- Email: [email protected]
Common genetic variants of GPC1 gene reduce risk of biliary atresia in a Chinese population
Journal of Pediatric Surgery xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Common genetic variants of GPC1 gene reduce risk of biliary atresia in a Chinese population Juntao Ke a,b, Shuaidan Zeng a, Jianxiong Mao a, Jianyao Wang a, Jiao Lou b, Jiaoyuan Li b, Xueqin Chen b, Cheng Liu b, Liu-Ming Huang c,⁎, Bin Wang a,⁎⁎, Lei Liu a,⁎⁎ a
Department of General Surgery, Shenzhen Children Hospital, Shenzhen, China State Key Laboratory of Environment Health (Incubation), MOE (Ministry of Education) Key Laboratory of Environment & Health, Ministry of Environmental Protection Key Laboratory of Environment and Health (Wuhan), and Department of Epidemiology and Biostatistics, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China c Department of Pediatric Surgery, BaYi Children's hospital, The military general hospital of Beijing, Beijing, China b
a r t i c l e
i n f o
Article history: Received 8 January 2016 Received in revised form 18 April 2016 Accepted 11 May 2016 Available online xxxx Key words: Biliary atresia GPC1 Common genetic variants Chinese population
a b s t r a c t Background: Biliary atresia (BA) is a major neonatal cholestatic disease and main indication for pediatric liver transplantation in the world. Recently, GPC1 has been implicated as a risk gene for BA by genetic studies and follow-up functional experiments on zebrafish. Methods: Two common genetic variants of GPC1, rs2292832 and rs3828336, were selected systematically through ‘SNPinfo’, and were examined using TaqMan Genotyping Assays for association studies in a Chinese population containing 134 cases and 618 controls. Results: Of the two single nucleotide polymorphisms (SNPs), we found a significantly decreased BA risk associated with rs2292832 (additive model: OR = 0.638, 95% CI: 0.467–0.873, P = 0.005), and a marginal effect for rs3828336 (heterozygous model: OR = 0.564, 95% CI: 0.312–1.020, P = 0.058). The haplotype analysis indicated that either Crs2292832-Crs3828336&Trs3828336 or Trs2292832-Trs3828336 conferred a protective effect from BA (OR = 0.569, 95% CI = 0.414–0.783, P b 0.001; OR = 0.528, 95% CI: 0.301–0.926, P = 0.026). Moreover, bioinformatics analysis suggested that rs2292832 altered GPC1 expression via effect on transcription-factor-binding sites (TFBS) of upstream binding transcription factor (UBTF), as a regulatory DNA variation in Deoxyribonuclease I (DNase I) hypersensitive sites (DHSs). Conclusion: Common variants of GPC1 gene were genetically involved in BA risk. © 2016 Published by Elsevier Inc.
Biliary atresia (BA) is a devastating neonatal disease characterized by progressive fibroinflammatory obstruction of the extrahepatic biliary tree, leading to cholestasis, fibrosis, and cirrhosis. Without medical and surgical intervention such as Kasai portoenterostomy, an operation routinely performed to reconstruct biliary system but still failed in more than half of children [1], BA results in liver failure and death by the age of two years [2,3]. It is a rare disorder, with an incidence of 1:15,000–19,000 live births of Caucasian [4–7], whereas the Asian incidence is obviously higher, ranging from 1:5400 to 5800 live births in Chinese [8]. The etiology of BA is largely unknown. But it is supposed to be caused by exposure of genetically susceptible infant to environmental factors, exampled by virus infections [9] or toxins [10]. Several genes, including CFC1[11], intercellular adhesion molecule-1 (ICAM1) [12], macrophage
⁎ Correspondence to: L.M. Huang, Department of Pediatric Surgery, BaYi Children's hospital, The military general hospital of Beijing, 100700, Beijing, China. ⁎⁎ Corresponding authors at: Department of General Surgery, Shenzhen Children's Hospital, 518026 Shenzhen, Guangdong, China. E-mail addresses: [email protected] (L.-M. Huang), [email protected] (B. Wang), [email protected] (L. Liu).
migration inhibitory factor gene (MIF) [13], CD14 endotoxin receptor gene [14], and upstream stimulatory factor 2 (USF2) [15], have been suggested to play a role in BA pathogenesis. A recent genetic study of 35 BA cases and 2026 controls identified a potential region of interest at 2q37.3 [16]. Cui et al. [17] extended the study of copy number variations (CNV) in 61 cases and 5088 controls, detected heterozygous deletions spanning the same locus in 6 subjects and narrowed the region to include only one gene GPC1. Moreover, they followed up with functional analysis in an animal model zebrafish, which showed that disruption of GPC1 led to biliary defects involving overactivation of Hedgehog signaling. Given that GPC1 has been identified as a risk gene for BA by Cui et al., we hypothesized that not only rare genomic indels but also common variants (minor allele frequency, MAF N 0.05) of GPC1 could constitute a genetic basis for susceptibility to BA. So we selected two single nucleotide polymorphisms (SNPs), rs2292832 and rs3828336, in GPC1 gene through an integrated bioinformatics tool ‘SNPinfo’ [18] (http://snpinfo.niehs.nih. gov/snpinfo/snpfunc.htm) where researchers can choose SNPs based on predicted functional characteristics. Afterwards, we investigated their association with BA via a case–control study of 134 cases and 618 controls, and attempted to explain the potential pathogenic role of positive SNP by further bioinformatics analysis. To the best of our knowledge,
http://dx.doi.org/10.1016/j.jpedsurg.2016.05.009 0022-3468/© 2016 Published by Elsevier Inc.
Please cite this article as: Ke J, et al, Common genetic variants of GPC1 gene reduce risk of biliary atresia in a Chinese population, J Pediatr Surg (2016), http://dx.doi.org/10.1016/j.jpedsurg.2016.05.009
2
J. Ke et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
this is the first study on common genetic variants in GPC1 and BA risk in a Chinese population. 1. Material and methods 1.1. Study population This study contained 134 children (males = 65; females = 69) diagnosed with BA and without other associated congenital malformation by laparoscopic cholangiography and biopsy of liver and extrahepatic biliary tree, all of which were consecutively enrolled between 2010 and 2013 from Shenzhen Children's Hospital, China. As controls, 618 healthy individuals (males = 303; females = 315) of southern Chinese without diagnosis of BA, congenital disease or liver disease were included. All subjects were unrelated ethnic Han Chinese. Written informed consent has been obtained from each subject or their legal guardians during the enrollment. This study was approved by the institutional review board (IRB) of Shenzhen Children's Hospital. 1.2. SNP selection We screened SNPs in the GPC1 gene by a web-base SNP selection tool called ‘SNPinfo’ which integrates GWAS and candidate gene information into functional SNP selection for association studies [18], as the procedures described blow. Firstly, we extracted the range of the physical position of GPC1 (chr 2: 241,023,788 ~ 241,056,165) and its upstream and downstream 2Kb range (chr 2: 241,021,788 ~ 241,058,165) from HapMap database (http://hapmap.ncbi.nlm.nih.gov/, HapMap Data Rel 24/phaseII Nov08, on NCBI B36 assembly, dbSNP b126). Secondly, we input the extensive range of the gene into ‘SNPinfo’ (http://snpinfo. niehs.nih.gov/snpinfo/snpfunc.htm) and received a list of SNPs with possible functions. Thirdly, we sifted out the SNP whose minor allele frequency (MAF) of Han Chinese in Beijing, China (CHB) is greater than 5%, and got a total of 7 SNPs (Table S1 in the online version at http:// dx.doi.org/10.1016/j.jpedsurg.2016.05.009). Fourthly, we prioritized 3 most possibly functional SNPs (rs3828336, rs2292832, rs2228331) which harbor more than one functional motifs (marked with a Y letter standing for ‘Yes’), such as transcription-factor-binding sites (TFBS), splicing site, microRNA-binding site, non-synonymous SNP (nsSNP). Finally, after we deleted rs2228331, an nsSNP that was classified as ‘benign’ by Polyphen [19], rs2292832 and rs3828336 were left for genotyping in our case–control study. To avoid redundancy, we analyzed the LD between these two SNPs by Haploview v4.2 [20], and confirmed that rs2292832 and rs3828336 are not in LD (r2 = 0.000). 1.3. Genotyping Genomic DNA was extracted from peripheral blood leukocytes or liver tissues of BA children, and from peripheral blood of healthy controls, using DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany) by reference to the manufacturer's instructions. Both SNPs were genotyped with the TaqMan SNP Genotyping Assay (Applied Biosystems, Foster City, CA, USA) on a 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). 5% duplicated samples were randomly selected to assess the reproducibility for quality control, with a concordance rate of 100%. 1.4. Statistical analysis The χ2 test was applied to estimated differences in variables and distributions of genotypes between cases and controls. Hardy–Weinberg equilibrium was evaluated using the goodness-of-fit χ 2 test in controls and a value of P b 0.05 was considered as significant disequilibrium. The association between the case–control status and each SNP was measured by the odds ratio (OR) and its corresponding 95% confidence interval (95% CI). In order to avoid the assumption of genetic models,
codominant, dominant and additive models were analyzed. The LD of candidate SNPs was analyzed using Haploview v4.2 [20] and haplotype were constructed by the PHASE v2.1 software [21]. The ORs and corresponding 95% CIs, adjusted by gender, were calculated by unconditional multivariate logistic regression. Statistical analyses were performed using SPSS Software v20.0 (SPSS, Chicago, Illinois, USA), P b 0.05 was considered statistically significant. 1.5. Bioinformatics analysis To try to explain the biological possibilities underlying significant SNPs associated with BA in details, bioinformatics analysis was conducted on a online database ‘UCSC’ [22] (http://genome.ucsc.edu/) that has been updated with many new annotation data sets including ‘Transcription Factor ChIP-seq Clusters’ and the ‘DNaseI Hypersensitivity Clusters’ from ENCODE [23]. 2. Results 2.1. Population characteristic A total of 134 children with BA and 618 controls were enrolled in this study. 130 (97%) cases for rs2292832, 133 (99%) cases for rs3828336 and all controls were genotyped successfully. The genotype distributions of rs2292832 and rs3828336 in our controls are both in Hardy–Weinberg equilibrium (HWE, P = 0.52 and P = 0.14). And there was no statistically significant difference in gender distribution between cases and controls (P = 0.913, Pearson χ2 = 0.012). 2.2. Association analysis The rs2292832 showed significant association in four genetic models, the other rs3828336 presented no obvious association with BA under any model. Detailed genotype frequencies of rs2292832 and rs3828336 are shown in Table 1. Under multivariate logistic regression model adjusted for sex, individuals with CT genotype of rs2292832 had a significantly decreased risk of BA (OR = 0.339, 95% CI = 0.214–0.535, P b 0.001) compared to those with TT homozygote. A dominant model was performed to increase statistical power by combining the CT with CC into a C-carrier group (CT plus CC), and it showed that the allele C carriers got an obviously lower risk (OR = 0.43, 95% CI = 0.289–0.641, P b 0.001). Likewise, the C allele presented protective effect for BA (OR = 0.629, 95% CI = 0.459–0.863, P = 0.004), while the C allele frequency in our controls was 31.3%, approximate to the MAF 26.7% of CHB from HapMap. Another positive result was found under additive model, with per-C-allele OR of 0.638 (95% CI = 0.467–0.873, P = 0.005). As for rs3828336, we observed no statistically significant difference in genotype (P = 0.125, Pearson χ 2 = 4.162) and allele (P = 0.115, Pearson χ2 = 2.487) distributions when comparing cases with controls, and found null associations with BA in all genetic models we studied (Table 1). Still, a marginal effect was observed in the heterozygous model (OR = 0.564, 95% CI: 0.312–1.020, P = 0.058). 2.3. Haplotypes analysis As shown in Table 2, there is significant difference in the distribution of haplotype frequencies between cases and controls (P b 0.001, Pearson χ 2 = 15.423). Compared with the Trs2292832-Crs3828336 that consists of the two wild alleles, haplotypes containing Crs2292832 or Trs3828336 allele were associated with decreased risk of BA. Of note, we combined Crs2292832-Crs3828336 and Crs2292832-Trs3828336 into a subgroup that containing the significant risk allele Crs2292832 to increase statistical power, because of the extremely low frequencies of the Crs2292832-Trs3828336 in both cases and controls (0.4% and 0.2%). The sex-adjusted ORs calculated by logistic regression for the Crs2292832-Crs3828336&Trs3828336 and
Please cite this article as: Ke J, et al, Common genetic variants of GPC1 gene reduce risk of biliary atresia in a Chinese population, J Pediatr Surg (2016), http://dx.doi.org/10.1016/j.jpedsurg.2016.05.009
J. Ke et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
3
Table 1 Association between SNPs and BA risk in a Chinese population. Case (%)
P#
OR (95% CI)+
288 (46.6) 273 (44.2) 57 (9.2) 330 (53.4) 849 (68.7) 387 (31.3)
87 (66.9) 28 (21.5) 15 (11.5) 43 (33.1) 202 (77.7) 58 (22.3)
b0.001
1.000 (reference) 0.339 (0.214–0.535) 0.868 (0.468–1.611) 0.430 (0.289–0.641) 1.000 (reference) 0.629 (0.459–0.863) 0.638 (0.467–0.873)
509 (82.4) 107 (17.3) 2 (0.3) 109 (17.6) 1125 (91.0) 111 (9.0)
118 (88.7) 14 (10.5) 1 (0.8) 15 (11.3) 250 (94.0) 16 (6.0)
Controls (%) rs2292832 TT CT CC Dominant model Allele T Allele C Additive model rs3828336 CC CT TT Dominant model Allele C Allele T Additive model # ¶ +
0.004
0.145¶
0.115
1.000 (reference) 0.564 (0.312–1.020) 2.161 (0.194–24.049) 0.594 (0.334–1.056) 1.000 (reference) 0.649 (0.377–1.115) 0.638 (0.368–1.107)
P+
b0.001 0.654 b0.001 0.004 0.005
0.058 0.531 0.076 0.117 0.110
P values were computed by Pearson Chi-Square test. P values were computed by Fisher's exact test. Data were calculated by logistic regression after adjusting for sex.
Trs2292832-Trs3828336 were 0.569 (95% CI = 0.414–0.783) and 0.528 (95% CI = 0.301–0.926), respectively. 2.4. Bioinformatics analysis According to UCSC [22], rs2292832 was situated within the first intron of GPC1 and the binding site of a transcription factor UBTF (upstream binding transcription factor, RNA polymerase I). At the same time, it overlapped Deoxyribonuclease I (DNase I) hypersensitive sites (DHSs) in 40 cell lines including one primary hepatocytes cell line ‘Hepatocytes’ and three hepatocellular carcinoma cell lines ‘HepG2’, ‘Huh-7’ and ‘Huh-7.5’, with a highest cluster score 1000 (where score is based on signal strength). 3. Discussion BA is a destructive inflammatory obliterative cholangiopathy [2] occurring exclusively in the neonatal period, and it has been the leading cause of liver transplantation in children for the past 20 years. Meanwhile, uncovering the cause of BA remains a major challenge. Recently, several inspiring findings were reported by Cui et al. [17], who convincingly specified GPC1 as a risk gene for BA. Analysis of copy number variant (CNV) differing between BA patients and controls revealed a notable region 2q37.3, which corresponded to the loss of GPC1. In addition, they took advantage of zebrafish which were knockdown of gpc1, and determined the mechanistic insight that deletion of GPC1 over-activating Hedgehog signaling resulted in biliary development defects. Accordingly, we selected two SNPs of GPC1 systematically and performed a case–control study. The results of our study indicated that the minor allele Crs2292832 was significantly associated with reduced Table 2 Risk estimates for extended GPC1 haplotypes in BA cases and controls. Haplotype
Ta-Cb Ca-Cb&Tb Ta-Tb P for trend
Control
Case
No. of Chromosomes (%)
No. of Chromosomes (%)
741 (60.0) 387 (31.3) 108(8.7)
195 (72.8) 58 (21.6) 15 (5.6)
P#
b0.001
OR (95% CI)+
P+
1.000(reference) 0.569(0.414–0.783) 0.528(0.301–0.926) b0.001
0.001 0.026
a: rs2292832. b: rs3828336. # P values were computed by the Pearson chi-square test. + Data were calculated by logistic regression after adjusting for sex.
risk of BA in heterozygous, dominant, allelic and additive genetic models, but not in the homozygous possibly because of our relative small sample size. And the over representation of rs2292832 homozygotes in our cases was likely because of the loss of heterozygosity (LOH) of 2q27.3. And it might reflect the presence of LOH in some of our samples as Cui et al. reported. On the other side, we found no positive association for rs3828336, but still a marginal effect in CT genotype, which implied the possibility of protective effect of Trs3828336. In the haplotypes Crs2292832-Crs3828336&Trs3828336 and Trs2292832-Trs3828336, BA risk reduced by approximately 50% referring to the haplotype of two major alleles. Our findings are biologically plausible. GPC1 encodes glypican 1, which belongs to glypican family of heparan sulfate proteoglycans, adjoins to the cell membrane by a glycosyl-phosphatidylinositol linkage [24]. Glypicans (GPCs) could modulate several signals containing Hedgehog (Hh), Wingless (Wnt), fibroblast growth factor, and bone morphogenetic protein [25], each of which play an important role in hepatogenesis. Cui et al. observed no apical GPC1 immunostaining in cholangiocytes and suggested the decreased amount of GPC1 in the BA liver samples [17], which might reflect the expression or functional abnormality of GPC1 in patients. Given that GPCs are known to regulate Hh activity and increased Hh activity is observed in BA patients [26], Cui et al. further confirmed that over-activated Hh was the basis of the GPC1-deficient phenotype and responsible for biliary defects [17]. Therefore, we speculated that the investigated polymorphisms might upregulate expression of GPC1, thereby being antagonistic to Hh signaling and, as a consequence, impairing the occurrence and development of BA. This presumption was supported by bioinformatics analysis in UCSC [22] integrating with Encode data [23]: rs2292832 was located in the binding site of UBTF, which played a critical role in ribosomal RNA transcription as a key component of the pre-initiation complex, mediating the recruitment of RNA polymerase I to ribosomal RNA promoter regions [27]. Under the consideration of that binding site variations contributed extensively on transcription factor binding and often correlated with differences in gene expression [28], we assumed that rs2292832 could alter the transcription and expression of GPC1 through the affection of UBTF's binding. In addition, it is strongly suggested that rs2292832 lying in DHSs functions as a regulatory DNA variation [29], whereas DHSs are sensitive and precise markers of actuated cis-regulatory elements [30]. However, the assumption above needs relevant experiments to be verified. Several limitations of this study should be noted. First of all, we neglected the possibility that the reported SNP association was partially synthesizing the rare deletion effect. Second, the strategy of screening candidate common polymorphisms depended on the prediction from
Please cite this article as: Ke J, et al, Common genetic variants of GPC1 gene reduce risk of biliary atresia in a Chinese population, J Pediatr Surg (2016), http://dx.doi.org/10.1016/j.jpedsurg.2016.05.009
4
J. Ke et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
SNPinfo on the basic of SNPs' position, which was not rigorous enough to discover all possibly functional SNPs including rare variants. Third, the sample size of our case–control study was relatively small. Forth, insufficient environmental information limited us to further investigate the gene–environment interactions, considering that BA is a complex trait. Fifth, lacking of functional experiments on relevant tissues and cells, biological reality of the statistically significant association we reported is still uncertain. In sum, our independent case–control study implicated that common SNPs in GPC1 gene contributed to BA susceptibility in Chinese population, in addition to the large genomic indels discovered before. Comprehensive researches with more representative SNPs, different races, greater sample sizes and follow-up functional analyses are warranted to identify causal variants and elaborate the biological mechanism of genetic etiology. Supplementary data to this article can be found online at http://dx.doi. org/10.1016/j.jpedsurg.2016.05.009. Acknowledgements The authors wish to thank the numerous patients and their families who have participated in these studies, and all the study participants, research staff and students who participated in this work. The studies reported here were supported by Shenzhen Science and Technology Project to Shenzhen Children's Hospital (JCYJ20150403100317072). References [1] Superina R, Magee JC, Brandt ML, et al. The anatomic pattern of biliary atresia identified at time of Kasai Hepatoportoenterostomy and early postoperative clearance of jaundice are significant predictors of transplant-free survival. Ann Surg 2011;254:577–85. [2] Hartley JL, Davenport M, Kelly DA. Biliary atresia. Lancet 2009;374:1704–13. [3] Bassett MD, Murray KF. Biliary atresia: recent progress. J Clin Gastroenterol 2008;42: 720–9. [4] Houwen RH, Kerremans II, van Steensel-Moll HA, et al. Time–space distribution of extrahepatic biliary atresia in the Netherlands and West Germany. Z Kinderchir 1988;43:68–71. [5] Chardot C, Carton M, Spire-Bendelac N, et al. Epidemiology of biliary atresia in France: a national study 1986-96. J Hepatol 1999;31:1006–13. [6] Yoon PW, Bresee JS, Olney RS, et al. Epidemiology of biliary atresia: a populationbased study. Pediatrics 1997;99:376–82. [7] McKiernan PJ, Baker AJ, Kelly DA. The frequency and outcome of biliary atresia in the UK and Ireland. Lancet 2000;355:25–9.
[8] Hsiao CH, Chang MH, Chen HL, et al. Universal screening for biliary atresia using an infant stool color card in Taiwan. Hepatology 2008;47:1233–40. [9] Mack CL. The pathogenesis of biliary atresia: evidence for a virus-induced autoimmune disease. Semin Liver Dis 2007;27:233–42. [10] Harper P, Plant JW, Unger DB. Congenital biliary atresia and jaundice in lambs and calves. Aust Vet J 1990;67:18–22. [11] Davit-Spraul A, Baussan C, Hermeziu B, et al. CFC1 gene involvement in biliary atresia with polysplenia syndrome. J Pediatr Gastroenterol Nutr 2008;46:111–2. [12] Arikan C, Berdeli A, Kilic M, et al. Polymorphisms of the ICAM-1 gene are associated with biliary atresia. Dig Dis Sci 2008;53:2000–4. [13] Arikan C, Berdeli A, Ozgenc F, et al. Positive association of macrophage migration inhibitory factor gene-173G/C polymorphism with biliary atresia. J Pediatr Gastroenterol Nutr 2006;42:77–82. [14] Shih HH, Lin TM, Chuang JH, et al. Promoter polymorphism of the CD14 endotoxin receptor gene is associated with biliary atresia and idiopathic neonatal cholestasis. Pediatrics 2005;116:437–41. [15] Huang YH, Huang CC, Chuang JH, et al. Upstream stimulatory factor 2 is implicated in the progression of biliary atresia by regulation of hepcidin expression. J Pediatr Surg 2008;43:2016–23. [16] Leyva-Vega M, Gerfen J, Thiel BD, et al. Genomic alterations in biliary atresia suggest region of potential disease susceptibility in 2q37.3. Am J Med Genet A 2010;152A: 886–95. [17] Cui S, Leyva–Vega M, Tsai EA, et al. Evidence from human and zebrafish that GPC1 is a biliary atresia susceptibility Gene. Gastroenterology 2013;144:1107–1115.e1103. [18] Xu Z, Taylor JA. SNPinfo: integrating GWAS and candidate gene information into functional SNP selection for genetic association studies. Nucleic Acids Res 2009;37: W600–5. [19] Sunyaev S, Ramensky V, Koch I, et al. Prediction of deleterious human alleles. Hum Mol Genet 2001;10:591–7. [20] Barrett JC, Fry B, Maller J, et al. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005;21:263–5. [21] Stephens M, Smith NJ, Donnelly P. A new statistical method for haplotype reconstruction from population data. Am J Hum Genet 2001;68:978–89. [22] Karolchik D, Barber GP, Casper J, et al. The UCSC genome browser database: 2014 update. Nucleic Acids Res 2014;42:D764–70. [23] Bernstein BE, Birney E, Dunham I, et al. An integrated encyclopedia of DNA elements in the human genome. Nature 2012;489:57–74. [24] Kirn-Safran C, Farach-Carson MC, Carson DD. Multifunctionality of extracellular and cell surface heparan sulfate proteoglycans. Cell Mol Life Sci 2009;66:3421–34. [25] Yan D, Lin X. Shaping morphogen gradients by proteoglycans. Cold Spring Harb Perspect Biol 2009;1:a002493. [26] Omenetti A, Bass LM, Anders RA, et al. Hedgehog activity, epithelial-mesenchymal transitions, and biliary dysmorphogenesis in biliary atresia. Hepatology 2011;53: 1246–58. [27] Pruitt KD, Brown GR, Hiatt SM, et al. RefSeq: an update on mammalian reference sequences. Nucleic Acids Res 2014;42:D756–63. [28] Kasowski M, Grubert F, Heffelfinger C, et al. Variation in transcription factor binding among humans. Science 2010;328:232–5. [29] Maurano MT, Humbert R, Rynes E, et al. Systematic localization of common diseaseassociated variation in regulatory DNA. Science 2012;337:1190–5. [30] Thurman RE, Rynes E, Humbert R, et al. The accessible chromatin landscape of the human genome. Nature 2012;489:75–82.
Please cite this article as: Ke J, et al, Common genetic variants of GPC1 gene reduce risk of biliary atresia in a Chinese population, J Pediatr Surg (2016), http://dx.doi.org/10.1016/j.jpedsurg.2016.05.009
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Common genetic variants of GPC1 gene reduce risk of biliary atresia in a Chinese population Juntao Ke a,b, Shuaidan Zeng a, Jianxiong Mao a, Jianyao Wang a, Jiao Lou b, Jiaoyuan Li b, Xueqin Chen b, Cheng Liu b, Liu-Ming Huang c,⁎, Bin Wang a,⁎⁎, Lei Liu a,⁎⁎ a
Department of General Surgery, Shenzhen Children Hospital, Shenzhen, China State Key Laboratory of Environment Health (Incubation), MOE (Ministry of Education) Key Laboratory of Environment & Health, Ministry of Environmental Protection Key Laboratory of Environment and Health (Wuhan), and Department of Epidemiology and Biostatistics, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China c Department of Pediatric Surgery, BaYi Children's hospital, The military general hospital of Beijing, Beijing, China b
a r t i c l e
i n f o
Article history: Received 8 January 2016 Received in revised form 18 April 2016 Accepted 11 May 2016 Available online xxxx Key words: Biliary atresia GPC1 Common genetic variants Chinese population
a b s t r a c t Background: Biliary atresia (BA) is a major neonatal cholestatic disease and main indication for pediatric liver transplantation in the world. Recently, GPC1 has been implicated as a risk gene for BA by genetic studies and follow-up functional experiments on zebrafish. Methods: Two common genetic variants of GPC1, rs2292832 and rs3828336, were selected systematically through ‘SNPinfo’, and were examined using TaqMan Genotyping Assays for association studies in a Chinese population containing 134 cases and 618 controls. Results: Of the two single nucleotide polymorphisms (SNPs), we found a significantly decreased BA risk associated with rs2292832 (additive model: OR = 0.638, 95% CI: 0.467–0.873, P = 0.005), and a marginal effect for rs3828336 (heterozygous model: OR = 0.564, 95% CI: 0.312–1.020, P = 0.058). The haplotype analysis indicated that either Crs2292832-Crs3828336&Trs3828336 or Trs2292832-Trs3828336 conferred a protective effect from BA (OR = 0.569, 95% CI = 0.414–0.783, P b 0.001; OR = 0.528, 95% CI: 0.301–0.926, P = 0.026). Moreover, bioinformatics analysis suggested that rs2292832 altered GPC1 expression via effect on transcription-factor-binding sites (TFBS) of upstream binding transcription factor (UBTF), as a regulatory DNA variation in Deoxyribonuclease I (DNase I) hypersensitive sites (DHSs). Conclusion: Common variants of GPC1 gene were genetically involved in BA risk. © 2016 Published by Elsevier Inc.
Biliary atresia (BA) is a devastating neonatal disease characterized by progressive fibroinflammatory obstruction of the extrahepatic biliary tree, leading to cholestasis, fibrosis, and cirrhosis. Without medical and surgical intervention such as Kasai portoenterostomy, an operation routinely performed to reconstruct biliary system but still failed in more than half of children [1], BA results in liver failure and death by the age of two years [2,3]. It is a rare disorder, with an incidence of 1:15,000–19,000 live births of Caucasian [4–7], whereas the Asian incidence is obviously higher, ranging from 1:5400 to 5800 live births in Chinese [8]. The etiology of BA is largely unknown. But it is supposed to be caused by exposure of genetically susceptible infant to environmental factors, exampled by virus infections [9] or toxins [10]. Several genes, including CFC1[11], intercellular adhesion molecule-1 (ICAM1) [12], macrophage
⁎ Correspondence to: L.M. Huang, Department of Pediatric Surgery, BaYi Children's hospital, The military general hospital of Beijing, 100700, Beijing, China. ⁎⁎ Corresponding authors at: Department of General Surgery, Shenzhen Children's Hospital, 518026 Shenzhen, Guangdong, China. E-mail addresses: [email protected] (L.-M. Huang), [email protected] (B. Wang), [email protected] (L. Liu).
migration inhibitory factor gene (MIF) [13], CD14 endotoxin receptor gene [14], and upstream stimulatory factor 2 (USF2) [15], have been suggested to play a role in BA pathogenesis. A recent genetic study of 35 BA cases and 2026 controls identified a potential region of interest at 2q37.3 [16]. Cui et al. [17] extended the study of copy number variations (CNV) in 61 cases and 5088 controls, detected heterozygous deletions spanning the same locus in 6 subjects and narrowed the region to include only one gene GPC1. Moreover, they followed up with functional analysis in an animal model zebrafish, which showed that disruption of GPC1 led to biliary defects involving overactivation of Hedgehog signaling. Given that GPC1 has been identified as a risk gene for BA by Cui et al., we hypothesized that not only rare genomic indels but also common variants (minor allele frequency, MAF N 0.05) of GPC1 could constitute a genetic basis for susceptibility to BA. So we selected two single nucleotide polymorphisms (SNPs), rs2292832 and rs3828336, in GPC1 gene through an integrated bioinformatics tool ‘SNPinfo’ [18] (http://snpinfo.niehs.nih. gov/snpinfo/snpfunc.htm) where researchers can choose SNPs based on predicted functional characteristics. Afterwards, we investigated their association with BA via a case–control study of 134 cases and 618 controls, and attempted to explain the potential pathogenic role of positive SNP by further bioinformatics analysis. To the best of our knowledge,
http://dx.doi.org/10.1016/j.jpedsurg.2016.05.009 0022-3468/© 2016 Published by Elsevier Inc.
Please cite this article as: Ke J, et al, Common genetic variants of GPC1 gene reduce risk of biliary atresia in a Chinese population, J Pediatr Surg (2016), http://dx.doi.org/10.1016/j.jpedsurg.2016.05.009
2
J. Ke et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
this is the first study on common genetic variants in GPC1 and BA risk in a Chinese population. 1. Material and methods 1.1. Study population This study contained 134 children (males = 65; females = 69) diagnosed with BA and without other associated congenital malformation by laparoscopic cholangiography and biopsy of liver and extrahepatic biliary tree, all of which were consecutively enrolled between 2010 and 2013 from Shenzhen Children's Hospital, China. As controls, 618 healthy individuals (males = 303; females = 315) of southern Chinese without diagnosis of BA, congenital disease or liver disease were included. All subjects were unrelated ethnic Han Chinese. Written informed consent has been obtained from each subject or their legal guardians during the enrollment. This study was approved by the institutional review board (IRB) of Shenzhen Children's Hospital. 1.2. SNP selection We screened SNPs in the GPC1 gene by a web-base SNP selection tool called ‘SNPinfo’ which integrates GWAS and candidate gene information into functional SNP selection for association studies [18], as the procedures described blow. Firstly, we extracted the range of the physical position of GPC1 (chr 2: 241,023,788 ~ 241,056,165) and its upstream and downstream 2Kb range (chr 2: 241,021,788 ~ 241,058,165) from HapMap database (http://hapmap.ncbi.nlm.nih.gov/, HapMap Data Rel 24/phaseII Nov08, on NCBI B36 assembly, dbSNP b126). Secondly, we input the extensive range of the gene into ‘SNPinfo’ (http://snpinfo. niehs.nih.gov/snpinfo/snpfunc.htm) and received a list of SNPs with possible functions. Thirdly, we sifted out the SNP whose minor allele frequency (MAF) of Han Chinese in Beijing, China (CHB) is greater than 5%, and got a total of 7 SNPs (Table S1 in the online version at http:// dx.doi.org/10.1016/j.jpedsurg.2016.05.009). Fourthly, we prioritized 3 most possibly functional SNPs (rs3828336, rs2292832, rs2228331) which harbor more than one functional motifs (marked with a Y letter standing for ‘Yes’), such as transcription-factor-binding sites (TFBS), splicing site, microRNA-binding site, non-synonymous SNP (nsSNP). Finally, after we deleted rs2228331, an nsSNP that was classified as ‘benign’ by Polyphen [19], rs2292832 and rs3828336 were left for genotyping in our case–control study. To avoid redundancy, we analyzed the LD between these two SNPs by Haploview v4.2 [20], and confirmed that rs2292832 and rs3828336 are not in LD (r2 = 0.000). 1.3. Genotyping Genomic DNA was extracted from peripheral blood leukocytes or liver tissues of BA children, and from peripheral blood of healthy controls, using DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany) by reference to the manufacturer's instructions. Both SNPs were genotyped with the TaqMan SNP Genotyping Assay (Applied Biosystems, Foster City, CA, USA) on a 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). 5% duplicated samples were randomly selected to assess the reproducibility for quality control, with a concordance rate of 100%. 1.4. Statistical analysis The χ2 test was applied to estimated differences in variables and distributions of genotypes between cases and controls. Hardy–Weinberg equilibrium was evaluated using the goodness-of-fit χ 2 test in controls and a value of P b 0.05 was considered as significant disequilibrium. The association between the case–control status and each SNP was measured by the odds ratio (OR) and its corresponding 95% confidence interval (95% CI). In order to avoid the assumption of genetic models,
codominant, dominant and additive models were analyzed. The LD of candidate SNPs was analyzed using Haploview v4.2 [20] and haplotype were constructed by the PHASE v2.1 software [21]. The ORs and corresponding 95% CIs, adjusted by gender, were calculated by unconditional multivariate logistic regression. Statistical analyses were performed using SPSS Software v20.0 (SPSS, Chicago, Illinois, USA), P b 0.05 was considered statistically significant. 1.5. Bioinformatics analysis To try to explain the biological possibilities underlying significant SNPs associated with BA in details, bioinformatics analysis was conducted on a online database ‘UCSC’ [22] (http://genome.ucsc.edu/) that has been updated with many new annotation data sets including ‘Transcription Factor ChIP-seq Clusters’ and the ‘DNaseI Hypersensitivity Clusters’ from ENCODE [23]. 2. Results 2.1. Population characteristic A total of 134 children with BA and 618 controls were enrolled in this study. 130 (97%) cases for rs2292832, 133 (99%) cases for rs3828336 and all controls were genotyped successfully. The genotype distributions of rs2292832 and rs3828336 in our controls are both in Hardy–Weinberg equilibrium (HWE, P = 0.52 and P = 0.14). And there was no statistically significant difference in gender distribution between cases and controls (P = 0.913, Pearson χ2 = 0.012). 2.2. Association analysis The rs2292832 showed significant association in four genetic models, the other rs3828336 presented no obvious association with BA under any model. Detailed genotype frequencies of rs2292832 and rs3828336 are shown in Table 1. Under multivariate logistic regression model adjusted for sex, individuals with CT genotype of rs2292832 had a significantly decreased risk of BA (OR = 0.339, 95% CI = 0.214–0.535, P b 0.001) compared to those with TT homozygote. A dominant model was performed to increase statistical power by combining the CT with CC into a C-carrier group (CT plus CC), and it showed that the allele C carriers got an obviously lower risk (OR = 0.43, 95% CI = 0.289–0.641, P b 0.001). Likewise, the C allele presented protective effect for BA (OR = 0.629, 95% CI = 0.459–0.863, P = 0.004), while the C allele frequency in our controls was 31.3%, approximate to the MAF 26.7% of CHB from HapMap. Another positive result was found under additive model, with per-C-allele OR of 0.638 (95% CI = 0.467–0.873, P = 0.005). As for rs3828336, we observed no statistically significant difference in genotype (P = 0.125, Pearson χ 2 = 4.162) and allele (P = 0.115, Pearson χ2 = 2.487) distributions when comparing cases with controls, and found null associations with BA in all genetic models we studied (Table 1). Still, a marginal effect was observed in the heterozygous model (OR = 0.564, 95% CI: 0.312–1.020, P = 0.058). 2.3. Haplotypes analysis As shown in Table 2, there is significant difference in the distribution of haplotype frequencies between cases and controls (P b 0.001, Pearson χ 2 = 15.423). Compared with the Trs2292832-Crs3828336 that consists of the two wild alleles, haplotypes containing Crs2292832 or Trs3828336 allele were associated with decreased risk of BA. Of note, we combined Crs2292832-Crs3828336 and Crs2292832-Trs3828336 into a subgroup that containing the significant risk allele Crs2292832 to increase statistical power, because of the extremely low frequencies of the Crs2292832-Trs3828336 in both cases and controls (0.4% and 0.2%). The sex-adjusted ORs calculated by logistic regression for the Crs2292832-Crs3828336&Trs3828336 and
Please cite this article as: Ke J, et al, Common genetic variants of GPC1 gene reduce risk of biliary atresia in a Chinese population, J Pediatr Surg (2016), http://dx.doi.org/10.1016/j.jpedsurg.2016.05.009
J. Ke et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
3
Table 1 Association between SNPs and BA risk in a Chinese population. Case (%)
P#
OR (95% CI)+
288 (46.6) 273 (44.2) 57 (9.2) 330 (53.4) 849 (68.7) 387 (31.3)
87 (66.9) 28 (21.5) 15 (11.5) 43 (33.1) 202 (77.7) 58 (22.3)
b0.001
1.000 (reference) 0.339 (0.214–0.535) 0.868 (0.468–1.611) 0.430 (0.289–0.641) 1.000 (reference) 0.629 (0.459–0.863) 0.638 (0.467–0.873)
509 (82.4) 107 (17.3) 2 (0.3) 109 (17.6) 1125 (91.0) 111 (9.0)
118 (88.7) 14 (10.5) 1 (0.8) 15 (11.3) 250 (94.0) 16 (6.0)
Controls (%) rs2292832 TT CT CC Dominant model Allele T Allele C Additive model rs3828336 CC CT TT Dominant model Allele C Allele T Additive model # ¶ +
0.004
0.145¶
0.115
1.000 (reference) 0.564 (0.312–1.020) 2.161 (0.194–24.049) 0.594 (0.334–1.056) 1.000 (reference) 0.649 (0.377–1.115) 0.638 (0.368–1.107)
P+
b0.001 0.654 b0.001 0.004 0.005
0.058 0.531 0.076 0.117 0.110
P values were computed by Pearson Chi-Square test. P values were computed by Fisher's exact test. Data were calculated by logistic regression after adjusting for sex.
Trs2292832-Trs3828336 were 0.569 (95% CI = 0.414–0.783) and 0.528 (95% CI = 0.301–0.926), respectively. 2.4. Bioinformatics analysis According to UCSC [22], rs2292832 was situated within the first intron of GPC1 and the binding site of a transcription factor UBTF (upstream binding transcription factor, RNA polymerase I). At the same time, it overlapped Deoxyribonuclease I (DNase I) hypersensitive sites (DHSs) in 40 cell lines including one primary hepatocytes cell line ‘Hepatocytes’ and three hepatocellular carcinoma cell lines ‘HepG2’, ‘Huh-7’ and ‘Huh-7.5’, with a highest cluster score 1000 (where score is based on signal strength). 3. Discussion BA is a destructive inflammatory obliterative cholangiopathy [2] occurring exclusively in the neonatal period, and it has been the leading cause of liver transplantation in children for the past 20 years. Meanwhile, uncovering the cause of BA remains a major challenge. Recently, several inspiring findings were reported by Cui et al. [17], who convincingly specified GPC1 as a risk gene for BA. Analysis of copy number variant (CNV) differing between BA patients and controls revealed a notable region 2q37.3, which corresponded to the loss of GPC1. In addition, they took advantage of zebrafish which were knockdown of gpc1, and determined the mechanistic insight that deletion of GPC1 over-activating Hedgehog signaling resulted in biliary development defects. Accordingly, we selected two SNPs of GPC1 systematically and performed a case–control study. The results of our study indicated that the minor allele Crs2292832 was significantly associated with reduced Table 2 Risk estimates for extended GPC1 haplotypes in BA cases and controls. Haplotype
Ta-Cb Ca-Cb&Tb Ta-Tb P for trend
Control
Case
No. of Chromosomes (%)
No. of Chromosomes (%)
741 (60.0) 387 (31.3) 108(8.7)
195 (72.8) 58 (21.6) 15 (5.6)
P#
b0.001
OR (95% CI)+
P+
1.000(reference) 0.569(0.414–0.783) 0.528(0.301–0.926) b0.001
0.001 0.026
a: rs2292832. b: rs3828336. # P values were computed by the Pearson chi-square test. + Data were calculated by logistic regression after adjusting for sex.
risk of BA in heterozygous, dominant, allelic and additive genetic models, but not in the homozygous possibly because of our relative small sample size. And the over representation of rs2292832 homozygotes in our cases was likely because of the loss of heterozygosity (LOH) of 2q27.3. And it might reflect the presence of LOH in some of our samples as Cui et al. reported. On the other side, we found no positive association for rs3828336, but still a marginal effect in CT genotype, which implied the possibility of protective effect of Trs3828336. In the haplotypes Crs2292832-Crs3828336&Trs3828336 and Trs2292832-Trs3828336, BA risk reduced by approximately 50% referring to the haplotype of two major alleles. Our findings are biologically plausible. GPC1 encodes glypican 1, which belongs to glypican family of heparan sulfate proteoglycans, adjoins to the cell membrane by a glycosyl-phosphatidylinositol linkage [24]. Glypicans (GPCs) could modulate several signals containing Hedgehog (Hh), Wingless (Wnt), fibroblast growth factor, and bone morphogenetic protein [25], each of which play an important role in hepatogenesis. Cui et al. observed no apical GPC1 immunostaining in cholangiocytes and suggested the decreased amount of GPC1 in the BA liver samples [17], which might reflect the expression or functional abnormality of GPC1 in patients. Given that GPCs are known to regulate Hh activity and increased Hh activity is observed in BA patients [26], Cui et al. further confirmed that over-activated Hh was the basis of the GPC1-deficient phenotype and responsible for biliary defects [17]. Therefore, we speculated that the investigated polymorphisms might upregulate expression of GPC1, thereby being antagonistic to Hh signaling and, as a consequence, impairing the occurrence and development of BA. This presumption was supported by bioinformatics analysis in UCSC [22] integrating with Encode data [23]: rs2292832 was located in the binding site of UBTF, which played a critical role in ribosomal RNA transcription as a key component of the pre-initiation complex, mediating the recruitment of RNA polymerase I to ribosomal RNA promoter regions [27]. Under the consideration of that binding site variations contributed extensively on transcription factor binding and often correlated with differences in gene expression [28], we assumed that rs2292832 could alter the transcription and expression of GPC1 through the affection of UBTF's binding. In addition, it is strongly suggested that rs2292832 lying in DHSs functions as a regulatory DNA variation [29], whereas DHSs are sensitive and precise markers of actuated cis-regulatory elements [30]. However, the assumption above needs relevant experiments to be verified. Several limitations of this study should be noted. First of all, we neglected the possibility that the reported SNP association was partially synthesizing the rare deletion effect. Second, the strategy of screening candidate common polymorphisms depended on the prediction from
Please cite this article as: Ke J, et al, Common genetic variants of GPC1 gene reduce risk of biliary atresia in a Chinese population, J Pediatr Surg (2016), http://dx.doi.org/10.1016/j.jpedsurg.2016.05.009
4
J. Ke et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
SNPinfo on the basic of SNPs' position, which was not rigorous enough to discover all possibly functional SNPs including rare variants. Third, the sample size of our case–control study was relatively small. Forth, insufficient environmental information limited us to further investigate the gene–environment interactions, considering that BA is a complex trait. Fifth, lacking of functional experiments on relevant tissues and cells, biological reality of the statistically significant association we reported is still uncertain. In sum, our independent case–control study implicated that common SNPs in GPC1 gene contributed to BA susceptibility in Chinese population, in addition to the large genomic indels discovered before. Comprehensive researches with more representative SNPs, different races, greater sample sizes and follow-up functional analyses are warranted to identify causal variants and elaborate the biological mechanism of genetic etiology. Supplementary data to this article can be found online at http://dx.doi. org/10.1016/j.jpedsurg.2016.05.009. Acknowledgements The authors wish to thank the numerous patients and their families who have participated in these studies, and all the study participants, research staff and students who participated in this work. The studies reported here were supported by Shenzhen Science and Technology Project to Shenzhen Children's Hospital (JCYJ20150403100317072). References [1] Superina R, Magee JC, Brandt ML, et al. The anatomic pattern of biliary atresia identified at time of Kasai Hepatoportoenterostomy and early postoperative clearance of jaundice are significant predictors of transplant-free survival. Ann Surg 2011;254:577–85. [2] Hartley JL, Davenport M, Kelly DA. Biliary atresia. Lancet 2009;374:1704–13. [3] Bassett MD, Murray KF. Biliary atresia: recent progress. J Clin Gastroenterol 2008;42: 720–9. [4] Houwen RH, Kerremans II, van Steensel-Moll HA, et al. Time–space distribution of extrahepatic biliary atresia in the Netherlands and West Germany. Z Kinderchir 1988;43:68–71. [5] Chardot C, Carton M, Spire-Bendelac N, et al. Epidemiology of biliary atresia in France: a national study 1986-96. J Hepatol 1999;31:1006–13. [6] Yoon PW, Bresee JS, Olney RS, et al. Epidemiology of biliary atresia: a populationbased study. Pediatrics 1997;99:376–82. [7] McKiernan PJ, Baker AJ, Kelly DA. The frequency and outcome of biliary atresia in the UK and Ireland. Lancet 2000;355:25–9.
[8] Hsiao CH, Chang MH, Chen HL, et al. Universal screening for biliary atresia using an infant stool color card in Taiwan. Hepatology 2008;47:1233–40. [9] Mack CL. The pathogenesis of biliary atresia: evidence for a virus-induced autoimmune disease. Semin Liver Dis 2007;27:233–42. [10] Harper P, Plant JW, Unger DB. Congenital biliary atresia and jaundice in lambs and calves. Aust Vet J 1990;67:18–22. [11] Davit-Spraul A, Baussan C, Hermeziu B, et al. CFC1 gene involvement in biliary atresia with polysplenia syndrome. J Pediatr Gastroenterol Nutr 2008;46:111–2. [12] Arikan C, Berdeli A, Kilic M, et al. Polymorphisms of the ICAM-1 gene are associated with biliary atresia. Dig Dis Sci 2008;53:2000–4. [13] Arikan C, Berdeli A, Ozgenc F, et al. Positive association of macrophage migration inhibitory factor gene-173G/C polymorphism with biliary atresia. J Pediatr Gastroenterol Nutr 2006;42:77–82. [14] Shih HH, Lin TM, Chuang JH, et al. Promoter polymorphism of the CD14 endotoxin receptor gene is associated with biliary atresia and idiopathic neonatal cholestasis. Pediatrics 2005;116:437–41. [15] Huang YH, Huang CC, Chuang JH, et al. Upstream stimulatory factor 2 is implicated in the progression of biliary atresia by regulation of hepcidin expression. J Pediatr Surg 2008;43:2016–23. [16] Leyva-Vega M, Gerfen J, Thiel BD, et al. Genomic alterations in biliary atresia suggest region of potential disease susceptibility in 2q37.3. Am J Med Genet A 2010;152A: 886–95. [17] Cui S, Leyva–Vega M, Tsai EA, et al. Evidence from human and zebrafish that GPC1 is a biliary atresia susceptibility Gene. Gastroenterology 2013;144:1107–1115.e1103. [18] Xu Z, Taylor JA. SNPinfo: integrating GWAS and candidate gene information into functional SNP selection for genetic association studies. Nucleic Acids Res 2009;37: W600–5. [19] Sunyaev S, Ramensky V, Koch I, et al. Prediction of deleterious human alleles. Hum Mol Genet 2001;10:591–7. [20] Barrett JC, Fry B, Maller J, et al. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005;21:263–5. [21] Stephens M, Smith NJ, Donnelly P. A new statistical method for haplotype reconstruction from population data. Am J Hum Genet 2001;68:978–89. [22] Karolchik D, Barber GP, Casper J, et al. The UCSC genome browser database: 2014 update. Nucleic Acids Res 2014;42:D764–70. [23] Bernstein BE, Birney E, Dunham I, et al. An integrated encyclopedia of DNA elements in the human genome. Nature 2012;489:57–74. [24] Kirn-Safran C, Farach-Carson MC, Carson DD. Multifunctionality of extracellular and cell surface heparan sulfate proteoglycans. Cell Mol Life Sci 2009;66:3421–34. [25] Yan D, Lin X. Shaping morphogen gradients by proteoglycans. Cold Spring Harb Perspect Biol 2009;1:a002493. [26] Omenetti A, Bass LM, Anders RA, et al. Hedgehog activity, epithelial-mesenchymal transitions, and biliary dysmorphogenesis in biliary atresia. Hepatology 2011;53: 1246–58. [27] Pruitt KD, Brown GR, Hiatt SM, et al. RefSeq: an update on mammalian reference sequences. Nucleic Acids Res 2014;42:D756–63. [28] Kasowski M, Grubert F, Heffelfinger C, et al. Variation in transcription factor binding among humans. Science 2010;328:232–5. [29] Maurano MT, Humbert R, Rynes E, et al. Systematic localization of common diseaseassociated variation in regulatory DNA. Science 2012;337:1190–5. [30] Thurman RE, Rynes E, Humbert R, et al. The accessible chromatin landscape of the human genome. Nature 2012;489:75–82.
Please cite this article as: Ke J, et al, Common genetic variants of GPC1 gene reduce risk of biliary atresia in a Chinese population, J Pediatr Surg (2016), http://dx.doi.org/10.1016/j.jpedsurg.2016.05.009