DISCUSSION

To our knowledge, this is the first study to look at PSW phenotypes andARG gene polymorphisms. We hypothesized that there would be relationships between a PSW phenotype and polymorphisms of theARG1 and ARG2 genes. Our study revealed that the frequency of the homozygote C-allele rs2781667 in ARG1 was significantly higher in the MW and asthma groups than in the EW group. Li et al.13 reported that four SNPs in ARG1(rs2781659, rs2781665, rs2781667, rs2246012) and rs 3742880 inARG2 were associated with the number of positive skin tests, and carrying two copies of the minor allele for rs3759757 and rs6573788 SNPs in ARG2 were found to be associated with the presence and severity of asthma. Litonjua et al.14 reported that rs2781659 in ARG1 was significantly associated with BDR in asthmatic children. In another study15, each variant allele of rs3742879 in ARG2 was found to be associated with increased asthma risk. Vonk et al.16 reported that two polymorphisms in ARG2 (rs17249437 and rs3742879) were associated with both asthma and more severe airway obstruction, and increased airway hyperreactivity (AHR) and lower beta-2-agonist reversibility were reported as associated with both ARG1 and ARG2. Hing Yee et al.20 reported that rs7216389 in ORMDL3 and rs3756780 in ARG1 interactions might be associated with a risk of asthma, and rs2749935 in ARG1 and rs2190242 in CRHR2interactions might be associated with BDR. A recent study from India showed that high arginase activity and IL-13 concentration in serum and the ARG1 rs2781666 G/T genotype might increase the risk of asthma21. Vonk et al.16 reported that the protective effect of inhaled corticosteroids (ICS) against annual FEV1 decline was significantly lower in homozygote carriers of the C-allele of rs2781667 in ARG1 in a longitudinal cohort of 200 adult asthma patients. Based on this, our finding of a higher frequency of homozygote C-allele rs2781667 in ARG1 in the MW and asthma groups may be important for the management of patients with PSW. We think it might be a useful marker for the differentiation of the PSW phenotype. In addition, the association between this SNP and ICS or BDR could be addressed in future studies.
We also found that the homozygote CC genotype of the ARG1rs2781667T>C SNP was more frequent in patients with allergic rhinitis than in those without allergic rhinitis (p = 0.004). While it is known that the arginase-NOS pathway is associated with allergic rhinitis 23,24, we were unable to find any study related to the association of allergic rhinitis and ARGgenes until now. Furthermore, we have no data for whether ARG1rs2781667 is a functional polymorphism and associated with changes in arginase expression in the nasal mucosa.
Our haplotype analyses revealed that the frequencies of the ARG1haplotype 2 and haplotype 5 were significantly higher in the EW group than in the control group. Two recent studies suggest an interaction between arginase expression and viral airway infections25,26. Santiago-Olivares et al.25 investigated the impact of Respiratory Syncytial Virus (RSV) infection on nitric oxide production in an experimental model and found that ARG1 overexpression contributes to the maintenance of the RSV genome in the host in persistent infection. Wieczfinska et al.26 found that the expression of TGF-β1 and arginase was increased due to rhinovirus infection. To date, we have no data for viral-induced wheezing and genetic variations inARG genes, but new studies could advance this. The ARG1haplotype 4 was more frequent in the MW group than in the EW group. Interestingly, ARG1 haplotype 4 carried the variant C-allele for rs2781667 in ARG1 , so it may be useful for differentiating between EW and MW. Salam et al.15 reported that the association between arginase variants and childhood asthma may be related with haplotypes rather than SNPs. Duan et al.19 studied haplotypes were identified according to the SNPs (rs2781659, rs2781663, rs2781665, and rs60389358) in theARG1 and they have reported that two haplotypes may alter BDR and differentially regulate gene expression. The haplotypes that we estimated according to the studied SNPs were different from those used in these studies, so it is impossible to make a comparison between haplotype-based results.
The diagnoses of asthma cases and the classification of patients according to a PSW phenotype were performed by pediatric allergists at a tertiary pediatric allergy specialty clinic, these were strengths of the our study. However, there were some limitations. First, the sample size was smaller than previous studies have referred to; additionally, it does not represent all Turkish children. Second, this was a case-control study; thus, we have no longitudinal data about the progress and bronchodilator or ICS responses of patients with a homozygote CC genotype of the ARG1 rs2781667T>C and/or ARG1haplotype 4. Third, we included no information about the functionality of the homozygote CC genotype of the ARG1rs2781667T>C and/or ARG 1 haplotype 4 in lung and nasal mucosa.
In conclusion, our study revealed that ARG1rs2781667T>C and/or ARG 1 haplotype 4 were more frequent in patients with MW and asthma than in the EW group. The frequencies of the ARG1 haplotype 2 and haplotype 5 were significantly higher in the EW group, while haplotype 9 was more frequent in the healthy controls. Taken together, we conclude that variations in ARG1 gene may be potentially important for differentiating PSW phenotypes. Our findings need to be confirmed in larger, different PSW populations, preferably with different ethnic backgrounds.