Gene/locus and variation/allele
Population
Cases
Controls
Odds ratio
P-Value
Reference
HLA-B (rs3099844)
Norwegian
285 PSC
298
4.8
2.6 × 10−26
Karlsen et al. [19]
GPC6 (rs9524260)
Norwegian
285 PSC
298
0.67
8.1 × 10−04
Karlsen et al. [19]
MST1 (rs3197999)
aFrom four geographic regions
1,740 PSC
5,136
1.39
1.1 × 10−16
Melum et al. [20]
BCL2L11 (rs6720394)
aFrom four geographic regions
1,740 PSC
5,136
1.29
4.1 × 10−08
Melum et al. [20]
FUT2 (rs602662)
–
1,936 PSC
6,470
Not available
1.9 × 10−06
Folseraas et al. [21]
MMEL1 (rs3748816)
European Ancestry
3,789 PSC
25,079
1.21
7.41 × 10−12
Liu et al. [22]
2q33 (rs7426056)
European Ancestry
3,789 PSC
25,079
1.3
1.89 × 10−20
Liu et al. [22]
MST1 (rs3197999)
European Ancestry
3,789 PSC
25,079
1.33
2.45 × 10−26
Liu et al. [22]
4q27 (rs13140464)
European Ancestry
3,789 PSC
25,079
1.3
8.87 × 10−13
Liu et al. [22]
6q15 (rs56258221)
European Ancestry
3,789 PSC
25,079
1.23
8.36 × 10−12
Liu et al. [22]
10p15 (rs4147359)
European Ancestry
3,789 PSC
25,079
1.24
8.19 × 10−17
Liu et al. [22]
PPP2R1B (rs7937682)
European Ancestry
3,789 PSC
25,079
1.17
3.17 × 10−09
Liu et al. [22]
HDAC7 (rs11168249)
European Ancestry
3,789 PSC
25,079
1.15
5.49 × 10−09
Liu et al. [22]
SH2B3 (rs3184504)
European Ancestry
3,789 PSC
25,079
1.18
5.91 × 10−11
Liu et al. [22]
CD226 (rs1788097)
European Ancestry
3,789 PSC
25,079
1.15
3.06 × 10−08
Liu et al. [22]
PRKD2 (rs60652743)
European Ancestry
3,789 PSC
25,079
1.25
6.51 × 10−10
Liu et al. [22]
21q22 (rs2836883)
European Ancestry
3,789 PSC
25,079
1.28
3.19 × 10−17
Liu et al. [22]
MMP-2 (rs243865)
–
132 PSC
Not available
Not available
0.031
Liu et al. [22]
GPR35 (rs3749171)
Northern European Ancestry
1,012 PSC
11,659
1.39
3.0 × 10−09
Ellinghaus et al. [23]
TCF4 (rs1452787)
Northern European Ancestry
1,012 PSC
11,659
0.75
2.61 × 10−08
Ellinghaus et al. [23]
TGR5 (rs11554825)
Norwegian
1,109 PSC
3,593
1.14
0.010
Hov et al. [24]
KIAA1109 (rs13151961)
Caucasian
41 PSC
1,487
0.34
0.013
Stallhofer et al. [25]
KIAA1109 (rs13119723)
Caucasian
41 PSC
1,487
0.40
0.023
Stallhofer et al. [25]
IL21-AS1 (rs6840978)
Caucasian
41 PSC
1,487
0.46
0.043
Stallhofer et al. [25]
HLA-DRB1*0405
Japanese
40 AIP
201
4.97
2.9 × 10−06
Kawa et al. [26]
HLA-DQB1*0401
Japanese
40 AIP
201
5.12
2.0 × 10−06
Kawa et al. [26]
HLA-DRB1*0405
Japanese
43 AIP
213
3.20
6.3 × 10−05
Ota et al. [27]
HLA-DQBI*0401
Japanese
43 AIP
213
3.29
4.6 × 10−05
Ota et al. [27]
ABCF1 (C3-2-11 microsatellite, allele 219)
Japanese
43 AIP
213
2.96
0.0076
Ota et al. [27]
FCRL3-110
Japanese
59 AIP
97
7.45
0.012
Umemura et al. [28]
CTLA4 (49A)
Chinese
46 AIP
200
7.20
0.0001
Chang et al. [29]
CTLA4 (−318C/+49A/CT60G)
Chinese
46 AIP
200
8.53
0.001
Chang et al. [29]
CTLA4 (+6230 G/G)
Japanese
59 AIP
102
2.48
0.011
Umemura et al. [30]
HLA-A2
Caucasian
19 MF
21,086
3.32
0.027
Peebles et al. [31]
HLA-B*08
Italian
35 CP
350
3.08
0.026
Martorana et al. [4]
HLA-DRB1*03
Italian
35 CP
350
3.18
0.0012
Martorana et al. [4]
CCL11 (haplotype TTCCAT)
Italian
142 CP
214
NA
0.00048
Mangieri et al. [32]
CCR5 (Δ32)
Italian
100 CP
180
2.8
0.017
Boiardi et al. [33]
DRB1*1501
–
Two Identical siblings (IAC)
No controls
Not available
Not available
Dastis et al. [10]
The first GWAS in PSC was performed in 2010, genotyping 443,816 SNPs in 285 Norwegian patients and 298 controls. Results from this discovery cohort were replicated in subjects from Scandinavia (137 cases and 368 controls), Belgium/the Netherlands (229 cases and 735 controls), and Germany (400 cases and 1832 controls). The strongest association identified was close to the HLA-B locus (SNP rs3099844). In non-HLA regions, genes involved in bile homeostasis and other inflammatory conditions were demonstrated to be associated with PSC [19].
In 2011 a GWAS of 2,466,182 genotyped SNPs in PSC patients was performed. The population studied consisted of 715 patients and 2,962 controls, followed by replication in 1,025 cases and 2,174 controls. No HLA regions were identified as disease associated, while non-HLA associations at rs3197999 in MST1 and rs6720394 near BCL2L11 were demonstrated [20].
After the first GWASs, replication studies of the most significant SNPs in other cohorts have been performed. In a single study, 45 SNPs in 1221 PSC cases and 3508 controls were analyzed. The results have been compared in a meta-analysis with previous studies, with 1936 PSC cases and 6470 controls. Furthermore, the researchers analyzed bile microbial community composition in 39 PSC patients by 16S rRNA sequencing. FUT2 gene rs602662 SNP in the chromosome 19q13 was able to influence susceptibility to infectious agents. As a result, multiple new PSC risk loci were demonstrated [21].
In 2013, in a large multicenter study, 3,789 PSC cases of European ancestry and 25,079 controls were compared, genotyping 130,422 SNPs using Immunochip platform. Twelve genome-wide significant associations outside the HLA complex were demonstrated; association of nine of them was described for the first time, increasing the number of the associated loci. Six of the 12 loci are shared with inflammatory bowel disease, suggesting an overlap between these two disorders [22].
Genetic studies performed in PSC have suggested to stratify patients in four pathogenic groups, depending on inflammation, cholangiocyte function, fibrosis, and carcinogenesis. This subclassification promises to be an important tool in the management of patients in the clinic and in the development of new potential treatments [34].
Based on the data that increased IgG4 serum levels have been reported in 9–15 % of patients with PSC, a genetic analysis of the HLA complex was performed in patients from Norway, Sweden, and the United States. The authors found that in patients with low IgG4 level, a diminished frequency of the HLA-B*08 and increased frequencies of HLA-B*07 and HLA-DRB1*15 were shown, suggesting HLA complex may influence the expression of IgG4 levels [35].
Autoimmune pancreatitis (AIP), a distinct disease entity whose type 1 subtype is characterized by high serum immunoglobulin G4 concentrations and tissue infiltration by IgG4+ plasma cells, has been investigated in Japanese patients. HLA-A, −B, −C, −DR, and -DQ gene typing and HLA-DRB1, −DQB1, and −DPB1 allele typing were performed in 40 patients compared with 201 controls. The authors found a significant association with DRB1*0405 and DQB1*0401 alleles and with the DRB1*0405-DQB1*0401 haplotype [26].
The same authors, 5 years later, investigated more deeply the HLA region in 43 AIP Japanese patients and 213 controls. The HLA-associated regions were restricted to HLA-DRB1*0405-DQB1*0401 in the HLA class II and ABCF1 proximal to C3-2-11, telomeric of HLA-E in the HLA class I regions [27].
In 2010, the role of single HLA genes in the development of AIP in transgenic mice was studied. HLA-DR*0405 transgenic mice developed AIP, whereas HLA-DR*0401, HLA-DQ8, and HLA-DR*0405/DQ8 transgenic controls did not. At the tissue level, the pancreas in HLA-DR*0405 transgenic mice showed destructive infiltration of the exocrine tissue with CD4(+) and CD8(+) T cells, B cells, and macrophages. Furthermore, mice with complete pancreatic atrophy lost weight, developed fatty stools, and had reduced levels of serum lipase activity. Finally, HLA-DR*0405 expression failed to protect mice from AIP. As a result, the HLA-DRB1*0405 allele was suggested to represent an important risk factor for AIP on the HLA-DRB1*0405-DQB1*0401 haplotype [36].
In a study of 40 AIP patients, genetic predictors of relapse were evaluated. The substitution of aspartic acid with another residue in position 57 of HLA-DQB1 showed a significant association with relapse of AIP, concluding that DQB57 may be an important genetic factor for relapse of AIP [37].
In chronic periaortitis (CP), a disease spectrum encompassing idiopathic retroperitoneal fibrosis and inflammatory abdominal aortic aneurysms, genetic association studies are difficult to perform because of the low incidence of the disease. The largest HLA study was performed in 35 Italian patients compared with 350 controls [4]. The results showed an increased HLA-DRB1*03 and HLA-B*08 allele frequencies in CP patients as compared with controls. These two alleles are linked to several autoimmune disorders, adding further evidence to the autoimmune origin of CP. This study needs to be replicated in larger populations, also in order to stratify the disease in different subsets.
In addition, a study that analyzed the perianeurysmal forms of the disease (i.e., inflammatory abdominal aortic aneurysms) found the HLA-DRB1 as a genetic risk factor. The HLA-DRB1*15 and DRB1*0404 alleles were described as predisposing alleles [38, 39]. Taken together, these findings provide evidence for a role of the HLA system in conferring susceptibility to CP.
The HLA locus was also investigated in mediastinal fibrosis. Nineteen Caucasian patients were compared with 21,086 white controls (kidney donors). The authors found a statistically significant association with the HLA-A2 gene. Several diseases have been associated with this gene, such as rheumatoid arthritis, juvenile rheumatoid arthritis, and psoriatic arthritis. Given the high prevalence of the HLA-A2 in the general population, the predictive value of this finding in the susceptibility to the disease is low [31].
1.5 Associations with Non-HLA Genetic Variants
Non-HLA genes have been poorly investigated in FIDs, because of the genetic heterogeneity of these diseases (Table 1.1).
In PSC, several genetic polymorphisms have been investigated. Matrix metalloproteinase-2 (−1306 C/T) and MMP-9 (−1562 C/T) gene promoter polymorphisms have been investigated in 132 PSC patients in a recent study, relating them to the severity of the disease [40]. MMP-9 genotype was demonstrated not to be associated with disease severity in PSC, while MMP-2 1306C>T gene promoter polymorphism has been postulated as an independent risk factor for PSC disease severity.
In a large GWAS, the authors investigated a first cohort of 392 PSC patients and 987 ulcerative colitis (UC) patients, compared with 3000 controls; the results were replicated in a cohort of 1,012 PSC patients, 4,444 UC patients, and 11,659 controls. Two novel SNPs were found to be disease associated: rs3749171 at 2q37 located in the G-protein-coupled receptor 35 (GPR35) gene and rs1452787 at 18q21 located in the transcription factor 4 (TCF4) gene. The first SNP causes a threonine to methionine amino acid change. Structural model showed that this SNP may alter the efficiency of signaling through the GPR35 receptor. GPR35 shows associations with both UC and PSC, while TCF4 SNP is a PSC risk locus not associated with UC. This study elucidated the difference in the genetic background between PSC and UC [23].
Based on the findings of previous studies, the TGR5 gene was investigated in PSC patients. TGR5 is a G-protein-coupled bile acid receptor 1 (GPBAR1), already linked to inflammatory pathways. The authors sequenced the TGR5 gene in 267 PSC patients and 274 healthy controls. Furthermore, they functionally analyzed the discovered variants, finding six nonsynonymous DNA variants linked to the disease. In a functional study using confocal microscopy, flow cytometry, and a cAMP-sensitive luciferase assay, a receptor model was studied, introducing the mutated TGR5 constructs into human epithelial cell lines. Five nonsynonymous variants (W83R, V178M, A217P, S272G, and Q296X) were found to reduce or abolish TGR5 function. This study demonstrated that TGR5 gene may represent a strong candidate gene for PSC [24]. The TGR5 gene maps to the chromosome 2q35, close to the rs3749171 SNP (located in 2q37), previously associated in a large GWAS [41].
The IL-2/IL-21 region was also investigated in PSC, because this region has been previously associated with UC and several autoimmune diseases. Four SNPs in the KIAA1109/TENR/IL-2/IL-21 linkage disequilibrium block were genotyped in 41 PSC subjects. The minor alleles of all four markers were associated with a decreased susceptibility to PSC. A haplotype of the four major alleles was independently associated with PSC when excluding the patients with concomitant inflammatory bowel disease. The authors concluded the IL-2/IL-21 region may be a susceptibility factor of PSC [25].
Another investigated condition was AIP. A group of genes called Fc receptor-like genes (FCRLs), which have high structural homology with classical Fc receptor genes, has been shown to be associated with several autoimmune diseases, such as rheumatoid arthritis, autoimmune thyroid disease, and systemic lupus erythematosus in Japanese populations; the authors found the FCRL3-110 alleles are related to susceptibility for AIP [28].
Another gene role investigated in AIP was cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) gene, which is a negative regulator of the T-cell immune response; the gene is highly polymorphic. Many positive associations between CTLA4 SNP and various autoimmune diseases have been identified. In a study of Chinese patients, 46 patients with AIP were investigated and the CTLA4 49A variant was found to be a susceptibility factor in AIP patients [29]. In another study, the authors replicated the association with CTLA4 gene [30]. Another gene investigated in Japanese AIP patients has been the protease serine 1 (PRSS1). Two novel mutations (p.81Leu → Met and p.91Ala → Ala) were found in PRSS1 gene, suggesting that this gene is also linked to the susceptibility to AIP [42].
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
