was mapped to the 6q23 locus (rs6920220), in an intergenic region between the OLIG3 and TNFAIP3 genes. TNFAIP3 acted as a negative regulator of NF-kB and, therefore, was the most plausible candidate gene in the region.
The aim of the project was to follow up the results of the WTCCC RA GWAS to identify and functionally characterise a RA disease causal locus mapping to 6q23.
The project first phase was devoted to the identification of the RA causal mutation mapping to 6q23. I performed a fine mapping study of the OLIG3 and TNFAIP3 locus. I found 18 Single nucleotide polymorphisms (SNPs) significantly associated with RA. A logistic regression analysis showed that only rs6920220, rs13207033 and rs5029937 were independently associated with RA. The combination of the carriage of both risk alleles of rs6920220 and rs5029937 along with the absence of the protective allele of rs13207033 was strongly associated with RA when compared to carriage of none. This equated to an effect size of 1.50 (95% CI 1.21-1.85) compared to controls and was higher than that obtained for any SNP individually. This study was the first to show that confirmed loci from GWA studies, which conferred only a modest effect size, could harbour a significantly greater effect once the effect of additional risk variants was accounted for.
In conclusion, there were three independent SNPs associated with RA. These markers, or any of their proxies, were likely to be the causal variants in this region.
During the second phase of the project the way in which polymorphisms within the RA disease region affected the function of the gene and, in turn, how this related to RA pathogenesis was investigated.
I first aimed at enhancing TNFAIP3 feasibility as a RA candidate gene by testing whether the protein was expressed in the main active site of the disease pathogenesis, the synovium. I performed an immunohistochemistry assay on synovial tissue sections from RA patients and Osteoarthritis (OA) patients. Expression of TNFAIP3 protein was detected in RA and OA synovium samples.
In addition, I sought to determine the functional role of RA associated SNPs. In the first place I focussed functional studies on the rs6920220 SNP and its seven proxies, since they showed the strongest association with RA. Bioinformatic analysis pointed to rs6927172 as the most likely functional marker. An initial luciferase reporter gene assay was performed. Three of the SNPs (rs6920220, rs6933404 and rs6927172) showed evidence of repressor activity of TNFAIP3 expression in CEMC7A T lymphoblast cells. A shortened 70 bp sequence around rs6927172 was still able to inhibit TNFAIP3 promoter activity.
Electrophoretic mobility shift assay (EMSA) showed that both alleles of rs6927172 were able to bind nuclear complexes in vitro. In addition, competition assays revealed a differential binding ability of the cytosine (C) and guanine (G) allele variants of the SNP.
By the time of this final report I was taking the experiments further in order to identify the transcription factors and protein complexes that bound rs6927172 in an allele-specific manner. For that purpose I was carrying out chromatin immunoprecipitation assays, using B lymphoblastoid cell lines from the HapMap repository. I was also determining allele-specific binding of different transcription factors, such as Ets-1 and NF-kB, to chromatin and analysing the mechanism of the regulatory effect of rs6927172 in TNFAIP3 expression through chromatin conformation capture.
The arthritis research United Kingdom epidemiology unit was involved in the ground-breaking 'Wellcome Trust Case Control Consortium' (WTCCC) study in which RA was one of seven diseases investigated using genome-wide association studies (GWAS). After human leukocyte antigen (HLA) and protein tyrosine phosphatase non-receptor type 22 (PTPN22), the next strongest association with RA was mapped to the 6q23 locus (rs6920220), in an intergenic region between the OLIG3 and TNFAIP3 genes. TNFAIP3 acted as a negative regulator of NF-kB and, therefore, was the most plausible candidate gene in the region.
The aim of the project was to follow up the results of the WTCCC RA GWAS to identify and functionally characterise a RA disease causal locus mapping to 6q23.
The project first phase was devoted to the identification of the RA causal mutation mapping to 6q23. I performed a fine mapping study of the OLIG3 and TNFAIP3 locus. I found 18 Single nucleotide polymorphisms (SNPs) significantly associated with RA. A logistic regression analysis showed that only rs6920220, rs13207033 and rs5029937 were independently associated with RA. The combination of the carriage of both risk alleles of rs6920220 and rs5029937 along with the absence of the protective allele of rs13207033 was strongly associated with RA when compared to carriage of none. This equated to an effect size of 1.50 (95% CI 1.21-1.85) compared to controls and was higher than that obtained for any SNP individually. This study was the first to show that confirmed loci from GWA studies, which conferred only a modest effect size, could harbour a significantly greater effect once the effect of additional risk variants was accounted for.
In conclusion, there were three independent SNPs associated with RA. These markers, or any of their proxies, were likely to be the causal variants in this region.
During the second phase of the project the way in which polymorphisms within the RA disease region affected the function of the gene and, in turn, how this related to RA pathogenesis was investigated.
I first aimed at enhancing TNFAIP3 feasibility as a RA candidate gene by testing whether the protein was expressed in the main active site of the disease pathogenesis, the synovium. I performed an immunohistochemistry assay on synovial tissue sections from RA patients and Osteoarthritis (OA) patients. Expression of TNFAIP3 protein was detected in RA and OA synovium samples.
In addition, I sought to determine the functional role of RA associated SNPs. In the first place I focussed functional studies on the rs6920220 SNP and its seven proxies, since they showed the strongest association with RA. Bioinformatic analysis pointed to rs6927172 as the most likely functional marker. An initial luciferase reporter gene assay was performed. Three of the SNPs (rs6920220, rs6933404 and rs6927172) showed evidence of repressor activity of TNFAIP3 expression in CEMC7A T lymphoblast cells. A shortened 70 bp sequence around rs6927172 was still able to inhibit TNFAIP3 promoter activity.
Electrophoretic mobility shift assay (EMSA) showed that both alleles of rs6927172 were able to bind nuclear complexes in vitro. In addition, competition assays revealed a differential binding ability of the cytosine (C) and guanine (G) allele variants of the SNP.
By the time of this final report I was taking the experiments further in order to identify the transcription factors and protein complexes that bound rs6927172 in an allele-specific manner. For that purpose I was carrying out chromatin immunoprecipitation assays, using B lymphoblastoid cell lines from the HapMap repository. I was also determining allele-specific binding of different transcription factors, such as Ets-1 and NF-kB, to chromatin and analysing the mechanism of the regulatory effect of rs6927172 in TNFAIP3 expression through chromatin conformation capture.