Asthma is one of the most common types of chronic respiratory diseases, affecting ∼300 million children and adults worldwide (1). It is well-known that asthma is affected by genetic as well as environmental factors and several genes have been shown to be associated with asthma (2,3). A disintegrin and metalloprotease domain 33 (ADAM33) has been reported as a new member of the ADAM gene family (4). ADAM33 was the first putative asthma susceptibility gene to be identified by positional cloning in two Caucasian populations; one in the United Kingdom (UK) and the other in the United States of America (USA) (5). Since the first study on the association between ADAM33 polymorphisms and asthma, several replication studies have been published with conflicting results (6,7).

However, these studies indicated a diversity in the relationship between ADAM33 polymorphisms and asthma. Possible reasons for such diversity may be the differences in the genetic background or environmental exposure that may affect the development of asthma. Thus, asthma susceptibility genes may differ among different nationalities. The majority of data currently available on ADAM33 and its correlation to asthma have been derived from Caucasian populations. Therefore, it is important to determine whether the findings in Caucasians may be replicated in different asthmatic populations. The Uygur population is located in the northwest of China and comprises ∼11,000,000 people, with a relatively stable area of residence and little reported migration. The Uygur people are rarely associated with ethnic intermarriages and their lifestyles are consistent. Therefore, this is an ideal population for the conduction of genetic studies. The aim of the present study was to determine the association of ADAM33 single-nucleotide polymorphisms (SNPs) with asthma and asthma-related phenotypes in the Uygur population of China.

The ADAM gene family, of which there currently are 34 members, is a subgroup of the zinc-dependent metalloproteinase superfamily (8,9). Previous studies have demonstrated that ADAM family members are involved in fundamental processes, such as cell adhesion, fusion, migration, membrane protein shedding and proteolysis (8,10). ADAM33, which was recently identified as a novel asthma susceptibility gene, maps to human chromosome 20p13, consists of 22 exons and a total of 37 SNPs have been identified (5). Various associations of ADAM33 SNPs with asthma, chronic obstructive pulmonary disease (COPD) and bronchial hyperresponsiveness (BHR) have been reported by previous studies (11–13). ADAM33 is a complex molecule, the expression of which is largely restricted to mesenchymal cells, including smooth muscle cells of the bronchi and pulmonary fibroblasts; thus, it is important in airway remodeling (14,15).

The three SNPs used in this study have been previously reported to be associated with asthma and BHR in Caucasian populations in the UK or the USA; however, the association was not significant in the two samples combined (5). Associations with asthma were then replicated in US Caucasians (T1 and T2), in a German family-based study sample (F+1) (16) and in pediatric and adult Japanese populations (17). However, no SNP has been consistently associated with asthma across ethnically diverse groups. Therefore, different SNPs of the ADAM33 gene may contribute to asthma susceptibility in specific ethnic populations.

SNP T1 (Met764Thr) in exon 20 encodes the cytoplasmic domain. The substitution of thymine for cytosine in the T1 SNP results in the substitution of methionine for threonine in the cytoplasmic domain of the protein and this may alter intracellular signaling, resulting in increased fibroblast and smooth muscle cell proliferation.

The rare allele of T1 (T1 C) exhibited increased frequency among asthmatic patients compared to healthy controls in the previously studied Caucasian (24.1 vs. 7.4%) (5), Japanese (15.3 vs. 5.3%) and Korean populations (9 vs. 8%) (17,18). Our study also demonstrated similar results (20.6 vs. 11.1%). There were significant differences in the allele frequency distribution of the T1 locus between asthmatic patients and healthy controls (χ²=8.559, P=0.003). Compared to the C allele, the T allele demonstrated a 2.08-fold increased risk of asthma (OR=2.08; 95% CI, 1.273–3.398). Our results revealed that the ADAM33 T1 TC genotype had an OR of 2.118 (P=0.016) compared to the ADAM33 T1 CC genotype, without adjustment for other confounding factors. Multiple logistic regression analysis demonstrated that the T1 TC genotype is an independent genetic factor for asthma (OR=2.317, P=0.031) compared to the TT genotype, following adjustment for other confounding factors. Similarly, significant associations with asthma were demonstrated for T1 (Met764Thr), T2 (Pro774Ser), S2 and V-3 (with the lowest P-value for T1, P=0.0015; OR 0.63) in the Japanese population (17). Differences in the frequency of SNPs in asthmatic patients and healthy controls as reported by different studies have been summarized by Postma and Howard (19). We demonstrated significant associations with asthma in SNPs (T1). Our results were consistent with those of a study conducted by Schedel et al on a German population (20).

Foley et al(21) demonstrated an increase in the expression of the ADAM33 gene associated with the increase in the severity of asthma. This gene may contribute to the remodeling process that occurs during the progression of asthma. In addition, recent studies demonstrated that SNPs in ADAM33 are associated with an accelerated decline in pulmonary function in the general population and were identified as a risk factor for COPD (11,22). The present study demonstrated a significant difference in FEV₁ levels among different T1 genotypes. Lower FEV₁ levels were observed in the TC compared to the TT genotype, a finding consistent with those of Simpson et al(23), who verified that ADAM33 polymorphisms affect pulmonary function during childhood and early life and may increase the risk of asthma. In the present study, there were no differences in the levels of total serum IgE and ECP among the T1, S+1 and F+1 genotypes. We also observed no association between ADAM33 and allergy. In agreement with the findings of Umland et al(24), this study has demonstrated that ADAM33 mRNA is significantly expressed in smooth muscle-containing organs and minimally expressed in immune organs, suggesting a role for ADAM33 in remodeling, rather than in the immune response.

Two of the SNPs we investigated have been previously reported to be associated with asthma and BHR in Caucasian populations from the UK (intron 6, 7575/F+1) and the USA (exon 20, 12433/T1) (5). Werner et al(16) demonstrated that SNP F+1 of the ADAM33 gene is associated with asthma with the lowest P-value in that study. Simpson et al(23) demonstrated that carriers of the rare allele, F+1, exhibited reduced pulmonary function at 3 years of age (P=0.003). When the recessive model was considered, four SNPs (F1, S1, ST5 and V-4) had an association with specific airway resistance (sRaw) at 5 years of age (P<0.04). Polymorphisms in ADAM33 predict impaired pulmonary function in early life. The functionally relevant polymorphism is likely to be located at the 5′ end of the gene. A meta-analysis demonstrated that only 2 out of the 13 studied variants (F+1 and ST+7, in strong linkage disequilibrium), were significantly associated with asthma in the unrelated case-control data and the family-based data (25). Significant associations of asthma with the S+1, ST+4 and T2 sites and with the T1, T2, S2 and V-3 sites have been reported in pediatric and adult Japanese populations, respectively (17). Raby et al(26) reported 8 additional SNPs (S+1, G1, I1, KL+3, N1, T+2, V-2 and V-3). None of the additional SNPs had an association with asthma.

In the present study, we determined the association between the S+1 and F+1 polymorphisms and asthma in the Chinese Uygur population. The study did not demonstrate a difference in the alleles and genotypes between the two groups. Our results indicated that the S+1 and F+1 polymorphisms in the ADAM33 gene are not associated with susceptibility to asthma. Part of the results of our study are in agreement with the results of Schedel et al(20), while the findings suggest that the F+1 polymorphism of the ADAM33 gene is not associated with asthma.

It has been demonstrated that the ADAM33 gene is involved in the pathogenesis of asthma (27). The T1 polymorphic site of the ADAM33 gene is associated with susceptibility to asthma in the Uygur ethnic group. The T allele may be protective, whereas the C allele may increase susceptibility to asthma. The potential mechanism underlying the association of the T1 locus SNP to asthma may be that the T1 locus SNP leads to decreased FEV₁ levels and increased susceptibility to asthma. However, in a study of 290 patients and 270 controls in a Chinese Han population, Wang et al(28) did not identify a significant association of the F+1, S+1 and T1 SNPs with bronchial asthma. The Uygur population is a mixture of European (40–60%) and East Asian ancestry and is thus different from other Chinese populations. Several findings are not reproducible, due to a number of confounding variables (e.g., age, gender and ethnic background). If we determine the association of specific polymorphisms of a specific gene with asthma, the benefits of studying the association between gene polymorphisms and diseases are probably limited. In addition, the diversity of the results regarding common polymorphisms of the ADAM33 gene, such as F+1 or S+1, are possibly related to gene-gene and gene-environment interactions. Undarmaa et al(29) reported that interactions between TBXA2R and ADAM33 may increase the risk of childhood asthma and adult/childhood asthma combined. Stratification of the subjects by environmental or other confounding factors may be required to improve the sensitivity and reliability of the association results.

In a multi-locus haplotype analysis, the ADAM33 gene haplotype profile is significantly different between patients and controls in the Uygur ethnic group. The frequencies of the CAC and CAT haplotypes are higher among asthmatic patients compared to healthy controls, suggesting that these two haplotypes increase the risk of asthma. Furthermore, we observed that the frequency of the TAC haplotype was lower among asthmatic patients compared to healthy controls. This result suggests that there are ethnic-specific disease-predisposing or -protecting haplotypes of asthma in the Chinese Uygur population.

In conclusion, this study demonstrated the association between ADAM33 gene polymorphisms and the risk of asthma in the Chinese Uygur population. To confirm the significance of these observed polymorphisms, definitions and standardization of phenotypes, endorsement from several investigators is required. Large-scale testing platforms may offer new opportunities for discovering susceptibility gene variants; however, they should be coupled with careful study design, international collaboration and dissection of gene-environment interactions.