Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Asthma is a common respiratory disease characterized by intermittent airways obstruction and respiratory symptoms that are caused by acute and chronic bronchial inflammation. Bronchial hyperresponsiveness (BHR) and total serum immunoglobulin (Ig)E levels are closely associated with the asthma phenotype and have a strong genetic component (1). It has been well documented that the presence of atopy and BHR may precede the development of clinical asthma (1, 2). The development of asthma appears to be determined by the interaction between host susceptibility (genetics) and a variety of environmental exposures.

Numerous genetic studies have mapped an asthma and/ or atopy susceptibility gene(s) to a region on chromosome 5q31-q33 in several populations (Dutch [5, 7], Amish [8], American Caucasian [9], Hutterite [10], and British [11, 12]). This region contains a cluster of proinflammatory cytokines important in immune regulation. Two members of this cluster, interleukin (IL)-4 and IL-13, have been both genetically and functionally implicated in the pathogenesis of asthma and atopy (13). These cytokines are produced by T helper (Th) 2 cells and are capable of inducing isotype class-switching of B cells to produce IgE (18). They also share a receptor component, IL4R, which has been shown to be an important factor in the development or expression of atopy and asthma (19). Further, both IL-4 and IL-13 messenger RNA (mRNA) and protein have been localized to the airways in allergic asthma (23, 24).

The IL-13 receptor consists of one IL4R subunit and either a low-affinity IL13R1 (25) or a high-affinity IL13R2 subunit (26). The complete receptor for IL-4 is composed of one IL4R subunit and an IL4R subunit. Therefore, it is possible that different polymorphisms in these receptors, as well as in the IL-4 and IL-13 cytokines, contribute to the complex regulation of atopy or asthma phenotypes. Association studies with polymorphisms in IL-13 have been performed using various atopy and asthma phenotypes in several populations. A promoter polymorphism was identified at position 1111 (referred to as position 1055 in the report by Van der Pouw Kraan and colleagues [14]) adjacent to the nuclear factor of activated T cells site and reported to be associated with allergic asthma in a Dutch population (14). In addition, an Arg130Gln polymorphism in exon 4 has been shown to be associated with high total serum IgE levels (15, 16), atopic dermatitis (15), and asthma (17) in German (15, 16), American (16), British (17), and Japanese (17) populations. In an effort to further understand the contribution of IL-13 to asthma and atopy phenotypes, we have sequenced the IL-13 gene in probands with asthma and their unaffected spouses to identify new sequence variants. We also performed case-control association studies with three of these polymorphisms in this Dutch population, in which we have previously obtained evidence for linkage on chromosome 5q31 to bronchial responsiveness and total serum IgE levels (5, 6).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Population

This genetically homogeneous population has been described in detail previously (5, 7, 27). Probands (local Caucasian Dutch patients with asthma) were originally studied between 1962 and 1975 at Beatrixoord Hospital, Haren, the Netherlands. At that time patients were diagnosed with asthma by the presence of characteristic symptoms, airways hyperresponsiveness (AHR), and reversibility of airway obstruction. Between 1990 and 1999, probands with asthma were restudied, together with their spouses (all Caucasians), children, and available grandchildren. Briefly, all individuals underwent baseline spirometry and reversibility to 800 µg albuterol; bronchial responsiveness testing to histamine was performed using a 30-s inhalation protocol (27, 28). A subject was considered to display BHR if the provocative concentration of histamine producing a 20% fall in forced expiratory volume in 1 s (FEV₁) (PC₂₀) was  32 mg/ml histamine. For atopy, subjects had intracutaneous skin testing with 16 common aeroallergens, which was considered positive if the maximum wheal diameter was  5 mm. In the first 92 families, total serum IgE levels were measured by solid-phase immunoassay (Pharmacia Diagnostics, Uppsala, Sweden). Duplicate measurements were made and the mean for each subject was used. If the duplicate samples differed by more than 5% the test was repeated. In the second group, consisting of 108 families, total IgE levels were measured by enzyme-linked fluorescence assay (Mini Vidas; Biomerieux, Inc., Marcy, France). Although the entire families were ascertained for genetic linkage studies, the probands and spouses represent an appropriate cohort for case-control association studies, especially because they are of comparable ages and have experienced similar environmental exposures, which accounts for some of the known age-related differences in the frequency of BHR, serum IgE levels, and skin-test responsiveness. A total of 184 probands and their spouses were used for this IL-13 association study. This study was approved by the Medical Ethics Committee at the University of Groningen. All subjects provided written informed consent. DNA was isolated from lymphocytes using standard procedures.

DNA Resequencing of IL-13

Resequencing of the IL-13 gene in 20 affected (probands) and 20 unaffected (spouses) individuals was performed by cycle sequencing of overlapping polymerase chain reaction (PCR)-amplified DNA fragments covering the 5'-flanking region, exons 1 to 4 and intron 1. We decided on this sequencing strategy to identify single nucleotide (nt) polymorphisms (SNPs) that are most likely to alter the regulation or functional activity of IL-13. Primers and fragment sizes are shown in Table 1 (PCR primer pairs were designed using Primer Express, version 1.0 [Perkin-Elmer Applied Biosystems, Warrington, U.K.]). PCR amplifications were carried out in 10-µl volumes containing 1× GeneAmp PCR buffer (Perkin-Elmer), 20 ng of genomic DNA, 30 ng of each forward and reverse primer, 400 µM of each deoxynucleotide triphosphate (dNTP) (Amersham-Pharmacia Biotech, Piscataway, New Jersey), 1.5 to 3 mM MgCl₂ (primer-dependent), and 0.5 units AmpliTaq Gold (Perkin-Elmer). Before sequencing, amplification products were incubated with shrimp alkaline phosphatase (0.5 units; Amersham-Pharmacia Biotech) and exonuclease I (5 units; Amersham-Pharmacia Biotech) at 37°C for 30 min, followed by heat inactivation at 80°C for 15 min. Amplification products were double-strand sequenced using the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction Kit (Perkin-Elmer). After cycle sequencing, unincorporated BigDye terminators were removed using Sephadex G-50 (Sigma, St. Louis, Missouri) in Millipore multiscreen HV plates. Sequence reactions were analyzed using an ABI PRISM 377 sequencer (Perkin-Elmer) and the resultant chromatograms were aligned and viewed using Phred/Phrap (29) and Consed (30) software. The GenBank accession numbers for the reference genomic sequence used for the IL-13 gene is reported in the footnote to Table 1 and legend to Figure 2.

View larger version (5K): [in this window] [in a new window]   View larger version (22K): [in this window] [in a new window]   Figure 2.   SNPs detected by resequencing of the IL-13 gene in this Dutch population. A total of 18 overlapping PCR-amplified DNA fragments covering 2.214 kb of genomic DNA sequence 5' to +1 of the open reading frame, exons 1 to 4 (including intron- exon boundaries), and intron 1 of the IL-13 gene were cycle sequenced in 20 affected (probands) and 20 unaffected (spouses) individuals as described in MATERIALS AND METHODS. The allotype for each SNP was determined by visual inspection of the sequence traces. (A) Graphic representation of SNP locations within the IL-13 gene. (B) Location of IL-13 SNPs with allotype numbers and affection status of sequenced individuals. *GenBank accession No. U31120 was used as the reference sequence. The coordinates for exons 1 thru 4 were nucleotides 2158-2345, 3403-3456, 3709-3813, and 4160-5095, respectively. ¶39 patients genotyped. The position of SNPs 2 and 3 relative to +1 differs between this report and that of Graves and colleagues (16) because GenBank accession number AC004039 was used as the reference sequence by the latter. The differences between U31120 and AC004039 are due to C insertions at positions 49963 and 49578 in AC004039 which are not present in U31120 (corresponding to positions 752-753 and 1136- 1137, respectively). Corresponds to the Gln110Arg of Heinzmann and associates (17) who used the amino acid coordinates of the mature protein (cytokine-web; http://www.psynix. co.uk/cytweb/http://targets/index.html).

Genotyping of IL-13 Polymorphisms

Three SNPs were genotyped in the IL-13 gene using the following methods. SNP 3 (Arg130Gln) was genotyped in PCR-amplified genomic DNA by allelic discrimination using TaqMan technology (Perkin-Elmer) on the ABI PRISM 7700 sequence detector (Perkin-Elmer). Oligonucleotide probes homologous to the wild-type (5'-TCGCGAGGGACGGTTCAACTGAAA-3'; labeled with carboxyfluorescein (FAM) 5'-reporter dye) and SNP (5'-TCGCGAGGGACAGTTCAACTGAAA-3'; labeled with tetrachloro-fluorescein (TET) 5'-reporter dye) sequences, and forward (5'-TAAAGGACCTGCTCTTACAT TTAAAGAAA-3') and reverse (5'-TCGAAAGCATCATTAT TTGCAGAGACAGG-3') PCR primers were designed using Primer Express (version 1.0; Perkin-Elmer) and synthesized by Perkin-Elmer. Allelic discrimination reactions were carried out on 20-ng samples of genomic DNA in a 25-µl reaction containing 50 to 900 nM of each forward and reverse PCR primer, 50 to 200 nM of each FAM and TET probe, and 1× TaqMan Universal PCR Master Mix (Perkin-Elmer). PCR cycling conditions on the ABI PRISM 7700 were as follows: 50°C for 2 min; 95°C for 10 min; followed immediately by 40 cycles of 95°C for 15 s and 60°C for 1 min.

For SNP 6 (1111 promoter) and SNP 7 (nt 4738, 3'-untranslated region [UTR]), PCR was performed in a 10-µl volume consisting of 60 ng DNA, 0.4 µM of each primer, 50 mM KCl, 10 mM Tris, 0.2 mM of each dNTP, 1.5 mM MgCl₂, and 0.1 U Taq DNA polymerase. The promoter polymorphism was amplified using the primers 5'-ATGCCTTGTGAGGAGGGTCAC-3' and 5'-CCA GTCTCTGCAGGATCAACC-3'. The PCR products were then purified with Qiagen PCR purification kit and sequenced using the given primers with the BigDye kit and the 3700 DNA Analyzer (ABI). The sequence was performed in both directions and analyzed and viewed with Phred/Phrap (29) and Consed (30). Genotypes were determined by visual inspection of the sequence files.

The 3'-UTR polymorphism was amplified using the primers 5'-CTTTGCTAACATATTTAATATTTAAATACG-3' and 5'-GTCACCGTTGGGGATTGGGGAAG-3'. PCR cycling conditions were as follows: 94°C for 4 min; 30 cycles at 94°C for 30 s; 68°C for 30 s; and 72°C for 30 s; with a final extension step of 72°C for 6 min. PCR products were digested with NheI (New England Biolabs, Beverly, MA) and the alleles resolved by electrophoresis on a 2% agarose gel. The fragment sizes were 289 base pairs (bp) for the G allele, and 252 and 37 bp for the A allele.

Genetic Analysis

Analysis was performed for four phenotypes including asthma and associated phenotypes: BHR, total serum IgE levels, and skin-test responsiveness. Genetic analysis was conducted with each of the biallelic polymorphisms by comparing differences of allele and genotype frequencies between cases and controls. For comparing the allele frequencies between cases and controls, -square ² tests were used. When genotypic frequencies were compared between cases and controls, ² tests assuming a dominant model were performed (due to the small number of homozygotes for the rare allele). No corrections were made for multiple comparisons, for two reasons. First, because the phenotypes tested (asthma, BHR, total serum IgE levels, and skin-test response) are strongly associated with each other in this population, the statistical analyses do not represent independent tests. Second, we performed tests for association with phenotypes that have been observed by other investigators, both to confirm previous results and to better characterize asthma susceptibility in our population. As described previously, all of the probands fit published criteria for an asthma diagnosis (27). For BHR, cases were defined as probands and spouses with a PC₂₀  32 mg/ml histamine. The control group for the analysis of both "asthma" and BHR cases was BHR-negative (BHR) spouses. Individuals were considered skin test-positive if one or more skin tests showed a maximum wheal diameter of  5 mm. Individuals with high total serum IgE levels were defined as having total serum IgE  100 IU/ ml, because this value best distinguished individuals with high versus low levels after examining the overall frequency distribution for the group (5, 6). Total serum IgE was also analyzed as a quantitative trait following logarithm-transformation to approximate a normal distribution.

The linkage disequilibrium test between pairs of SNPs was based on an exact test assuming multinominal probability of the multilocus genotype, conditional on the single-locus genotype (31). A Monte Carlo simulation was used to assess the significance, by permuting the single-locus genotypes among individuals in the sample to simulate the null distribution. The empirical P values of the linkage disequilibrium for each pair of SNPs was based on 10,000 replicate samples.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Characteristics of Population Sample

As illustrated in Table 2, the probands and spouses were of similar ages (mean = 52 and 51 yr, respectively). All spouses are included in the table, but only BHR spouses (PC₂₀ > 32 mg/ml histamine) were used for comparison with the probands for the asthma and hyperresponsiveness phenotypes. All probands and BHR-positive (BHR+) spouses were included in the BHR+ group for the BHR comparison (with BHR spouses). All probands were BHR+ at the time of initial testing. A total of 171 of the 184 subjects were retested because 13 had an FEV₁ that was too low to be retested safely (FEV₁  40% predicted); 10% of those retested were no longer BHR+. A high proportion of the probands (40.3%) were very hyperresponsive (PC₂₀  2 mg/ml). Although the probands were not selected for atopy, 80.9% had  1 positive skin tests compared with 29.9% of the spouses. The degree of overlap between the asthma and atopy phenotypes in this population is shown in Figure 1.

View larger version (13K): [in this window] [in a new window]   Figure 1.   Relationship of BHR, skin-test response, and total serum IgE levels in Dutch probands and spouses genotyped for IL-13 polymorphisms.

Resequencing of the IL-13 Gene to Identify SNPs

The human IL-13 gene is composed of four exons spanning 2.938 kb of genomic DNA (32). SNPs in the IL-13 gene were identified by resequencing genomic DNA from 20 probands and 20 control subjects (spouses) from the Dutch families. In addition to exons 1 to 4, over 2 kb of the 5'-flanking region and the first intron were resequenced because potential recognition sites for a number of transcription factors, interferon-inducible elements, and enhancer elements have been localized to these regions (33).

A total of 10 SNPs were identified by resequencing (Figure 2), with only one (SNP 6, located in exon 4) leading to a predicted amino-acid change in the IL-13 protein (glutamine for arginine amino-acid substitution; Arg130Gln). Six of the SNPs have previously been identified in other population groups: SNP 3 (5'-promoter [14, 16, 34]); SNP 6 (exon 4 [15-17]); SNP 2 (5'-promoter [16]); and SNPs 7, 8, and 9 (exon 4, 3'-UTR; [16]). SNPs 1 (5'-promoter), 4 and 5 (intron 1), and 10 (exon 4, 3'-UTR) are novel.

Association Analysis of IL-13 with Asthma and Atopy Phenotypes

A total of 184 Dutch probands and spouses were genotyped and analyzed with SNPs 3, 6, and 7. These SNPs were located in the promoter region, exon 4, and the 3'-UTR (Figure 3). All SNPs were in Hardy-Weinberg equilibrium and the allele frequencies for each SNP are reported in Table 3. These frequencies are similar to those reported in other populations from the United States, the Netherlands, and Germany (16), but differ from those from the United Kingdom and Japan (17). Significant linkage disequilibrium was observed between the 1111 promoter and the 3'-UTR SNP (P < 10⁴) and between the Arg130Gln and 3'-SNP (P = 10⁴), but not between the 1111 promoter and the Arg130Gln.

View larger version (17K): [in this window] [in a new window]   Figure 3.   Schematic representation of the IL-13 gene showing the genomic structure ( filled boxes represent the four exons) and the location of SNPs genotyped in this study and others. Phenotypes with a significant association (P < 0.05) are shown in the table below, aligned with the associated polymorphism.

The results of this study reveal significant associations between specific asthma and atopy phenotypes for the 5' promoter and 3'-UTR SNPs (Table 4). The most significant associations were observed with the promoter polymorphism and the presence of both asthma and skin-test sensitivity ( 1 positive skin test) to common aeroallergens. The promoter "T" allele was significantly more prevalent in the cases with asthma than in controls (24 versus 14%; p = 0.004). This association was also present to a lesser degree in individuals with one or more positive skin tests (25 versus 17%; P = 0.01). When compared by genotype, the association with both asthma and allergy phenotypes was highly significant. TT homozygotes were much more common in cases than controls with regard to asthma (P = 0.008), BHR (P = 0.007), and asthma and skin-test sensitivity combined (P = 0.006). Although we did not observe a significant association of IL-13 polymorphisms with IgE levels, there was evidence of higher total serum IgE levels with the rare allele of the promoter polymorphism (P = 0.089; Table 5). When stratified by skin test-positive and skin test-negative individuals, we did not observe an association with total serum IgE levels in skin test-negative individuals, as previously reported with the Arg130Gln polymorphism (16). Lower levels of significance were also observed with the 3'-UTR SNP (Table 4).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Asthma is an inflammatory airways disease characterized by BHR and airways obstruction. Atopy traits, such as elevated total serum IgE levels and positive allergen skin-test responses, are also associated with this disease and may predict the development of symptomatic asthma (1, 2, 4). IL-13 is expressed in asthmatic airways and has an important role in the production of IgE, and is therefore an excellent biologic candidate gene for the development or expression of diseases with atopy components such as asthma.

In this study of the IL-13 gene, 10 SNPs were detected, of which six have been identified previously and four are novel. Of the former, SNPs 3 (1111 promoter) and 6 (Arg130Gln) have been associated with increased risk of allergic asthma (14) and higher total serum IgE levels (16) or asthma (17), respectively. In addition, SNP 3 appears to promote increased binding of nuclear proteins to the promoter region (14), whereas the amino-acid change resulting from SNP 6 could affect the interaction of IL-13 with IL13R1 (16, 17). In contrast, none of the remaining eight SNPs occur within identified regulatory elements in the IL-13 gene (e.g., 33), or alter the amino-acid sequence, so their functional relevance is unclear at the present time. We investigated the contribution of SNPs 3, 6, and 7 using a case-control study in a Dutch asthma population consisting of probands, selected on the basis of a diagnosis of asthma, and their unaffected spouses. One SNP was chosen in the 3'-UTR (SNP 7, nt 4738) to examine the potential regulatory elements in this region. These three SNPs encompassed the entire IL-13 gene so that the contribution of genetic variation could be detected. Consistent with other reports that have evaluated IL-13 polymorphisms, we observed a significant association of several polymorphisms in IL-13 with various atopy or asthma phenotypes.

There are several unique properties of this study. The cases and controls are the parents in the same families used to identify linkage to chromosome 5q31, allowing us to examine candidate genes within this region in linked families. The comprehensive clinical data collected in these families allowed us to examine multiple clinical phenotypes associated with both asthma and atopy. And finally, because of the study design, the controls were similar in regard to age and overall environmental exposures (allergens).

Both IL-4 and IL-13 can elevate baseline IgE levels. However, although an essential role for IL-4 in the induction of asthma has been proposed, murine models have demonstrated the critical nature of IL-13 independent of IL-4. In a murine model of allergic AHR, blockade of IL-13 reversed many of the characteristics found in allergic asthma, such as AHR, eosinophil infiltration, and mucous production (35). Further, the effects of IL-13 were shown to be mediated by a pathway dependent on the IL4R receptor (35). Transgenic mice expressing IL-13 specifically in the lungs exhibited increased BHR, bronchial eosinophilia, and increased mucous production (36).

Using a mouse model for asthma, Symula and colleagues (37) were able to demonstrate the effects of quantitatively changing mouse IL-4 and IL-13 gene expression using transgenic mice constructed with human yeast artificial chromosomes from the chromosome 5q3 region. Surprisingly, the transgenic mice had significantly lower total IgE levels due to decreased endogenous gene expression, which influenced the development of Th2 cells. (Human IL-4 and IL-13 appear to have minimal activity in mice.) Further, when a bacterial artificial chromosome containing mouse IL-4 and IL-13 genes was transfected into the mice, significant increases in IgE levels, BHR, and asthma were observed (37).

It has been suggested that IL-13 may be an important regulatory cytokine in the pathogenesis of asthma, whereas IL4R, which is required for the functioning of both IL-4 and IL-13, contributes primarily to atopy (38). Several IL4R polymorphisms have been associated with a higher risk of atopy (19, 21), atopic asthma (15), and variation in IgE levels (21). In addition, specific alleles of these variants were shown to modulate the activity of IL4R (19). In a recent study by Ober and associates (22), eight polymorphisms were studied in both inbred and outbred populations. Significant evidence for an association between these variants and the resulting haplotypes were observed for asthma and atopy. These studies indicate that biologic functions related to these disorders may involve this receptor.

The most significant associations with IL-13 were identified in individuals with a diagnosis of asthma (i.e., the original probands in this study) and in individuals with the BHR phenotype (including affected spouses) (Table 4). The study by Heinzmann and coworkers (17) also supports this role of IL-13 in asthma. In both the Japanese and the British populations, the most significant associations were observed between IL-13 (Gln110Arg) and atopy and non-atopy asthma. There was no evidence of an association with this polymorphism and total serum IgE levels. In the Japanese population, strong association was observed with the Ile50Val IL4R polymorphism and IgE levels, both total serum and allergen-specific (P < 0.0001). The coexistence of BHR and atopy characteristics (e.g., Figure 1) makes it difficult to discern the exact roles of IL4R and IL-13. However, the interaction of these two genes as important components of IgE-mediated inflammatory responses supports the role of these cytokine pathways in the development or expression of atopy conditions and asthma.

Association analyses with specific polymorphisms in IL-13 have produced varied results (Figure 3) (14). This may be due, in part, to the fact that each study is based on population samples that were ascertained differently. For example, the previous studies focused on recruitment of patients with allergic asthma (14, 17), atopy (15), or random ascertainment from longitudinal and cross-sectional groups (16). In our study design, we ascertained families on the basis of asthma, but were also able to examine both asthma and atopy phenotypes. In addition, these families showed evidence for linkage to the region of chromosome 5q where IL-13 is located (5).

Another potential cause of differences in the results of these studies is that they were performed in different population groups. Because linkage disequilibrium varies between populations, this would suggest one of two possibilities. First, it is possible that different polymorphisms or haplotypes within the IL-13 gene contribute to the allergy phenotype in each population. Therefore, each analysis may be identifying the specific allele or haplotype responsible for the phenotype in that specific population. This suggests that several of the polymorphisms identified are capable of significantly altering the function of IL-13, resulting in a predisposition to atopy, and that reported differences are primarily due to the founder allele in that specific population. A second explanation is that an additional, unidentified sequence variant is responsible for the phenotype, and the level of detection (i.e., significance of the association) is dependent on the degree of linkage disequilibrium in that population for this region of chromosome 5q. In this case, the true susceptibility allele would have to be a fairly distant enhancer or promoter element, inasmuch as IL-13 resequencing studies have most likely identified all of the common variations within or near the gene (this report, and refs. 14 and 16). One candidate for such a distal element would be the conserved noncoding sequence (CNS-1) recently identified between IL-13 and IL-4 (39). This sequence is approximately 3 kb from the 3' end of IL-13 and 10 kb from the 5' end of IL-4; is highly conserved between mice, humans, cows, dogs, and rabbits (~ 80% identity); and was shown to specifically regulate IL-4, IL-13, and IL-5 in human YAC transgenic mice. This region should be investigated to determine whether polymorphisms are present that may influence susceptibility or expression of atopy or asthma. A variation in CNS-1 may explain the different results from the IL-13 studies, and may also explain the conflicting results with a promoter polymorphism in IL-4 (13), inasmuch as various degrees of linkage disequilibrium with CNS-1 and the IL-13 promoter between populations may contribute to differences in reported associations.

Asthma and allergy are common diseases caused by an intricate interaction of genetic susceptibility and environmental exposure. This association study, as well as those reported by others, implicated IL-13 as a major component in the expression of these conditions. Functional studies examining the roles that these polymorphisms have on the activity or expression levels of IL-13 and, perhaps more importantly, the downstream responses to these changes will provide valuable insight into the overall mechanisms that cause susceptibility to asthma and atopy. As we define specific genes that are associated with allergy and asthma phenotypes, patterns are beginning to develop that may be useful in delineating important biologic pathways, leading to a better understanding of gene-gene and gene- phenotype relationships in asthma and allergy. The importance of IL-13 and its functional and genetic interactions with IL-4 and IL4R in the development and expression of asthma and atopy support this approach that combines molecular genetic techniques with clinical studies.