Genetic variants in Hippo pathway genes are associated with house dust mite‐induced allergic rhinitis in a Chinese population

Abstract Background House dust mite (HDM)‐induced allergic rhinitis (AR) is a highly prevalent disease with bothersome symptoms. Genetic variants of the Hippo pathway genes play a critical role in the respiratory disease. However, no study has reported associations between variants of the Hippo pathway genes and HDM‐induced AR risk. Methods Forty‐three key genes in the Hippo pathway were selected from the Kyoto Encyclopedia of Genes and Genomes (KEGG), Reactome pathway database, and previous reported studies. A case‐control study of 222 cases and 237 controls was performed to assess the associations between 121 genetic variants in these genes and HDM‐induced AR risk. DNeasy Blood & Tissues Kits were used for extracting genomic DNA from the venous blood and Infinium Asian Screening Array BeadChips for performing genotyping. A logistic regression model was applied to evaluate the effects of variants on HDM‐induced AR risk. The false discovery rate (FDR) method was utilized to correct for multiple testing. The receiver operating characteristic (ROC) curve was plotted to obtain the cut‐off value of total IgE for the diagnosis of HDM‐induced AR. Histone modification and transcription factor binding sites were visualized by UCSC genome browser. Moreover, expression qualitative trait loci (eQTL) analysis was obtained from Genotype‐Tissue Expression (GTEx) database. Results We found that rs754466 in DLG5 was significantly associated with a decreased HDM‐induced AR risk after FDR correction (adjusted odds ratio [OR] = 0.52, 95% confidence interval [CI] = 0.36–0.74, p = 3.25 × 10−4, P FDR = 3.93 × 10−2). The rs754466 A allele reduced the risk of HDM‐induced AR in the subgroup of moderate/severe total nasal symptom score (TNSS). Furthermore, rs754466 was associated with a high mRNA expression of DLG5. Additionally, histone modification and transcription factor binding sites were rich in the region containing rs754466. Conclusion Our findings indicated that rs754466 in DLG5 decreased the susceptibility to HDM‐induced AR.


| INTRODUCTION
Allergic rhinitis (AR) is often triggered after exposure to indoor or outdoor aeroallergens, such as house dust mite (HDM), pollens, animal dander and fungal allergens. 1 HDM is almost omnipresent and perennial indoors, leaving no effective measures to eliminate HDM.
Additionally, constant exposure to HDM, compared with other aeroallergens, may contribute to more chronic and severe symptoms that exert long-term impact on quality of life. [2][3][4] Moreover, HDMinduced AR can increase the risk of asthma. 5 Many studies have verified the critical role of genetic variants in the pathogenesis of AR. 6 This inspired us to delve into HDM-induced AR mechanism through genetic variations.
The Hippo pathway, first identified in Drosophila melanogaster, is an evolutionarily conserved pathway regulating organ size, tissue generation, and stem cell self-renewal. 7,8 Dysregulation of the Hippo pathway is widely implicated in cancers, 9 cardiac diseases, 10 renal diseases, 11 and respiratory diseases. [12][13][14] Previous studies suggested that single-nucleotide polymorphisms (SNPs) in YAP, FRMD6, BIRC5, and DLG2 of the Hippo pathway were associated with asthma risk. 13,15,16 To date, no study has been undertaken to investigate the associations between genetic variants in the Hippo pathway and HDM-induced AR risk.
In this study, we hypothesized that genetic variants in the Hippo pathway were associated with HDM-induced AR risk, and tested this hypothesis with a case-control study among Han Chinese.

| Study population
A total of 222 cases and 237 controls were included in this casecontrol study. The cases were recruited from the First Affiliated Hospital of Nanjing Medical University between May 2008 and October 2017. The diagnosis of HDM-induced AR was established according to Allergic Rhinitis and its Impact on Asthma criteria. 1

| Clinical evaluation
A total nasal symptom score (TNSS) on a scale of 0-12 evaluating sneezing, rhinorrhea, nasal itching and nasal obstruction was used to assess the disease severity. The score of each symptom is: 0 = no symptoms; 1 = mild symptoms (symptom clearly present, but easily tolerated); 2 = moderate symptoms (definite awareness of symptom, i.e., bothersome but tolerable); 3 = severe symptoms (symptom, i.e., hard to tolerate; causes interference with activities of daily living and/or sleeping). 17 Cases with TNSS of 0-4 and 5-12 were diagnosed with mild and moderate/severe HDM-induced AR, respectively. 18 The serum total IgE was measured by ImmunoCAP system (Ther-moFisher Scientific, Uppsala, Sweden). Serum specific IgEs to Der p, Der f, cat dander, dog dander, Blatella germanica, Alternaria alternate, Ambrosia elatior, and Artemisia vulgaris were also detected in cases by ImmunoCAP system. Phadiatop tests were conducted in controls.

| Selection of genes and genetic variants in the Hippo pathway
Pathway database, and reported studies. 15,16,[19][20][21][22][23] Genes located on sex chromosomes were excluded. Next, the data of Han Chinese in Beijing (CHB) and Japanese in Tokyo (JPT) from the 1000 Genomes Project (March 2012) was used to identify genetic variants. Then, variants matching the following quality control criteria were selected: minor allele frequency (MAF) > 0.05, call rate > 95%, and p value of Hardy-Weinberg equilibrium (HWE) > 10 −6 .

| In silico analysis
Histone modification and transcription factor binding sites were visualized by UCSC genome browser from chromatin immunoprecipitation sequencing (ChIP-Seq) data stored in ENCODE. The expression qualitative trait loci (eQTL) analysis was obtained from Genotype-Tissue Expression (GTEx) database (https://www.gtexportal.org/home/) for genetic variants.

| Characteristics of study population
The demographic and clinical characteristics of 222 cases and 237 healthy controls are summarized in Table 1. No significant differences in age (p = 0.739) and sex (p = 0.974) were found between cases and controls. The level of total IgE was significantly higher in cases than in controls (p < 0.001). According to the ROC curve ( Figure S1), the cut-off value of total IgE was set as 60.45 kU/L, with

| Genetic variant in DLG5 and HDM-induced AR risk
Of 49 candidate genes from KEGG and Reactome Pathway database, we teased out 43 genes having vital functions in diseases from reported studies ( Figure S2 and Table S1). The flow chart of selecting genetic variants is presented in Figure 1. We identified 5066 variants in 43 genes from the 1000 Genomes Project after quality control.
Next, 577 genetic variants were filtered out with LD analysis. Then, 121 variants in 33 genes were retained for further study after functional annotation predicted by SNPinfo, HaploReg, and Reg-ulomeDB. In this study, we found that 8 genetic variants (rs754466, rs11002309, rs7744287, rs6790596, rs2032, rs7650899, rs2425672, and rs2236947) were nominally associated with HDMinduced AR risk in the additive genetic model (all p < 0.05) ( Table 2). After FDR correction, only rs754466 in DLG5 was found to be associated with a decreased risk of HDM-induced AR To comprehensively investigate the association between rs754466 and HDM-induced AR risk, we used four genetic models (codominant, additive, dominant, and recessive models) to analyze the correlation of rs754466 with HDM-induced AR risk ( Table 3).

| Stratification analyses of rs754466 with HDMinduced AR risk
Subgroup analyses based on different demographic characteristics were performed ( Figure 2 and Table S2). The rs754466 A allele was associated with a lower risk in age and sex subgroups (all p < 0.05).
Additionally, the rs754466 A allele decreased the risk of HDMinduced AR in all subgroups of total IgE, specific IgE to Der p, and specific IgE to Der f (all p < 0.05). However, no significant heterogeneity was observed in all these subgroup analyses (all p > 0.05).
In the moderate/severe TNSS subgroup, the rs754466 A allele showed the most significant association with a reduced risk of HDMinduced AR (p = 9.70 � 10 −4 ). A stronger association was also observed between the rs754466 A allele and a decreased HDMinduced AR risk in moderate/severe rhinorrhea, moderate/severe nasal obstruction, mild sneezing and mild nasal itching subgroups ( Figure 3 and Table S3).
Moreover, we found that histone modification and transcription factor binding sites were rich in the region harboring rs754466 by UCSC genome browser ( Figure S3). In addition, the rs754466 T allele was associated with a high expression of DLG5 in whole blood, lung and cultured fibroblasts based on GTEx database. However, rs754466 mutated in a manner of A > T in the public database, but T > A in our study. This inconsistency may be explained by that the DNA strand detected in our study was complementary to that detected in public database. Moreover, similar results were reported by Wang et al. 24 and Feng et al. 25 Hence, the rs754466 A allele in our study increased the expression of DLG5 in whole blood, lung, and cultured fibroblasts ( Figure S4).

| DISCUSSION
In mammals, the Hippo pathway is mainly composed of mamma- proliferation of airway epithelial cells. [26][27][28][29] When the Hippo pathway is suppressed, YAP and TAZ translocate to the nucleus to induce expression of target genes, thus causing asthma, lung fibrosis and chronic rhinosinusitis with nasal polyps (CRSwNP). [12][13][14]22 It was reported that the susceptibility to HDMinduced AR was closely related to genetic variations. 30

lesions,
Previous study indicated that rs9671722 in the Hippo pathway increased the risk of AR with asthma. 13 To our knowledge, no study has reported the association between genetic variants in the Hippo pathway genes and HDM-induced AR risk. In the present study, we found that rs754466 in DLG5 was strongly correlated with a decreased risk of HDM-induced AR.  33 which might serve as a mechanism in the pathogenesis of AR. 34 Additionally, DLG5 maintains the integrity of epithelial barrier to curb the invasion of pathogens into the tissue interstitium. 35,36 As previously described, DLG5 lowly expressed in Crohn's disease, 35 breast cancer, 32 prostate cancer, 37 and hepatocellular carcinoma. 38

| CONCLUSIONS
We performed the first analysis for associations between genetic variations in the Hippo pathway genes and the risk of HDM-induced AR in a Chinese population. And we found that rs754466 in DLG5 had a significant decreased risk of HDM-induced AR. Our findings might provide new insight into the pathogenesis of HDM-induced AR and a new therapeutic target.

ACKNOWLEDGMENTS
We are grateful to Prof. Dr. Mei-Lin Wang for his valuable suggestions for improving the quality of this study. We also thank Associate Prof. Yong-Ke Cao for professional English-language proofreading of the manuscript. This study was supported by a grant (No. 81870733) from the National Natural Science Foundation of China.

CONSENT FOR PUBLICATION
All participants signed their written consent for publication.

CONFLICT OF INTEREST
The authors declare that they have no competing interests. F I G U R E 3 Stratification analyses of clinical features for the association between rs754466 and HDM-induced AR risk in the additive model. A logistic regression model with adjustments for age and sex was constructed to calculate adjusted odds ratios (ORs), and their 95% confidence intervals (CIs). P adj , p value adjusted for age and sex