Meta-Analysis of the Relation Between IL10 Promoter Polymorphisms and Autoimmune Liver Disease Risk

Abstract

Background:

Single nucleotide polymorphisms of the IL10 gene have been linked to the occurrence of autoimmune liver disease.

Methods:

We performed a meta-analysis to assess the association between three IL10 promoter polymorphisms (rs1800896, rs1800871, and rs1800872) and the risk of autoimmune hepatitis, primary biliary cholangitis, and primary sclerosing cholangitis.

Results:

In total, 1420 articles were initially identified through database retrieval. After screening, seven eligible articles were ultimately included in the meta-analysis. A fixed-effect model was used for all Mantel-Haenszel statistics due to the absence of large between-study heterogeneity (all I² < 50%, p > 0.1). No association between any of the studied polymorphisms and risk of autoimmune liver disease was detected in the allele, homozygote, heterozygote, dominant, recessive, or carrier genetic models (p_association > 0.05). Potential publication bias was excluded using Begg's and Egger's tests. Similar negative results were observed in subgroup analyses and in an analysis of the three haplotypes of rs1800896/rs1800871/rs1800872 (G/C/C, A/C/C, and A/T/A).

Conclusion:

Our meta-analysis strongly suggests that the IL10 rs1800896, rs1800871, and rs1800872 polymorphisms are not associated with the risk of autoimmune liver disease.

Introduction

Autoimmune liver diseases, such as autoimmune hepatitis (AIH), primary biliary cholangitis (PBC), and primary sclerosing cholangitis (PSC), are chronic liver ailments related to responses of the body's own immune system (Karlsen and Chung, 2015; Grant and Liberal, 2017). AIH is characterized by high levels of aminotransferases, circulating autoantibodies, hypergammaglobulinemia, and interface hepatitis (van Gerven et al., 2016; Aizawa and Hokari, 2017); PBC is characterized by autoimmune-associated inflammation and destruction within intrahepatic bile ducts (Floreani et al., 2017; Joshita et al., 2017); and PSC often results in the cirrhosis or liver failure (Kovac and Weber, 2016; Jiang and Karlsen, 2017). Both genetic and environmental factors contribute to the occurrence and progression of autoimmune liver disease (Karlsen and Chung, 2015). Related genes include human leukocyte antigen (HLA), cytotoxic T lymphocyte antigen-4 (CTLA-4), and protein tyrosine phosphatase N22 (PTPN22) (de Boer et al., 2014; Eskandari-Nasab et al., 2015; Umemura et al., 2016).

Interleukin 10, encoded by IL10 located on chromosome 1 (1q31-32) (Eskdale et al., 1997), is a multifunctional cytokine that has been implicated in various biological events and diseases (Walter, 2014; Saxena et al., 2015). Single nucleotide polymorphisms (SNPs) are variations in a single nucleotide within a gene exon or intron and may result in alteration of a protein's structure or function (Benitez et al., 2017; Matsuda, 2017). Three commonly occurring human IL10 SNPs, rs1800896 (-1082A/G), rs1800871 (-819 T/C), and rs1800872 (-592 T/C), are situated within the gene's promoter region (Matsushita et al., 2002; Liu et al., 2015; Liu and Zheng, 2016). These IL10 SNPs are reportedly associated with susceptibility to several clinical diseases, such as tuberculosis (Liu et al., 2015), type 2 diabetes mellitus (Tarabay et al., 2016), liver cancer (Wei et al., 2011), or autoimmune thyroid disease (AITD) (Jung et al., 2016). In the present study, we performed a meta-analysis to investigate the genetic relation between those IL10 SNPs and the risk of autoimmune liver disease. To our knowledge, no such meta-analysis has been reported previously.

Materials and Methods

Database retrieval

Based on preferred reporting items for systematic reviews and meta-analyses (PRISMA) (Moher et al., 2009), we carried out retrieval from the PubMed, Embase, Scopus and Web of Science (WOS) databases on November 23, 2017, without any restriction of publication language, region, and article type. The search terms used are listed in Supplementary Table S1 (Supplementary Data are available online at www.liebertpub.com/gtmb).

Article screening

After excluding duplicate articles by Endnote X7 software (Thomson Corporation, Thomson ResearchSoft), we further removed articles using the following exclusion criteria: (1) cell or animal experiments; (2) review articles or case studies; (3) meeting abstracts; (4) other genes/diseases; (5) meta-analyses; (6) unselected SNPs; (7) lack of genotype data. The included studies should report the genetic association between IL10 SNPs and autoimmune liver disease risk, and contain sufficient data of genotype frequency distribution in case and control groups.

Basic information

The specific data were separately collected from the eligible articles by three investigators and summarized as basic information, including first name, publication year, ethnicity, SNP, case/control number, specific type of autoimmune liver disease, age, gender, genotyping assay, source of control, and p-value for Hardy-Weinberg equilibrium (p_HWE) in control group. The genotype frequency distribution was considered to be in Hardy-Weinberg equilibrium, when p_HWE was larger than 0.05.

Quality assessment

We also assessed the quality of the studies using the Newcastle-Ottawa Scale (NOS) (Stang, 2010; Ye et al., 2016). The following items were evaluated: (1) Case Definition; (2) Representativeness of cases; (3) Selection of controls; (4) Definition of controls; (5) Important factors of comparability; (6) Other factors of comparability; (7) Secure record of exposure; (8) Same method of ascertainment; (9) Nonresponse rate. NOS scores larger than 7 indicate high quality. We will fully discuss the uncertain or controversial points.

Statistical analysis

STATA software (version 12.0; Stata Corporation, College Station, TX) was used for analysis. P_{heterogeneity} values derived from Cochran's Q statistic >0.1 or I² value <50% indicated the application of a fixed-effect model. Otherwise, a random-effect model was used.

Odds ratios, 95% confidence intervals, and p_association were obtained from Mantel-Haenszel statistics under the allele (G vs. A for rs1800896, C vs. T for rs1800871, C vs. A for rs1800872), homozygote (GG vs. AA, CC vs. TT, CC vs. AA), heterozygote (AG vs. AA, TC vs. TT, AC vs. AA), dominant (AG+GG vs. AA, TC+CC vs. TT, AC+CC vs. AA), recessive (GG vs. AA+AG, CC vs. TT+TC, CC vs. AA+AC), and carrier (G vs. A, C vs. T, C vs. A) models. A statistical significant difference between case and control was considered, when p_association was less than 0.05.

Begg's and Egger's tests were used to assess the potential publication bias. p-Value for Begg's and Egger's tests larger than 0.05 indicated the absence of publication bias. Subgroup meta-analyses based on ethnicity (Caucasian or Asian), p_HWE (>0.05 or <0.05), and disease type (AIH, PBC, or PSC) were also performed under the allele, homozygote, heterozygote, dominant, recessive, and carrier models, respectively.

Results

Identification of eligible case-control studies

Figure 1 shows our flow chart for identifying eligible case-control studies. In total, 1420 articles were initially identified through comprehensive retrieval from the PubMed (n = 228), Embase (n = 284), Scopus (n = 254), and WOS (n = 654) databases. We then excluded 515 articles as duplicates and another 896 articles based on our exclusion criteria. Evaluation of the full text of the remaining nine articles led to the removal of additional two articles due to unselected SNPs (Mitchell et al., 2001) or lack of genotype data (Ma and Qiu, 2004). As a result, seven articles (Zappala et al., 1998; Cookson et al., 1999; Bathgate et al., 2000; Donaldson et al., 2000; Matsushita et al., 2002; Chen et al., 2004; and Fan et al., 2005) were ultimately selected for meta-analysis. From the eligible articles, we obtained 11 case-control studies for rs1800896, 8 for rs1800871, and 9 for rs1800872. Table 1 and Supplementary Table S2 show the characteristics of the participants in the included studies and their genotype distributions. Notably, we found that the genotype distributions for some control groups of two studies were not in line with Hardy-Weinberg equilibrium (Supplementary Table S2, p_HWE <0.05) (Bathgate et al., 2000; Matsushita et al., 2002). Summarized in Supplementary Table S3 are results from NOS system evaluations, which showed that all the included studies were of high quality (NOS score >8).

FIG. 1.

Selection process for eligible case-control studies. WOS, Web of Science.

Table 1.

Basic Information About the Studies Included in Meta-Analysis

				Case group					Control group
Study	Region	Ethnicity	SNP	Number	Type	Age ^a	Female (%)	Number	Age	Female (%)	Genotyping assay
Bathgate (2000)	United Kingdom	Caucasian	rs1800896	61	PBC	54.5 ± 9.0	NR	Taq PCR	330	NR	NR
	United Kingdom	Caucasian	rs1800896	17	PSC	46.0 ± 11.0	NR	Taq PCR	330	NR	NR
	United Kingdom	Caucasian	rs1800896	8	AIH	36.9 ± 12.0	NR	Taq PCR	330	NR	NR
Chen (2004)	China	Asian	rs1800896	54	AIH	21-74	66.67	PCR-RFLP	160	NR	NR
	China	Asian	rs1800871	54	AIH	21-74	66.67	PCR-SSP	160	NR	NR
	China	Asian	rs1800872	54	AIH	21-74	66.67	PCR-RFLP	160	NR	NR
Cookson (1999)	United Kingdom	Caucasian	rs1800896	84	AIH	16-68	69.0	PCR-SSCP	78	NR	38.0
	United States	Caucasian	rs1800896	85	AIH	13-73	78.0	PCR-SSCP	101	26-68	58.0
	United Kingdom	Caucasian	rs1800871	84	AIH	16-68	69.0	PCR-SSCP	78	NR	38.0
	United States	Caucasian	rs1800871	85	AIH	13-73	78.0	PCR-SSCP	101	26-68	58.0
	United Kingdom	Caucasian	rs1800872	84	AIH	16-68	69.0	PCR-SSCP	78	NR	38.0
	United States	Caucasian	rs1800872	101	AIH	13-73	78.0	PCR-SSCP	85	26-68	58.0
Donaldson (2000)	United Kingdom	Caucasian	rs1800896	96	PSC	15-79	32.0	PCR-SSCP	78	NR	NR
	United Kingdom	Caucasian	rs1800871	96	PSC	15-79	32.0	PCR-SSCP	78	NR	NR
	United Kingdom	Caucasian	rs1800872	96	PSC	15-79	32.0	PCR-SSCP	78	NR	NR
Fan (2005)	China	Asian	rs1800896	62	AIH	16-79 (F)/21-62 (M)	71.0	PCR-RFLP	160	NR	NR
	China	Asian	rs1800896	77	PBC	32-79 (F)/38-60(M)	90.7	PCR-RFLP	160	NR	NR
	China	Asian	rs1800871	62	AIH	16-79 (F)/21-62 (M)	71.0	NR	160	NR	NR
	China	Asian	rs1800871	77	PBC	32-79	90.7	NR	160	NR	NR
	China	Asian	rs1800872	62	AIH	16-79 (F)/21-62 (M)	71.0	PCR-RFLP	160	NR	NR
	China	Asian	rs1800872	77	PBC	32-79	90.7	PCR-RFLP	160	NR	NR
Matsushita (2002)	Italy	Caucasian	rs1800896	94	PBC	56.6 ± 1.6	84.0	PCR-RFLP	72	60.0 ± 3.0	72.2
	Japan	Asian	rs1800896	65	PBC	62.3 ± 1.9	87.7	PCR-RFLP	71	63.5 ± 3.7	59.2
	Italy	Caucasian	rs1800871	94	PBC	56.6 ± 1.6	84.0	PCR-RFLP	72	60.0 ± 3.0	72.2
	Japan	Asian	rs1800871	65	PBC	62.3 ± 1.9	87.7	PCR-RFLP	71	63.5 ± 3.7	59.2
	Italy	Caucasian	rs1800872	94	PBC	56.6 ± 1.6	84.0	PCR-RFLP	72	60.0 ± 3.0	72.2
	Japan	Asian	rs1800872	65	PBC	62.3 ± 1.9	87.7	PCR-RFLP	71	63.5 ± 3.7	59.2
Zappala (1998)	United Kingdom	Caucasian	rs1800872	171	PBC	62.8 ± 11.0	87.7	PCR-RFLP	141	NR	NR

The data of age are shown as mean ± standard deviation, or age range.

AIH, autoimmune hepatitis; F, female; M, male; NR, not reported; PBC, primary biliary cholangitis; PSC, primary sclerosing cholangitis; PCR, polymerase chain reaction; RFLP, restriction fragment length polymorphism; SNP, single nucleotide polymorphism; SSCP, single-strand conformation polymorphism; SSP, sequence specific primer.

IL10 rs1800896

In the first meta-analysis, we analyzed the relationship between the IL10 rs1800896 A/G SNP and the risk of autoimmune liver disease. Enrolled were 703 cases and 1870 controls. As shown in Supplementary Table S4, there was not a high degree of heterogeneity across the studies (all I² < 50%, p_{heterogeneity} >0.1), and a fixed-effect model was used for all Mantel-Haenszel statistics. Pooled data (Table 2) showed that there was no difference in the risk of autoimmune liver disease between case and control groups (allele G vs. A, p_association = 0.207; GG vs. AA, p_association = 0.388; AG vs. AA, p_association = 0.081; AG+GG vs. AA, p_association = 0.081; GG vs. AA+AG, p_association = 0.938; carrier G vs. A, p_association = 0.351).

Table 2.

Meta Analysis of the Association Between IL10 Polymorphisms and Risk of Autoimmune Liver Disease

			Association test
SNP	Genetic models	Case-control studies (number)	OR (95% CI)	p_association	Sample size Case-control
rs1800896	Allele G vs. A	11	1.11 (0.94-1.31)	0.207	703/1870
	GG vs. AA	7	1.17 (0.82-1.66)	0.388	445/1319
	AG vs. AA	11	1.25 (0.97-1.60)	0.081	703/1870
	AG+GG vs. AA	11	1.24 (0.97-1.57)	0.081	703/1870
	GG vs. AA+AG	7	1.01 (0.76-1.36)	0.938	445/1319
	Carrier G vs. A	11	1.09 (0.91-1.32)	0.351	703/1870
rs1800871	Allele C vs. T	8	1.18 (1.00-1.39)	0.056	617/880
	CC vs. TT	8	1.41 (0.95-2.11)	0.091	617/880
	TC vs. TT	8	1.17 (0.87-1.56)	0.297	617/880
	TC+CC vs. TT	8	1.23 (0.93-1.61)	0.142	617/880
	CC vs. TT+TC	8	1.23 (0.95-1.59)	0.120	617/880
	Carrier C vs. T	8	1.12 (0.93-1.35)	0.230	617/880
rs1800872	Allele C vs. A	9	1.15 (0.99-1.34)	0.070	804/1005
	CC vs. AA	9	1.55 (1.07-2.24)	0.021	804/1005
	AC vs. AA	9	1.11 (0.83-1.49)	0.472	804/1005
	AC+CC vs. AA	9	1.21 (0.92-1.60)	0.167	804/1005
	CC vs. AA+AC	9	1.18 (0.95-1.47)	0.130	804/1005
	Carrier C vs. A	9	1.09 (0.92-1.29)	0.311	804/1005

CI, confidence interval; OR, odds ratio; p_association, p-value based on Mantel-Haenszel statistics of association test.

Subgroup analysis based on ethnicity (Caucasian or Asian), p_HWE (>0.05 or <0.05), and disease type (AIH, PBC, or PSC) yielded similar results in all genetic models for all subgroups (Table 3, all p_association >0.05). The forest plot in Figure 2A shows the subgroup analysis based on disease type in a G versus A model, whereas Supplementary Figure S1 shows data for the ethnicity subgroup. These data indicate that IL10 rs1800896 is not linked to the risk of autoimmune liver disease.

FIG. 2.

Forest plot and Begg's funnel plot for subgroup analysis of IL10 rs1800896 based on disease type under a G versus A model. (A) Forest plot; (B) Begg's funnel plot. Color images available online at www.liebertpub.com/gtmb

Table 3.

Subgroup Analysis for the Association Between IL10 RS1800896 and Risk of Autoimmune Liver Disease

			Association test
Genetic models	Subgroup	Case-control studies (number)	OR (95% CI)	p_association	Sample size Case-control
Allele G vs. A	Caucasian	7	1.08 (0.90-1.29)	0.413	445/1319
	Asian	4	1.35 (0.88-2.09)	0.172	258/551
	p_HWE >0.05	8	1.14 (0.94-1.38)	0.174	617/880
	p_HWE <0.05	3	1.03 (0.75-1.42)	0.850	86/990
	AIH	5	1.02 (0.78-1.32)	0.898	293/829
	PBC	4	1.30 (1.00-1.69)	0.053	297/633
	PSC	2	0.99 (0.69-1.41)	0.942	113/408
GG vs. AA	Caucasian	7	1.17 (0.82-1.66)	0.388	445/1319
	p_HWE >0.05	4	1.22 (0.80-1.86)	0.368	359/329
	p_HWE <0.05	3	1.07 (0.57-1.99)	0.834	86/990
	AIH	3	0.94 (0.53-1.66)	0.841	177/509
	PBC	2	1.60 (0.90-2.85)	0.106	155/402
	PSC	2	0.98 (0.48-2.03)	0.965	113/408
AG vs. AA	Caucasian	7	1.19 (0.89-1.61)	0.241	445/1319
	Asian	4	1.38 (0.88-2.17)	0.158	258/551
	p_HWE >0.05	8	1.23 (0.93-2.17)	0.145	617/880
	p_HWE <0.05	3	1.32 (0.75-2.31)	0.330	86/990
	AIH	5	1.07 (0.74-1.61)	0.646	293/829
	PBC	4	1.32 (0.90-1.93)	0.156	297/633
	PSC	2	1.51 (0.81-2.85)	0.197	113/408
AG+GG vs. AA	Caucasian	7	1.19 (0.90-1.57)	0.230	445/1319
	Asian	4	1.38 (0.88-2.17)	0.158	258/551
	p_HWE >0.05	8	1.24 (0.95-1.52)	0.112	617/880
	p_HWE <0.05	3	1.22 (0.72-2.05)	0.464	86/990
	AIH	5	1.07 (0.74-1.55)	0.722	293/829
	PBC	4	1.39 (0.96-1.55)	0.077	297/633
	PSC	2	1.30 (0.72-1.55)	0.382	113/408
GG vs. AA+AG	Caucasian	7	1.01 (0.76-1.36)	0.938	445/1319
	p_HWE >0.05	4	1.08 (0.75-1.55)	0.684	359/329
	p_HWE <0.05	3	0.90 (0.54-1.48)	0.671	86/990
	AIH	3	0.95 (0.59-1.53)	0.823	177/509
	PBC	2	1.33 (0.83-2.15)	0.240	155/402
	PSC	2	0.75 (0.41-1.34)	0.328	113/408
Carrier G vs. A	Caucasian	7	1.05 (0.85-1.29)	0.643	445/1319
	Asian	4	1.33 (0.85-2.08)	0.211	258/551
	p_HWE >0.05	8	1.12 (0.90-1.39)	0.307	617/880
	p_HWE <0.05	3	1.02 (0.70-1.48)	0.917	86/990
	AIH	5	1.09 (0.91-1.32)	0.825	293/829
	PBC	4	1.22 (0.90-1.65)	0.194	297/633
	PSC	2	1.03 (0.77-1.39)	0.962	113/408

p_HWE, p-value for Hardy-Weinberg equilibrium.

Begg's and Egger's tests showed an absence of potential publication bias (Supplementary Table S4, all p_Begg >0.05, p_Egger >0.05). The Begg's funnel plot in Figure 2B presents the pseudo 95% confidence limits in the G versus A model.

IL10 rs1800871

Our meta-analysis of IL10 rs1800871 included 8 case-control studies containing 424 cases and 400 controls. A fixed-effect model was used (all I² = 0.0%, p_{heterogeneity} >0.1) for this analysis (Supplementary Table S4). As with IL10 rs1800896, there was no significant difference between the cases and population-based control group under any genetic model (Table 2, all p_association >0.05). The subsequent subgroup analyses yielded similar negative results (Table 4, Fig. 3A, and Supplementary Fig. S2, all p_association >0.05). Begg's and Egger's tests showed no meaningful publication bias (Supplementary Table S4 and Fig. 3B, all p_Begg >0.05, p_Egger >0.05). This result provides no support for a positive correlation between IL10 rs1800871 and risk of autoimmune liver disease.

FIG. 3.

Forest plot and Begg's funnel plot for subgroup analysis of IL10 rs1800871 based on disease type under a C versus T model. (A) Forest plot; (B) Begg's funnel plot. Color images available online at www.liebertpub.com/gtmb

Table 4.

Subgroup Analysis of the Association Between IL10 RS1800871 and Risk of Autoimmune Liver Disease

			Association test		Sample size
Genetic models	Subgroup	Case-control studies (number)	OR (95% CI)	p_association	Case-control
Allele C vs. T	Caucasian	4	1.17 (0.91-1.50)	0.218	359/329
	Asian	4	1.18 (0.94-1.48)	0.144	258/551
	p_HWE >0.05	7	1.17 (0.98-1.40)	0.079	552/809
	p_HWE <0.05	1	1.22 (0.73-2.03)	0.446	65/71
	AIH	4	1.13 (0.89-1.43)	0.330	285/499
	PBC	3	1.27 (0.97-1.66)	0.079	236/303
	PSC	1	1.10 (0.68-1.79)	0.700	96/78
CC vs. TT	Caucasian	4	1.40 (0.72-2.69)	0.320	359/329
	Asian	4	1.42 (0.86-2.37)	0.171	258/551
	p_HWE >0.05	7	1.41 (0.91-2.20)	0.471	552/809
	p_HWE <0.05	1	1.43 (0.54-3.73)	0.127	65/71
	AIH	4	1.32 (0.73-2.38)	0.359	285/499
	PBC	3	1.55 (0.85-2.84)	0.154	236/303
	PSC	1	1.29 (0.35-4.73)	0.706	96/78
TC vs. TT	Caucasian	4	1.20 (0.61-2.37)	0.589	359/329
	Asian	4	1.16 (0.84-1.59)	0.370	258/551
	p_HWE >0.05	7	1.18 (0.86-1.61)	0.298	552/809
	p_HWE <0.05	1	1.08 (0.51-2.31)	0.834	65/71
	AIH	4	1.01 (0.67-1.53)	0.955	285/499
	PBC	3	1.36 (0.88-2.09)	0.162	236/303
	PSC	1	1.19 (0.32-4.51)	0.794	96/78
TC+CC vs. TT	Caucasian	4	1.32 (0.69-2.52)	0.398	359/329
	Asian	4	1.21 (0.89-1.63)	0.220	258/551
	p_HWE >0.05	7	1.24 (0.92-1.67)	0.166	552/809
	p_HWE <0.05	1	1.19 (0.61-2.33)	0.617	65/71
	AIH	4	1.08 (0.73-1.60)	0.689	285/499
	PBC	3	1.40 (0.94-2.09)	0.102	236/303
	PSC	1	1.25 (0.35-4.47)	0.735	96/78
CC vs. TT+TC	Caucasian	4	1.19 (0.88-1.61)	0.266	359/329
	Asian	4	1.33 (0.82-2.16)	0.244	258/551
	p_HWE >0.05	7	1.21 (0.93-1.59)	0.157	552/809
	p_HWE <0.05	1	1.38 (0.55-3.45)	0.490	65/71
	AIH	4	1.23 (0.86-1.78)	0.259	285/499
	PBC	3	1.29 (0.82-2.03)	0.270	236/303
	PSC	1	1.10 (0.60-2.01)	0.751	96/78
Carrier C vs. T	Caucasian	4	1.12 (0.85-1.48)	0.417	359/329
	Asian	4	1.12 (0.87-1.45)	0.377	258/551
	p_HWE >0.05	7	1.12 (0.92-1.36)	0.265	552/809
	p_HWE <0.05	1	1.15 (0.63-2.11)	0.650	65/71
	AIH	4	1.09 (0.83-1.42)	0.539	285/499
	PBC	3	1.19 (0.88-1.61)	0.268	236/303
	PSC	1	1.07 (0.62-1.84)	0.810	96/78

IL10 rs1800872

Nine case-control studies with 804 cases and 1005 controls were included in the meta-analysis of IL10 rs1800872. Again, a fixed-effect model was used (Supplementary Table S4, all I² = 0.0%, p_{heterogeneity} >0.1), and no significant difference was detected between the overall case and control groups (Table 2, all p_association >0.05). Similar negative results were observed in most subgroup analyses (Table 5, Fig. 4A, and Supplementary Fig. S3, p_association >0.05), apart from the subgroup of Asian (p_association = 0.036) and AIH (p_association = 0.046) in the CC versus AA genetic model. Begg's and Egger's tests indicated an absence of publication bias (Supplementary Table S4 and Fig. 4B, all p_Begg >0.05, p_Egger >0.05). Thus, IL10 rs1800872 seems not to be associated with the risk of autoimmune liver disease.

FIG. 4.

Forest plot and Begg's funnel plot for subgroup analysis of IL10 rs1800872 based on disease type under a C versus A model. (A) Forest plot; (B) Begg's funnel plot. Color images available online at www.liebertpub.com/gtmb

Table 5.

Subgroup Analysis for the Association Between IL10 RS1800872 and Risk of Autoimmune Liver Disease

			Association test
Genetic models	Subgroup	Case-control studies (number)	OR (95% CI)	p_association	Sample size Case-control
Allele C vs. A	Caucasian	5	1.09 (0.88-1.34)	0.419	546/454
	Asian	4	1.22 (0.98-1.52)	0.075	258/551
	p_HWE >0.05	8	1.14 (0.98-1.34)	0.097	739/934
	p_HWE <0.05	1	1.22 (0.73-2.03)	0.446	65/71
	AIH	4	1.22 (0.96-1.53)	0.099	301/483
	PBC	4	1.11 (0.89-1.38)	0.345	407/444
	PSC	1	1.07 (0.66-1.73)	0.785	96/78
CC vs. AA	Caucasian	5	1.39 (0.76-2.55)	0.284	546/454
	Asian	4	1.65 (1.03-2.64)	0.036	258/551
	p_HWE >0.05	8	1.57 (1.05-2.35)	0.471	739/934
	p_HWE <0.05	1	1.43 (0.54-3.73)	0.028	65/71
	AIH	4	1.76 (1.01-3.08)	0.046	301/483
	PBC	4	1.47 (0.85-2.53)	0.165	407/444
	PSC	1	1.07 (0.31-3.75)	0.914	96/78
AC vs. AA	Caucasian	5	1.25 (0.67-2.34)	0.474	546/454
	Asian	4	1.08 (0.77-1.50)	0.664	258/551
	p_HWE >0.05	8	1.12 (0.81-1.53)	0.489	739/934
	p_HWE <0.05	1	1.08 (0.51-2.31)	0.834	65/71
	AIH	4	1.08 (0.70-1.64)	0.736	301/483
	PBC	4	1.17 (0.77-1.79)	0.469	407/444
	PSC	1	0.97 (0.27-3.48)	0.960	96/78
AC+CC vs. AA	Caucasian	5	1.34 (0.74-2.43)	0.333	546/454
	Asian	4	1.18 (0.87-1.61)	0.291	258/551
	p_HWE >0.05	8	1.22 (0.90-1.65)	0.197	739/934
	p_HWE <0.05	1	1.19 (0.61-2.33)	0.617	65/71
	AIH	4	1.20 (0.80-1.80)	0.376	301/483
	PBC	4	1.25 (0.84-1.86)	0.271	407/444
	PSC	1	1.03 (0.30-3.50)	0.966	96/78
CC vs. AA+AC	Caucasian	5	1.08 (0.84-1.38)	0.570	546/454
	Asian	4	1.59 (1.03-2.46)	0.036	258/551
	p_HWE >0.05	8	1.17 (0.94-1.47)	0.165	739/934
	p_HWE <0.05	1	1.38 (0.55-3.45)	0.490	65/71
	AIH	4	1.36 (0.96-1.94)	0.084	301/483
	PBC	4	1.08 (0.79-1.48)	0.631	407/444
	PSC	1	1.10 (0.60-2.01)	0.751	96/78
Carrier C vs. A	Caucasian	5	1.06 (0.84-1.33)	0.640	546/454
	Asian	4	1.14 (0.88-1.46)	0.322	258/551
	p_HWE >0.05	8	1.09 (0.91-1.30)	0.356	739/934
	p_HWE <0.05	1	1.15 (0.63-2.11)	0.650	65/71
	AIH	4	1.13 (0.87-1.47)	0.344	301/483
	PBC	4	1.06 (0.83-1.36)	0.627	407/444
	PSC	1	1.06 (0.61-1.82)	0.842	96/78

IL10 haplotypes

Finally, we investigated the role of IL10 haplotypes in the risk of autoimmune liver disease. Three promoter haplotypes, “G/C/C” (rs1800896G/rs1800871C/rs1800872C), “A/C/C” (rs1800896A/rs1800871C/rs1800872C), and “A/T/A” (rs1800896A/rs1800871T/rs1800872A), were studied using a fixed-effect model (Supplementary Table S5, all I² = 0.0%, p_{heterogeneity} >0.1). No genetic relationship between the G/C/C, A/C/C, or A/T/A haplotype and risk of autoimmune liver disease was identified in the overall meta-analysis or in the subgroup analysis of Caucasians (Supplementary Table S5, all p_association >0.05). However, p_Egger for “A/C/C” equaled 0.013, suggesting possible potential publication bias. Other p_Egger and p_Begg values were larger than 0.05 (Supplementary Table S5), suggesting no large publication bias.

Discussion

Genome-wide association studies (GWAS) were previously performed to identify the susceptibility foci for AIH, PBC, and PSC in different populations. For instance, GWAS data from a Chinese population that included 1122 PBC cases and 4036 controls identified risk loci within two cytokine genes, IL21 and IL16 (Qiu et al., 2017). In addition, GWAS data from a European population that included 536 PBS cases and 1536 controls revealed susceptibility foci at rs6441286 and rs574808 within IL12α and rs3790567 in IL12Rβ2 (Hirschfield et al., 2009). Likewise, a potential link between SNPs and the risk of PSC was observed in GWAS data from a Caucasian population that included 4796 cases and 19,955 controls (Ji et al., 2017). Nevertheless, we found no significant association between cytokine gene SNPs and autoimmune liver disease in the results of several GWAS (de Boer et al., 2014; Webb and Hirschfield, 2016; Higuchi et al., 2017). In spite of this, the potential contributions to cytokine gene polymorphisms and relevant immunoregulatory signaling axes to the risk of autoimmune liver disease cannot be excluded. In this study, we examined three polymorphisms within the IL10 gene promoter: rs1800896, rs1800871, and rs1800872.

Although a link between IL-10 and autoimmune responses has been reported (Saxena et al., 2015), results have been inconsistent. For example, the IL10 rs1800896 G/G genotype may be associated with susceptibility to PBS in Italians, but not Japanese (Matsushita et al., 2002). The IL10 rs1800896 A/G SNP was also reported to be a risk factor for PSC in a population from the United Kingdom (Bathgate et al., 2000). The results of the present meta-analysis point to a lack of involvement of IL10 promoter variants (rs1800896, rs1800871, and rs1800872) in the risk of autoimmune liver disease. We also observed a similar negative result for the three IL10 haplotypes (“G/C/C,” “A/C/C,” and “A/T/A”).

The case-control studies included in our analysis were screened using strict inclusion and exclusion criteria. All the included studies contained population-based control data and were of high quality. In addition, no heterogeneity was observed in any of the Mantel-Haenszel statistics, and large publication bias was excluded. Furthermore, the stability of the statistical results was confirmed through sensitivity analysis (data no shown).

Nonetheless, there are several limitations to this study. The first is the small sample size in the included case-control studies. Only 11 case-control studies for rs1800896, 8 for rs1800871, and 9 for rs1800872 were extracted from the 7 eligible articles (Zappala et al., 1998; Cookson et al., 1999; Bathgate et al., 2000; Donaldson et al., 2000; Matsushita et al., 2002; Chen et al., 2004; and Fan et al., 2005). Moreover, the sample size for the subgroup meta-analyses based on disease type (AIH, PBC, PSC) and ethnicity (Caucasian/Asian) were even smaller. Indeed, only one case-control study (Donaldson et al., 2000) was included in the subgroup analysis of the association between IL10 rs1800871 and the risk of autoimmune liver disease. Second, despite the absence of a significant effect in all subgroup meta-analyses, the departure from Hardy-Weinberg equilibrium in some studies weakens the power of the statistical analysis. Third, we only investigated three SNPs of IL10 in our meta-analysis, in that sufficient data were essential for the successful implementation of meta-analysis. The genetic effects of other SNPs (e.g., rs3135932, rs1554286, and rs3024490) or the effect of IL10 in combination with other cytokine genes should be analyzed in the future.

Fourth, AIH, PBC, and PSC all have complex and different etiologies. Additional confounding factors such as age, sex, and patient characteristics should be taken into consideration in future analyses. Fifth, IL10 rs1800896 is reportedly linked to the risk of other autoimmune diseases, including AITD (Jung et al., 2016) and systemic lupus erythematosus (Liu et al., 2013), while IL10 rs1800871 and rs1800872 are reportedly associated with risk of ulcerative colitis (Zou et al., 2014). In that context, and given the essential function of IL10 in autoimmune diseases (Miyagaki et al., 2015), additional studies using both in vitro and in vivo models are needed so as to reach a definite conclusion regarding the negative associations observed here.

In summary, our meta-analysis demonstrated the absence of a relationship between IL10 SNPs (rs1800896, rs1800871, rs1800872) and the risk of AIH, PBC, and PSC.

Footnotes

Acknowledgments

This work was supported in part by grants from the National Natural Science Foundation of China (31370749 and 31670759), the Key Science Foundation of Tianjin Health and Family Planning Commission (16KG151), and National Science and Technology major projects (2017ZX10302202).

Author Disclosure Statement

No competing financial interests exist.

References

Aizawa

, Hokari

(2017) Autoimmune hepatitis: current challenges and future prospects. Clin Exp Gastroenterol, 10:9-18.

Bathgate

, Pravica

, Perrey

, et al. (2000) Polymorphisms in tumour necrosis factor alpha, interleukin-10 and transforming growth factor beta1 genes and end-stage liver disease. Eur J Gastroenterol Hepatol, 12:1329-1333.

Benitez

, Cheng

, Deng

(2017) Revealing allele-specific gene expression by single-cell transcriptomics. Int J Biochem Cell Biol, 90:155-160.

Chen

, Fan

, Zhong

, et al. (2004) [A study on the relationship between interleukin-10 promoter polymorphism and autoimmune liver disease]. Zhonghua Gan Zang Bing Za Zhi, 12:356-358.

Cookson

, Constantini

, Clare

, et al. (1999) Frequency and nature of cytokine gene polymorphisms in type 1 autoimmune hepatitis. Hepatology, 30:851-856.

de Boer

, van Gerven

, Zwiers

, et al. (2014) Genome-wide association study identifies variants associated with autoimmune hepatitis type 1. Gastroenterology, 147:443-452.e5.

Donaldson

, Norris

, Constantini

, et al. (2000) The interleukin-1 and interleukin-10 gene polymorphisms in primary sclerosing cholangitis: no associations with disease susceptibility/resistance. J Hepatol, 32:882-886.

Eskandari-Nasab

, Tahmasebi

, Hashemi

(2015) Meta-analysis: the relationship between CTLA-4 + 49 A/G polymorphism and primary biliary cirrhosis and type I autoimmune hepatitis. Immunol Invest, 44:331-348.

Eskdale

, Kube

, Tesch

, et al. (1997) Mapping of the human IL10 gene and further characterization of the 5′ flanking sequence. Immunogenetics, 46:120-128.

10.

Fan

, Tu

, Zhu

, et al. (2005) Genetic association of cytokines polymorphisms with autoimmune hepatitis and primary biliary cirrhosis in the Chinese. World J Gastroenterol, 11:2768-2772.

11.

Floreani

, Tanaka

, Bowlus

, et al. (2017) Geoepidemiology and changing mortality in primary biliary cholangitis. J Gastroenterol, 52:655-662.

12.

Grant

, Liberal

(2017) Liver immunology: how to reconcile tolerance with autoimmunity. Clin Res Hepatol Gastroenterol, 41:6-16.

13.

Higuchi

, Oka

, Furukawa

, et al. (2017) Association of a single nucleotide polymorphism upstream of ICOS with Japanese autoimmune hepatitis type 1. J Hum Genet, 62:481-484.

14.

Hirschfield

, Liu

, Xu

, et al. (2009) Primary biliary cirrhosis associated with HLA, IL12A, and IL12RB2 variants. N Engl J Med, 360:2544-2555.

15.

, Juran

, Mucha

, et al. (2017) Genome-wide association study of primary sclerosing cholangitis identifies new risk loci and quantifies the genetic relationship with inflammatory bowel disease. Nat Genet, 49:269-273.

16.

Jiang

, Karlsen

(2017) Genetics of primary sclerosing cholangitis and pathophysiological implications. Nat Rev Gastroenterol Hepatol, 14:279-295.

17.

Joshita

, Umemura

, Tanaka

, et al. (2017) Genetic contribution to the pathogenesis of primary biliary cholangitis. J Immunol Res, 2017:3073504.

18.

Jung

, Song

, Kim

, et al. (2016) Association of interleukin 10 gene polymorphisms with autoimmune thyroid disease: meta-analysis. Scand J Immunol, 84:272-277.

19.

Karlsen

, Chung

(2015) Genetic risk and the development of autoimmune liver disease. Dig Dis, 33 Suppl 2:13-24.

20.

Kovac

, Weber

(2016) Primary biliary cirrhosis and primary sclerosing cholangitis: an update on MR imaging findings with recent developments. J Gastrointestin Liver Dis, 25:517-524.

21.

Liu

, Zheng

(2016) IL-10 -1082A/G, -592C/A, and -819T/C polymorphisms in association with lung cancer susceptibility: a meta-analysis. Onco Targets Ther, 9:6083-6091.

22.

Liu

, Song

, Su

, et al. (2013) IL-10 gene polymorphisms and susceptibility to systemic lupus erythematosus: a meta-analysis. PLoS One, 8:e69547.

23.

Liu

, Li

, et al. (2015) The association of interleukin-10 -1082, -819, -592 polymorphisms and tuberculosis risk. Saudi Med J, 36:407-417.

24.

, Qiu

(2004) [Cytokine gene polymorphisms in Chinese patients with type I autoimmune hepatitis]. Zhonghua Gan Zang Bing Za Zhi, 12:296-298.

25.

Matsuda

(2017) PCR-based detection methods for single-nucleotide polymorphism or mutation: real-time PCR and its substantial contribution toward technological refinement. Adv Clin Chem, 80:45-72.

26.

Matsushita

, Tanaka

, Kikuchi

, et al. (2002) Association of single nucleotide polymorphisms of the interleukin-10 promoter gene and susceptibility to primary biliary cirrhosis: immunogenetic differences in Italian and Japanese patients. Autoimmunity, 35:531-536.

27.

Mitchell

, Grove

, Spurkland

, et al. (2001) Association of the tumour necrosis factor alpha -308 but not the interleukin 10 -627 promoter polymorphism with genetic susceptibility to primary sclerosing cholangitis. Gut, 49:288-294.

28.

Miyagaki

, Fujimoto

, Sato

(2015) Regulatory B cells in human inflammatory and autoimmune diseases: from mouse models to clinical research. Int Immunol, 27:495-504.

29.

Moher

, Liberati

, Tetzlaff

, et al. (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med, 6:e1000097.

30.

Qiu

, Tang

, Zuo

, et al. (2017) A genome-wide association study identifies six novel risk loci for primary biliary cholangitis. Nat Commun, 8:14828.

31.

Saxena

, Khosraviani

, Noel

, et al. (2015) Interleukin-10 paradox: A potent immunoregulatory cytokine that has been difficult to harness for immunotherapy. Cytokine, 74:27-34.

32.

Stang

(2010) Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol, 25:603-605.

33.

Tarabay

, Elshazli

, Settin

(2016) African vs. Caucasian and Asian difference for the association of interleukin-10 promotor polymorphisms with type 2 diabetes mellitus (a meta-analysis study). Meta Gene, 9:10-17.

34.

Umemura

, Joshita

, Yamazaki

, et al. (2016) Genetic association of PTPN22 polymorphisms with autoimmune hepatitis and primary biliary cholangitis in Japan. Sci Rep, 6:29770.

35.

van Gerven

, de Boer

, Mulder

, et al. (2016) Auto immune hepatitis. World J Gastroenterol, 22:4651-4661.

36.

Walter

(2014) The molecular basis of IL-10 function: from receptor structure to the onset of signaling. Curr Top Microbiol Immunol, 380:191-212.

37.

Webb

, Hirschfield

(2016) Using GWAS to identify genetic predisposition in hepatic autoimmunity. J Autoimmun, 66:25-39.

38.

Wei

, Liu

, Li

, et al. (2011) Interleukin-10 gene polymorphisms and hepatocellular carcinoma susceptibility: a meta-analysis. World J Gastroenterol, 17:3941-3947.

39.

, Qian

, Yin

, et al. (2016) Association between the HFE C282Y, H63D polymorphisms and the risks of non-alcoholic fatty liver disease, liver cirrhosis and hepatocellular carcinoma: an updated systematic review and meta-analysis of 5,758 cases and 14,741 controls. PLoS One, 11:e0163423.

40.

Zappala

, Grove

, Watt

, et al. (1998) No evidence for involvement of the interleukin-10 -592 promoter polymorphism in genetic susceptibility to primary biliary cirrhosis. J Hepatol, 28:820-823.

41.

Zou

, Wang

, Gong

, et al. (2014) The association between three promoter polymorphisms of IL-10 and inflammatory bowel diseases (IBD): a meta-analysis. Autoimmunity, 47:27-39.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.51 MB