GENOME-WIDE ASSOCIATION STUDY OF MITOCHONDRIAL DNA IN CHINESE MEN IDENTIFIES SEVEN NEW SUSCEPTIBILITY LOCI FOR HIGH-ALTITUDE PULMONARY OEDEMA

Background / Aim: High-altitude pulmonary oedema (HAPE), which normally occurs at altitudes in excess of 3,000 m, is a potentially fatal disease due to hypoxia. The role of mitochondrial genomes in determining an individual's susceptibility to HAPE has not been determined. However, a number of genetic polymorphisms have recently been found to be overrepresented in HAPE patients. Most published genome-wide association studies (GWASs) have investigated only a small number of top-ranking single-nucleotide polymorphisms (SNPs)/genes by overview of nuclear DNA and considered each of the identified SNPs/genes independently. Little research has been conducted on mitochondrial genomes in relapsing HAPE patients by GWASs. Methods: To identify biological pathways important to HAPE occurrence, we examined approximately 500,000 SNPs genome-wide from 10 unrelated cases of relapsing HAPE, and we compared the SNPS in these cases with those in the CHB population (45 controls) to find the association between genotypes and HAPE susceptibility among the mitochondrial function-related genes. We used the FUMA platform to expand those SNPs to selected candidate SNPs. Results: A total of 369 candidate SNPs, 4 lead SNPs, 4 genomic risk loci and 5 mapped genes were obtained. The 7 mapped genes were ADAMTS9-AS2, NEK1, CLCN3, C4orf27(HPF1), RP11-219J21.2, ANKRD26 and YME1L1. Conclusions: This study confirms the association of ADAMTS9-AS2, NEK1, CLCN3, C4orf27(HPF1), RP11-219J21.2, ANKRD26 and YME1L1 with HAPE, which may provide future targets for the treatment of this disease.


Introduction
High-altitude pulmonary oedema (HAPE) is a kind of pulmonary oedema that occurs primarily in the hypoxic environment at high altitude. HAPE occurs mostly among residents of low-lying areas who enter the plateau for the first time or when the inhabitants of the plateau enter the higher-altitude areas. The incidence rate is 0.4%~2%. Because HAPE has acute onset and rapid progress and causes considerable harm to the body, if the treatment is not timely, it can develop to coma or even death in a relatively short time, which seriously threatens life and health [1][2][3][4]. High-altitude pulmonary oedema has an obvious susceptibility tendency.
Previous studies have shown that there are significant individual differences in susceptibility to HAPE in the same high-altitude hypoxia environment [5][6]. Accumulated evidence has suggested that a large number of genetic factors are associated with genetic susceptibility to HAPE, including nitric oxide synthase 3 (NOS3), cytochrome b-245 (CYBA), angiotensin converting enzyme (ACE), surfactants A1 and A2, and hypoxiainducible factor-1 (HIF-1) [5][6][7][8]. The genetic analysis of these studies was based on an overview of nuclear DNA. However, the role of mitochondria and their genomes is an area of genetic investigation that has been neglected.
Mitochondria are organelles that produce energy in aerobic cells and contain their own genome. Maintaining a sufficient quantity of mitochondrial DNA (mtDNA) in specific tissues is essential for cell viability. Therefore, many common human diseases, such as cancer[9, 10], cardiomyopathy [11] and liver disease [12], are associated with changing mtDNA levels. In a previous study, we sequenced the mtDNA of Ochotona curzoniae (Chinese red pika) and identified 15 novel mtDNA-encoded amino acid changes, including 3 in the subunits of cytochrome c oxidase. These amino acid substitutions may modulate mitochondrial complexes and electron transport efficiency during cold weather conditions and hypoxia adaptation [7]. In another study, we found that the sperm mtDNA copy number for those living at high altitude (5,300 m) for one month was significantly higher than for those at the lower altitude (1,400 m) or in donors who had been living at the 5,300-m altitude for 1 year [13]. However, the association between mitochondria and HAPE occurrence has not been determined.
In addition, with the emergence of genome-wide linkage disequilibrium (LD)-based marker panels and improvements in high-throughput genotyping technology, genome-wide association studies (GWAS) have become feasible [14]. GWAS can systematically survey the whole genome for causal genetic variants for complex traits/diseases and is a powerful tool for dissecting the genetic basis for HAPE. Combining the modest association signals in the GWAS data with information on biological pathways and networks, the emerging pathway-based approaches can be designed to utilize the GWAS data to a greater extent and are likely to yield new insights into HAPE aetiology.
To identify the important aetiology mechanism of HAPE occurrence more systematically and comprehensively, we used a novel pathway-based GWAS to approximately 871166 SNPs from 10 unrelated re-occurrence HAPE, which is different from other studies based on GWAS [15]. Those studies chose patients occurring for only one time, which cannot demonstrate that these patients have HAPE susceptibility compared with the data of CHB (Chinese in Beijing, China). Although these patients did not go to high-altitude areas, the incidence rate of HAPE is too low (0.4%~2%) to affect CHB as a control group; therefore, we investigated the association between mtDNA function-related genes and HAPE susceptibility.

Patients and controls
Relapsing HAPE patients (n=10) were recruited from the Han ethnic group in China.
We compared the allele frequency of HAPEs with the CHB (Chinese in Beijing, China) population (control=45) to exclude 185646 SNPs with minimum allele frequency (MAF) <0.01. The SNPs with the last successful assay were 673843. The recurrent HAPE patients consisted of 10 individuals (25.01±10.70 years old) who had at least two episodes of HAPE, as determined by the standard diagnostic criteria[16], including cough and dyspnea at rest, with pulmonary rales, cyanosis, and patchy shadows detected using chest X-ray.

Results
In the GWAS, we genotyped a total of 871,166 SNPs, and 673,843 SNPs were successfully genotyped (77.35%). We ranked genotyped SNPs based on the strength of association using the allelic association test. Nominally significant results were detected for 1558 SNPs (p<5×10-8) (Supplementary Table 1). This analysis indicates that HAPE cases are genetically similar to the combined CHB population. HapMap populations provide context for the patterns of variation observed among these populations. Genotyping data yielded an average call rate of 96.6%, and apparent inheritance errors in trio samples were detected in <0.2% of all SNPs. A Manhattan plot was generated for the SNPs in patients with recurrent HAPE in Figure Ι. A quantile-quantile (QQ) plot for association results is provided in Figure Ⅱ for all SNPs. The group of SNPs that slightly deviated from a diagonal straight line in the QQ plot are considered to reflect SNPs with weak genetic effects, and from the plot, it seems that there is not gross inflation of false-positive results derived from genotyping errors.
We used the FUMA platform to expand those of SNP p <5×10 -8 to SNPs that included their linkage disequilibrium (r 2 ≥ 0.6). After the data were imported into FUMA, we chose the East Asian population (EAS, consistent with the GWAS population), selected the SNP minimum allele frequency (MAF≥ 0.01) and r 2 (minimum r 2 ≥ 0.6). A total of 369 candidate SNPs (Supplementary Table 2), 4 lead SNPs, 4 genomic risk loci and 5 mapped genes were obtained. The 7 mapped genes were ADAMTS9-AS2, NEK1, CLCN3, C4orf27(HPF1), RP11-219J21.2, ANKRD26 and YME1L1 (Table 1).  Red plots are the cases for all loci, and blue plots are the cases after removing the significant locus.

Acknowledgements
Not applicable.

Funding
We are grateful to all the people who participated in this study. This work was supported

Authors' contributions
Yongjun Luo participated in the design of the present study and performed the statistical analysis. Caizhi Tang, Yu Chen and Xinyuan Liu conducted the study and analyses and collected patient information. We also appreciate assistance in data analysis from Dr.

Liyuchun in State Key Laboratory of Genetic Resources and Evolution, Kunming Institute
of Zoology, Chinese Academy of Sciences, Kunming, China.

Ethics approval and consent to participate
Not applicable.

Patient consent for publication
Not applicable.