Introduction

Atopic diseases such as asthma, atopic dermatitis and allergic rhinitis are some of the most common diseases in the world, and their prevalence is still increasing in some countries.1 Atopy is defined as a genetic predisposition toward the development of immediate hypersensitivity reactions against common environmental antigens such as house-dust mites and pollen, with the production of specific immunoglobulins E (IgEs) against these allergens; atopy is considered the strongest predisposing factor for allergic diseases.2 A study of twins from the general Australian population revealed that the association for atopy was greater in monozygous pairs (odds ratio 14.6; 95% confidence interval, 7.1–30.1) than dizygous pairs (odds ratio 2.5; 1.4–44.5).3 Other studies have shown that heritability for atopic traits is between 0.6 and 0.7, and monozygous twins have always a higher concordance of the traits under study than dizygotic twins.4

Meta-analysis of genome-wide linkage studies of asthma and related traits revealed that 3p25.3–q24 and 17q12–q24 reached genome-wide significance for skin prick test positivity, one of the atopic phenotypes in European families.5 In candidate gene approaches, interlukin-4 (IL4), IL4 receptor, IL13, the beta-chain of the high-affinity receptor for IgE (FcɛRI) and the major histocompatibility complex have been shown to be associated with atopy and its related phenotypes.6, 7 A recent meta-analysis of genome-wide association studies (GWAS) on allergic sensitization identified 10 loci that reached the genome-wide significance level, including single-nucleotide polymorphisms (SNPs) in TLR6, C11orf30, STAT6, SLC25A46, HLA-DQB1, IL1RL1, LPP, MYC, IL2 and HLA-B.8 In addition, GWAS of self-reported pollen, dust mite, and cat allergy revealed genome-wide significant association at TLR1, TLR6, TLR10, HLA-C, MICA, PTGER4, PLCL1, LPP, NFATC2, ADAD1, FOXA1 and TTC6.9

We recently identified MILR1 (synonyms, Allergin-1), mast cell immunoglobulin-like receptor, that contains an immunoreceptor tyrosine-based inhibitory motif-like domain10 and showed that coligation of MILR1 and FcɛRI suppressed IgE-mediated degranulation of bone marrow-derived cultured mast cells and that mice deficient in MILR1 developed enhanced passive systemic and cutaneous anaphylaxis. Because MILR1 is located on chromosome 17q23.3 in humans, one of the linkage region for atopy defined by skin prick test,5, 11 we hypothesized that variants in MILR1 may be associated with atopy. In the present study, we screened for mutations in the 5′-flanking region and the coding region of MILR1 and conducted an association study for atopy in the general Japanese population. We found that a promoter polymorphism, rs6504234, affects the level of MILR1 expression and that rs6504234 in MILR1 is associated with atopy.

Materials and methods

Subjects for genetic analysis

A cross-sectional analysis of 1505 Japanese men and women aged 22–81 years was carried out for subjects who visited the Tsukuba Medical Center for an annual medical check-up between June 2008 and May 2009.12 All the participants were asked about their respiratory health, medical history, lifestyle and exposure to environmental irritants (for example, cigarette smoke, allergens and air pollution). Total serum IgE levels were measured by fluorescence-enzyme immunoassay (SRL, Tokyo, Japan). Specific serum IgE antibody was measured with the multiple allergen simultaneous test (MAST)-26 chemiluminescent assay systems (Hitachi Chemical Company, Tokyo, Japan). Of the 1505 subjects, 1464 subjects completed both the questionnaires and the MAST-26 analysis. Genomic DNA was extracted from peripheral blood samples from all participants by an automated DNA extraction system (QuickGene-610L; Fujifilm, Tokyo, Japan). The characteristics of the study population are shown in Table 1. Positive sensitization (atopy) was defined as the demonstration of specific IgE levels greater than or equal to a lumicount 2.76 (class 2) for any allergens included in MAST-26, and those subjects whose lumicount was <2.76 for all 26 allergens included in MAST-26 (Supplementary Table 1) and who had no history of any allergic diseases served as the controls for the study. MAST-26 has been reported to produce results that correlate well with those produced by capsulated hydrophilic carrier polymer radioallergosorbent test (CAP RAST) and by the intradermal skin test.13

Table 1 Characteristics of the study population

Written informed consent was obtained from each participant in accordance with institutional requirements and the principles of the Declaration of Helsinki. This study was approved by the Ethical Committee of the University of Tsukuba (No 103-5).

Mutation screening and genotyping

All exons and exon–intron junctions as well as the 5′-flanking region (348 bp) and 3′-flanking region (84 bp) of MILR1 were amplified by PCR with the genomic DNA obtained from 146 unrelated Japanese subjects. Direct sequencing was carried out using the Big Dye Terminator Cycle Sequencing Ready Reaction Kit (Life Technologies, Carlsbad, CA, USA) on an ABI 3130 autosequencer (Life Technologies) using primers described in Supplementary Table 2. Primers were designed using the human genome reference sequence (NM_001085423, hg18/GRCh36). The genotype data for MILR1 in the general Japanese population were obtained from the HapMap website (http://www.hapmap.org/), and tag SNPs were selected using the Tagger software implemented in the Haploview software (version 3, release 27), with an r2 threshold of 0.8 and allele frequencies of 0.1.14, 15 Three SNPs, rs6504230, rs8071835 and rs12936887, were selected as tag SNPs.

Genotyping of subjects was done with the TaqMan SNP Genotyping Assay (Life Technologies) using primer pairs described in Supplementary Table 3. The reactions were performed in a 384-well format using a total reaction volume of 5 μl with 5 ng of genomic DNA. The plates were then placed on a thermal cycler (GeneAmp 9700 PCR systems, Life Technologies) and heated to 95 °C for 10 min, followed by 60 cycles of 92 °C for 15 s and 60 °C for 1 min. The end point fluorescence readings were performed using an ABI Prism 7900 HT System (Life Technologies). Genomic regions around c.−170_−166delAGGAA polymorphisms were amplified with primers 5′-FAM-CACCAAACTCCTTTTGGCATA-3′ and 5′-GGTTTGGATGAGGTTTGCAT-3′. PCR product (wild type; 242 bp, c.−170_−166delAGGAA; 237 bp) were electrophoresed on an ABI 3130 autosequencer (Life Technologies), and c.−170_−166delAGGAA genotypes were determined using GeneMapper v4.1 (Life Technologies). Genomic regions around the rs143526766 were amplified with primers 5′-CGTGGCCTTTTATATTCTCTCTTG-3′ and 5′-ACAGTGCTGACCAGTGGAAA-3′. PCR products were subjected to direct sequencing using the BigDye Terminator v.1.1 Cycle Sequencing Kit (Life Technologies) on an ABI PRISM 3130 Genetic Analyzer (Life Technologies).

In silico analysis

GeneVar (http://www.sanger.ac.uk/resources/software/genevar/) was used to access HapMap expression quantitative trait loci (eQTL) data, and the association between SNP genotype and gene expression was estimated using the Spearman’s rank correlation coefficient.16, 17

Construction of luciferase reporter plasmids

Real-time (RT)–PCR was performed to examine MILR1 expression using cDNA derived from the K562 human leukemia cell lines and human embryonic kidney 293T cell line. The primers used for RT–PCR are given in Supplementary Table 2. Genomic fragments of the 5′-flanking region of MILR1 were amplified with the primers shown in Supplementary Table 4 to construct plasmids encompassing −1472 to +28 bp, −759 to +28 bp, −318 to +28 bp and −117 to +28 bp of the human MILR1 promoter relative to the translation initiation site. The PCR product was digested with XhoI and HindIII and ligated into a pGL4.10 vector (Promega, Madison, WI, USA) containing the firefly luciferase gene as a reporter and then transformed into Escherichia coli DH5α (TOYOBO, Osaka, Japan). Plasmid DNA was extracted using the QIAGEN EndoFree Plasmid Maxi Kit (QIAGEN, Hilden, Germany). In order to generate vectors containing region rs6504230:T>C (c.−171T>C), a plasmid construct containing −318 to +28 bp of the human MILR1 (wild-type rs6504230:T) was subjected to site-directed mutagenesis with the PrimeSTAR mutagenesis kit (Takara, Tokyo, Japan) using primers 5′-TCTTCTTACAGGAACTCACAGAGGAAGACAGAT-3′ and 5′-GTTCCTGTAAGAAGAGAGATATTTGGTAGATTG-3′ to introduce rs6504230:C. All reactions were performed according to the manufacturer's instructions, and the accuracy of the plasmid constructs was verified by direct sequencing.

Transient transfections and luciferase assays

We seeded 1 × 105 K562 cells per well in 24-well plates and transfected them with pGL4.10 containing the MILR1 promoter. The pRL-TK plasmid (Promega) was cotransfected as a normalizing control. After 24 h of incubation, cells were collected, and luciferase activity were determined using 5 μl of lysate sample with the Dual-Luciferase Assay System (Promega) and a TD 20/20 luminometer (Turner Designs, Sunnyvale, CA, USA).

Statistical methods

Deviation from the Hardy–Weinberg equilibrium was examined using the χ2 test. Logistic regression analysis was performed under a log-additive model adjusted for sex, age, smoking status and asthma by using the SNPassoc package from R software.18 P-values were corrected by Bonferroni corrections for allelic association study, and P<0.01 (0.05/5) were considered statistically significant. Power calculation was performed with the Genetic Power Calculator.19

Results

Mutation search and case–control analysis

We screened for MILR1 mutations in DNA obtained from 146 unrelated Japanese and found five polymorphisms (rs6504230, rs8071835, rs12936887, rs143526766 and c.−170_−166delAGGAA ) and two rare missense variants (Val273Ala for 1 in 146 subjects and Leu311Val for 1 in 146 subjects; Figure 1 and Supplementary Table 5). c.−170_−166delAGGAA was in complete linkage disequilibrium with rs6504230:C (D′=1.0). None of the polymorphisms deviated from the expected values based on the Hardy–Weinberg equilibrium (P>0.05). The characteristic differences between atopic and non-atopic subjects (controls) are shown in Table 2. Atopic subjects were significantly younger than non-atopic subjects, and male subjects were more susceptible to atopy than female subjects. To control for confounders, logistic regression analysis was performed, adjusting for sex, age, smoking status and asthma. Our analysis showed that rs6504230 was significantly associated with atopy, suggesting that the C allele of rs6504230 had a protective effect against allergic sensitization (Table 3).

Figure 1
figure 1

Polymorphisms and variants identified in MILR1. Closed boxes represent exons, and arrows indicate the positions of the variants.

Table 2 Characteristics of atopic and non-atopic subjects
Table 3 Case–control study for Allergin-1 polymorphisms

Publically available eQTL data showed that the C allele of rs6504230 was associated with increased expression levels of MILR1 (P<0.0001; Supplementary Figure 1).

Transcriptional activities in the 5′-flanking region of MILR1

We first selected cell lines expressing MILR1 using RT-PCR. The PCR product sizes of MILR1 isoforms20 (Allergin-1L and Allergin-1S) were 1104 and 820 bp, respectively. MILR1 was expressed in K562 cells but not in 293T cells (Supplementary Figure 2). To examine transcriptional activity in the 5′-flanking region of the MILR1 gene, we constructed plasmids that contained sequences from −1472 bp, −759 bp, −318 bp and −117 bp to +28 bp, relative to the translation initiation site (NM_001085423). The expression of MILR1 in K562 cells was confirmed by reverse transcription-PCR (data not shown). These constructs were then transiently transfected into K562 cells. Deletion analysis revealed that the region driving maximal reporter gene expression in K562 cells was located −318 bp proximal to the translation initiation site (Figure 2). To determine whether rs6504230:T>C (c.−171T>C) affected the MILR1 expression levels, we generated luciferase reporter gene constructs that contained sequences from −318 bp to +28 bp and differed from wild type only at rs6504230:T>C. The construct containing rs6504230:C had higher transcriptional activity than that containing rs6504230:T (P=1.2 × 10−7; Figure 3).

Figure 2
figure 2

Luciferase activity of K562 cells transfected with plasmids. Each construct was transfected into K562 cells as described in Material and methods. Luciferase activity is relative to Renilla activity, and data represent the average±s.d. obtained from five samples. The numbers represent positions relative to the translation initiation site as +1.

Figure 3
figure 3

Effect of the three polymorphisms on MILR1 promoter activity. Luciferase activity is relative to Renilla activity, and data represent the average±s.d. obtained from five samples. The numbers represent positions relative to the translation initiation site as +1. *P<0.0001.

Discussion

In this study, we identified four polymorphisms and two rare missense variants in MILR1. Of them, rs6504230, which is located in the promoter region, was associated with atopy and expression levels of MILR1. The C allele of rs6504230 was associated with increased expression levels of MILR1, suggesting that increased expression of MILR1 has a protective role against the development of atopy. In the present study, the statistical powers of the case–control analysis for rs6504230, c.-170_-166delAGGAA, rs8071835, rs12936887 and rs143526766 were 0.99, 0.25, 0.78, 0.99 and 0.58, respectively, at the α level of 0.05 and the genotype relative risk of the risk allele of 1.3 for homozygote and 1.2 for heterozygote.19 Thus, our sample size was relatively small to sufficiently detect the variants with minor allele frequencies <0.05. Therefore, it is probable that the association of atopy with c.-170_-166delAGGAA, rs8071835 or rs143526766 could not be detected because of the small sample size.

MILR1 was first identified in 201110 and was reported to mediate an inhibitory signal for FcɛRI-mediated degranulation. Mice lacking MILR1 showed more severe anaphylaxis than wild-type mice in the passive cutaneous anaphylaxis model. In humans, MILR1 inhibits IgE-mediated activation of human primary mast cells in bronchoalveolar lavage fluid.20 Our data showing that increased expression of MILR1 has a protective role against the development of atopy is in agreement with the functional role for MILR1 in suppressing FcɛRI-mediated activation.

Recently conducted GWASs identified several loci for allergic sensitization,8, 9 but none of the loci was located in chromosome 17q. Allergic sensitization is believed to be influenced by exposure to allergens and the genetic background. The majority of the study populations described in the two GWAS were Caucasian, and the inconsistent results of these studies can be attributed to the differences in the genetic backgrounds and the types of allergens the subjects were exposed to. Wood et al.21 performed a genome-wide SNP analysis to identify eQTLs using whole blood from 712 individuals, and the results showed that rs6504230 was significantly associated with expression levels of MILR1 (synonyms, C17orf60; P=3.78 × 10−17). Together with the results from our experiment using the luciferase reporter assay and other publically available eQTL data (Supplementary Figure), these observations show that rs6504230 affects MILR1 expression levels in humans, leading to a susceptibility to producing specific IgEs against common allergens.