Potential conflict of interest: Nothing to report.
Genome-wide association studies in primary sclerosing cholangitis: Still more questions than answers?†
Article first published online: 25 MAY 2011
Copyright © 2011 American Association for the Study of Liver Diseases
Volume 53, Issue 6, pages 2133–2135, June 2011
How to Cite
Chapman, R. (2011), Genome-wide association studies in primary sclerosing cholangitis: Still more questions than answers?. Hepatology, 53: 2133–2135. doi: 10.1002/hep.24333
- Issue published online: 25 MAY 2011
- Article first published online: 25 MAY 2011
- Manuscript Accepted: 27 MAR 2011
- Manuscript Received: 25 MAR 2011
Melum E, Franke A, Schramm C, Weismüller TJ, Gotthardt DN, Offner FA, et al. Genome-wide association analysis in primary sclerosing cholangitis identifies two non-HLA susceptibility loci. Nat Genet 2011;43: 17-19. (Reprinted with permission.)
Primary sclerosing cholangitis (PSC) is a chronic bile duct disease affecting 2.4-7.5% of individuals with inflammatory bowel disease. We performed a genome-wide association analysis of 2,466,182 SNPs in 715 individuals with PSC and 2,962 controls, followed by replication in 1,025 PSC cases and 2,174 controls. We detected non-HLA associations at rs3197999 in MST1 and rs6720394 near BCL2L11 (combined P = 1.1 × 10−16 and P = 4.1 × 10−8, respectively).
The etiopathogenesis of primary sclerosing cholangitis (PSC) remains unknown, although it is now accepted that genetic factors play a major role in the development of the disease.1 First-degree relatives have an 80-fold–increased risk of developing PSC.2 Moreover, studies that were first carried out 30 years ago established that there are close associations with the human leukocyte antigen (HLA) complex on chromosome 6p21.3 Surprisingly, the exact gene or genes responsible for the association in this highly polymorphic region have not been identified.1, 2 The heritability of PSC has an estimated relative sibling risk of approximately 10, which is in the range of other HLA-associated conditions.2 PSC is likely to be a complex disease in which different environmental factors interact with multiple genetic factors and contribute to both the pathogenesis and progression of this chronic cholestatic biliary disease.
PSC is characterized by a close association with inflammatory bowel disease and particularly ulcerative colitis, which coexists in approximately three-quarters of Northern European patients with PSC.3 In addition, approximately 5% to 10% of patients with total ulcerative colitis will have or will develop PSC during the course of their illness. Intriguingly, the clinical phenotype of ulcerative colitis associated with PSC (inflammatory bowel disease with PSC) exhibits significant differences with the ulcerative colitis phenotype without PSC,4 and this raises the possibility of significant genotypic differences between the patient groups.
A recent meta-analysis of six genome-wide association study (GWAS) data sets for ulcerative colitis compared 6687 ulcerative colitis patients with 19,718 controls.5 The report identified 29 additional risk loci not previously identified for ulcerative colitis and thereby increased the number of ulcerative colitis– associated loci to 47. The authors documented that the number of confirmed risk loci in inflammatory bowel disease is 99; this number includes at least 28 association signals shared by ulcerative colitis and Crohn's disease.
In contrast, GWASs of liver disease in general and PSC in particular are in their infancy.6 A recent GWAS in Northern European patients with PSC from the Norwegian PSC group confirmed that the strongest associations are on chromosome 6p21, which is closest to the HLA-B region.7 These HLA associations appear to be specific to PSC.
For PSC, just as for the majority of HLA-associated diseases, significant associations have been demonstrated outside the HLA complex with a gene present on chromosome 13q31. However, this gene is not specific to PSC because significant associations at this locus have been found for ulcerative colitis and multiple sclerosis. This suggests that the gene may be involved in the pathogenesis of inflammation rather than specific disease susceptibility. Further significant associations that were also previously established as ulcerative colitis susceptibility loci were detected on chromosomes 2q35 and 3p21.7 The authors suggested G-protein–coupled bile acid receptor 1 and macrophage stimulating 1 (MST1), respectively, as the likely genes involved in the disease process.
In the latest study,8 the same group repeated and extended the GWAS by increasing the size of the Scandinavian and German study population and the replication cohort to 715 and 1025 PSC patients, respectively, although the numbers of patients included in the study are still relatively small in comparison with the much larger numbers included in GWASs of inflammatory bowel disease.
In agreement with the original study, the strongest associations were detected in the HLA complex. An analysis revealed a complex association signal in the class II region and confirmed the strong association with the HLA-B*08 locus in the class I region. This suggests multiple causative loci within the region. Clear differences were found in the HLA complex in comparison with ulcerative colitis (for which the association signal is less extensive), and associated single-nucleotide polymorphisms (SNPs) were observed near the HLA class II region. This is a potentially important observation that requires further study in the search for the elusive PSC susceptibility gene or genes in this region.
Outside the HLA complex, the findings of the first study were confirmed and extended. Multiple SNPs in strong linkage equilibrium at chromosome 3p21 were associated at a genome-wide significance level.8 Further analysis using the replication cohort demonstrated the most prominent association at MST1. The amino acid change at this locus, which is now associated with PSC and was previously demonstrated in patients with ulcerative colitis and Crohn's disease,9 has been proposed to affect the MST1 receptor interaction. MST1 is known to encode macrophage stimulating protein, which regulates innate immune responses to bacterial ligands. The variant (rs3197999, R689C) identified in this study and previously in patients with inflammatory bowel disease9 has been predicted to interfere with the binding of macrophage stimulating protein to its receptor. The MST1 protein is known to be expressed at high levels in gallbladder epithelium; it is possible, therefore, that the disease-associated variant may influence biliary and intestinal inflammation in these diseases. This potentially important finding clearly warrants further study.
In addition, a replicated associated signal outside the HLA complex was demonstrated on chromosome 2q13 by multiple SNPs encompassing the B cell lymphoma 2–like 11 (BCL2L11) locus. BCL2L11 is known to encode the B cell lymphoma 2 interacting protein (Bim), which has a major influence on the maintenance of immune tolerance.10 Bim is a member of an apoptotic subgroup of proteins called BH3-only proteins, which are critical to the initiation of apoptosis in response to many death stimuli.10, 11 The protein induces tolerance by the regulation of the negative selection of B lymphocytes in the bone marrow and by the induction of the apoptosis of autoreactive T lymphocytes and the deletion of activated T lymphocytes after an immune response in both the thymus and the periphery.10, 11 Mice lacking Bim may develop a systemic lupus erythematosus–like autoimmune disease.11 Preliminary studies in Bcl2l11−/− mice demonstrated the infiltration of mononuclear cell around some intrahepatic bile ducts, which were not seen in wild-type mice.8 The potentially important role of this regulatory protein in the immune response of the liver and biliary system in patients with PSC remains to be established.
Although it was not significantly replicated, a highly significant association was shown at the interleukin-2 receptor alpha locus for several SNPs that were previously shown to influence the risk of developing type 1 diabetes and multiple sclerosis.
What does the future hold for GWASs of PSC? The evidence from the GWAS results for inflammatory bowel disease has shown that “one GWAS is never enough.”6 At least two larger GWASs of PSC are currently in progress in the United Kingdom and United States, and they may detect unsuspected loci and confirm previous findings. However, it now seems unlikely that the specific disease susceptibility gene or genes for PSC will be detected by GWASs. The strength of GWASs lies in the detection of genetic variants commonly found in the general population (>5%); these can be used to detect general disease genes influencing, for example, the immune response and carcinogenesis and hence can provide potentially important insights into the pathogenesis of disease.
After 3 decades of research, the region of the HLA complex on chromosome 6p21 still appears to be the most likely site of genetic susceptibility. Other techniques that are designed to detect uncommon genes and to explore the HLA complex in more detail, such as exome sequencing, may prove to more rewarding in the future.