Volume 117, Issue 6 , Pages 1306-1313, June 2006
Association of IL13 with total IgE: Evidence against an inverse association of atopy and diabetes
Article Outline
- Abstract
- Methods
- Results
- Association of total serum IgE levels and IL13 variants
- Association of IgE levels with variation in atopy-related candidate genes IL4, IL4RA, IL12B, FCER1B, and TBET
- Association of IgE levels with variation in type 1 diabetes–associated genes, CTLA4 and PTPN22, and the type 1 diabetes candidate gene IL2RA/CD25
- Discussion
- Acknowledgment
- References
- Copyright
Background
Atopic illnesses, related to high circulating IgE levels, and the autoimmune disease type 1 diabetes, have been reported to be inversely associated. One possible explanation is that susceptibility alleles for one disease provide protection for the other.
Objective
Using the largest sample sizes reported so far for the identification of genetic determinants of circulating IgE levels, we investigated associations between total serum IgE (log-transformed) and single nucleotide polymorphisms in 8 genes that are candidate susceptibility loci for IgE levels/atopic illness (IL13, IL4, IL4RA, FCER1B, IL12B, TBET) and/or type 1 diabetes (CTLA4, PTPN22, IL2RA).
Methods
As many as 4570 DNA samples obtained from members of the British 1958 Birth Cohort were genotyped for 51 candidate variants, and the associations of alleles and genotypes with log-transformed serum IgE levels were evaluated by regression modeling.
Results
We obtained evidence of association between IL13 variants and total serum IgE levels (P = .00002, explaining 0.59% of phenotypic variance). However, there was no evidence of association of the confirmed type 1 diabetes susceptibility genes CTLA4 and PTPN22 and the candidate gene IL2RA with IgE levels.
Conclusion
Allelic variation in the IL-13 gene is robustly confirmed as a contributor to the variance of IgE levels but has no detectable effect in type 1 diabetes.
Clinical implications
Although the allelic variation at the confirmed IL-13 locus explains too little of the between-individual variation of circulating IgE to be of use for clinical prediction on its own, the discovery of additional susceptibility loci in the future may aid in the stratification of atopic subjects and improve risk assessment.
Key words: IL-13, gene, IgE, type 1 diabetes, atopy
Abbreviations used: DIP, Deletion/insertion polymorphism, SNP, Single nucleotide polymorphism
The familial aggregation of asthma and allergic diseases is well recognized, and twin studies strongly suggest a heritable component to both asthma and allergic disease.1 The heritability of asthma has been estimated as approximately 60%,2 and a similar figure can be derived for the heritability of total circulating IgE.1, 2 Several studies have reported an inverse relationship between atopic diseases, characterized by a TH2-associated pathway, and type 1 diabetes, which involves cell-mediated, TH1 destruction of the insulin-producing cells of the pancreas.3, 4, 5, 6 The possibility of an inverse association between the 2 immune-mediated disorders stems from the TH1/TH2 paradigm: in response to exposure to different kinds of pathogens, counterregulatory activities in the immune system control the TH1 cells (secreting IL-2 and IFN-γ) and TH2 cells (secreting IL-4, IL-5, and IL-13) while maintaining immunologic self-tolerance and suppression of autoimmune disease. Some pathways driving TH1 or TH2 differentiation can be mutually exclusive in that activators of one pathway are considered to be negative regulators of the other.7 For example, in experiments in animal models used to study allergic inflammatory diseases and in human beings with chronic inflammatory disease, TH1 T-cell–induced actions such as those via the TH1 cytokine IFN-γ ameliorated TH2 T-cell responses.8, 9
A meta-analysis of studies investigating the association between type 1 diabetes and atopic disease suggested a significant, albeit small, reduction in the frequency of asthma among children with type 1 diabetes (OR, 0.82; 95% CI, 0.68-0.99).10 However, a study of as many as 20,050 American adults found no evidence of an inverse association between atopy and autoimmune disorders.11 The discovery of genes with alleles exerting opposite effects in the 2 disorders would provide support for some of the clinical and epidemiologic observations.
Therefore, we investigated whether the TH2-related phenotype total circulating IgE and the TH1-mediated disease type 1 diabetes share genetic loci. Testing whether total circulating IgE and type 1 diabetes share genetic loci requires that susceptibility loci for both phenotypes have unequivocal statistical support for their associations. Aside from various associations in the HLA region, both diseases have been associated with several non-HLA loci. In type 1 diabetes, 2 immunoregulatory genes, CTLA4 and PTPN22, both of which influence T-cell activity and expansion, are established susceptibility loci for the disease. Another candidate gene region, containing IL2RA, which encodes an essential surface receptor (CD25) for the differentiation and activity of T-regulatory cells, has recently been associated with type 1 diabetes.12 In atopic illness and IgE production, several genes have been implicated, most notably those encoding the cytokine IL-13 (Table I), which is a positive regulator of IgE production, and also IL-4 and its receptor IL-4Rα. Our study aims to find genetic markers that are associated with total serum IgE levels. We note that such markers, although useful in aiding prediction of total serum IgE levels that are commonly observed in atopy, may not be sufficient in the prediction of incidence of clinical disease on their own.
Table I. Polymorphisms in IL13 previously found to be associated with allergic disease, IgE levels, skin test (ST) reactivity, or bronchial hyperresponsiveness
| SNP | Reference | Association study | Sample size and population |
|---|---|---|---|
| −1512A>C (rs1881457) | 18 | Associated with higher IgE levels in skin test positive children (P ≤ .005) | 1399 white children (United States, Germany) |
| −1024C>T (rs1800925) | 18 | Associated with higher IgE levels in skin-test positive children (P ≤ .005) | 1399 white children (United States, Germany) |
| 65 | T allele associated with asthma (P = .004), bronchial hyperresponsiveness (P = .003), positive ST (P = .03) | 184 probands and spouses (The Netherlands) | |
| 45 | NA | 188 patients with allergic rhinitis (China) | |
| 44 | T allele associated with higher total serum IgE (P = .0002) | 823 children and 382 trios (Germany) | |
| 20 | TT genotype with allergic asthma (P = .002), TT genotype associated with reduced inhibition of IL-13 production (P < .002) | 101 patients with allergic asthma (The Netherlands) | |
| 66 | T allele associated with atopy (P = .0087) | 238 atopic patients and 104 controls (Denmark) | |
| 67 | NA | 329 children (Denmark) | |
| 46 | NA | 207 European American children | |
| +1923C>T (rs1295686) | 46 | NA | 207 European American children |
| +2044G>A Arg110Gln (rs20541) | 18 | A allele associated with higher IgE levels in skin test positive (P = .00002) and skin test negative children (P = .05) | 1399 white children (United States, Germany) |
| 65 | NA | 184 probands and spouses (The Netherlands) | |
| 68 | NA | 83 nuclear families (Costa Rica) | |
| 44 | A allele associated with higher IgE levels (P < .0001 for AA) in 823 children and in 382 trios (P = .02) | 823 children and 382 trios (Germany) | |
| 19 | A allele associated with higher IgE levels (P = .006) and bronchial asthma (P = .04) | 267 patients with atopy and 228 with bronchial asthma (Germany) | |
| 45 | A allele associated with higher IgE levels (P = .04) | 188 patients with allergic rhinitis (China) | |
| 67 | AA genotype associated with atopic dermatitis (P = .01) and positive ST (P = .05) | 326 children (Denmark) | |
| 46 | A allele associated with increased total IgE (P = .0026) | 207 European American children | |
Polymorphisms in IL13 were considered major candidates for our association study, not only because 10 investigations have published associations between IgE levels and IL13 variants (Table I), but also because of the availability of some functional data.13, 14, 15 IL-13 plays a crucial role in TH2-cytokine–mediated immune responses, and its signaling pathway is important in the pathogenesis of allergic disease.16, 17 Previous studies showed that the Arg110Gln variant in IL13 is associated with raised total serum IgE level18 and asthma.19 In vitro studies show that the Gln110 variant represents a more active form than the Arg110 variant in promoting effector pathways of allergic inflammation.13, 14 Functional properties have also been attributed to the IL13 promoter single nucleotide polymorphism (SNP) at position –1024 C>T (also referred to as –1055, –1111, –1112), located in a region containing a nuclear factor of activated T cells transcription factor binding site; the T allele of this –1024 C>T SNP is associated with greater IL-13 production by stimulated T cells.20
Polymorphisms in IL4RA have been associated with atopic phenotypes: the Ile50Val (+148 A>T), Cys406Arg (+1216 T>C), Ser478Pro (+1682 T>C), and Gln551Arg (+1902 G>T) variants have shown association with increased risk for asthma21, 22 and atopy,21, 23 but a large study cohort of German children reported that IL4RA polymorphisms only have a minor effect on total serum IgE levels.24
We were also interested in whether the previously published association between IgE levels and the +49A>G/rs231775 SNP in CTLA4 using a Dutch population with asthma25 could be replicated in an independent data set. The same group also studied CTLA4 −1147C>T/rs16840252 SNP, which was previously associated with bronchial hyperresponsiveness and asthma.25 In contrast with the study by Howard et al,25 Munthe-Kaas et al26 could not detect associations between IgE levels and the +49A>G/rs231775 and –1147C>T/rs16840252 variants, but found P values < .05 between IgE levels and the CTLA4 variants MH30/C (rs231806), CT60/A (rs3087243), JO31/T (rs11571302), JO30/A (rs7565213), and JO27_1/C (rs11571297). Intriguingly, the alleles positively associated in the study by Munthe-Kaas et al26 are negatively associated with type 1 diabetes and other autoimmune-mediated diseases,27 providing suggestive evidence of a genetic basis for the proposed inverse relationship between atopic illness and type 1 diabetes.
We also analyzed a further 4 loci for association with total IgE levels: the β-chain of the high-affinity receptor for IgE (FcɛRI-β) gene (FCER1B, MS4A2) on chromosome 11q13 was previously associated with atopy28 and asthma29; variants in IL12B on chromosome 5q31-q33 have also been studied for association with asthma and IgE levels30, 31, 32; and a nonsynonymous SNP in the TBX21 (also called TBET) gene (His33Glu, rs2240017), the gene product of which is implicated in the initiation of TH1 lineage development and is capable of switching cytokine profiles from TH2 polarized immune responses toward a TH1-like profile.33
We also included the coding SNP in the hematopoietic-specific tyrosine phosphatase LYP (PTPN22; Arg620Trp, rs2476601), associated with several autoimmune diseases, including type 1 diabetes.34, 35 In addition, we speculate that the major regulatory mechanisms of the peripheral immune system, T regulatory lymphocytes, could well influence susceptibility of both type 1 diabetes and atopy illness. Therefore, we have also tested allelic variation in the IL2RA/CD25 gene12 for association with IgE levels.
Methods
Description of the British 1958 Birth Cohort
The British 1958 Birth Cohort (also known as the National Child Development Study) includes all births in England, Wales, and Scotland during 1 week in 1958 (http://www.cls.ioe.ac.uk/studies.asp?section=000100020003). Survivors have been followed up by parental interview and school medical examination at ages 7, 11, and 16 years, and by cohort member interview at 23, 33, and 41 years. Immigrants of the same dates of birth were identified at ages 7, 11, and 16 years and followed into adulthood, but adult immigrants (after age 16 years) have not been included. The cohort has previously been influential in advancing understanding of the natural history36, 37 and epidemiologic features36, 38, 39, 40 of allergic diseases.
This work relates to analyses of DNA from as many as 4570 white subjects recruited in the first half of the fieldwork period. These were equally distributed by sex. Forty-six percent have always been nonsmokers, 26% were exsmokers, 4% were occasional smokers, and 24% were current daily smokers. Subjects were distributed throughout England, Wales, and Scotland. Because of changing DNA resources in the laboratory over time, certain DNA samples were unavailable for some SNP assays and available for others.
A biomedical assessment of cohort members at age 44 to 45 years included measurement of total serum IgE. Total IgE was assayed by the HYTEC automated enzyme immunoassay41 (Hycor Biomedical, Edinburgh, United Kingdom) in a single laboratory (Royal Victoria Hospital, Newcastle). Specimens were processed in monthly batches over a period of 18 months of fieldwork, and the results are adjusted for batch in the statistical analysis. The lower limit of detection was 1 kU/L (14 individuals), and the upper limit was 2000 kU/L (17 individuals).
Genotyping
Invader (Third Wave Technologies, Madison, Wis) and TaqMan (Applied Biosystems, Warrington, United Kingdom) genotyping technologies were used. All genotyping data were scored by 2 independent researchers. Sequences of primers and probes used for genotyping are available on request. Note that the IL12B deletion/insertion polymorphism (DIP) originally described by Morahan et al32 is not a 4-bp deletion but a CTCTAA/CG 6 or 2-bp polymorphism as determined by direct sequencing of this complex variant (Payne F, Todd JA, Unpublished data, March 2006). All SNP genotypes were in Hardy-Weinberg equilibrium (P > .05).
Statistical analysis
Statistical analyses were performed within STATA version 8 (www.stata.com). Particular use was made of routines available from http://www-gene.cimr.cam.ac.uk/clayton/software/stata. Reported P values are not corrected for multiple testing because of a previous hypothesis and nonindependence of tests.
Total IgE was log10 transformed, thus ensuring it was approximately normally distributed. The censored values (n = 31) were set to the mean of their expected values given the observed data, assuming a log normal distribution for log10(total circulating IgE).
Regression models were used to test for an association between log-transformed total IgE levels and genotype at each locus. Analyses were restricted to white subjects and adjusted for sex, laboratory batch, geographic region (12 categories), and smoking habit (4 categories).
For most purposes, analyses included genotype at each locus as 3 groups, deriving 1 df and 2 df significance tests. The 1 df test assumes a linear relationship between the number of minor alleles and log total IgE, whereas the 2 df test allows for dominant or recessive effects. For IL12B and FCER1B, sets of tag SNPs were used to capture the common variation throughout the gene (Payne F, Todd JA, Unpublished data, March 2006), and a multilocus test was used to assess statistical significance.12, 42
Stepwise regression allowed us to test whether a single SNP could explain the association of other SNPs in the same gene.43 To investigate haplotype effects at IL13, haplotypes were generated by using snphap (www-gene.cimr.cam.ac.uk/clayton/software/) and then analyzed by regression using the posterior probabilities as weights.
Results
Association of total serum IgE levels and IL13 variants
Genotyping was performed in as many as 3615 DNA samples of subjects with total IgE level measurements. We genotyped 4 SNPs in IL13: 5′-1512A>C/rs1881457, 5′-1024C>T/rs1800925, +1923C>T/rs1295686, and +2044G>A/Arg110Gln/rs20541. Evidence for associations of log IgE with all 4 IL13 SNPs was obtained, with P values in the range of .001 to .00002 (Table II). For all 4 SNPs, the minor allele was associated with higher IgE levels than the major allele. The frequencies for the 4 main haplotypes (AG, AA, CG, CA) correspond to 75.6%, 6.6%, 6.4%, and 11.3%, respectively. Tests investigating the possibility of a haplotype that shows stronger association than either of the 2 most associated SNPs (5′-1512A>C/rs1881457 and +2044G>A/Arg110Gln/rs20541) did not provide evidence of a haplotype effect (P = .76). The 5′-1512/rs1881457 SNP alone accounted for 0.59% of the phenotypic variance, compared with 0.69% for all 4 SNPs combined, and in multiple regression analyses explained the associations of the other 3 SNPs with log IgE.
Table II. Association of IL13 candidate SNPs with total serum IgE levels in as many as 3615 British adults
| Variant | DNAs tested | MAF (%) | Genotypes | N (%) | Mean log10(IgE) | SD | P (1 df)∗ | P (2 df)∗ |
|---|---|---|---|---|---|---|---|---|
| 5′-1512A>C | 3615 | 17.8 | AA | 2451 (67.8) | 1.43 | 0.62 | .00003 | .00002 |
| rs1881457 | AC | 1047 (29.0) | 1.48 | 0.66 | ||||
| CC | 117 (3.2) | 1.67 | 0.64 | |||||
| 5′-1024C>T | 3595 | 18.0 | CC | 2428 (67.5) | 1.44 | 0.63 | .0009 | .0009 |
| rs1800925 | CT | 1049 (29.2) | 1.48 | 0.66 | ||||
| TT | 118 (3.3) | 1.65 | 0.62 | |||||
| +1923C>T | 3576 | 18.1 | CC | 2416 (67.6) | 1.43 | 0.63 | .0007 | .0011 |
| rs1295686 | CT | 1052 (29.4) | 1.47 | 0.65 | ||||
| TT | 108 (3.0) | 1.62 | 0.68 | |||||
| +2044G>A | 3493 | 18.0 | GG | 2427 (67.6) | 1.44 | 0.63 | .0003 | .0005 |
| Arg110Gln | GA | 1047 (29.1) | 1.48 | 0.65 | ||||
| rs20541 | AA | 119 (3.3) | 1.61 | 0.67 |
∗1 df and 2 df test of heterogeneity, adjusted for batch, sex, region, and smoking status. |
Association of IgE levels with variation in atopy-related candidate genes IL4, IL4RA, IL12B, FCER1B, and TBET
We genotyped 8 SNPs in the IL4RA gene on chromosome 16p11-p12 (IL4RA; 5′-3223/rs2057768, Ile50Val/rs1805010, Glu375Ala/rs1805011, Cys406Arg/rs1805012, Ser411Leu/ rs1805013, Ser478Pro/rs1805015, Gln551Arg/rs1801275, Ser761Pro/rs1805014), 1 SNP in the IL-4 gene on chromosome 5q31 (IL4; 5′-524/rs2243250), and 2 previously studied variants in IL12B (promoter/5′DIP/ss28514857, 3′untranslated region/TaqI site/rs3212227). In addition, 6 tag SNPs were genotyped in IL12B and 5 tag SNPs for FCER1B/MS4A2 (Payne F, Todd JA, Unpublished data, March 2006).
The Gln551Arg variant in IL4RA showed some evidence of association with IgE levels (Table III; P = .01), but there was no support for an association of IL12B and TBET polymorphisms with IgE levels (Table III). Multilocus tests for association of log IgE with variation in IL12B and FCER1B provided no support for association: IL12B, 6 tag SNPs, P = .74, (6 df); FCER1B, 5 tag SNPs, P = .61 (5 df).
Table III. Association of variants in atopy candidate genes with IgE levels
| Gene and variant | DNAs tested | MAF (%) | Genotypes | N (%) | Mean log10(IgE) | SD | P (1 df)∗ | P (2 df)∗ |
|---|---|---|---|---|---|---|---|---|
| IL4 | 1612 | 13.4 | TT | 1225 (76.0) | 1.44 | 0.64 | .09 | .24 |
| 5′-524T>C | TC | 354 (22.0) | 1.50 | 0.67 | ||||
| rs2243250 | CC | 33 (2.0) | 1.56 | 0.61 | ||||
| IL4RA | 1613 | 29.2 | CC | 784 (48.6) | 1.45 | 0.64 | .57 | .55 |
| 5′-3223C>T | CT | 656 (41.97) | 1.45 | 0.67 | ||||
| rs2057768 | TT | 126 (8.18) | 1.51 | 0.54 | ||||
| IL4RA | 1569 | 45.0 | AA | 460 (29.3) | 1.43 | 0.66 | .33 | .62 |
| Ile50Val | AG | 805 (51.3) | 1.46 | 0.65 | ||||
| rs1805010 | GG | 304 (19.4) | 1.49 | 0.60 | ||||
| IL4RA | 1604 | 11.9 | AA | 1248 (77.8) | 1.47 | 0.65 | .18 | .35 |
| Glu375Ala | AC | 331 (20.6) | 1.43 | 0.66 | ||||
| rs1805011 | CC | 25 (1.6) | 1.45 | 0.55 | ||||
| IL4RA | 1665 | 11.2 | TT | 1309 (78.6) | 1.47 | 0.65 | .20 | .33 |
| Cys406Arg | TC | 334 (20.1) | 1.42 | 0.66 | ||||
| rs1805012 | CC | 22 (13.2) | 1.46 | 0.59 | ||||
| IL4RA | 1642 | 5.8 | CC | 1491 (90.8) | 1.46 | 0.65 | .62 | .70 |
| Ser411Leu | CT | 146 (8.9) | 1.45 | 0.66 | ||||
| rs1805013 | TT | 5 (0.3) | 1.35 | 0.29 | ||||
| IL4RA | 1624 | 17.2 | TT | 1103 (67.9) | 1.47 | 0.65 | .24 | .50 |
| Ser478Pro | TC | 478 (29.4) | 1.44 | 0.65 | ||||
| rs1805015 | CC | 43 (2.7) | 1.42 | 0.58 | ||||
| IL4RA | 1660 | 22.6 | AA | 984 (59.3) | 1.49 | 0.65 | .01 | .04 |
| Gln551Arg | AG | 604 (36.4) | 1.42 | 0.65 | ||||
| rs1801275 | GG | 72 (4.3) | 1.40 | 0.63 | ||||
| IL4RA | 1632 | 0.8 | TT | 1606 (98.45) | 1.46 | 0.65 | .69 | .69 |
| Ser761Pro | TC | 26 (1.55) | 1.50 | 0.60 | ||||
| rs1805014 | CC | 0 (0) | — | — | ||||
| IL12B | 4570 | 49.7 | 6bp/6bp | 1124 (24.6) | 1.44 | 0.61 | .36 | .09 |
| Promoter DIP | 6bp/2bp | 2297 (50.3) | 1.49 | 0.63 | ||||
| rs17875299 | 2bp/2bp | 1149 (25.1) | 1.46 | 0.64 | ||||
| IL12B | 4568 | 19.8 | AA | 2950 (64.6) | 1.46 | 0.62 | .17 | .39 |
| 3′UTR/Taq1 (A>C) | AC | 1446 (31.7) | 1.49 | 0.63 | ||||
| rs3212227 | CC | 172 (3.8) | 0.64 | 0.64 | ||||
| TBET | 3647 | 2.6 | CC | 3463 (94.9) | 1.45 | 0.64 | .91 | .45 |
| His33Glu | CG | 182 (5.0) | 1.48 | 0.67 | ||||
| rs2240017 | GG | 2 (0.1) | 1.00 | 0.06 | ||||
∗1 df and 2 df test of heterogeneity, adjusted for batch, sex, region, and smoking status. |
Association of IgE levels with variation in type 1 diabetes–associated genes, CTLA4 and PTPN22, and the type 1 diabetes candidate gene IL2RA/CD25
There was no evidence of association of the type 1 diabetes susceptibility genes CTLA4 and PTPN22 with IgE levels (Table IV). In addition, genotyping of 20 tag SNPs in the type 1 diabetes candidate gene CD2512 did not provide evidence of association (P = .12; 20 df).
Table IV. Association of variants in type 1 diabetes-related genes, CTLA4 and PTPN22
| Gene and variant | DNAs tested | MAF (%) | Genotypes | N (%) | Mean log10(IgE) | SD | P (1 df)∗ | P (2 df)∗ |
|---|---|---|---|---|---|---|---|---|
| CTLA4 | 1607 | 14.7 | CC | 1178 (73.3) | 1.46 | 0.65 | .98 | .44 |
| −1147C>T | CT | 388 (24.1) | 1.44 | 0.65 | ||||
| rs16840252 | TT | 41 (2.6) | 1.58 | 0.70 | ||||
| CTLA4 | 1631 | 38.3 | GG | 622 (38.1) | 1.43 | 0.65 | .24 | .35 |
| +49G>A | AG | 763 (46.8) | 1.48 | 0.65 | ||||
| rs231775 | AA | 246 (15.1) | 1.47 | 0.64 | ||||
| CTLA4 | 3598 | 46.1 | GG | 1077 (29.9) | 1.46 | 0.63 | .42 | .65 |
| +6230G>A | GA | 1731 (48.1) | 1.45 | 0.63 | ||||
| rs3087243 | AA | 790 (22.0) | 1.44 | 0.64 | ||||
| PTPN22 | 3562 | 9.7 | CC | 2898 (81.4) | 1.45 | 0.64 | .79 | .83 |
| Arg620Trp (C>T) | CT | 631 (17.7) | 1.44 | 0.64 | ||||
| rs2476601 | TT | 33 (0.9) | 1.47 | 0.56 |
∗1 df and 2 df test of heterogeneity, adjusted for batch, sex, region, and smoking status. |
Discussion
Our results confirm the association of the IL-13 gene with IgE levels (Table II) and concur with previous reports of the variant allele being associated with higher IgE levels.18, 19, 44, 45, 46 For the first time, we provide evidence that the 5′-1512A>C/rs1881457 SNP might alone account for the association of the IL-13 gene with total IgE in this national sample of the middle-aged British population. We note that functional studies that investigated IL13 polymorphisms have not studied the potential effects of the 5′-1512A>C/rs1881457 SNP. The SNPs with functional data, namely 5′-1024C>T/rs1800925 and Arg110Gln/rs20541,13, 14, 15, 20 are associated with total IgE in our study but do not appear to have genetic effects in addition to that of 5′-1512A>C/rs1881457. However, in the absence of functional studies investigating all IL13 SNPs, it is impossible to exclude the presence of several functional SNPs in the region. It is also possible that the causal variant remains untyped or even undiscovered, lying elsewhere in this chromosome region, perhaps regulating IL-13 expression, but is in linkage disequilibrium with the currently typed SNPs. This possibility can be evaluated only by assembling and analyzing a complete polymorphism and linkage disequilibrium map of the entire region in a large sample set, thereby defining subregions in which a causal variant is unlikely to exist. Nevertheless, the previous functional studies of IL13 and its SNPs strongly suggest that IL13 is the causal gene for raised IgE levels and atopy risk.
We hypothesized that IL13 alleles predisposing to high IgE levels, which are part of TH2 responses, might show an inverse effect on TH1-associated type 1 diabetes. Our data do not lend support for such a mechanism, because we had previously shown that these IL13 SNPs are not associated with type 1 diabetes in as many as 3266 type 1 diabetes cases and controls and as many as 748 type 1 diabetes families, including taking into account genotypes at the IL4 and IL4RA loci.47
Similarly, there is unequivocal evidence that CTLA4 and PTPN22 are associated with type 1 diabetes (and indeed many other autoimmune diseases, because both loci have been identified as autoimmune loci rather than being specific to type 1 diabetes).27, 35, 48, 49, 50 Neither of these genes or their SNPs showed any effect on IgE levels in our sample, nor did the type 1 diabetes candidate gene IL2RA/CD25. Furthermore, the genes encoding IL-4 and its receptor, IL-4Rα, are not associated with IgE levels (Table III) or with type 1 diabetes.47 This also applied to the 3 other candidate genes, IL12B, TBET, and FCER1B (Table III). It is noted that associations between type 1 diabetes and SNPs in IL12B have been reported,51, 52, 53 but these failed to be replicated by others,54, 55, 56, 57 including our unpublished observations. TBET and FCER1B are also not associated with type 1 diabetes (Bailey R, Todd JA, Unpublished data, March 2006; Payne F, Todd JA, Unpublished data, March 2006).
Inconsistencies with previously published studies of the role of IL4RA,24 CTLA4,25, 26 and IL12B30, 31, 32 variants in the control of total serum IgE levels may be caused by several factors. Twin studies on the heritability of total IgE levels have reported that 40% of the variance in IgE is a result of environmental factors,1 implying that allergens and pharmacologic, dietary, and infectious stimuli are likely to present parameters to consider for comparisons of studies investigating different populations in which environmental factors are important. In addition, heterogeneity in allele frequencies, true differences in IgE associations between populations, ascertainment bias, and methods of analyses present plausible explanations for nonreplication of genetic association studies, because these were performed in different countries across Europe, Asia, and North America. Potentially the most important factor described in the literature, however, is the use of small sample sizes that are statistically inadequate to detect a true genetic effect.54, 58, 59, 60, 61, 62, 63 For a complex trait such as total serum IgE levels, studies using small sample sizes have minimal statistical power to detect plausible genetic effects. An effect size of a 5% increase in log(IgE), for example, will need a sample size in excess of 2000 individuals to ensure a low type 2 error probability. A convincing genetic susceptibility finding requires replication in independent, large sample sizes from the same population and ethnic group. With a size of as many as 4570 samples, our study is the largest so far attempting replication of previously reported SNPs associated with total serum IgE levels.
Finally, we remark on the size of the IL13 effect and its implications. The IL13 association with IgE and atopic illness is widely published, and is confirmed here, indicative that it is a true effect and not a false-positive result,59, 61, 64 and that SNPs within or very near the IL-13 gene are causal. Yet despite their prominence in the literature and their known direct biological effects on IgE levels, the IL13 SNPs account for only 0.69% of the phenotypic (ie, log IgE) variance. We would, therefore, not recommend too large an investment in exploratory genotyping in sample sizes smaller than were used here.
In conclusion, our results have significant implications for the design of reliable studies to locate and confirm other genes regulating IgE levels and contributing to atopy, not least that sample sizes will have to be large, consisting of several thousands of subjects. Because we have previously reported that IL13 variants are not associated with type 1 diabetes, and as shown here, other type 1 diabetes susceptibility variants do not influence IgE levels, we conclude that the proposed inverse association of atopic illness and type 1 diabetes is not explained by these genetic variants. We note that our negative findings for other cytokine genes do not rule out the possibility that application of these molecules clinically might have therapeutic benefits.
We acknowledge the Juvenile Diabetes Research Foundation and the Wellcome Trust for support. We thank Mr R. Ward and staff in the Regional Immunology Laboratory Newcastle for IgE measurements. Dr Maier was a Wellcome Trust Prize student.
References
- . Concordance and interrelationship of atopic diseases and markers of allergic sensitization among adult female twins. J Allergy Clin Immunol. 2001;108:901–907
- . Genetic analysis of allergic disease in twins. J Allergy Clin Immunol. 1984;73:265–270
- . Induction of Th1 and Th2 CD4+ T cell responses: the alternative approaches. Annu Rev Immunol. 1997;15:297–322
- Pancreatic biopsy as a procedure for detecting in situ autoimmune phenomena in type 1 diabetes: close correlation between serological markers and histological evidence of cellular autoimmunity. Diabetes. 2001;50:1269–1273
- . Pancreatic beta cells in insulin-dependent diabetes. Diabetes Metab Rev. 1992;8:209–227
- . Genetic protection from the inflammatory disease type 1 diabetes in humans and animal models. Immunity. 2001;15:387–395
- . An epigenetic view of helper T cell differentiation. Nat Immunol. 2003;4:616–623
- Allergen-specific Th1 cells counteract efferent Th2 cell-dependent bronchial hyperresponsiveness and eosinophilic inflammation partly via IFN-gamma. J Immunol. 2001;166:207–217
- . Interferon-gamma therapy reduces blood leukocyte levels in patients with atopic dermatitis: correlation with clinical improvement. Clin Immunol. 1999;92:49–55
- . A meta-analysis of the association between childhood type 1 diabetes and atopic disease. Diabetes Care. 2003;26:2568–2574
- . There is no evidence of an inverse relationship between TH2-mediated atopy and TH1-mediated autoimmune disorders: lack of support for the hygiene hypothesis. J Allergy Clin Immunol. 2003;111:131–135
- Localization of a type 1 diabetes locus in the IL2RA/CD25 region by use of tag single-nucleotide polymorphisms. Am J Hum Genet. 2005;76:773–779
- Upregulation of IL-13 concentration in vivo by the IL13 variant associated with bronchial asthma. J Allergy Clin Immunol. 2002;109:980–987
- . IL-13 R130Q, a common variant associated with allergy and asthma, enhances effector mechanisms essential for human allergic inflammation. J Clin Invest. 2005;115:747–754
- . Functional effect of the R110Q IL13 genetic variant alone and in combination with IL4RA genetic variants. J Allergy Clin Immunol. 2004;114:553–560
- . Interleukin-13 in asthma pathogenesis. Immunol Rev. 2004;202:175–190
- . The epidemic of allergy and asthma. Nature. 1999;402:B2–B4
- A cluster of seven tightly linked polymorphisms in the IL-13 gene is associated with total serum IgE levels in three populations of white children. J Allergy Clin Immunol. 2000;105:506–513
- Genetic variants of IL-13 signalling and human asthma and atopy. Hum Mol Genet. 2000;9:549–559
- An IL-13 promoter polymorphism associated with increased risk of allergic asthma. Genes Immun. 1999;1:61–65
- . The association of atopy with a gain-of-function mutation in the alpha subunit of the interleukin-4 receptor. N Engl J Med. 1997;337:1720–1725
- Variation in the interleukin 4-receptor alpha gene confers susceptibility to asthma and atopy in ethnically diverse populations. Am J Hum Genet. 2000;66:517–526
- A novel polymorphism in the 5′ promoter region of the human interleukin-4 receptor alpha-chain gene is associated with decreased soluble interleukin-4 receptor protein levels. Immunogenetics. 2001;53:264–269
- A comprehensive analysis of interleukin-4 receptor polymorphisms and their association with atopy and IgE regulation in childhood. Int Arch Allergy Immunol. 2004;135:319–324
- Fine mapping of an IgE-controlling gene on chromosome 2q: analysis of CTLA4 and CD28. J Allergy Clin Immunol. 2002;110:743–751
- CTLA-4 polymorphisms in allergy and asthma and the TH1/ TH2 paradigm. J Allergy Clin Immunol. 2004;114:280–287
- Association of the T-cell regulatory gene CTLA4 with susceptibility to autoimmune disease. Nature. 2003;423:506–511
- Fc epsilon RI-beta polymorphism and risk of atopy in a general population sample. BMJ. 1995;311:776–779
- Association between atopic asthma and a coding variant of Fc epsilon RI beta in a Japanese population. Hum Mol Genet. 1996;5:1129–1130
- The IL12B gene is associated with asthma. Am J Hum Genet. 2004;75:709–715
- Associations of the IL12B promoter polymorphism in longitudinal data from asthmatic patients 7 to 42 years of age. J Allergy Clin Immunol. 2004;113:475–481
- Association of IL12B promoter polymorphism with severity of atopic and non-atopic asthma in children. Lancet. 2002;360:455–459
- Sustained T-bet expression confers polarized human TH2 cells with TH1-like cytokine production and migratory capacities. J Allergy Clin Immunol. 2004;113:987–994
- A functional variant of lymphoid tyrosine phosphatase is associated with type I diabetes. Nat Genet. 2004;36:337–338
- Replication of an association between the lymphoid tyrosine phosphatase locus (LYP/PTPN22) with type 1 diabetes, and evidence for its role as a general autoimmunity locus. Diabetes. 2004;53:3020–3023
- . Incidence and prognosis of asthma and wheezing illness from early childhood to age 33 in a national British cohort. BMJ. 1996;312:1195–1199
- . The natural history of childhood eczema: observations from the British 1958 birth cohort study. Br J Dermatol. 1998;139:834–839
- . Hay fever, hygiene and household size. BMJ. 1989;299:1259–1260
- . The epidemiology of childhood asthma. Allergy. 1999;54(suppl 49):7–11
- . Childhood eczema: disease of the advantaged?. BMJ. 1994;308:1132–1135
- . Performance characteristics of a new automated enzyme immunoassay for the measurement of allergen-specific IgE: summary of the probability outcomes comparing results of allergen skin testing to results obtained with the HYTEC system and CAP system. Ann Allergy Asthma Immunol. 1997;79:27–34
- . Detecting disease associations due to linkage disequilibrium using haplotype tags: a class of tests and the determinants of statistical power. Hum Hered. 2003;56:18–31
- . A unified stepwise regression procedure for evaluating the relative effects of polymorphisms within a gene using case/control or family data: application to HLA in type 1 diabetes. Am J Hum Genet. 2002;70:124–141
- Associations between total serum IgE levels and the 6 potentially functional variants within the genes IL4, IL13, and IL4RA in German children: the German Multicenter Atopy Study. J Allergy Clin Immunol. 2003;112:382–388
- A common IL-13 Arg130Gln single nucleotide polymorphism among Chinese atopy patients with allergic rhinitis. Hum Genet. 2003;113:387–390
- Genetic variation in immunoregulatory pathways and atopic phenotypes in infancy. J Allergy Clin Immunol. 2004;113:511–518
- No evidence of association or interaction between the IL4RA, IL4, and IL13 genes in type 1 diabetes. Am J Hum Genet. 2005;76:517–521
- Association of the CT60 marker of the CTLA4 gene with systemic lupus erythematosus. Arthritis Rheum. 2004;50:2211–2215
- A missense single-nucleotide polymorphism in a gene encoding a protein tyrosine phosphatase (PTPN22) is associated with rheumatoid arthritis. Am J Hum Genet. 2004;75:330–337
- Association between the PTPN22 gene and rheumatoid arthritis and juvenile idiopathic arthritis in a UK population: further support that PTPN22 is an autoimmunity gene. Arthritis Rheum. 2005;52:1694–1699
- Linkage disequilibrium of a type 1 diabetes susceptibility locus with a regulatory IL12B allele. Nat Genet. 2001;27:218–221
- . IL-12p40 is associated with type 1 diabetes in Caucasian-American families. Diabetes. 2002;51:2334–2336
- Alleles of the IL12B 3′UTR associate with late onset of type 1 diabetes. Hum Immunol. 2004;65:1432–1436
- Parameters for reliable results in genetic association studies in common disease. Nat Genet. 2002;30:149–150
- . The IL12B 3′ untranslated region DNA polymorphism is not associated with early-onset type 1 diabetes. Genes Immun. 2002;3:433–435
- IL12B polymorphism and type 1 diabetes in the Italian population: a case-control study. Diabetes. 2002;51:1649–1650
- Genetic and functional evaluation of an interleukin-12 polymorphism (IDDM18) in families with type 1 diabetes. J Med Genet. 2004;41:e39
- . Epidemiological methods for studying genes and environmental factors in complex diseases. Lancet. 2001;358:1356–1360
- . A comprehensive review of genetic association studies. Genet Med. 2002;4:45–61
- . Problems of reporting genetic associations with complex outcomes. Lancet. 2003;361:865–872
- . Genetic associations in large versus small studies: an empirical assessment. Lancet. 2003;361:567–571
- . Meta-analysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nat Genet. 2003;33:177–182
- . Betting odds and genetic associations. J Natl Cancer Inst. 2004;96:421–423
- . Genome-wide association studies: theoretical and practical concerns. Nat Rev Genet. 2005;6:109–118
- Identification and association of polymorphisms in the interleukin-13 gene with asthma and atopy in a Dutch population. Am J Respir Cell Mol Biol. 2001;25:377–384
- Association between an interleukin-13 promoter polymorphism and atopy. Eur J Immunogenet. 2003;30:355–359
- Genetic variants of the IL13 and IL4 genes and atopic diseases in at-risk children. Genes Immun. 2003;4:385–389
- Lack of association between a polymorphism in the interleukin-13 gene and total serum immunoglobulin E level among nuclear families in Costa Rica. Clin Exp Allergy. 2002;32:387–390
Fieldwork and serum IgE analyses for the British 1958 Birth Cohort were supported by the United Kingdom Medical Research Council (grant G0000934). The British 1958 Birth Cohort DNA collection was funded by the Medical Research Council and the Wellcome Trust (grant 068545/Z/02). Genotyping and analyses were supported by a grant from the Juvenile Diabetes Research Foundation and the Wellcome Trust (grant 061858).Disclosure of potential conflict of interest: The authors have declared that they have no conflict of interest.
PII: S0091-6749(06)00172-2
doi:10.1016/j.jaci.2005.12.1354
© 2006 American Academy of Allergy, Asthma and Immunology. Published by Elsevier Inc. All rights reserved.
Volume 117, Issue 6 , Pages 1306-1313, June 2006
