An Intronic Variant Associated with Systemic Lupus Erythematosus Changes the Binding Afﬁnity of Yinyang1 to Downregulate WDFY4

ORIGINAL ARTICLE An intronic variant associated with systemic lupus erythematosus changes the binding afﬁnity of Yinyang1 to downregulate WDFY4

H Zhao1, W Yang2, R Qiu1,JLi1, Q Xin1, X Wang1, Y Feng1, S Shan1, Y Liu1, Y Gong1 and Q Liu1

Two recent genome-wide association studies of East Asian populations revealed three genetic variants in WDFY4/LRRC18 associated with systemic lupus erythematosus (SLE). To identify the gene contributing to this disease susceptibility, we examined the mRNA expression of WDFY4 and LRRC18 in patients with SLE and healthy controls. WDFY4 was significantly downregulated in SLE patients as compared with controls. We used allelic expression and dual-luciferase assays to identify the functional variant. Transcriptional activity was lower for the rs877819A than -G allele. Electrophoretic mobility shift and supershift assays revealed that the transcription factor Yinyang1 (YY1) binds to rs877819, with lower affinity to the A allele, which explained the reduced transcriptional activity. This effect was further confirmed by YY1 small interfering RNA knockdown, overexpression and chromatin immunoprecipitation experiments. rs877819 in WDFY4 might be the functional site associated with SLE by reduced binding of YY1 and downregulating WDFY4 expression.

Genes and Immunity (2012) 13, 536–542; doi:10.1038/gene.2012.33; published online 13 September 2012 Keywords: Systemic lupus erythematosus; WDFY4; LRRC18; susceptibility gene; single nucleotide polymorphism; causative variant

INTRODUCTION all 61 coding exons of WDFY4 in 70 normal individuals but found Systemic lupus erythematosus (SLE) is an autoimmune disease no novel variants in coding sequences. Our sample size has the characterized by autoantibody production, complement activation power of 76 and 94% for detecting the rare variants with minor and organ failure, and is influenced by both genetic and allele frequency 41% and 2%, respectively. environmental factors.1,2 The exact pathophysiological features of SLE are not completely understood.3 The probability of SLE occurrence is increased 20- to 50-fold among first-degree relatives Expression of WDFY4 and LRRC18 in peripheral blood as compared with the general population. The concordance of the mononuclear cells from patients with SLE and healthy controls disease in monozygotic twins is B24–69% and only 2–9% in The susceptibility SNP rs1913517 is located in an intron in LRRC18, dizygotic twins.4–6 Therefore, the genetic factors have a significant which itself is located in an intron of WDFY4 (Figure 1a); thus role in the pathogenesis of SLE. The genes related to SLE are which gene is directly associated with SLE is unclear. To identify increasingly being identified in various populations,7–9 although the susceptibility gene, we examined the mRNA levels of the two research into their functional aspects is rare. genes in peripheral blood mononuclear cell (PBMC) samples from Recently, two genome-wide association studies (GWAS) of Asian 58 SLE patients and 58 healthy controls. Relative WDFY4 mRNA populations identified three single-nucleotide polymorphisms expression was lower in patients with SLE than controls, with no (SNPs) in WDFY4, rs877819, rs7097397 and rs1913517, which was significant difference between the groups in LRRC18 expression found to be related to SLE.10,11 rs1913517 is located in LRRC18, in (Figures 1b and c). an intron of WDFY4. However, little is known about the role of WDFY4 and LRRC18 as the risk factors of susceptibility to SLE. Here, we confirmed the decreased expression of WDFY4 in SLE Identifying potential causal variant among rs1913517, rs877819 patients, and demonstrated that the intronic rs877819 may be the and rs7097397 causal variant that changes the binding affinity of a regulatory The relative positions of rs1913517, rs877819 and rs7097397 transcription factor to downregulate WDFY4 expression. showing positive association with SLE in two GWAS are in Figure 1a. To identify the functional variant among these SNPs, we conducted an independent contribution analysis and ASE assay. As compared with rs1913517, rs877819, associated with SLE RESULTS with P ¼ 2.96 Â 10 À 7, might have an independent effect with Resequencing coding sequence of WDFY4 for rare variant P ¼ 5.35 Â 10 À 4 in the same study.11 Conditional logistic Although GWAS have identified common variants for complex regression revealed rs877819 with an independent contribution, diseases, the common variants explain only a small fraction of the at P ¼ 2.5 Â 10 À 4, when controlling for rs1913517. However, phenotypic variation, and the rare variants with a large effect rs1913517 did not have an independent effect with P ¼ 0.7758 might help predict individual risk. Because a rare variant in WDFY4 when controlling for rs877819. Also, imputation results showed might have a functional effect on SLE susceptibility, we sequenced that rs877819 has one of the most significant association of

1Key Laboratory for Experimental Teratology of the Ministry of Education and Department of Medical Genetics, Shandong University School of Medicine, Shandong, PRC and 2Department of Paediatrics and Adolescent Medicine, LKS Faculty of Medicine, The University of Hong Kong, Pokfulam, Hong Kong. Correspondence: Dr Q Liu, Key Laboratory for Experimental Teratology of the Ministry of Education and Department of Medical Genetics, Shandong University School of Medicine, Jinan, Shandong 250012, China. E-mail: [email protected] Received 15 May 2012; revised and accepted 16 July 2012; published online 13 September 2012 An intronic variant associated with SLE H Zhao et al 537

Figure 1. WDFY4 and LRRC18 mRNA expression in peripheral blood mononuclear cells from patients with SLE (n ¼ 60) and healthy controls (n ¼ 60). (a) Relative chromosome position of WDFY4 and LRRC18 and linkage disequilibrium patterns for some of the SNPs in this region. The SNP is marked with vertical lines. (b) Relative expression of WDFY4 and (c) LRRC18 in cases and healthy controls (Ctrl). Each dot represents the mean of a duplicated assay; bars represent mean values of four replicates for each subject.

P-values, which is much better than those from the other two pyrosequencing, for it is located at the transcript sequence. Most SNPs described in this study as well as previous studies on this of the mean values of A/(G þ A) ratios were between 0.4 and 0.6, locus (Supplementary Figure 1). with no signiﬁcant difference in expression between the two Next, we genotyped healthy control samples for rs877819 alleles of rs7097397 (Figure 2a). and rs7097397. To detect whether WDFY4 expression was To determine whether intronic rs877819 was associated with associated with rs877819 or rs7097397, we performed ASE assay WDFY4 and LRRC18 expression, we compared the expression of of 31 healthy subjects heterozygous for rs7097397 by using subjects with different genotypes on the SNP. Because of its low

& 2012 Macmillan Publishers Limited Genes and Immunity (2012) 536 – 542 An intronic variant associated with SLE H Zhao et al 538

Figure 2. ASE assays of SNP rs877819 in WDFY4 and LRRC18 gene, and rs7097397 in WDFY4 gene. (a) Reverse transcriptase–PCR (RT–PCR) analysis of relative mRNA level of WDFY4 in 40 individuals with GG and 15 with AG alleles from rs877819. (b) X axis represents the allele frequencies for rs7097397 corrected by the genomic DNA ratio for 31 heterozygous individuals. (c) RT–PCR analysis of relative mRNA level of LRRC18 in 47 individuals with GG alleles and 25 individuals with AG alleles of rs877819. Each dot represents the mean of a duplicated assay; bars represent mean of four replicates for each subject.

Figure 3. Sequence conservation of the intronic variation rs877819 between human, gorilla and mouse.

frequency, the AA genotype (only two samples) was not included in the analysis. Quantitative PCR analysis showed no difference on LRRC18 expression, but signiﬁcantly lower WDFY4 expression was seen in AG heterozygotes than in GG homozygotes (Po0.0001, Figures 2b and c). Thus, subjects with SLE carrying risk allele A may have a signiﬁcantly lower WDFY4 expression than those with the G allele. In addition, the ancestral allele G of rs877819 is conserved among human, gorilla and mouse, which also suggests that the variant site might be functional (Figure 3).

Comparison of the relative transcription activity of the two rs877819 alleles Although rs877819 is located in an intron, an intronic SNP might still affect gene expression through transcriptional regulation.12 To examine this possibility, we generated two luciferase reporter constructs: pGL3Pro-G or -A, which contained the pGL3-promoter, and the genomic fragment that carries, rs877819A or -G allele. HeLa and HEK-293T cells were transfected with these reporter plasmids. At 24 h after transfection, we determined luciferase activity for each reporter relative to the basal empty reporter activity. The activity was signiﬁcantly lower for the construct carrying the A allele than G allele (Figure 4). This ﬁnding is consistent with our observation of the rs877819 risk allele A associated with low-transcript activity and reduced WDFY4 expression.

Transcription factors binding to rs877819 and allelic difference in Figure 4. Dual-luciferase assay for allele A and G of rs877819. HeLa nuclear protein-binding afﬁnity (a) and HEK-293T (b) cells were transfected with pGL3-promoter Gene transcription is modulated by regulatory-factor binding sites (pGL3Pro), pGL3Pro-G and -A, with a modiﬁed pRL-TK vector. Data in intronic regions.12,13 We used electrophoretic mobility shift are mean±s.d. ratios from four experiments. **Po0.01, ***Po0.001 assay (EMSA) to investigate whether the allelic difference in compared with pGL3Pro-G. transcription activity from intronic SNP might be explained by their different binding capacities for nuclear factors. We synthesized two biotin-labeled probes corresponding to the Probe G could form DNA–protein complexes with the nuclear 27-bp sequence in rs877819, and used unlabeled oligo- extracts (Figure 5). With the addition of excessive allelic nucleotides as ‘competitors’ (cold probe). Nuclear extracts were competition probes in different concentrations, the band intensity prepared from HeLa cells. was reduced in a concentration-dependent manner and abolished

Genes and Immunity (2012) 536 – 542 & 2012 Macmillan Publishers Limited An intronic variant associated with SLE H Zhao et al 539 by 50-fold excess G, whereas inclusion of 200-fold excess previously, and neither was affected by the mutant probe that competition probe A did not significantly alter the band intensity demonstrates the binding specificity (Figure 5c). (Figure 5a). Therefore, rs877819 could bind with some nucleopro- These DNA binding analyses indicate that endogenous YY1 teins, with the affinity of G higher than A. We used MatInspector protein specifically interacted with intronic rs877819 within (www.genomatix.de) to identify which protein(s) bind at or near WDFY4, with the risk-associated A allele having lower affinity than rs877819 to regulate the transcription of WDFY4, and identified the G allele in binding to YY1, which might explain why allele A the transcription-factor candidates Freac2, ZBTB7 and Yinyang 1 was associated with reduced WDFY4 transcription. (YY1). To establish whether the binding also occurs in vivo,we To verify the predicted transcription factors, we synthesized performed chromatin immunoprecipitation assay with the HeLa conserved motifs for the three factors as competition probes. cell line and YY1 antibody. A pair of primers was designed to The detected complex was completely abrogated only by the amplify a 254-bp fragment containing rs877819. YY1 bound to the conserved YY1 binding motif, thus YY1 might be the transcription rs877819 site in vivo confirms the EMSA data (Figure 5e). factor binding to the rs877819 site (Figure 5b). To further examine whether YY1 could bind to this site, we performed supershift assay with the addition of YY1 antibody, and the band disappeared The effect of transcription factor YY1 on different alleles of (Figure 5d). To test the specificity of this binding, we biotin-labeled rs877819 the conserved YY1 binding motif as a digoxin-labeled probe, and Because YY1 could bind to intronic rs877819 that was associated the YY1 mutation probe with a three-base mutation that was with WDFY4 transcription, we then wondered whether the allelic synthesized as a competition sequence. As expected, the YY1 ‘hot’ difference of rs877819 in regulatory activity was attributable to probe produced a band similar to that of the band described different binding affinity for YY1. We investigated whether YY1

Figure 5. Comparison of allele differential DNA–protein interaction in HeLa cells. EMSA was used to test DNA–protein interaction on alleles G and A of rs877819 (a). Conserved binding sequences of transcription factor YY1, Freac2 and ZBTB7 were used to compete for binding bands in (b) and (c). (d) Supershift assay of YY1. Nuclear extracts were obtained from HeLa cells and competitor oligonucleotides are indicated above each lane. (e) Chromatin immunoprecipitation assay of YY1 and control Immunoglobulin G; non-immunoprecipitated input DNA is shown on the left.

& 2012 Macmillan Publishers Limited Genes and Immunity (2012) 536 – 542 An intronic variant associated with SLE H Zhao et al 540 could regulate WDFY4 transcriptional activity by binding to this expression construct (CMV-YY1). Cells transfected with CMV-YY1 SNP. YY1 small interfering RNA (siRNA) targets, and normal constructs had a high level of YY1, for an B1.5-fold increase in controls were used to transiently transfect HeLa cells, which were relative luciferase activity with pGL3Enh-G. However, for the A transfected together with the constructed vectors pGL3Enh-G and -A. allele, the relative luciferase activity did not change significantly, As compared with the normal controls, HeLa cells with YY1 siRNA which was similar to the results obtained from the knockdown showed significantly reduced relative luciferase activity of experiments (Figure 6b). In addition, we cotransfected HeLa cells pGL3Enh-G (Figure 6a). However, the risk allele was not with pGL3Enh-G and YY1 expression constructs at different significantly affected by YY1 knockdown, which suggests that concentrations. The relative activity of G allele increased with YY1 is a positive regulator of the G allele with almost no effect on the CMV-YY1 concentration from 20–200 ng, which further the A allele. supports the role of YY1 as a positive regulation factor by binding Next, to measure the effect of YY1 overexpression on rs877819 to the G allele (Figure 6c). Thus, YY1 had a positive role in activity, we cotransfected HeLa cells with the constructed vectors transcription activity of the G allele from rs877819, with little role pGL3Enh-G or -A and an empty vector control (pcDNA3.1) or a YY1 in that of the A allele. Moreover, we also instantaneously

Figure 6. The effect of transcription factor YY1 on luciferase activity driven by rs877819G and rs877819A. (a) HeLa cells were transfected with pGL3Enh-G or -A pcDNA3.1 empty vector and cytomegalovirus (CMV)-YY1. (b) Cells were cotransfected with pGL3Enh-G or -A and siRNA normal control (siNC) or siRNA for YY1. (c) Cells were co-transfected with 200 ng pGL3Enh-G and CMV-YY1 vector from 20–200 ng. (d, e) The effect on the relative luciferase activity of pGL3Enh-G or -A and (f, g) W-pro-G or -A was respectively detected with YY1 knockdown or YY1 overexpression. Data are mean±s.d. from three experiments. *P ¼ 0.05, **Po0.01, ***Po0.001. N.S., not signiﬁcant.

Genes and Immunity (2012) 536 – 542 & 2012 Macmillan Publishers Limited An intronic variant associated with SLE H Zhao et al 541 transfected HeLa cells using siYY1 or CMV-YY1, together with ampliﬁcation conditions were 94 1C for 5 min, 35 cycles at 94 1C for 30 s, pGL3Pro-G or -A or W-pro-G or -A. A similar result was obtained as optimal annealing temperature for 30 s, 72 1C for 30 s and a ﬁnal extension previously, which showed that transcription factor YY1 had a at 72 1C for 10 min. All sequencing data were produced by the Genomic much greater impact on the relative luciferase activity of G allele Analysis Facility (Huada, Shenzhen, China) and analyzed with the use of than A allele (Figures 6d–g). Chromas 2.33 (Technelysium Pty, South Brisbane, QLD, Australia).

Genotyping DISCUSSION Genomic DNA was extracted from peripheral blood leukocytes by a Previous studies have shown SNPs in WDFY4 or LRRC18 associated standard phenol–chloroform method. Sanger sequencing was used for with SLE,10,11 but the causative gene or functional variants genotyping. Primers flanking the variant were designed with the use of Primer 5.0. PCR reaction and sequencing were performed as described remained unclear. Here, we demonstrated that a common previously. intronic WDFY4 variant, rs877819 G 4A, influenced WDFY4 gene expression by affecting the binding of YY1. This finding may explain the mechanism of this SNP conferring susceptibility to SLE. Reverse transcriptase–PCR The zinc finger protein YY1 is a transcriptional activator or Total peripheral leukocyte RNA was extracted from 58 SLE patients and 58 repressor of several genes, including some critical genes in B and T healthy controls by the Trizol method (Invitrogen, Carlsbad, CA, USA), then cells, which are central to the pathogenesis of autoimmune underwent reverse transcription by the use of a kit (Takara, Otsu, Shiga, 14–16 Japan). ASE for WDFY4 and LRRC18 were analyzed only in healthy controls. response. For example, YY1 can enhance FCGR2B promoter Complementary DNA (cDNA) was amplified by real-time PCR with an ABI activity by a functional promoter haplotype, which is also 7500 real-time PCR system. The expression of GAPDH was measured as an 17 associated with SLE. A genome-wide expression analysis internal control. Four duplicate wells were used for each subject. revealed that YY1 may have a key role in the pathogenesis of 18 multiple sclerosis or its subtypes. In our study, we also showed Pyrosequencing that YY1 could have a role in transcription of WDFY4, with different affinity for a functional variant in the gene. However, whether and To test the ASE of rs7097397, we extracted total peripheral leukocyte RNA from 31 healthy controls heterozygous for rs7097397 by the Trizol method how the decreased expression of WDFY4 contributes to the (Invitrogen), which underwent reverse transcription by the use of a kit pathogenesis of SLE still requires further research. An intron is not (Takara). RNA samples were treated with RNase-free DNAase to eliminate part of mature mRNA or a promoter complex; but it may still genomic DNA contamination before being reverse-transcribed into cDNA. modulate DNA–protein interactions and affect gene expres- cDNA was amplified by PCR with transcript-specific primers. The average sion.12,19,20 Our study provides experimental evidence for the A/G cDNA expression ratio of each individual was normalized by role of YY1 in autoimmune disease, and shows that YY1, a the genomic DNA ratio from the same sample. Pyrosequencing involved ubiquitously expressed transcription factor, can bind to the the use of a PyroMark ID machine for three times for each sample. The A/(A þ G) ratio result of 0.5 indicated ASE negative ( À ), and 0.4–0.6, ASE intronic SNP to affect the expression of WDFY4. 22 WDFY4 has multiple transcripts with unknown function, the positive ( þ ). longest transcript encodes 3184 amino acid residues. The human protein and that from other species show high sequence Plasmid construction conservation (84% similarity with Bos taurus and Canis lupus To compare the relative expression of different alleles on rs877819, familiaris, and 42% with Danio rerio).11 Similar to its closest genomic fragments around the variant were amplified from subjects paralog, WDFY3, WDFY4 contains WD40 domains and a BEACH homozygous for rs877819, and cloned into the pGL3-promoter or with a truncated FYVE zinc finger domain. Although little is known -enhancer luciferase vector (Promega, Fitchburg, WI, USA) as pGL3Pro-G about this well-conserved protein, the gene is predominantly or -A and pGL3Enh-G. We also generated pGL3-basic-WDFY4-G or -A constructs (W-pro-G or -A) that contained the confirmed WDFY4 promotor expressed in immune tissues such as lymph node, spleen, thymus sequence (800 bp from the transcription start site) and the genomic and tonsil, thus it may be important in autoimmune disease. fragment of rs877819A or G. The above genomic fragments were inserted In conclusion, we performed a functional analysis of WDFY4 and into a pGL3-basic vector as a negative control. The YY1 expression vector demonstrated that its downregulation may explain the increased was kindly provided by Dr Naoko Tanese (NYU School of Medicine). risk of SLE. In addition, an intronic SNP, rs877819, may be the All constructs were verified by direct sequencing. functional variant with the minor allele A, which is associated with decreased expression of WDFY4. Our findings may lend significant Cell culture and transfection insight into the search for functional variants for understanding HeLa and 293T cells were maintained in Dulbecco’s modified Eagle’s complex diseases. medium supplemented with 10% fetal bovine serum. To detect allelic activity difference, the pGL3Pro-G or -A constructs were transiently transfected into the two cell lines with the use of Lipofectamine 2000, MATERIALS AND METHODS with a modified pRL-TK vector. Patients and controls To determine the effect on allelic transcription activity of rs877819, HeLa cells were cotransfected by the use of Lipofectamine 2000 with the We recruited 70 patients with SLE and 108 sex- and age-matched healthy YY1 expressing vector or YY1 siRNA alone(50-GCU AGA AUG AAG CCA controls, all from a Chinese Han population living in Jinan, China. All of the AGA ATT-30), or combined with pGL3Enh-G or -A and a modified pRL-TK SLE patients fulfilled the American College of Rheumatology 1997 revised 0 21 plasmid. A universal negative siRNA (5 -UUC UCC GAA CGU GUC ACG criteria for SLE. Participants with other diseases or concurrent infection UTT-30) was used as control. were excluded. Blood samples were obtained from all participants in this study after informed consent. The protocol of this study was approved by the ethics review committee for human studies at the School of Medicine, Dual-luciferase reporter assay Shandong University. After transfection for 24 or 48 h, cells were extracted with lysis buffer, and relative dual-luciferase activity was measured by a dual-luciferase reporter assay system. The ratio of firefly to renilla luciferase was calculated for each Sequencing well, with four duplicates for each sample. All 61 exons of WDFY4 gene were sequenced in DNA samples from the 70 SLE cases to screen for genetic variants. The PCR reaction was run in a total volume of 50 ml solution containing 5 Â PCR buffer, 10 ml; 5 Â dNTP Electrophoretic mobility shift assay (1 mmol l À 1 each of dATP, dTTP, dGTP and dCTP), 10 ml; 50 ng genomic HeLa cells (1 Â 107) were used for preparing nuclear protein extracts. DNA; 0.4 mM of each primer and 1 U Taq DNA polymerase. The PCR For EMSA, double-stranded allelic probes containing rs877819 were

& 2012 Macmillan Publishers Limited Genes and Immunity (2012) 536 – 542 An intronic variant associated with SLE H Zhao et al 542 synthesized and labeled with digoxin. Conservative oligonucleotides for 4 Deapen D, Escalante A, Weinrib L, Horwitz D, Bachman B, Roy-Burman P et al. the transcription factors Freac2, ZBTB7 and YY1, and YY1 mutation probe A revised estimate of twin concordance in systemic lupus erythematosus. Arthritis were synthesized for competition assay. EMSA involved a gel-shift kit Rheum 1992; 35: 311–318. (Roche, Mannheim, Germany). 5 Sestak AL, Shaver TS, Moser KL, Neas BR, Harley JB. Familial aggregation of lupus For supershift assay, a specific YY1 antibody (ab12132; Abcam, UK) was and autoimmunity in an unusual multiplex pedigree. J Rheumatol 1999; 26: added to the binding reaction mixture. The YY1 consensus oligonucleotide 1495–1499. was used as a labeled probe and as an unlabeled competitor in EMSA. 6 Vyse TJ, Kotzin BL. Genetic susceptibility to systemic lupus erythematosus. Annu Rev Immunol 1998; 16: 261–292. Chromatin immunoprecipitation assay 7 Hom G, Graham RR, Modrek B, Taylor KE, Ortmann W, Garnier S et al. Association of systemic lupus erythematosus with C8orf13-BLK and ITGAM-ITGAX. N Engl J HeLa cells were crosslinked with 1.5% formaldehyde. The specific YY1 Med 2008; 358: 900–909. antibody or normal immunoglobulin G was used to immunoprecipitate 8 Gateva V, Sandling JK, Hom G, Taylor KE, Chung SA, Sun X et al. A large-scale protein–DNA complexes. Immunoprecipitated DNA was amplified by PCR replication study identifies TNIP1, PRDM1, JAZF1, UHRF1BP1 and IL10 as risk loci with the following primer sequences: forward, 50-TGGGAGCTAGGGAAG for systemic lupus erythematosus. Nat Genet 2009; 41: 1228–1233. GA-30, and reverse, 50-TCAATGTGAGCCAAGAGC-30, to produce an B200-bp 9 Harley JB, Alarcon-Riquelme ME, Criswell LA, Jacob CO, Kimberly RP, Moser KL fragment. et al. Genome-wide association scan in women with systemic lupus erythematosus identifies susceptibility variants in ITGAM, PXK, KIAA1542 and other loci. Western blot analysis Nat Genet 2008; 40: 204–210. Equal amounts (40 mg) of extracts underwent 10% SDS–polyacrylamide gel 10 Han JW, Zheng HF, Cui Y, Sun LD, Ye DQ, Hu Z et al. Genome-wide association electrophoresis, and then were transferred to a polyvinylidene difluoride study in a Chinese Han population identifies nine new susceptibility loci for membrane, which were incubated in nonfat milk in tris-buffered saline systemic lupus erythematosus. Nat Genet 2009; 41: 1234–1237. with Tween-20 (TBST; 0.1%) at room temperature for 1 h, and then 11 Yang W, Shen N, Ye DQ, Liu Q, Zhang Y, Qian XX et al. Genome-wide association immunoblotted with the primary antibodies for rabbit YY1 (ab109237; study in Asian populations identifies variants in ETS1 and WDFY4 associated with Abcam) and mouse anti-b-actin (Sigma) overnight at 4 1C. After a washing systemic lupus erythematosus. PLoS Genet 2010; 6: e1000841. with TBST, membranes were incubated with secondary antibodies for 1 h 12 Choi JW, Park CS, Hwang M, Nam HY, Chang HS, Park SG et al. A common intronic at room temperature. Chemiluminescence detection involved the ECL variant of CXCR3 is functionally associated with gene expression levels and the PLUS kit (Amersham Biosciences, Sunnyvale, CA, USA). polymorphic immune cell responses to stimuli. J Allergy Clin Immunol 2008; 122: 1119–1126, e1117. 13 Meyer KB, Maia AT, O’Reilly M, Teschendorff AE, Chin SF, Caldas C et al. Allele- Data analysis specific up-regulation of FGFR2 increases susceptibility to breast cancer. PLoS Biol Pairwise linkage disequilibrium (LD) was calculated by using the Haploview 2008; 6: e108. 4.2 program (http://www.broad.mit.edu/mpg/haploview/). Additional SNPs 14 Berndt JA, Kim JG, Tosic M, Kim C, Hudson LD. The transcriptional regulator Yin were imputed by IMPUTE2 using data from two Chinese GWAS. Data are Yang 1 activates the myelin PLP gene. J Neurochem 2001; 77: 935–942. expressed as mean±s.d., and analyzed by Mann–Whitney U-test with the 15 Hines RN, Luo Z, Hopp KA, Cabacungan ET, Koukouritaki SB, McCarver DG. Genetic use of SPSS v10.0 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 4, v4.03 variability at the human FMO1 locus: significance of a basal promoter yin yang 1 (GraphPad Software, La Jolla, CA, USA). The relative luciferase activities and element polymorphism (FMO1*6). J Pharmacol Exp Ther 2003; 306: 1210–1218. mRNA expression levels were performed with Student’s t-test. Po0.05 was 16 Zalani S, Coppage A, Holley-Guthrie E, Kenney S. The cellular YY1 transcription considered statistically significant. factor binds a cis-acting, negatively regulating element in the Epstein-Barr virus BRLF1 promoter. J Virol 1997; 71: 3268–3274. 17 Su K, Li X, Edberg JC, Wu J, Ferguson P, Kimberly RP. A promoter haplotype of the CONFLICT OF INTEREST immunoreceptor tyrosine-based inhibitory motif-bearing FcgammaRIIb alters The authors declare no conflict of interest. receptor expression and associates with autoimmunity. II. Differential binding of GATA4 and Yin-Yang1 transcription factors and correlated receptor expression and function. J Immunol 2004; 172: 7192–7199. ACKNOWLEDGEMENTS 18 Riveros C, Mellor D, Gandhi KS, McKay FC, Cox MB, Berretta R et al. A transcription factor map as revealed by a genome-wide gene expression analysis of whole- We thank all the participants in the study. This work was supported by grants from blood mRNA transcriptome in multiple sclerosis. PLoS One 2010; 5: e14176. the national Natural Science Foundation of China (grant no. 30771201, 81072452). 19 Alberobello AT, Congedo V, Liu H, Cochran C, Skarulis MC, Forrest D et al. An intronic SNP in the thyroid hormone receptor beta gene is associated with pituitary cell-specific over-expression of a mutant thyroid hormone receptor REFERENCES beta2 (R338W) in the index case of pituitary-selective resistance to thyroid hor- 1 Arbuckle MR, McClain MT, Rubertone MV, Scofield RH, Dennis GJ, James JA et al. mone. J Transl Med 2011; 9:144. Development of autoantibodies before the clinical onset of systemic lupus ery- 20 Williams HJ, Moskvina V, Smith RL, Dwyer S, Russo G, Owen MJ et al. Association thematosus. N Engl J Med 2003; 349: 1526–1533. between TCF4 and schizophrenia does not exert its effect by common non- 2 Ramos-Niembro F, Alarcon-Segovia D. Familial aspects of mixed connective tissue synonymous variation or by influencing cis-acting regulation of mRNA expression disease (MCTD). I. Occurrence of systemic lupus erythematosus in another in adult human brain. Am J Med Genet B Neuropsychiatr Genet 2011; 156B: 781–784. member in two families and aggregation of MCTD in another family. J Rheumatol 21 Hochberg MC. Updating the American College of Rheumatology revised criteria for 1978; 5: 433–440. the classification of systemic lupus erythematosus. Arthritis Rheum 1997; 40: 1725. 3 Loor F, Jachez B, Montecino-Rodriguez E, Klein AS, Kuntz L, Pflumio F et al. 22 Quinn EM, Hill M, Anney R, Gill M, Corvin AP, Morris DW. Evidence for cis-acting Radiation therapy of spontaneous autoimmunity: a review of mouse models. Int J regulation of ANK3 and CACNA1C gene expression. Bipolar Disord 2010; 12: Radiat Biol Relat Stud Phys Chem Med 1988; 53: 119–136. 440–445.

Supplementary information accompanies the paper on Genes and Immunity website (http://www.nature.com/gene)

Genes and Immunity (2012) 536 – 542 & 2012 Macmillan Publishers Limited