Genes and Immunity (2003) 4, 385–389 & 2003 Nature Publishing Group All rights reserved 1466-4879/03 $25.00 www.nature.com/gene Genetic variants of the IL13 and IL4 genes and atopic diseases in at-risk children

J-Q He1, M Chan-Yeung2, AB Becker3, H Dimich-Ward2, AC Ferguson4, J Manfreda3, WTA Watson3 and AJ Sandford1 1McDonald Research Laboratories/iCAPTURE Center, St Paul’s Hospital, University of British Columbia, Canada; 2The Occupational and Environmental Lung Diseases Unit, Department of Medicine, UBC, Canada; 3The Section of and Clinical , Department of Pediatrics, University of Manitoba, Canada; 4The Division of Allergy, Department of Paediatrics, University of British Columbia, Canada

We studied a cohort containing 368 children at high risk of developing and atopic disorders and 540 parents of those children to investigate whether the IL13 Arg130Gln and CÀ1112 T polymorphisms were associated with these outcomes. We also investigated whether haplotypes consisting of any two polymorphisms of IL13 Arg130Gln, IL13 CÀ1112 T and IL4 CÀ589 T were associated with these phenotypes. In 288 white children, the IL13 130Gln allele was associated with atopy (RR ¼ 1.9, P ¼ 0.047), and with (RR ¼ 2.5, P ¼ 0.014). The associations were confirmed using a family-based test of association (P ¼ 0.027 and 0.030, respectively) in all subjects. In white subjects there were associations of haplotypes consisting of IL13 Arg130Gln and IL4 CÀ589 T with atopic dermatitis (P ¼ 0.006) and with atopy (P ¼ 0.009). Our data suggest that the IL13 Arg130Gln polymorphism and haplotypes consisting of IL13 Arg130Gln and IL4 CÀ589 T were associated with the development of atopy and atopic dermatitis at 24 months of age. Genes and Immunity (2003) 4, 385–389. doi:10.1038/sj.gene.6363985

Keywords: interleukin-13; interleukin-4; polymorphism; haplotype; atopy; atopic dermatitis

Atopy is an immune disorder characterized by Two independent groups have demonstrated that IL13 increased (IgE) production and plays a critical role in animal models of .14,15 skin-prick test responsiveness after exposure to aller- Subsequently, several association studies have reported gens. Atopy can exist without clinical manifestations, but associations of three polymorphisms of IL13 with atopy it also underlies diseases such as asthma, and atopic phenotypes.16–21 One polymorphism is lo- and atopic dermatitis. It has been well documented cated in the promoter region, at position À1112 that atopy is caused by the interaction of genetic and (CÀ1112 T) relative to the transcription start site,16 which environmental factors. Some authors believe that the was referred to as position À105517,22 or À111118 in genetic contribution is more important than environ- previous reports. The second polymorphism is located in mental factors. the coding region, that is a G-A transition at nucleotide Genomewide searches have identified strong linkage +2044 that leads to the Arg130Gln amino-acid substitu- of a region on chromosome 5q31–33 with atopy and tion,16,19 which was referred to as Arg110Gln in the atopic diseases in different populations,1–3 although the reports of Heinzman et al20 and Arima et al.21 The third results of linkage studies of total IgE with 5q31–33 have polymorphism is located at the +2525 position (G2525A), not always been consistent.4,5 This region is rich in which is in the 30 untranslated region,16 and is the same candidate genes involved in the IgE-mediated inflam- as the +4738 polymorphism in the report by Howard matory response, in particular the genes encoding et al.18 Significant linkage disequilibrium was observed interleukin-4 (IL4) and interleukin-13 (IL13).6–8 A T allele between the G2525A and Arg130Gln polymorphisms in in the CÀ589 T polymorphism in the IL4 promoter,9 both reports.16,18 referred as IL4 CÀ590 T polymorphism in different However, the results of these studies are inconsistent reports,10–12 was associated with increased luciferase with regard to the associated phenotypes. In addition, reporter gene activity.13 The T allele was also associated association of phenotypes with haplotypes of IL4 and with increased total IgE, asthma or asthma severity in IL13 has only rarely been analyzed.23 Furthermore, the other studies.9,12,13 However, these associations were not previous studies were cross-sectional family and case– replicated by other investigators.10,11 control studies and consequently the relative risks associated with the polymorphisms and their haplotypes have not been investigated in a prospective cohort. Therefore, we performed a prospective study in a Correspondence: Dr AJ Sandford, UBC McDonald Research Laboratories/ cohort of children at high risk for atopy and atopic iCAPTURE Center, St Paul’s Hospital, 1081 Burrard Street, Vancouver, diseases. In this cohort, we have previously shown that BC, Canada V6Z 1Y6. E-mail: [email protected] the IL4-589 T allele is a risk factor for the development of Polymorphisms, haplotypes and atopic diseases J-Q He et al 386 atopy, asthma and rhinitis by 12 months of age.24 In the typical areas, for example, face, postauricular area, scalp, present report, we determined whether the CÀ1112 T extensor surface of arms and legs, and flexures of neck, and Arg130Gln polymorphisms in the IL13 gene were elbows, etc according to the definition of Hanifin and associated with the development of atopy and atopic Lobitz.26 Atopy was defined by the presence of a positive diseases at 24 months of age. We also determined skin-prick test reaction (3 mm or greater mean wheal whether haplotypes consisting of any two polymorph- diameter than the negative control) to one or more of the isms among IL13 Arg130Gln, IL13 CÀ1112 T and IL4 following inhalant : cat, dog, house dust mites CÀ589 T were associated with these outcomes. (extracts of Dermatophagoides pteronyssinus and Dermato- High-risk children, defined as those with at least one phagoides farinae), cockroach, Cladosporium, Alternaria, and first-degree relative with asthma or two first-degree to the following ingestant allergens: cow’s milk, egg, soy, relatives with other IgE-mediated allergic diseases, were wheat and peanut. recruited from two centers (Vancouver and Winnipeg) In our previous study, there were genotypic data for for a study designed to assess the effectiveness of a the IL4 CÀ589 T polymorphism on 376 children.24 In this multifaceted intervention program in the primary pre- study, we genotyped 326 and 329 children for the IL13 vention of asthma and other atopic disorders.25 The Arg130Gln and CÀ1112 T polymorphisms, respectively. intervention program includes avoidance of house dust These are lower numbers than the previous study mite and pet allergens and environmental tobacco because eight children dropped out of this study after smoke, encouragement of breastfeeding, and supple- 12 months, and we did not genotype nonwhite, non- mentation with a partially hydrolyzed formula.25 Asian children for whom parental DNA was unavailable. Among 545 families enrolled initially, 64 families The rationale for the latter exclusion was that data from dropped out of the study for various reasons by the these children would not have been included in any of end of 24 months of follow-up. DNA samples were the analyses presented here. The allele frequency of IL13 available for 368 children and 540 parents of these À1112 T was not significantly different between white children. Of the 368 children, 288 were whites and 26 and Asian children (19.6 and 7.5%, respectively, were of Asian ancestry, the rest were of other ancestry. P ¼ 0.058). However, the allele frequencies of IL13 Two pediatric allergists performed assessment of the 130Gln and IL4 À589 T were significantly different children at 24 months for atopy and atopic diseases such between white and Asian groups (20.1 vs 34.2% and as asthma, rhinitis without colds and atopic dermatitis. 17.7 vs 82.7%, respectively, P ¼ 0.039 and o0.0001, The definitions of ‘probable’ asthma, rhinitis without respectively); therefore, we limited our association study colds, atopic dermatitis, and atopy have been described to white subjects, of which we had the largest number of previously.24 In detail, because of the difficulty of subjects. All of the three polymorphisms studied were in diagnosing asthma at 24 months, we used a conservative Hardy–Weinberg equilibrium in the white population. definition of ‘probable’ asthma. ‘Probable’ asthma was The prevalence of ‘probable’ asthma, rhinitis without defined as at least two distinct episodes of cough, each colds, atopic dermatitis and atopy at 24 months was lasting for 2 or more weeks; or at least two distinct compared in white children for genotype groups of the episodes of wheeze, each lasting for one or more weeks; IL13 Arg130Gln, CÀ1112 T and IL4 CÀ589 T polymorph- plus at least one of the following: nocturnal cough at isms (Table 1). least once a week in the absence of a cold, hyperpnea- Significant associations were found for the IL13 induced cough or exercise-induced cough or wheeze at Arg130Gln polymorphism in a dominant model for the any time, response to treatment with beta-agonist and/or IL13 130Gln allele with atopy and with atopic dermatitis. anti-inflammatory drugs. The definition of rhinitis with- The relative risks of IL13 Gln/Gln and Arg/Gln out colds (‘noninfectious’ rhinitis) was two or more genotypes compared with the Arg/Arg genotype for episodes of runny nose and sneezing without apparent atopic dermatitis and atopy were 2.5 (P ¼ 0.014) and 1.9 colds. Atopic dermatitis was defined as characteristic (P ¼ 0.047), respectively. Binary logistic regression was skin lesion (scaly, erythematous and itchy) in at least two used to calculate P-values adjusted for the presence/

Table 1 Prevalence of atopy and atopic diseases at 24 months of age in different genotype groups

Phenotype IL13Arg130Gln IL13CÀ1112T IL4CÀ589T

Arg/Arg Arg/Gln Gln/Gln CC CT TT CC CT TT

Probable asthma 7 (12/182) 9 (8/85) 14 (2/14) 7 (13/181) 9 (8/93) 11 (1/9) 6 (12/198) 13 (10/78) 8 (1/12) Rhinitis without colds 14 (25/182) 7 (6/85) 21 (3/14) 14 (25/181) 9 (8/93) 0 (0/9) 10 (20/198) 9 (7/78) 33a (4/12) Atopic dermatitis 6 (11/179) 16 (14/85) 7b (1/14) 9 (16/180) 11 (10/91) 0 (0/9) 8 (15/197) 13 (10/76) 17 (2/12) Positive skin test 9 (17/182) 17 (14/83) 21c (3/14) 11 (19/180) 16 (15/92) 0 (0/9) 12 (23/197) 13 (10/77) 17 (2/12)

bRelative risk Gln/Gln and Arg/Gln vs Arg/Arg=2.5 (95% CI=1.2–5.2, P=0.014). cRelative risk Gln/Gln and Arg/Gln vs Arg/Arg=1.9 (95% CI=1.0–3.5, P=0.047). aRelative risk TT vs CT and CC=3.4 (95% CI=1.4–8.2, P=0.010). Data are shown as % (n). The IL13 Arg130Gln, CÀ1112T and IL4 CÀ1112T polymorphisms were defined by PCR-RFLP as described.16,24 The frequencies of atopy and atopic diseases within each genotypic group were compared by w2 analyses or by Fisher’s exact test when applicable. w2 tests assuming both dominant and recessive models for the rare allele were performed. Relative risks (RR) and 95% confidence intervals (CI) were calculated as previously described.27

Genes and Immunity Polymorphisms, haplotypes and atopic diseases J-Q He et al 387 absence of the intervention (a potential confounding affected child approximately twice as often as the 130Arg factor) using the JMP Statistics software package (SAS allele. However, because of the small number of families Institute Inc.). The associations remained significant after in these analyses these distortions were not significant correction for intervention group with P-values of 0.017 (P ¼ 0.074 and 0.103, respectively). We also calculated the and 0.049, respectively, although the association with P-value for the TDT using both affected and unaffected atopy became borderline. However, no significant offspring and their parents, which can increase the difference was found for phenotypic prevalence in power of the test.32 In this test, the IL13 130Gln allele was genotypic groups of the IL13 C–1112 T polymorphism transmitted preferentially to children who had atopic in either dominant or recessive models for IL13 À1112 T. dermatitis (14 transmissions vs six nontransmissions) For the IL4 CÀ589 T polymorphism, we found that the and was transmitted less often to children without atopic relative risk of À589TT compared with other genotypes dermatitis at the same time point (63 transmissions vs 85 for rhinitis was 3.4 (P ¼ 0.010), which was similar to nontransmissions) (P ¼ 0.027). Similarly, the IL13 130Gln that we reported previously at 12 months of age.24 The allele was transmitted preferentially to children who had association remained significant after correction for a positive skin test (16 transmissions vs eight nontrans- intervention group with a P-value of 0.021. Unlike missions) and was transmitted less often to children previous results at 12 months of age, we did not find without atopic dermatitis at the same time point (62 any association of the IL4 À589TT genotype with transmissions vs 83 nontransmissions) (P ¼ 0.030). ‘probable asthma’ at 24 months of age. The results of theoretical studies indicate that includ- In previous studies, IL13 130Gln has been associated ing genotyped, unaffected offspring in association tests with total serum IgE level, atopy, asthma and atopic can increase power, especially with more common dermatitis.16,18,19,28 Our results were consistent with the diseases.32,33 In our study, the associations of IL13 previous studies that showed that IL13 130Gln was 130Gln with atopic dermatitis and atopy were confirmed associated with atopic dermatitis19 and with atopy.28 A by TDT using both unaffected and affected offspring, recent study showed that IL13 130Gln might act as a which were not significant by TDT using only affected functional genetic factor in the context of asthma with a offspring. unique mechanism to upregulate local and systemic IL13 There are several reasons for discrepancies between concentration in vivo.21 However, we did not find an association studies of polymorphisms and atopy and association of IL13 130Gln with asthma, as was reported atopic diseases. Firstly, the study population is an in case–control populations from both Britain and important factor affecting the result of a genetic study. Japan,20 but not in a Chinese population.28 Some genetic associations may be ethnicity dependent.34 The IL13 À1112TT genotype was associated with As shown in this and other studies, the frequency of the serum IgE in three populations,16 allergic asthma in a IL4 À589T allele in Asians (83%) is much higher than in Dutch population and altered regulation of IL13 produc- white populations (18%). This may explain why the tion.17 Howard et al.18 found that the IL13 À1112T allele association of IL4 À589T with atopic dermatitis found in was associated with asthma, bronchial hyperresponsive- Japanese individuals was not found in white indivi- ness and atopy in another Dutch population. However, duals.29,35 Even in the same ethnic group, there are in this study we found no association of the IL13 various degrees of linkage disequilibrium. Significant CÀ1112T polymorphism with atopy and atopic diseases. linkage disequilibrium was observed between two IL13 We have extended our previous IL4 CÀ589T report24 polymorphisms in this study, but the D’ of 0.46 was less by using phenotypic data from 24 months of age. The than that (0.65) in the report of Graves et al.16 In this association of IL4 À589T with rhinitis remained similar at report, linkage disequilibrium was also observed for IL4 24 months of age. However, the association of IL4 À589T CÀ589T polymorphism with either of the IL13 poly- with ‘probable’ asthma disappeared at 24 months of age. morphisms. Although the D’ values were only 0.37 for We did not find an association of IL4 À589T with atopic IL4 CÀ589T and IL13 Arg130Gln, and 0.23 for IL4 dermatitis as previously reported.29 CÀ589T and IL13 CÀ1112T, they were statistically Prospective studies such as this are generally less significant with Po0.0001. However, no significant prone to type 1 error due to population stratification linkage disequilibrium between these two IL13 poly- because of the uniform method of ascertainment that is morphisms was found in the report of Howard et al.18 employed. However, there still remains the possibility Secondly, the age of the study subjects is also very that unidentified population structure is responsible important. It is difficult to diagnosis asthma in children for the associations with IL13 130Gln and IL4 À589T. at 24 months, so we used a definition of ‘probable Therefore, we used a family-based test of association asthma’. About two-thirds of wheezing in infants is implemented in the FBAT software.30 transient and is not associated with increased risk By genotyping 540 parents of the children in this study of asthma later in life.36 In addition, the prevalence of for the IL13 Arg130Gln, IL13 CÀ1112T and IL4 CÀ589T allergic sensitization changes with age. The peak age of polymorphisms, we were able to study 269 trios (ie allergic sensitization is between 15 and 25 years,37,38 so genotypic data on a child and both of his/her parents). skin tests in young children may not reflect their atopic We performed family-based tests of association for those status later in life. In a previous study, the prevalence of genotypes with each of the phenotypes in families from atopic dermatitis was shown to increase markedly all ethnic backgrounds. The tests using affected offspring during the first 12 months, while ‘asthma-like’ disease only are equivalent to the ‘traditional’ TDT, which was started to increase mainly after 12 months of age.39 In this proposed as a test for association in the presence of study, we found that 13 children were diagnosed as linkage.31 We did not find a significant transmission having ‘probable asthma’ at 12 months but not at 24 distortion using this test. In the case of atopic dermatitis months of age (although some of these children may or atopy, the IL13 130Gln allele was transmitted to the have recurrence of symptoms at a later date). In contrast,

Genes and Immunity Polymorphisms, haplotypes and atopic diseases J-Q He et al 388 Table 2 Frequencies of haplotypes of IL13 Arg130Gln (G2044A) and IL4 CÀ589T in different phenotypic and control groups in high-risk children calculated using the Arlequin software package40

Phenotype IL13 2044G/IL4 À589C IL13 2044G/IL4 À589T IL13 2044A/IL4 À589C IL13 2044A/IL4 À589T P-valuea

Probable asthma (n=44) 25 (56.8) 7 (15.9) 8 (18.2) 4 (9.1) 0.090 Control (n =510) 370 (72.6) 40 (7.8) 57 (11.2) 43 (8.4) Rhinitis (n =62) 41 (66.1) 9 (14.5) 6 (9.7) 6 (9.7) 0.276 Control (n =492) 355 (72.2) 37 (7.5) 58 (11.8) 42 (8.5) Atopic dermatitis (n =52) 25 (48.1) 10 (19.2) 11 (21.2) 6 (11.5) 0.006 Control (n =496) 363 (73.2) 37 (7.4) 55 (11.1) 41 (8.3) Skin test positive (n =68) 39 (57.4) 9 (13.2) 15 (22.1) 5 (7.3) 0.009 Control (n =482) 354 (73.4) 38(7.9) 49(10.2) 41(8.5)

aHaplotype frequencies between groups were compared by CLUMP software.41 Data are shown as n (%).

21 children developed ‘probable asthma’ after 12 months In this study, we have used four outcome variables and of age. Only six children were diagnosed as ‘probable genotypic data from three polymorphisms and therefore asthma’ at both 12 and 24 months of age. This age- there may be concern regarding the number of compar- dependent penetrance of ‘probable asthma’ may explain isons that were made. However, we chose not to adjust discrepant results from studies that utilized subjects the P-values for multiple comparisons because all of different ages. It may also explain why we found phenotypes and all genotypes were not independent, different results by using 24-month phenotypic data. and IL13 and IL4 are known atopy-susceptibility genes. This emphasizes the value of a prospective study design Since the prevalence of atopic phenotypes changes with and long-term follow-up. age in children, it is important to know which stage is Thirdly, the power of the study design to detect most predictive of adult atopy and atopic diseases. significant association varies in different reports. In our Several studies have shown that the presence of atopic study, the frequencies of genotypes containing at least dermatitis in infancy is an important risk factor in one risk allele were greater than 30% for each of the three subsequent development of asthma.42,43 A recent study polymorphisms. This translates to a power of 80% to has shown that skin sensitivity to egg or cow’s milk in detect an RR of 2.2 or greater (a ¼ 0.05, for a two-sided the first year, but not wheezing before the age of 2 years, test). In the study of Howard et al,18 the odds ratio for was predictive of adult asthma.44 Therefore, it is IL13 À1112T genotypes in subjects with asthma was only important to follow up this cohort to assess the outcomes 1.9. Therefore, our negative results for the IL13 CÀ1112T of atopy and atopic diseases. polymorphism may be due to lack of power. In summary, our data suggest that IL13 130Gln and IL4 We found significant linkage disequilibrium between À589T alleles are risk factors for the development of IL4 À589T and IL13 130Gln. This raises the possibility atopic dermatitis and atopy at 24 months of age. The that neither of these alleles is responsible for these haplotypes composed of these two polymorphisms were associations but instead they are in linkage disequili- associated with atopic dermatitis and atopy. brium with a third, causal allele. Therefore, we performed haplotype analysis among the different phenotypic groups. Haplotype analysis of haplotypes consisting of any two polymorphisms among IL13 Acknowledgements Arg130Gln, IL13 CÀ1112T and IL4 CÀ589T showed that This study was supported by the Canadian Institutes of haplotypes composed of two polymorphisms of the IL13 Health Research and by the Respiratory Health Network gene were not associated with any phenotypes at 24 of Centres of Excellence, Canada. Dr Sandford was months of age (data not shown). However, haplotypes supported by a Parker B Francis Fellowship and is a composed of IL13 Arg130Gln and IL4 CÀ589T poly- recipient of a Canada Research Chair in genetics. morphisms were associated with atopic dermatitis and We thank Dr Peter Pare´ for critical appraisal of the atopy with P-values of 0.006 and 0.009, respectively manuscript. (Table 2). Haplotypes consisting of the IL13 CÀ1112T and IL4 CÀ589T polymorphisms were associated with atopic dermatitis with a P-value of 0.011 (data not shown). All four phenotypic groups had a decreased percentage of References the IL13 130Arg/IL4 À589C haplotype, which is con- sistent with the data that these two alleles were lower 1 The Collaborative Study on the Genetics of Asthma (CSGA). A risk alleles in previous studies of atopy and atopic genome-wide search for asthma susceptibility loci in ethni- diseases. Either the IL13 130Gln/IL4 À589C haplotype or cally diverse populations. Nat Genet 1997; 15: 389–392. the IL13 130Arg/IL4 À589T haplotype was significantly 2 Xu J, Meyers DA, Ober C. Genomewide screen and identifica- increased in both the atopic dermatitis and atopy groups. tion of gene-gene interactions for asthma-susceptibility loci in three US populations: collaborative study on the genetics of The associations with single polymorphisms could not asthma. Am J Hum Genet 2001; 68: 1437–1446. explain these data. One explanation might be that there 3 Postma DS, Bleecker ER, Amelung PJ. Genetic susceptibility to is a protective allele on the IL13 130Arg/IL4 À589C asthmaFbronchial hyperresponsiveness coinherited with a haplotype that is not found in the other three haplotypes. major gene for atopy. N Engl J Med 1995; 333: 894–900.

Genes and Immunity Polymorphisms, haplotypes and atopic diseases J-Q He et al 389 4 Mathias RA, Freidhoff LR, Blumenthal MN. Genome-wide disorders in at-risk infants. Am J Respir Crit Care Med 2000; 161: linkage analyses of total serum IgE using variance compo- 1655–1659. nents analysis in asthmatic families. Genet Epidemiol 2001; 20: 25 Chan-Yeung M, Manfreda J, Dimich-Ward H, Ferguson A, 340–355. Watson W, Becker A. A randomized controlled study on the 5 Xu J, Postma DS, Howard TD. Major genes regulating total effectiveness of a multifaceted intervention program in the serum immunoglobulin E levels in families with asthma. Am primary prevention of asthma in high-risk infants. Arch Pediatr J Hum Genet 2000; 67: 1163–1173. Adolesc Med 2000; 154: 657–663. 6 Cookson WO, Moffatt MF. Genetics of asthma and allergic 26 Hanifin JM, Lobitz Jr WC. Newer concepts of atopic disease. Hum Mol Genet 2000; 9: 2359–2364. dermatitis. Arch Dermatol 1977; 113: 663–670. 7 Niimi T, Munakata M, Keck-Waggoner CL. A polymorphism 27 Morris JA, Gardner MJ. Calculating confidence intervals for in the human UGRP1 gene promoter that regulates transcrip- relative risks (odds ratios) and standardised ratios and rates. tion is associated with an increased risk of asthma. Am J Hum BMJ (Clin Res Ed) 1988; 296: 1313–1316. Genet 2002; 70: 718–725. 28 Leung TF, Tang NL, Chan IH, Li AM, Ha G, Lam CW. A 8 Marsh DG, Neely JD, Breazeale DR. Linkage analysis of polymorphism in the coding region of interleukin-13 gene is IL4 and other chromosome 5q31.1 markers and total associated with atopy but not asthma in Chinese children. Clin serum immunoglobulin E concentrations. Science 1994; 264: Exp Allergy 2001; 31: 1515–1521. 1152–1156. 29 Kawashima T, Noguchi E, Arinami T. Linkage and association 9 Sandford AJ, Chagani T, Zhu S. Polymorphisms in the IL4, of an interleukin 4 gene polymorphism with atopic dermatitis IL4RA, and FCERIB genes and asthma severity. J Allergy Clin in Japanese families. J Med Genet 1998; 35: 502–504. Immunol 2000; 106: 135–140. 30 Horvath S, Xu X, Laird NM. The family based association test 10 Walley AJ, Cookson WO. Investigation of an interleukin-4 method: strategies for studying general genotype–phenotype promoter polymorphism for associations with asthma and associations. Eur J Hum Genet 2001; 9: 301–306. atopy. J Med Genet 1996; 33: 689–692. 31 Spielman RS, McGinnis RE, Ewens WJ. Transmission test for 11 Hijazi Z, Haider MZ. Interleukin-4 gene promoter polymorph- linkage disequilibrium: the insulin gene region and insulin- ism [C590T] and asthma in Kuwaiti Arabs. Int Arch Allergy dependent mellitus (IDDM). Am J Hum Genet 1993; Immunol 2000; 122: 190–194. 52: 506–516. 12 Noguchi E, Shibasaki M, Arinami T. Association of asthma 32 Lunetta KL, Faraone SV, Biederman J, Laird NM. Family- and the interleukin-4 promoter gene in Japanese. Clin Exp based tests of association and linkage that use unaffected sibs, Allergy 1998; 28: 449–453. covariates, and interactions. Am J Hum Genet 2000; 66: 605–614. 13 Rosenwasser LJ, Klemm DJ, Dresback JK. Promoter poly- 33 Scott L, Rogus JJ. Using unaffected child trios to test for morphisms in the chromosome 5 gene cluster in asthma and transmission distortion. Genet Epidemiol 2000; 19: 381–394. atopy. Clin Exp Allergy 1995; 25: 74–78; discussion 95–96. 34 Osei-Hyiaman D, Hou L, Zhiyin R. Association of a novel 14 Wills-Karp M, Luyimbazi J, Xu X. Interleukin-13: central point mutation (C159G) of the CTLA4 gene with type 1 mediator of allergic asthma. Science 1998; 282: 2258–2261. diabetes in West Africans but not in Chinese. Diabetes 2001; 50: 15 Grunig G, Warnock M, Wakil AE. Requirement for IL-13 2169–2171. independently of IL-4 in experimental asthma. Science 1998; 35 Elliott K, Fitzpatrick E, Hill D. The À590C/T and À34C/T 282: 2261–2263. interleukin-4 promoter polymorphisms are not associated 16 Graves PE, Kabesch M, Halonen M. A cluster of seven tightly with atopic eczema in childhood. J Allergy Clin Immunol 2001; linked polymorphisms in the IL-13 gene is associated with 108: 285–287. total serum IgE levels in three populations of white children. 36 Martinez FD, Wright AL, Taussig LM, Holberg CJ, Halonen M, J Allergy Clin Immunol 2000; 105: 506–513. Morgan WJ. Asthma and wheezing in the first six years of life. 17 van der Pouw Kraan TC, van Veen A, Boeije LC. An IL-13 The Group Health Medical Associates. N Engl J Med 1995; 332: promoter polymorphism associated with increased risk of 133–138. allergic asthma. Genes Immun 1999; 1: 61–65. 37 Burrows B, Lebowitz MD, Barbee RA. Respiratory disorders 18 Howard TD, Whittaker PA, Zaiman AL. Identification and and allergy skin-test reactions. Ann Intern Med 1976; 84: association of polymorphisms in the interleukin-13 gene with 134–139. asthma and atopy in a Dutch population. Am J Respir Cell Mol 38 Niemeijer NR, de Monchy JG. Age-dependency of sensitiza- Biol 2001; 25: 377–384. tion to aero-allergens in asthmatics. Allergy 1992; 47: 431–435. 19 Liu X, Nickel R, Beyer K. An IL13 coding region variant is 39 Laan MP, Baert MR, Bijl AM. Markers for early sensitization associated with a high total serum IgE level and atopic and inflammation in relation to clinical manifestations of dermatitis in the German multicenter atopy study (MAS-90). atopic disease up to 2 years of age in 133 high-risk children. J Allergy Clin Immunol 2000; 106: 167–170. Clin Exp Allergy 2000; 30: 944–953. 20 Heinzmann A, Mao XQ, Akaiwa M. Genetic variants of IL-13 40 Schneider S, Roessli D, Excoffier L. Arlequin ver. 2.000: A signalling and human asthma and atopy. Hum Mol Genet 2000; Software for Population Genetics Data Analysis. Genetics and 9: 549–559. Biometry Laboratory, University of Geneva: Switzerland, 2000. 21 Arima K, Umeshita-Suyama R, Sakata Y. Upregulation of IL- 41 Sham PC, Curtis D. Monte Carlo tests for associations between 13 concentration in vivo by the IL13 variant associated with disease and alleles at highly polymorphic loci. Ann Hum Genet bronchial asthma. J Allergy Clin Immunol 2002; 109: 980–987. 1995; 59: 97–105. 22 Van Der Pouw Kraan TC, Kucukaycan M, Bakker AM. 42 Reijonen TM, Kotaniemi-Syrjanen A, Korhonen K, Korppi M. Chronic obstructive pulmonary disease is associated with Predictors of asthma three years after hospital admission for the À1055 IL-13 promoter polymorphism. Genes Immun 2002; wheezing in infancy. Pediatrics 2000; 106: 1406–1412. 3: 436–439. 43 Bergmann RL, Edenharter G, Bergmann KE. Atopic dermatitis 23 Noguchi E, Nukaga-Nishio Y, Jian Z. Haplotypes of the 5’ in early infancy predicts allergic airway disease at 5 years. Clin region of the IL-4 gene and SNPs in the intergene sequence Exp Allergy 1998; 28: 965–970. between the IL-4 and IL-13 genes are associated with atopic 44 Rhodes HL, Sporik R, Thomas P, Holgate ST, Cogswell asthma. Hum Immunol 2001; 62: 1251–1257. JJ. Early life risk factors for adult asthma: a birth cohort 24 Zhu S, Chan-Yeung M, Becker AB. Polymorphisms of the IL-4, study of subjects at risk. J Allergy Clin Immunol 2001; 108: TNF-alpha, and Fcepsilon RIbeta genes and the risk of allergic 720–725.

Genes and Immunity