Allergy

ORIGINAL ARTICLE

EPIDEMIOLOGY AND GENETICS

A polymorphism in the TH2 locus control region is associated with changes in DNA methylation and gene expression M. Schieck1,2,*, V. Sharma2,*, S. Michel1,2,*, A. A. Toncheva1,2, L. Worth1, D. P. Potaczek2, J. Genuneit 3, A. Kretschmer4,5, M. Depner 6,7, J.-C. Dalphin8, J. Riedler9, R. Frei10,11, J. Pekkanen12,13, J. Tost14 & M. Kabesch1,2,7 1

Department of Pediatric Pneumology and Allergy, University Children‘s Hospital Regensburg (KUNO), Regensburg; 2Department of Pediatric Pneumology, Allergy and Neonatology, Hannover Medical School, Hannover; 3Institute of Epidemiology and Medical Biometry, Ulm University, Ulm; 4Research Unit of Molecular Epidemiology, Helmholtz Center Munich, Neuherberg; 5Clinic for Dermatology, Venerology and Allergy, University Clinic Schleswig-Holstein, Kiel; 6Dr. von Hauner Children’s Hospital, Ludwig Maximilians University Munich, Munich; 7German Lung
de Franche-Comte
, Research Center (DZL), Germany; 8Department of Respiratory Diseases, UMR CNRS Chrono-Environment, Universite Besancßon, France; 9Children’s Hospital, Schwarzach, Austria; 10Swiss Institute of Allergy and Asthma Research (SIAF), University of Zurich; 11 €hne-Center for Allergy Research and Education (CK-CARE), Davos, Switzerland; 12Department of Environmental Health, National Christine Ku Institute for Health and Welfare (THL); 13Unit of Public Health and Clinical Nutrition, University of Eastern Finland, Kuopio, Finland; 14
notypage, CEA-Institut de Ge
nomique, Evry, France Laboratory for Epigenetics and Environment (LEE), Centre National de Ge

To cite this article: Schieck M, Sharma V, Michel S, Toncheva AA, Worth L, Potaczek DP, Genuneit J, Kretschmer A, Depner M, Dalphin J-C, Riedler J, Frei R, Pekkanen J, Tost J, Kabesch M. A polymorphism in the TH2 locus control region is associated with changes in DNA methylation and gene expression. Allergy 2014; 69: 1171–1180.

Keywords 5q31; epigenetics; IgE; locus control region; RAD50. Correspondence Michael Kabesch, MD, Department of Pediatric Pneumology and Allergy, University Children’s Hospital Regensburg (KUNO), Campus St. Hedwig, Steinmetzstr. 1-3, D-93049 Regensburg, Germany. Tel.: +49-941-369-5801 Fax: +49-941-369-5802 E-mail: [email protected] *These authors contributed equally to this work. Accepted for publication 23 May 2014 DOI:10.1111/all.12450 Edited by: Michael Wechsler

Abstract Background: Genomewide association and epigenetic studies found a region within the RAD50 gene on chromosome 5q31 to be associated with total serum IgE levels and asthma. In mice, this region harbors a locus control region for nearby TH2 cytokines, which is characterized by four Rad50 DNase I hypersensitive sites (RHS4–7). Among these, RHS7 seems to have the strongest impact on TH2 differentiation. We investigated whether within the human homolog of RHS7, functional polymorphisms exist, which could affect DNA methylation or gene expression in the 5q31 locus and might have an influence on asthma status or IgE regulation. Methods: The human RHS7 region was fine mapped using 1000 genomes database information. In silico analysis and electrophoretic mobility shift assays were used to assess SNP function. Allele-specific effects on DNA methylation were evaluated in cord blood (n = 73) and at age of 4.5 years (n = 61) by pyrosequencing. Allele-specific effects on RAD50, IL4, and IL13 expression were analyzed in 100 subjects. Associations with asthma and IgE levels were investigated in the MAGICS/ISAAC II population (n = 1145). Results: Polymorphism rs2240032 in the RHS7 region is suggestive of allele-specific transcription factor binding, affects methylation of the IL13 promoter region and influences RAD50 and IL4 expression (lowest P = 0.0027). It is also associated with total serum IgE levels (P = 0.0227). Conclusion: A functional relevant polymorphism in the TH2 locus control region, equivalent to RHS7 in mice, affects DNA methylation and gene expression within 5q31 and influences total serum IgE on the population level.

In recent genomewide association studies (GWAS), we and others found that polymorphisms located on chromosome 5q31 were associated with asthma and total serum IgE levels

(1, 2). The association signals peaked in the RAD50 gene, which is located in the center of the TH2 cytokine locus harboring interleukin 4 (IL4), IL5, and IL13.

Allergy 69 (2014) 1171–1180 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

1171

Schieck et al.

Effects of RHS7 polymorphisms on 5q31

In the 30 region of mouse Rad50, a locus control region (LCR) with influence on the expression of the neighboring cytokines was identified (3). Within this LCR, four clusters of DNase I hypersensitive sites are situated (RHS4-7: RAD50 DNase I hypersensitive sites 4, 5, 6, and 7), indicating a less condensed chromatin structure to enable the expression of nearby genes (4). Interestingly, RHS4, 5, and 7 were specifically present in TH2 cells, but not in na€ıve T or TH1 cells, which suggests a defined binding of regulatory elements within these sites in TH2 cells. In line with this, acetylation of histones in TH2 cells was remarkably increased compared with na€ıve T or TH1 cells in the region of all four RHS but only RHS7 displayed a TH2-specific demethylation upon Tcell receptor stimulation, while RHS4–6 were constitutively methylated (4). From these data, Lee and colleagues hypothesized that especially RHS7 has a vital function in the LCR, and in a RHS7-deficient mouse model, they could show that RHS7 deletion causes a major reduction in IL4 and IL13 levels (5). In our previous epigenetic studies in children, the 5q31 region revealed differences in methylation from birth to early childhood, suggesting that epigenetic regulation of the region exists also in humans (6). In that study, a CpG island upstream of the transcription start site of IL13, analyzed with an amplicon termed IL13_a, showed significant methylation effects in two independent cohorts. Considering the high sequence conservation between mouse and human RHS7 (4), we speculated that this site has a functional role in the human TH2 locus as well (Fig. 1). Therefore, we investigated genetic variation in the human RHS7 and whether it is associated with DNA methylation or gene expression from the locus as well as with asthma status or IgE levels.

Methods Selection of polymorphisms and in silico analysis A region in human RAD50 with 70% homology to mouse RHS7 has been described previously (4). This human homolog of RHS7 was queried for single nucleotide polymorphism (SNP) information in the CEU population of the 1000 genomes dataset (rel. 2010_07) (7). In silico predictions for allele-specific transcription factor (TF) binding were performed with PROMO (8) and MatInspector (9) to identify transcription factor binding as potential targets for electrophoretic mobility shift assays (EMSA).

Electrophoretic mobility shift assay Nuclear protein extracts were prepared from Jurkat T cells (DSMZ, Braunschweig, Germany) after 3 h stimulation with PMA (50 ng/ml)–ionomycin (1 lM). EMSA were carried out using double-stranded (ds) DNA probes. Sequences of oligonucleotides used for probe preparation were as follows: rs2240032_‘C’: 50 -ACTCCTGGCTCCA GACAGTCCCTTTCTGGCA-30 ; rs2240032_‘T’: 50 -ACT CCTGGCTCCAGATAGTCCCTTTCTGGCA-30 ; SMAD3_ (1) (10): 50 -TCGAGAGCCAGACAAAAAGCCAGACATT TAGCCAGACAC-30 ; and SMAD3_(2) (11): 50 -GG GTGTCTAGACGGCC-30 . Sequences of complementary oligonucleotides are not shown. dsDNA probes were 50 labeled with 32P-c-ATP (PerkinElmer, Rodgau, Germany) using T4 polynucleotide kinase (New England BioLabs, Frankfurt/Main, Germany). Unbound radioactivity was removed using spin columns (Roche Diagnostics, Mannheim, Germany), and labeling efficiency was validated with scintillation counting. For EMSA experiments, 32P-labeled probes were incubated with Jurkat nuclear protein extract and resulting DNA–protein complexes were resolved in polyacrylamide gel electrophoresis. Autoradiography on film was used for visualization. To test specificity of DNA–protein binding, unlabeled probe was used in a 100fold molar excess as a binding competitor before incubation of nuclear protein extracts with 32P-labeled probes. DNA methylation analysis Allele-specific effects on methylation were investigated in children from the PASTURE cohort (protection against allergy: study in rural environments) (12) who were genotyped using MALDI–TOF MS (matrix-assisted laser desorption/ionization time-of-flight mass spectrometry)based primer extension assay from SEQUENOM (Sequenom, Inc., San Diego, CA, USA). DNA methylation was assessed by pyrosequencing in two independent subsamples of the PASTURE study (Table 1; discovery cohort and replication cohort). DNA was extracted from cord blood and from whole blood taken at the age of 4.5 years (FlexiGene DNA kit; Qiagen, Hilden, Germany). Details of the sodium bisulfite conversion (Qiagen) and the pyrosequencing experiments have previously been published (6). Experiments were performed with the PSQ 96MD Pyrosequencing system using the PyroGold SQA Reagent kit (Qiagen) and the Q-CpG software (V.1.0.9; Biotage, Uppsala, Sweden).

Abbreviations 18S rRNA, 18S ribosomal RNA gene; CEU, Utah residents with Northern and Western European ancestry from the CEPH collection; ds, double-stranded; EMSA, electrophoretic mobility shift assays; EXACT, expression analysis cohort; GWAS, genomewide association study; IgE, immunoglobulin E; IL, interleukin; ISAAC II, International Study of Asthma and Allergy in Childhood, phase II; MAF, minor allele frequency; MAGICS, Multicentre Asthma Genetics in Childhood Study; n, number of subjects; PASTURE, protection against allergy: study in rural environments; qRT-PCR, quantitative real-time polymerase chain reaction; RHS, RAD50 DNase I hypersensitive site; SNP, single nucleotide polymorphism; TF, transcription factor.

1172

Allergy 69 (2014) 1171–1180 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Schieck et al.

Effects of RHS7 polymorphisms on 5q31

Figure 1 TH2 locus on human chromosome 5q31. The positions of the conserved RAD50 DNase I hypersensitive sites 5–7 between mouse and human are marked [RHS5–7, as described by Fields et al. (4)]. Positions are given for rs2240032 and amplicon IL13_a,

which displayed allele-specific methylation effects for this polymorphism in the PASTURE population. This observation points to a function of RHS7 as a cis-acting locus control element on the TH2 locus.

peripheral blood mononuclear cells (PBMC) were stimulated with Derp1 (30 lg/ml) for 48 h and protein levels in the supernatant were measured with the Bio-Plex Human Cytokine 17-plex panel (Bio-Rad, Munich, Germany). All samples were measured in duplicates.

Expression analyses Effects of polymorphisms on gene expression were investigated in the EXpression Analysis CohorT (EXACT; Table 1). Genomic DNA was isolated from EDTA whole blood (FlexiGene DNA kit; Qiagen) in 100 randomly selected adults from Munich (39 individuals; 18 males) and Hannover (61 individuals; 28 males), Germany. Genotyping of rs2240032 was performed by direct Sanger sequencing. Total RNA was extracted from 2 ml of venous blood (NucleoSpin RNA Blood Mini kit; Macherey-Nagel, Du¨ren, Germany). Complementary DNA (cDNA) synthesis was carried out using QuantiTect reverse transcription kit (Qiagen) with input of 1 lg RNA per sample. Gene expression assays with fluorescently labeled TaqMan probes (50 -6-FAM, 30 -TAMRA) were established for IL13 and 18S rRNA, and predesigned TaqMan assays were used for IL4 and RAD50 (Life Technologies, Carlsbad, CA, USA). For all quantitative realtime polymerase chain reactions (qRT-PCR), mi-real-time Probe Master mix (ROX+) was used (Metabion, Martinsried, Germany) and efficiency of PCR (90–110%) was verified in a cDNA dilution series. All primers and assay conditions are available from the authors upon request. Gene expression levels (mRNA) were evaluated based on the delta Ct (dCt) method (13). For each sample, the dCt was calculated by subtracting the Ct of the endogenous control 18S rRNA from the Ct of the gene of interest. The higher dCt values represent lower mRNA levels. Protein measurements of IL4 and IL13 were available for n = 61 subjects of the EXACT population. From these,

Population genetics Genetic associations with total serum IgE levels and asthma were assessed in individuals from MAGICS and ISAAC II, which represent two distinct cohorts from the same ethnic population source used for the first GWAS on asthma, as previously described (14). Asthma cases (n = 581) were recruited from the Multicentre Asthma Genetics In Childhood Study (MAGICS), while from the large (n = 5629) International Study of Asthma and Allergy in Childhood, phase II (ISAAC II) (15), a reference sample was drawn (n = 730; Table 1). Total serum IgE levels were measured in both cohorts using the IMMULITE 2000 system (Siemens Healthcare Diagnostics, Eschborn, Germany). The rs2240032 genotype information in MAGICS/ISAAC II was derived from an imputation based on HapMap II data. Details of the recruitment, chip-genotyping (Sentrix HumanHap300 BeadChip, Illumina Inc., San Diego, CA, USA), and imputation have previously been described (16, 17). Statistical analyses Linkage disequilibrium (LD) between polymorphisms was assessed with Haploview software version 4.2 (18). Additive

Table 1 Overview of populations used for specific analyses in this study

Population

Subjects

PASTURE cord blood

43 Discovery cohort 31 Discovery cohort 100 1,311

PASTURE age 4.5

EXACT MAGICS/ISAAC II

30 Replication cohort 30 Replication cohort

Mean age

rs2240032 MAF ‘T’

Methylation analysis

Expression analysis

Total IgE analysis

Doctor’s diagnosis of asthma

0

0.21

x

–

–

–

4.5

0.15

x

–

–

–

Adults 10.1

0.22 0.26

– –

x –

– x

– x

For all populations, the number of subjects and their mean age are given. The ‘x’ indicates that a specific analysis of this study was carried out in the respective population.

Allergy 69 (2014) 1171–1180 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

1173

Schieck et al.

Effects of RHS7 polymorphisms on 5q31

genetic effects on asthma and atopic asthma were modeled by logistic regression using Plink (version 1.07) (19). Allelespecific differences in DNA methylation, gene and protein expression, and total serum IgE were assessed by nonparametric Wilcoxon–Mann–Whitney U-tests using R software (20). For subjects of the EXACT population with no detectable IL13 gene expression in qRT-PCR (n = 25/100), the maximum PCR cycle of 40 was considered for dCt calculations. Consequently, to analyze these censored data, a generalized Wilcoxon test was applied (13). For all analyses, a P-value < 0.05 was considered to be significant. As the ‘TT’ genotype was rare in the cohorts, ‘CT’ and ‘TT’ carriers were combined and compared against ‘CC’ carriers. The DNA methylation was analyzed separately for cord blood and 4.5 year samples. Results Fine mapping in the 1000 genomes CEU dataset identified only one polymorphism, rs2240032, in the human equivalent of mouse RHS7. We applied bioinformatic tools to determine whether rs2240032 has an allele-specific functional effect. Using PROMO, the ‘T’-allele of rs2240032 was predicted to create a ‘GATAG’ consensus site for GATA1 and GATA3 binding, whereas MatInspector predicted SMAD3 to bind to the ‘C’-allele of rs2240032. Based on these in silico results, rs2240032 was selected for in vitro TF binding studies using electrophoretic mobility shift assays (EMSA) in a Jurkat T-cell line. A distinct DNA– protein complex was observed in EMSA for the ‘C’-allele of rs2240032 (Fig. 2, lanes 7 and 8, complex I), but not for the ‘T’-allele. Self-competition with a 100-fold molar excess of unlabeled ‘C’-allele probe led to a complete abrogation of this complex (lane 6). Cross-competition with the ‘T’-allele did not eliminate this complex (lane 5), which indicates high specificity of protein binding to the ‘C’-allele. Binding of SMAD3 to this allele was predicted by in silico analysis, and thus, two different SMAD3 consensus sites were tested in competition experiments (10, 11). SMAD3_(1) caused almost complete abrogation of the complex, while SMAD3_(2) showed no effect under the given experimental conditions (lanes 4 and 3, complex I). Both SMAD3 consensus sites can also bind other proteins, especially SMAD4. Two DNA–protein complexes present in both alleles (lanes 7 and 8, complex II and III) revealed quantitative differences in binding affinity. As expected, self-competition led to complete abrogation of the complexes (lanes 6 and 9). Cross-competition with unlabeled probes carrying the respective opposite allele of rs2240032 was successful against the ‘T’-allele (lane10), but incomplete against the ‘C’-allele (lane 5). Taken together, these results suggest allele-specific binding of SMAD proteins to the ‘C’-allele of rs2240032. Next, we investigated whether rs2240032 influences DNA methylation in the 5q31 locus. In our previous methylation study of the region, we had investigated ten distinct areas (amplicons) within the locus (three in RAD50, four in IL13, two in IL4, and one in a conserved noncoding sequence CNS1) in the PASTURE cohort in cord blood and at age

1174

4.5 years (6). In this previous study, the amplicon IL13_a (CpG island 50 of the IL13 transcription start site) showed significant differences in DNA methylation which were replicated in a second, independent population within PASTURE. Thus, we now investigated whether rs2240032 is associated with the methylation status of amplicon IL13_a in cord blood or at age 4.5 (Fig. 1). Homozygous ‘CC’ carriers revealed significantly lower methylation of amplicon IL13_a in comparison with pooled ‘CT’ and ‘TT’ carriers in cord blood (discovery cohort P = 0.0350; replication cohort P = 0.0234; combined P = 0.0032) and with a less pronounced effect at age 4.5 years (discovery cohort P = 0.2198; replication cohort P = 0.0400; combined P = 0.0320) (Fig. 3). We explored whether rs2240032 may also influence the expression of genes in the 5q31 locus. Therefore, we investigated the mRNA levels of RAD50, IL4, and IL13 in whole blood in a cohort of 100 adult volunteers. Significant allelespecific differences in gene expression were observed for RAD50 (P = 0.0425) and IL4 (P = 0.0027) (Fig. 4A). IL13 mRNA levels were at the border of the detection limit in many subjects; hence, allele-specific effects on IL13 expression may not be ruled out but were not detectable in our study. Correspondingly, protein measurements in cell culture supernatants of Derp1-stimulated PBMC suggested a trend for allele-specific influences of rs2240032 on IL4 secretion (P = 0.0737) and no effect on IL13 (Fig. 4B). Finally, we analyzed whether rs2240032 is associated with disease outcomes previously mapped to this polymorphism (total serum IgE and asthma) (1, 2). As we did not have a confirmed asthma diagnosis or total serum IgE measurements available in the PASTURE study at age 4.5 years, we queried our MAGICS/ISAAC II population. Indeed, rs2240032 was associated with total serum IgE levels in this population (P = 0.0227) (Fig. 5). However, no SNP association with asthma or atopic asthma was observed (data not shown). Discussion The human 5q31 locus has attracted the interest of immunologists and geneticists involved in allergy research alike. It is a peculiar region of the genome, as major TH2-associated cytokines are concentrated in close vicinity (Fig. 1). A number of mechanisms controlling the orchestrated expression of these cytokines have been identified in the past: Epigenetic regulation of the region has been elegantly shown by Anjana Rao and her group in many landmark studies (21–23). In mice, a LCR of this TH2 locus was identified in the 30 region of the Rad50 gene. Within this LCR, four DNase I hypersensitive sites (RHS4-7) are located, and especially in RHS7, DNA methylation and histone acetylation correlated with TH2 responses (3–5). This RHS7 contributes to the regulation of Il4 and Il13 expression as well, suggesting involvement in TH2-related mechanisms such as allergy (5). Due to the high sequence similarity of RHS7 between mouse and human (4), we speculated that polymorphisms mapping to this region in humans could influence atopic diseases via regulating the expression of IL4 or IL13 and

Allergy 69 (2014) 1171–1180 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Schieck et al.

Effects of RHS7 polymorphisms on 5q31

Figure 2 Distinct protein binding to the ‘C’-allele of rs2240032. EMSA was performed using 32P-labeled probes containing either the ‘C’- or ‘T’-allele of rs2240032 and nuclear protein extracts from Jurkat T cells (JTC). Complex I. Additional binding of protein to the ‘C’-allele was observed (lanes 7 and 8), and competition with unlabeled ‘T’-allele probe demonstrated allele-specificity of this protein

binding (lane 5). Competition with SMAD3 consensus site indicated involvement of SMAD3 (lane 4). However, a second SMAD3 consensus site showed no effect (lane 3). Complex II. and III. Quantitative differences in protein binding with a higher affinity to the ‘C’-allele were observed for two DNA–protein complexes (lanes 5 and 10).

potentially RAD50. Indeed, our fine mapping in the 1000 genomes CEU dataset revealed that only one polymorphism, rs2240032, is located in RHS7. This polymorphism has previously been investigated and displayed an association with total serum IgE levels and a strong association with adult severe asthma (1, 2). We observed significantly lower IgE levels in homozygous ‘CC’ carriers of rs2240032. However, a strong LD (r2 > 0.9

in our dataset) between rs2240032 and four other SNPs in RAD50 (rs2706347; rs3798135; rs2040704; rs7737470) was already described (24). All these polymorphisms could potentially influence IgE levels by so far unidentified mechanisms (2). Additionally, moderate LD (r2 = 0.4) exists with IL13 SNP rs20541, which was also reported to influence IgE (25). Thus, the investigated SNP in RHS7 may not be the only polymorphism in the region affecting IgE levels.

Allergy 69 (2014) 1171–1180 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

1175

Schieck et al.

Effects of RHS7 polymorphisms on 5q31

A

B

Figure 3 Allele-specific effect of rs2240032 on DNA methylation of a CpG island 50 of the transcription start site of IL13 (amplicon IL13_a). Methylation analyses were carried out in two independent subsamples of the PASTURE study (discovery and replication

cohort). For both cohorts, DNA methylation measurements were available (A) in cord blood and (B) at the age of 4.5 years. Due to low numbers of ‘TT’ carriers (n = 4), ‘CT’ and ‘TT’ carriers were combined for analysis.

Regarding the previously described association of rs2240032 with adult severe asthma (1), we observed no effect on asthma or atopic asthma in our childhood asthma population. Failure to replicate this association may indicate distinctive effects of SNPs in the TH2 locus on adult and childhood onset asthma. Moreover, the 5q31 region was not identified to be a major susceptibility locus for childhood onset asthma in previous GWAS, neither in our, nor in other populations (14, 26). In silico analysis predicted functional relevance of rs2240032 due to allele-specific binding of GATA1 and GATA3 to the ‘T’-allele and SMAD3 to the ‘C’-allele. While no allele-specific binding of GATA3, a well-known master regulator of TH2 cells (27, 28) and epigenetic control mechanisms (22), could be verified in our EMSA experiments, the specific binding of a SMAD family TF to the ‘C’-allele was suggested by the results from our competition experiments in Jurkat T cells. Further studies with SMAD-specific

antibodies would be a next step to verify SMAD3 binding, which itself is an asthma candidate gene identified in one of our consortium-based GWAS (26). SMAD3 is a downstream factor of transforming growth factor b-signaling, and Smad3/ mice displayed increased proinflammatory cytokine levels in lung tissue (29). The absence of GATA3 binding to the ‘T’-allele in our experiments could be explained by a lack of TH2-specific TF expression in Jurkat T cells. The absence of TF binding due to the state of T-cell development has previously been shown for the nearby IL13 promoter (30). Methylation analyses also revealed the association of rs2240032 with a distal CpG island in the IL13 promoter (amplicon IL13_a; Fig. 3), which seems stronger at birth than at the age of 4.5 years. This is interesting as in previous investigations and ongoing studies, it appears that most changes in methylation, associated with the development of atopic diseases, are more prominent in early childhood than

1176

Allergy 69 (2014) 1171–1180 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Schieck et al.

Effects of RHS7 polymorphisms on 5q31

A

at birth, which might be due to the fact that onset of disease itself induces changes in methylation patterns. The observation that rs2240032 in the putative LCR of 5q31 leads to small but measurable changes in distinct (but not all) methylation sites in the region already at birth could indicate that it affects very early mechanisms, for example, of T-cell differentiation after birth. In line with such a hypothesis, the amplicon RAD50_b, covering two CpG sites located within RHS7 in close vicinity to rs2240032, displayed no changes in methylation (data not shown). This would imply that RHS7 is part of a cis-acting LCR affecting TH2 cytokines expression by methylation changes rather than RAD50, as suggested by experiments in mice where the presence or absence of RHS7 had no effect on basal Rad50 expression levels (5). However, this may be somewhat different in humans, as our gene expression data exhibit an association of rs2240032 with RAD50 expression. Homozygous ‘CC’ carriers displayed increased IL4 mRNA and protein levels and increased mRNA levels of RAD50 in the EXACT population, but decreased IgE levels in the MAGICS/ISAAC II population. This result is surprising and needs further investigation as IL4 is a known inducer of IgE production (31). Unfortunately, the number of subjects with available IgE measurements in the EXACT population was too little to investigate these allele-specific effects. For IL13, such an effect could not be shown, but this could be due to low IL13 levels present in the samples analyzed in our cohorts. Therefore, future studies should focus on purified T cells or T-cell subsets as these may be more suitable to detect allele-specific effects on gene expression and methylation in this locus. The small changes in DNA methylation observed in our study were in line with previous results in complex diseases where differences in DNA methylation of the NPSR1 promoter between asthmatics and controls of 1% were described (32). Interestingly, there EMSA experiments showed differential binding of transcription factors depending on the methylation status of a certain CpG, indicating a potential functional relevance. In summary and in accordance with data from mice, our results suggest the existence of a locus control region situated within the 30 region of human RAD50. Our study demonstrates that genetic variation in this LCR can influence DNA methylation and gene expression in the 5q31 TH2 locus and has an effect on IgE regulation in children, but not on

B

Figure 4 Allele-specific effects of rs2240032 on RAD50, IL4, and IL13 expression. (A) Expression of RAD50, IL4, and IL13 was measured in cDNA from whole blood derived from 100 adult subjects of the EXACT population. A higher dCt value corresponds to lower gene expression; therefore, the y-axis was inverted. IL13 levels were generally low in whole blood, and for subjects with no detectable IL13, the CtIL13 was set to 40 to enable dCt calculations. Due to low numbers of ‘TT’ carriers (n = 5), ‘CT’ and ‘TT’ were combined (n = 39) and compared against ‘CC’ carriers (n = 61). (B) Protein measurements for IL4 in supernatants of PBMC after stimulation with Derp1 were available for n = 61 subjects of the EXACT population (‘TT’ n = 4, ‘CT’ n = 24, and ‘CC’ n = 33).

Allergy 69 (2014) 1171–1180 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

1177

Schieck et al.

Effects of RHS7 polymorphisms on 5q31

Figure 5 Allele-specific effect of rs2240032 on total serum IgE levels in the MAGICS/ISAAC II population. IgE measurements and rs2240032 genotype information were available for 1,145 of 1,311

children in the MAGICS/ISAAC II population. IgE values have been log-transformed. To be uniform with Figs 3 and 4, ‘CT’ (n = 439) and ‘TT’ (n = 76) carriers were combined for analysis.

childhood onset asthma. The polymorphism may affect gene expression, methylation, and IgE production not in an uniform manner, which highlights the complex genetic architecture within the 5q31 locus in respect to allergy and needs to be dissected further. The 5q31 locus is peculiar, as it harbors a number of genes which may have very similar functions such as IL4 and IL13; genes such as IL5 displaying somewhat different properties; and genes such as RAD50 whose role in allergy may still have to be determined. A LCR exerting its influence across all genes tightly packed in that region may activate genes in the region, but this may not necessarily lead to a synergistic effect on a complex and multidimensional outcome such as allergy. Moreover, the SNP we studied here is not the only variant influencing allergy in 5q31 as further allergy-associated SNPs in IL4 and IL13 have previously been described by us and others (33). Also, we cannot yet finally determine the role that RAD50 may play in IgE regulation. Various polymorphisms located in RAD50 were found to be associated with atopic traits and other diseases in humans (1, 2, 34). If and how RAD50 may influence these diseases, for example, by affecting IgE switching, or if these SNPs are only linked to cis-acting mechanisms on nearby cytokines is not yet clarified. Based on the conclusive effects of RHS7 on TH2 differentiation in mice (4, 5), our study focused on genetic variation within the human

homolog of this region. However, additional polymorphisms outside of RHS7 might be relevant for LCR function, including SNPs in LD with rs2240032 but also independent variants. In conclusion, our findings add another piece to the puzzle of complex regulation found in the 5q31 locus and may indicate a target which could be manipulated by epigenetic strategies aimed at interfering with exaggerated TH2 development.

1178

Acknowledgments The authors are grateful to all participants in the study. Author contributions MS contributed to the EXACT study, qRT-PCR and data interpretation, performed EMSA experiments and BioPlex measurements, and drafted the manuscript. VS contributed to the EXACT study, qRT-PCR, EMSA preparation, and drafting of the manuscript and performed genotyping. SM performed methylation experiments and statistical analyses and contributed to data interpretation and drafting of the manuscript. AT contributed to the EXACT study, qRTPCR, and drafting of the manuscript. LW performed qRTPCR and BioPlex measurements. DPP contributed to the

Allergy 69 (2014) 1171–1180 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Schieck et al.

Effects of RHS7 polymorphisms on 5q31

EXACT study, the study design, and drafting of the manuscript. AK contributed to EMSA experiments and manuscript preparation. JG, MD, JCD, JR, RF, and JP contributed to PASTURE sample collection. JT supervised pyrosequencing and genotyping in PASTURE and contributed to the final version of the manuscript. MK designed and supervised the study, participated in data interpretation, and wrote the final version of the manuscript.

Research Network (NGFN grant 01GS0810) and the European Union, EU FP7 KBBE-2007-2-2-06 (EFRAIM). Vishwas Sharma was supported by the German Academic Exchange Service (DAAD grant A/09/74779). Maximilian Schieck received funding from the German Research Foundation (DFG) through the cluster of excellence REBIRTH (EXC 62/1). Sven Michel was supported by a PINA fellowship.

Funding

Conflicts of interest

This work was funded by the German Ministry of Education and Research (BMBF) as part of the National Genome

The authors declare that there are no conflicts of interest.

References 1. Li X, Howard TD, Zheng SL, Haselkorn T, Peters SP, Meyers DA et al. Genome-wide association study of asthma identifies RAD50-IL13 and HLA-DR/DQ regions. J Allergy Clin Immunol 2010;125:328–335. 2. Weidinger S, Gieger C, Rodriguez E, Baurecht H, Mempel M, Klopp N et al. Genome-wide scan on total serum IgE levels identifies FCER1A as novel susceptibility locus. PLoS Genet 2008;4:e1000166. 3. Lee GR, Fields PE, Griffin TJ, Flavell RA. Regulation of the Th2 cytokine locus by a locus control region. Immunity 2003;19:145– 153. 4. Fields PE, Lee GR, Kim ST, Bartsevich VV, Flavell RA. Th2-specific chromatin remodeling and enhancer activity in the Th2 cytokine locus control region. Immunity 2004;21:865–876. 5. Lee GR, Spilianakis CG, Flavell RA. Hypersensitive site 7 of the TH2 locus control region is essential for expressing TH2 cytokine genes and for long-range intrachromosomal interactions. Nat Immunol 2005;6:42– 48. 6. Michel S, Busato F, Genuneit J, Pekkanen J, Dalphin JC, Riedler J et al. Farm exposure and time trends in early childhood may influence DNA methylation in genes related to asthma and allergy. Allergy 2013;68:355– 364. 7. Abecasis GR, Altshuler D, Auton A, Brooks LD, Durbin RM, Gibbs RA et al. A map of human genome variation from populationscale sequencing. Nature 2010;467:1061– 1073. 8. Messeguer X, Escudero R, Farre D, Nunez O, Martinez J, Alba MM. PROMO: detection of known transcription regulatory elements using species-tailored searches. Bioinformatics 2002;18:333–334. 9. Cartharius K, Frech K, Grote K, Klocke B, Haltmeier M, Klingenhoff A et al. MatInspector and beyond: promoter analysis based on transcription factor binding sites. Bioinformatics 2005;21:2933–2942.

10. Dennler S, Itoh S, Vivien D, ten Dijke P, Huet S, Gauthier JM. Direct binding of Smad3 and Smad4 to critical TGF betainducible elements in the promoter of human plasminogen activator inhibitor-type 1 gene. EMBO J 1998;17:3091–3100. 11. Denissova NG, Pouponnot C, Long J, He D, Liu F. Transforming growth factor beta -inducible independent binding of SMAD to the Smad7 promoter. Proc Natl Acad Sci USA 2000;97:6397–6402. 12. von Mutius E, Schmid S. The PASTURE project: EU support for the improvement of knowledge about risk factors and preventive factors for atopy in Europe. Allergy 2006;61:407–413. 13. Ballenberger N, Lluis A, von Mutius E, Illi S, Schaub B. Novel statistical approaches for non-normal censored immunological data: analysis of cytokine and gene expression data. PLoS ONE 2012;7:e46423. 14. Moffatt MF, Kabesch M, Liang L, Dixon AL, Strachan D, Heath S et al. Genetic variants regulating ORMDL3 expression contribute to the risk of childhood asthma. Nature 2007;448:470–473. 15. Weiland SK, von Mutius E, Hirsch T, Duhme H, Fritzsch C, Werner B et al. Prevalence of respiratory and atopic disorders among children in the East and West of Germany five years after unification. Eur Respir J 1999;14:862–870. 16. Michel S, Liang L, Depner M, Klopp N, Ruether A, Kumar A et al. Unifying candidate gene and GWAS approaches in Asthma. PLoS ONE 2010;5:e13894. 17. Potaczek DP, Michel S, Sharma V, Zeilinger S, Vogelberg C, von Berg A et al. Different FCER1A polymorphisms influence IgE levels in asthmatics and non-asthmatics. Pediatr Allergy Immunol 2013;24:441–449. 18. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005;21:263–265.

Allergy 69 (2014) 1171–1180 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

19. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 2007;81:559–575. 20. R Development Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing, 2013. 21. Agarwal S, Rao A. Modulation of chromatin structure regulates cytokine gene expression during T cell differentiation. Immunity 1998;9:765–775. 22. Avni O, Lee D, Macian F, Szabo SJ, Glimcher LH, Rao A. T(H) cell differentiation is accompanied by dynamic changes in histone acetylation of cytokine genes. Nat Immunol 2002;3:643–651. 23. Lee DU, Agarwal S, Rao A. Th2 lineage commitment and efficient IL-4 production involves extended demethylation of the IL-4 gene. Immunity 2002;16:649–660. 24. Kretschmer A, Moller G, Lee H, Laumen H, von Toerne C, Schramm K et al. A common atopy-associated variant in the Th2 cytokine locus control region impacts transcriptional regulation and alters SMAD3 and SP1 binding. Allergy 2014;69:632–642. 25. Granada M, Wilk JB, Tuzova M, Strachan DP, Weidinger S, Albrecht E et al. A genome-wide association study of plasma total IgE concentrations in the Framingham Heart Study. J Allergy Clin Immunol 2012;129:840–845. 26. Moffatt MF, Gut IG, Demenais F, Strachan DP, Bouzigon E, Heath S et al. A largescale, consortium-based genomewide association study of asthma. N Engl J Med 2010;363:1211–1221. 27. Zheng W, Flavell RA. The transcription factor GATA-3 is necessary and sufficient for Th2 cytokine gene expression in CD4 T cells. Cell 1997;89:587–596. 28. Lee HJ, Takemoto N, Kurata H, Kamogawa Y, Miyatake S, O’Garra A et al. GATA3 induces T helper cell type 2 (Th2) cytokine

1179

Schieck et al.

Effects of RHS7 polymorphisms on 5q31

expression and chromatin remodeling in committed Th1 cells. J Exp Med 2000;192:105–115. 29. Anthoni M, Wang G, Leino MS, Lauerma AI, Alenius HT, Wolff HJ. Smad3 -signalling and Th2 cytokines in normal mouse airways and in a mouse model of asthma. Int J Biol Sci 2007;3:477–485. 30. Cameron L, Webster RB, Strempel JM, Kiesler P, Kabesch M, Ramachandran H et al. Th2 cell-selective enhancement of human IL13 transcription by IL13-1112C>T,

1180

a polymorphism associated with allergic inflammation. J Immunol 2006;177:8633– 8642. 31. Pene J, Rousset F, Briere F, Chretien I, Bonnefoy JY, Spits H et al. IgE production by normal human lymphocytes is induced by interleukin 4 and suppressed by interferons gamma and alpha and prostaglandin E2. Proc Natl Acad Sci USA 1988;85:6880–6884. 32. Reinius LE, Acevedo N, Joerink M, Pershagen G, Dahlen SE, Greco D et al. Differential DNA methylation in purified human

blood cells: implications for cell lineage and studies on disease susceptibility. PLoS ONE 2012;7:e41361. 33. Vercelli D. Discovering susceptibility genes for asthma and allergy. Nat Rev Immunol 2008;8:169–182. 34. Heikkinen K, Rapakko K, Karppinen SM, Erkko H, Knuutila S, Lundan T et al. RAD50 and NBS1 are breast cancer susceptibility genes associated with genomic instability. Carcinogenesis 2006;27: 1593–1599.

Allergy 69 (2014) 1171–1180 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd