MIR17HG polymorphisms contribute to high-altitude pulmonary edema susceptibility in the Chinese population

High-altitude pulmonary edema (HAPE) is a common acute altitude sickness. This study was designed to investigate the effect of MIR17HG polymorphisms on HAPE risk in the Chinese population. The Agena MassARRAY platform was used to genotype six single-nucleotide polymorphisms (SNPs) in the MIR17HG gene in 244 HAPE patients and 243 non-HAPE controls. The odds ratio (OR) and 95% confidence interval were used to evaluate the association between each MIR17HG polymorphisms and the risk of HAPE under a polygenetic model. Statistical analysis was performed using the χ2 test. Multifactor dimensionality reduction (MDR) analysis was used to analyze the impacts of SNP–SNP interactions on the risk of HAPE. According to the allele model, the HAPE risk of people with the rs7318578 A allele of MIR17HG was lower than that of people with the C allele (OR 0.74, p = 0.036).Logistic regression analysis of four models for all selected MIR17HG SNPs showed significant differences in the frequencies of rs7318578 (OR 0.74, p = 0.037) and rs17735387 (OR 1.51, p = 0.036) between cases and controls. The results of the sex stratification analysis showed that among males, rs17735387 in the MIR17HG gene is associated with an increased risk of HAPE. MDR analysis showed that the best combination model was a three-locus model incorporating rs72640334, rs7318578, and rs7336610. This study revealed the correlations between rs7318578 and rs17735387 on the MIR17HG gene and the risk of HAPE in the Chinese population, providing a theoretical basis for the early screening, prevention, and diagnosis of HAPE in high-risk populations.


Associations of genetic polymorphisms with HAPE risk.
shows the basic information of the six SNPs in the MIR17HG gene identified in this study. All SNPs were found to be at Hardy-Weinberg equilibrium (HWE) based on exact tests. In the allele model, we observed a lower risk of HAPE (odds ratio (OR) 0.74, 95% confidence interval (95% CI) 0.56-0.98, p = 0.036) associated with the A allele of MIR17HG SNP rs7318578 than with the C allele. Logistic regression analysis of the four models for all of the selected MIR17HG SNPs between cases and controls showed significant differences for rs7318578 and rs17735387 in the HAPE population. For rs7318578, under the log-additive model, we found that people carrying the A allele had a decreased risk of HAPE compared with C allele carriers, which with the A allele playing a protective role (OR 0.74, 95% CI 0.56-0.98, p = 0.037; Table 3). Individuals carrying the rs17735387 A allele had a 51% higher risk of HAPE than those with the G allele (OR 1.51, 95% CI 1.03-2.21, p = 0.036; Table 3).

Stratification analysis of the associations of SNPs with HAPE risk.
To further clarify the associations between MIR17HG polymorphisms and HAPE risk, we conducted age-and sex-stratified analyses. In the age stratification analysis, we found no association of any of the six polymorphic loci (rs75267932, rs72640334, rs7336610, rs7318578, rs17735387, and rs1428) in the MIR17HG gene with the risk of HAPE (Supplementary Table 2). The results of the sex stratification analysis showed that among men, individuals with the rs17735387 genotype AG had an increased risk of HAPE compared with GG individuals (OR 1.55, 95% CI 1.03-2.32, p = 0.036; Table 4). www.nature.com/scientificreports/ SNP-SNP interactions. As shown in Fig. 1A, B a dendrogram and a Fruchterman Reingold algorithm were used to describe the interactions among these SNPs. We evaluated all three-way to five-way models (Table 5). We chose a three-loci model as the best model considering the small sample size. The overall best model included rs72640334, rs7318578, and rs7336610, with a testing accuracy of 56.58%, a maximum crossvalidation consistency (CVC) of 10/10 and p value of < 0.001 obtained by the χ 2 test. Compared with the lowrisk genotype combination, the generated combination of high-risk genotypes was associated with an increased HAPE risk (OR 2.43, 95% CI: 1.61-3.68, p < 0.001; Table 5). The MDR approach uses entropy to evaluate the correlation between variable combinations and attribute interactions (IG) 17 . The strongest interaction effect was found between rs7318578 and rs7336610, with an IG value of 0.59%.
Haplotype analysis with the risk of HAPE. In addition, we conducted linkage disequilibrium (LD) and haplotype analysis of MIR17HG polymorphisms. The constructed LD map is shown in Fig. 2. The LD block consists of four SNPs, rs7336610, rs7318578, rs17735387 and rs1428, in the MIR17HG gene. The frequency distributions of the haplotypes in the case group and the control group are shown in Table 6. However, there was no significant association between haplotype and HAPE risk.    www.nature.com/scientificreports/

Discussion
HAPE is a very harmful disease to humans. This study explored the impact of the MIR17HG polymorphisms on the risk of HAPE in the Chinese population. In our study, we found that MIR17HG rs7318578 minimized the risk and that rs17735387 increased the risk. MIR17HG is a noncoding RNA that affects protein synthesis by regulating gene expression, and its sole function is to produce six miRNAs (miR-17, miR-18a, miR-19a, miR-20a, miR-19b-1, and miR-92-1) 18,19 . Yin et al. 20 showed that miR-17 is involved in the regulation of EGFR-TKI resistance and can be used as a predictive biomarker for NSCLC EGFR-TKI resistance. Li et al. 21 showed that the expression of miR-18a affects the occurrence of lung fibroblast fibrosis. In addition, Ren et al. 22 showed that the downregulation of miR-19a-3p expression is related to the reduction of sepsis-induced lung injury. These findings indicate that MIR17HG plays an important role in the occurrence and development of a variety of lung diseases. Multiple reports have revealed that MIR17HG polymorphisms are associated with various disease risks 23,24 , however, there is no report about the polymorphisms of MIR17HG and the risk of HAPE.. Our research revealed that people carrying the rs7318578 A allele had a reduced risk of HAPE compared with C allele carriers, and that individuals carrying the rs17735387 A allele had a 51% increased risk of HAPE compared with G-allele carriers. Meta-analysis also showed that rs7318578 is related to the reduction of HAPE risk. Our study further revealed that the MIR17HG polymorphisms are associated with the risk of HAPE. The incidence of HAPE differs between the sexes. The results of the sex stratification analysis showed that among males, individuals with the rs17735387 AG genotype had an increased risk of HAPE compared with GG individuals, which indicated that the rs17735387 polymorphism of the MIR17HG gene affects the risk of HAPE. This result has implications for the screening of biomarkers of HAPE risk in male populations.
Studies have confirmed that p53 and HIF-1α can inhibit the expression of MIR17HG 25 . In addition, studies have shown that VEGF-induced MIR17HG expression is involved in the regulation of angiogenesis 26 . Chamorro-Jorganes et al. 27 showed that the downstream effector of HIF-1α, VEGF, affects the occurrence of HAPE by regulating vascular permeability. However, there is no report on the association between MIR17HG and HAPE, and we hypothesize that there is a correlation between the two. The promoter of the miR-17-92 cluster host gene MIR17HG is highly conserved, and the promoter region has multiple conserved transcription factor binding sites 28 .SNPs might impact biological processes involved in the production or functional effects of the miR-17-92 cluster host gene. We found that MIR17HG rs7318578 minimized the risk of HAPE and that rs17735387 increased the risk. These results provide some empirical support for our hypothesis, however, to clarify the associations between MIR17HG and HAPE, we conducted an MDR analysis. MDR is a recently developed method for analyzing interactions. It can simultaneously detect and describe the effects of multifactor interactions on diseases and has been widely used in scientific research 29,30 . This study clarified the influences of MIR17HG polymorphism site interactions on the risk of HAPE through MDR. The results showed that rs7318578 and rs7336610 had a strong synergistic effect on HAPE sensitivity and that rs7336610 and rs1428 had a strong antagonistic effect on HAPE sensitivity (Fig. 1B). The best model combination for HAPE sensitivity prediction was rs72640334, rs7318578, and rs7336610. All of these results upport our hypothesis, which MIR17HG polymorphisms are related to HAPE.
This study has limitations. First, the sample size was small. Second, although the correlation between six SNPs in MIR17HG and the risk of HAPE in the Chinese population was explored in this study, the MIR17HG gene includes many other sites. In our future work, we will explore the correlations between other sites on MIR17HG and the risk of HAPE. Third, due to a lack of cell biology methods, we did not explore the molecular mechanisms underlying the influences of MIR17HG polymorphisms on the occurrence of HAPE. In the future, we will explore the specific role of MIR17HG in HAPE from these perspectives.
In conclusion, this study clarified that MIR17HG polymorphisms are associated with the risk of HAPE in the Chinese population, which lays a foundation for further research on the mechanism of MIR17HG action on HAPE and provides a scientific basis for early screening and prevention of HAPE in high-risk population.

Methods
Participants. Based on the case-control strategy for exploring polymorphisms in the MIR17HG gene, we recruited 487 participants from the Affiliated Hospital of Qinghai University. All participants lived in low-altitude areas below 2500 m and reached high-altitude areas above 3000 m due to work or travel. By examining the symptoms (dyspnea), signs (cyanosis at rest) and imaging results (such as X-ray radiography and computed tomography (CT) of the patient chest) for all participants, and adhering to with the diagnostic criteria of HAPE 31 , we found that 244 participants had HAPE, forming the case group, while 243 participants had no symptoms of HAPE, forming the control group. There were 225 males and 19 females in the case group, and 224 and 19 females in the control group.
Genotyping. To ensure the successful statistical analysis of genotypes in the research population, SNPs with a minor allele frequency of at least 5% in the Han Chinese from Beijing (CHB) population were initially screened using the Thousand Genomes Project database (http:// www. inter natio nalge nome. org/). Haploview software (version 4.2) (https:// www. broad insti tute. org/ haplo view/ downl oads# JAR) was then used to conduct LD haplotype analysis 32 . Subsequently, function prediction analysis (https:// www. ncbi. nlm. nih. gov/ snp/) of tag single nucleotide polymorphisms (tagSNP) was performed online with the dbSNP database and showed that the tagSNPs are introns that affect the activity of splice sites. Following further analysis, six SNPs in the MIR17HG gene (rs75267932, rs72640334, rs7336610, rs7318578, rs17735387 and rs1428) were selected and explored for their correlations with the risk of HAPE in the Chinese population. Agena MassARRAY Assay Design 3.0 software (https:// seqpw s1. agena bio. com/ Assay Desig nerSu ite. html) was used to design the primers (Supplementary Meta-analysis. This study obtained genotyping data of six SNPs (rs75267932, rs72640334, rs7336610, rs7318578, rs17735387, rs1428) on the MIR17HG of the control group from the 1000 Genome Projects (http:// www. inter natio nalge nome. org/). The genotyping data of the six SNPs in the case group was obtained through the interpolation procedure. The data obtained above was analyzed by STATA 11.0 software (Stata Corp., College Station, TX, USA) 34 .
Statistical analysis. The analysis of the general characteristics of the subjects, the chi-square test was used to compare the categorical variable between the case group and the control group. The t-test was used to compare the continuous variable (age) between the case and control groups. The goodness of fit chi-square test was used to calculate the HWE of each polymorphism in the control group to test the representativeness and randomness of the study population. The chi-square test was used to detect the distribution differences of genotype and allele frequency of each polymorphism between the case group and the control group. Logistic regression was used to calculate OR values and 95% CIs to evaluate the strength of the association between each genotype and HAPE by PLINK software. All statistical tests were bilateral, with a significance threshold of 0.05. In this study, the minimum sample size and proportions of the experimental group and the control group were determined by G * Power 3.1.9.2 software 35 . The analysis results indicated that the total sample size should be greater than 388 and that the sample size of each of the control group and case group should be greater than 194. The total sample size in this study was 487, with 243 subjects in the control group and 244 in the case group thus, the sample sizes of this study met the statistical requirements. In addition, the statistical power of this study to detect an effect at the specified level of significance was 99.98%. Multifactor dimensionality reduction (MDR) analysis was performed using MDR_3.0.2 software (http:// sourc eforge. net/ proje cts/ mdr/) to determine a highorder interaction model for the risk of HAPE 36 .
Ethics approval and consent to participate. This study was approved by the Ethics Committee of the Affiliated Hospital of Qinghai University and was conducted in compliance with the Declaration of Helsinki. The purpose of this study was well conveyed to all participants, and written informed consent was obtained from all participants prior to biological material collection in this study. All subsequent study analyses were conducted in accordance with the approved guidelines and regulations.
Consent for publication. Written informed consent was obtained from the patients for publication of this report.

Data availability
The datasets used or analyzed during the current study are available from the corresponding author on reasonable request.