Systematic identification of disease genes for alopecia areata
Final Report Abstract
This project aimed to identify new causative genes for the genetically complex inherited disorder AA, circumscribed hair loss), with the aim of gaining a comprehensive understanding of AA. We have the largest sample of AA patients available worldwide, which includes a current total of almost 2,000 individuals of middle European origin. We have been able to demonstrate the contribution of the HLA-complex and the genes PTPN22, TRAF1/C5, CTLA4, IL-2RA, and the TNF/LTA locus to disease risk using candidate gene studies. In a genome-wide association study (GWAS) with pooled DNA, we identified SPATA5 as a potential susceptibility gene for AA. In order to continue the systematic identification of disease genes for AA, we performed a GWAS of 800 patients with a positive family history and/or with a severe disease course (Sample I) using the genotyping platform at the Life & Brain Center, Bonn, and the Illumina HumanOmni1-Quad BeadChip. Our other sample of 800 patients was genotyped on Illumina Human660W-Quad BeadChips, within the context of our collaboration with Professor A.M. Christiano (Columbia University, New York) (Sample II). After performing the genotyping procedures and applying state-of-the art quality criteria, statistical analyses have been performed. In single marker analysis we identified 148 SNPs (Sample I) and 12 variants (Sample II) with P-values ≤ 5 x 10-8. All from the markers identified in Sample I and 8 from Sample II were localised to the major histocompatibility (HLA) region on chromosome 6p21. We suspected that the remaining four variants from Sample II were a result of genotyping artifacts and did not pursue them further. For the meta-analysis, the data of our combined samples (Samples I + II) were analysed together with data from more than 1,000 US-American patients. In a first step, we obtained 52 variants with genome-wide significance, among them 41 in the HLA-region that we decided not follow-up further in this initial phase. The other 11 variants had been observed in the former US-GWAS or in another recent laboratory study from our group. As a large number of variants showed borderline significance, both groups (we, and the US group) decided to perform a replication step including additional patients that had not been included in the initial GWAS. With this replication step, we found the so far unknown variant rs3789129 with a genome-wide significance of P = 1.51 x 10-8, located in an intronic region of the gene ACOXL on chromosome 2. ACOXL was formerly identified in GWA-studies of other diseases and is known to be involved in acyl-CoA dehydrogenase activity. Five additional variants showed borderline significance (P-values < 5 x 10-7 > 5 x 10-8), among these, were variants in close vicinity to known autoimmune genes, e.g., IRF4, PTPN22, and TNFRSF1A. Recently, the so-called “immunochip” was developed and promising results have been reported for other autoimmune diseases. We therefore decided to genotype 778 patients with AA on the Illumina Immunochip. In a first analysis, no genome-wide significant results were obtained. Therefore, we performed a replication study including 66 variants with borderline significance, using a sample of 1,053 AA patients and 1,200 controls. The calculations for the replication study are currently ongoing; however, preliminary data point to a new genome-wide significant association at the gene locus for the variant rs4916209 with a P-value of 4.63 x 10-8. This variant is not localised within a gene, being located approximately 20 kb upstream of TNFS4. This gene is a cytokine and a member of the TNF ligand family. It is involved in T cell antigen-presenting cell interactions. Further functional studies are required to clarify the influence of this variant on TNFS4. Recently, the first GWAS of alopecia areata (AA) was conducted in a North-American sample. We therefore aimed to perform a follow-up association analysis to confirm five of these eight loci (SNPs from three regions were not included as they had been identified in previous results from our laboratory) and to test 12 other loci from the GWAS, which did not surpass the threshold for genome-wide significance. Association was confirmed for all of the five loci with previously reported genome-wide significance. To detect robust evidence of association among the 12 other loci, a meta-analysis of the present association data and those of the recent GWAS was performed. Genome-wide significant association was found for rs20541 (Pcomb = 7.52 x 10-10; odds ratio (OR) = 1.30 (1.23–1.38)) and rs998592 (Pcomb = 1.11 x 10-11; OR = 1.28 (1.21–1.36)), thus establishing IL-13 and KIAA0350/CLEC16A as new susceptibility loci for AA with genomewide significance. Interestingly, both genes are known as susceptibility loci for other autoimmune diseases, supporting the hypothesis of shared pathways for autoimmune susceptibility. Further genetic analyses concerning the HLA-region and using immunochips, in addition to functional studies and enlargement of GWAS datasets, are planned for the future to provide even deeper insights into the complex pathomechanisms of alopecia areata.
Publications
- A follow-up study of a genome-wide association scan in alopecia areata: replication of previously identified loci and identification of IL13 and KIAA0350 as new susceptibility loci supported with genome-wide significance. J Invest Dermatol 132:2192-2197, 2012
Jagielska D, Redler S, Brockschmidt FF, Herold C, Garcia Bartels N, Hanneken S, Eigelshoven S, Refke M, Barth S, Giehl KA, Kruse R, Lutz G, Wolff H, Blaumeiser B, Böhm M, Blume-Peytavi U, Becker T, Nöthen MM, Betz RC
- Genome-wide pooling approach identifies SPATA5 as a new susceptibility locus for alopecia areata. Eur J Hum Genet 20:326-332, 2012
Forstbauer LM, Brockschmidt FF, Moskvina V, Herold C, Redler S, Herzog A, Hillmer AM, Meesters C, Heilmann S, Albert F, Alblas M, Hanneken S, Eigelshoven S, Giehl KA, Jagielska D, Blume-Peytavi U, Garcia Bartels N, Kuhn J, Hennies HC, Goebeler M, Jung A, Peitsch WK, Kortüm A-K, Moll I, Kruse R, Lutz G, Wolff H, Blaumeiser B, Böhm M, Kirov G, Becker T, Nöthen MM, Betz RC
(See online at https://doi.org/10.1038/ejhg.2011.185) - Investigation of selected cytokine genes suggests that IL-2RA and the TNF/LTA locus are risk factors for severe alopecia areata. Br J Dermatol 167:1360-1365, 2012
Redler S, Albert F, Brockschmidt FF, Herold C, Hanneken S, Eigelshoven S, Giehl KA, Kruse R, Lutz G, Wolff H, Blaumeiser B, Böhm M, Becker T, Nöthen MM, Betz RC
