Association between EPAS1 and ATP6V1E2 polymorphisms and susceptibility to high altitude polycythemia in Chinese Tibetan population

High altitude polycythemia (HAPC) is an important public health problem at high altitude, and genetic factors play a key role in hypoxia adaptation in Tibetan populations. The aim of this study was to investigate the association between EPAS1 and ATP6V1E2 gene locus polymorphisms and genetic susceptibility to HAPC in Chinese Tibetan population. This study included 78 HAPC patients and 85 healthy controls and genotyped the EPAS1 gene single nucleotide polymorphism loci (rs1868092, rs4953396, and rs4953354) and ATP6V1E2 rs896210. We analysed the association between EPAS1 and ATP6V1E2 genes and HAPC using logistic regression analysis Multifactorial dimensionality reduction, protein–protein interaction and KEGG pathway. This study found that ATP6V1E2 rs896210 and EPAS1 rs1868092, rs4953396, rs4953354 were significantly associated with genetic susceptibility to HAPC in the Chinese Tibetan population, and synergistic effects existed among these genetic loci. This provides new evidence for the genetic mechanism of high altitude adapted diseases in Tibetan populations, which is valuable for individualized risk assessment and exploration of potential therapeutic targets for HAPC at high altitude.

1 Introduction

Globally, more than 140 million people live at altitudes above 2,500 m, mainly in the Tibetan Plateau, the Andes and the Ethiopian Plateau (1). When humans live at high altitude, they may develop high altitude polycythemia (HAPC) due to overcompensated proliferation of red blood cells caused by hypoxia, which is one of the typical chronic alpine diseases and has been a serious public problem at high altitudes (2). HAPC leads to a significant increase in blood viscosity, which results in microcirculatory damage and immune response disturbances such as vascular thrombosis, widespread organ damage and sleep disorders (3, 4). Central to the development of HAPC is the over compensatory proliferation of erythrocytes, and erythropoiesis is largely dependent on erythropoietin (EPO), a glycoprotein hormone regulated by hypoxia-inducible factor (HIF) (5). It is noteworthy that the risk of developing HAPC varies significantly among different indigenous high-altitude populations. For instance, compared to residents of the South American Andes, the indigenous Tibetan population of the Qinghai-Tibet Plateau exhibits superior hypoxic adaptation, with a relatively lower incidence of HAPC (6–8). This disparity suggests that different populations may have evolved distinct genetic adaptation mechanisms to high-altitude environments. It has been shown that EPAS1, ATP6V1E2 and other genes regulate the HIF pathway, and mutations in these loci have been shown to correlate with genetic susceptibility to HAPC in Tibetan populations (9–11).

Yi X et al. found population-specific variations in allele frequencies of several genes by whole exome sequencing of 50 Tibetan individuals, with the endothelial PAS structural domain protein 1 (EPAS1) gene showing the strongest signal of natural selection (12). The gene is located on chromosome 2p16-21 and is expressed mainly in tissues and organs involved in metabolism and oxygen supply, such as the placenta, vascular endothelium, and kidney (13–15). Previous studies have demonstrated that multiple single nucleotide polymorphisms (SNPs) in the EPAS1 gene—including rs1868092, rs4953396, and rs4953354—are significantly associated with hypoxic adaptation in the Tibetan population. However, the mechanisms underlying their association with high-altitude hypoxic anemia (HAPC) require further elucidation (16, 17).

ATP6V1E2 (also known as ATP6E1 or VMA4) is located on chromosome 2p21,and encodes the E subunit of the V-ATPase complex, which is a proton pump that regulates intracellular acid–base homeostasis (18). There are fewer direct studies on ATP6V1E2 and hypoxia, but several lines of evidence suggest that this gene may be involved in high altitude hypoxia acclimatization. Whole-exome sequencing of the Tibetan population revealed significant racial differences in allele frequencies of ATP6V1E2 (12),and significant correlations between the ATP6V1E2 locus and levels of red blood cell counts (RBC), hemoglobin (HGB), and hematocrit (HTC), which are important features of hypoxic acclimatization in high altitude populations (9). Notably, ATP6V1E2 is located in close proximity to the EPAS1 gene, which is a known key gene for high altitude acclimatization, suggesting a possible synergistic regulatory relationship.

Genetic studies of Tibetans at high altitude in China have shown that the EPAS1 and ATP6V1E2 genes exhibit significant adaptive evolution under the selective pressure of prolonged high altitude. Although existing studies have confirmed the association of polymorphisms in these two genes with the risk of HAPC in Tibetan populations, the relevant loci are still understudied. In this study, we analyzed the genetic polymorphisms of EPAS1 rs1868092, rs4953396, rs4953354 and ATP6V1E2 rs896210 in the Chinese Tibetan population and their associations with the susceptibility to HAPC, which will provide more theoretical basis for early screening and individualized prevention and treatment of HAPC in the Tibetan population in this region.

2 Materials and methods

2.1 Study subjects

A total of 163 subjects who visited Tibet Autonomous Region People’s Hospital participated in this study. Samples data were accessed between 01/01/2018 and 31/12/2022 for the purpose of our study. Authors who were involved in recruitment, screening and conducting the experiments had access to information that could identify individual participants during data collection. This study included 78 patients with HAPC in the case group and 85 healthy individuals in the control group. The criteria for inclusion in the case group were (19): (1) Hb ≥ 21 g/dl for men and ≥ 19 g/dl for women; (2) Long-term residence in a high altitude area at an altitude of more than 3,000 m; (3) Three or more of the following symptoms: headache, dizziness, fatigue, cyanosis, sleep disturbance, conjunctival congestion, and purplish skin; (4) The study population excludes true cytokinesis as well as other secondary erythropoiesis. None of the study subjects had cardiovascular diseases, autoimmune diseases, malignant tumors, immune system diseases, or neurological diseases. Clinical data related to the two groups were also collected: gender, age, RBC, HGB, and HCT. All subjects are the Tibetan population in Lhasa, Tibet, China. The study was conducted in accordance with the principles of the Declaration of Helsinki and approved by the Research Ethics Committee of the First Affiliated Hospital of Dali University (Approval No. DFY20171210002, Date: December 10, 2017) and all participants provided written informed consent. The flowchart of this study is shown in Figure 1.

Figure 1

Figure 1. Flow chart of this study.

2.2 Experimental methods

2.2.1 Reagents and instruments

DNA extraction kit (QIAGEN, lot 166,034,547); PCR primers (Life); Taq enzyme (Fermentas); qPCR premix (General Biologicals, lot FP208); PCR 96-well plate (Axygen, PCR-96-FLT-C); real-time PCR (Axygen, lot 166,034,547); and a PCR plate (Axygen, lot 166,034,547). PCR (BioRad, model: CFX96); ND5000 ultra-micro spectrophotometer (Thermo, model: NanoDrop 5,000); electrophoresis instrument (Major Science, model: Mini Pro 300 V Power Supply).

2.2.2 Selection of SNP

The screening of SNPs in this study was based on reference information from the National Center for Biotechnology Information (NCBI, https://www.ncbi.nlm.nih.gov/) database. We focused on the following loci selected for analysis: ATP6V1E2 rs896210 (A/G) and EPAS1 rs1868092 (A/G), rs4953396 (A/C), rs4953354 (A/G). It should be noted in particular that although the gene annotation information for the rs4953396 locus in the NCBI database is incomplete, it has been confirmed that this locus is located in the EPAS1 gene region (20, 21), and therefore we included it in the locus analysis of the EPAS1 gene.

2.2.3 DNA extraction and quality control

Collect 2 mL of fasting venous blood from subjects into EDTA anticoagulant tubes and store at −20 °C for later use. Extract genomic DNA using a DNA extraction kit, then assess integrity via agarose gel electrophoresis and determine concentration and purity using a micro-spectrophotometer. Genotyping samples must meet quality control standards: concentration ≥20 ng/μL, A₂₆₀/A₂₈₀ ratio 1.8–2.0. The measured DNA concentrations in this study ranged from 0.9 to 1171.0 ng/μL, with A₂₆₀/A₂₈₀ ratios between 0.68 and 2.36. All samples ultimately included met the criteria, with ratios concentrated between 1.7 and 1.9. A few samples initially deviating from the criteria were included after purification and retesting.

2.2.4 SNP detection

PCR amplification was first performed and the primer probe sequences for EPAS1 rs1868092, rs4953396, rs4953354 and ATP6V1E2 rs896210 are shown in Table 1. The reaction conditions were as follows: pre-denaturation at 95 °C for 5 min, 40 cycles (denaturation at 95 °C for 10 s, annealing at 60 °C for 30 s, extension at 72 °C for 2 min), and finally extension at 16 °C for 5 min. Then TaqMan fluorescent probe technology was used for genotyping polymorphic loci, and different fluorescence signals were detected by real-time fluorescence quantitative PCR.

Table 1

Table 1. Primer probe sequences.

2.3 Statistical analysis

Statistical analysis was performed using SPSS 25.0 software (p < 0.05 statistically different). In the analysis of baseline information, numerical variables were compared between groups using the t-test, described as (x² ± s), and categorical variables were compared using the X² test, expressed as frequencies and percentages. To evaluate the association between EPAS1 and ATP6V1E2 gene polymorphisms and the risk of developing HAPC, this study constructed five genetic models (codominant, dominant, recessive, superdominant, and allele models). Multivariate logistic regression analysis was performed, adjusting for age and sex to control for potential confounding factors. The Hardy–Weinberg equilibrium (HWE) test was used to assess the Population genetics of control group and if p > 0.05, it indicated that the data in this study followed the population genetics pattern and was representative of the population.

2.4 Other analyses

Linkage disequilibrium determination and haplotype analysis using online SHEsis¹ to analyze the association between genetic haplotypes and the risk of developing (22). The interactions between the studied loci were analyzed using the multifactor downscaling software MDR 3.0.2 to calculate the cross-validation consistency (CVC) and testing accuracy of each model. A network map of EPAS1 and ATP6V1E2 protein-protein interaction (PPI) was constructed using the STRING website² and Cytoscape software, followed by KEGG pathway analysis using the clusterProfiler package in R software (version 4.4.1) to further explore EPAS1 and ATP6V1E2 biological functions. Given the sample size of this study, to assess the reliability of variables that failed to reach statistical significance in the logistic regression analysis, we conducted a post-hoc power analysis. This analysis was performed using the pwr package in R software (version 4.4.1).

3 Results

3.1 Clinical characteristics of study subjects

The clinical characteristics of these participants are shown in Table 2. The mean age of the control and HAPC patients was 45.76 ± 18.14 and 48.41 ± 14.67 years, respectively, with no significant difference between the groups (p = 0.306). The percentage of males in HAPC patients was 76%, which was significantly higher than the percentage of males in the control group (38%), with a significant difference between the groups (p < 0.001). In addition, the differences in RBC, HGB and HCT were significantly different between the two groups (p < 0.001).

Table 2

Table 2. Analysis of the basic clinical characteristics of the study population.

3.2 Hardy Weinberg equilibrium (HWE) test analysis

The HWE-P for ATP6V1E2 rs896210 and EPAS1 rs1868092, rs4953396, and rs4953354 in both the control and case groups was greater than 0.05, indicating that the gene frequencies observed in this study population were representative of the gene distributions observed in the general population, as shown in Table 3.

Table 3

Table 3. HWE balance of gene loci.

3.3 Association analysis of EPAS1 and ATP6V1E2 gene locus polymorphisms with HAPC susceptibility

We assessed the association of SNPs in the ATP6V1E2 and EPAS1 genes with HAPC by constructing multiple genetic models, including codominant, dominant, recessive, and overdominant models, followed by logistic regression analysis. The final results showed ATP6V1E2 rs896210 and EPAS1 rs1868092, rs4953396, rs4953354 were all significantly correlated with HAPC. Genotyping and alleles in the control and HAPC groups are shown in Table 4, and the percentage of genotypes in each gene model is shown in Figure 2.

Table 4

Table 4. Association analysis of gene locus polymorphisms with the risk of developing HAPC.

Figure 2

Figure 2. Horizontal stacked bar chart of genetic models; * for HAPC group.

The distribution of the ATP6V1E2 rs896210 genotype is as follows: in the control group, GG accounted for 58.8%, AG for 36.5%, and AA for 4.7%; whereas in the HAPC group, GG accounted for 48.7%, AG for 39.7%, and AA for 11.5%. Compared with allele G, allele A [OR = 1.748 (1.032–2.959), p = 0.038] was associated with an increased risk of HAPC. In the codominant model, the AA genotype [OR = 4.686 (1.160–18.935), p = 0.030] significantly increased HAPC risk compared to the GG genotype. In the recessive model, individuals carrying the AA genotype [OR = 4.194 (1.073–16.398), p = 0.039] also exhibited increased susceptibility to HAPC.

The genotype distribution of EPAS1 rs1868092 was as follows: in the control group, 69.4% AA, 28.2% AG, and 2.4% GG, whereas in the HAPC group, 46.2% AA, 42.3% AG, and 11.5% GG. Using allele A as a reference, allele G [OR = 2.554 (1.429–4.564), p = 0.002] was associated with the risk of developing HAPC. In a codominant model with the AA genotype as reference, the GG genotype increases the risk of HAPC [OR = 10.070 (1.765–57.451), p = 0.009]. Individuals carrying the G genotype (AG+GG) [OR = 2.551 (1.269–5.129), p = 0.009] in the dominant model and the GG [OR = 7.858 (1.397–44.194), p = 0.019] genotype in the recessive model were both more likely to develop HAPC.

The genotypes of EPAS1 rs4953396 were distributed as follows: in the control group, 56.5% were AA, 38.8% were AC, and 4.7% were CC, whereas in the HAPC group, 41.0% were AA, 42.3% were AG, and 16.7% were GG. Using allele A as a reference, allele C [OR = 2.069 (1.261–3.395), p = 0.004] was associated with an increased risk of developing HAPC. The CC [OR = 6.400 (1.689–24.246), p = 0.006] genotype in the codominant model could be significantly associated with an increased risk of HAPC compared to the AA genotype. Individuals carrying the CC [OR = 5.427 (1.490–19.760), p = 0.010] genotype in the recessive model were both more likely to develop HAPC.

The genotype distribution of EPAS1 rs4953354 was as follows: in the control group, GG accounted for 77.6%, AG for 18.8%, and AA for 3.5%, whereas GG accounted for 43.6%, AG for 43.6%, and AA for 12.8% in the HAPC group. The allele A [OR = 3.806 (2.003–7.231), p < 0.001]for this SNP, AG [OR = 9.137 (2.075–40.222) and AA [OR = 3.648 (1.652–8.056), p = 0.001] genotypes in the codominant model, and AG+AA in the dominant model [OR = 4.380 (2.085–9.197), p < 0.001]genotype, AA [OR = 2.852 (1.327–6.130), p = 0.007]genotype in the recessive model, and AG [OR = 6.136 (1.418–26.550), p = 0.015] genotype in the overdominant model significantly increased HAPC susceptibility.

3.4 Haplotype and linkage disequilibrium analysis of EPAS1 and ATP6V1E2 gene SNPs with HAPC

We performed haplotype analysis of these four SNPs using online SHEsis, with the SNPs in the order of rs896210, rs1868092, rs4953396, and rs4953354. After removing the haplotypes with frequencies lower than 3%, a total of six haplotypes were obtained, GAAG, AGCA, GGCA, GAAA, AACG, and AGCG (in descending order of frequency in the HAPC group), as shown in Table 5. Among them, GAAG (OR = 0.437 [0.268–0.715], p < 0.001), AGCA (OR = 2.871 [1.485–5.552], p = 0.001), and GGCA (OR = 9.973 [1.713–58.064], p = 0.012) differed significantly between the HAPC and control groups. GAAG appeared significantly less frequently in the HAPC group than in the control group, so the risk of HAPC was significantly lower in GAAG haplotype carriers, whereas the risk of HAPC would be significantly higher in AGCA and GGCA haplotype carriers. In addition, a high degree of linkage disequilibrium between rs896210 and rs4953396 can be seen in Figure 3 (D′ = 0.98, r² = 0.80).

Table 5

Table 5. Haplotype analysis of gene locus polymorphisms and risk of HAPC occurrence.

Figure 3

Figure 3. Linkage disequilibrium (LD) of 5 SNPs. (A) The numbers inside the diamonds indicate the D’ for pairwise analyses. (B) The numbers inside the diamonds indicate the r² for pairwise analyses.

3.5 SNP-SNP interaction analysis

We performed MDR analysis on these four SNPs, to explore the interactions between these loci and their associations with HAPC, and the optimal models obtained from 1st-3rd order interactions are shown in Figure 4 and Table 6. The association between the interactions among these SNPs and the risk of HAPC prevalence is statistically significant (p < 0.01). The rs4953354 model (testing accuracy: 0.6703, OR = 4.4954, CVC: 10/10) showed that risk allele A at the rs4953354 locus elevates the risk of developing HAPC up to 4.4954-fold. The rs4953396, rs4953354 model (testing accuracy: 0.6772, OR = 4.9934, CVC: 9/10) suggests that when an individual carries the risk allele for both loci, the risk of developing HAPC is 4.9934 times higher than that of an individual who does not carry the allele. The rs896210, rs1868092, rs4953354 model (testing accuracy: 0.6329, OR = 6.032, CVC: 7/10) had moderate test accuracy, there was a high stability (CVC = 7/10) and a strong effect size, which still accounted for the risk of developing HAPC when the risk alleles at the 3 loci coexistedis elevated by a factor of 6.0392. The results showed that the interaction of these SNPs significantly increased the risk of developing HAPC, with rs4953354 being the core risk locus. Figure 5 shows the obtained dendrogram, which shows that redundant effects of rs4953354, rs1868092, and rs4953396 in regulating HAPC risk, and both synergistic and redundant effects between rs896210 and these three loci. And rs896210 was more strongly associated with rs4953354 than the other two loci.

Figure 4

Figure 4. Cell diagram of the optimal model. (A) rs4953354 model; (B) rs4953396, rs4953354 model; (C) rs1868092, rs4953396, rs4953354 model. Black bars on the left side indicate the case group, black bars on the right side indicate the control group, one cell represents one interaction combination, light gray cells indicate that the ratio of the combination does not exceed the ratio threshold and is low-risk, dark gray indicates that the ratio threshold is exceeded and is high-risk, and white indicates that there is no data on the combination.

Table 6

Table 6. Multi-factor dimensionality reduction analysis.

Figure 5

Figure 5. Dendrogram of SNP-SNP interactions. The blue line indicates that the SNPs have a redundancy effect in regulating the risk of HAPC and the green line represents the intermediate point between synergistic and redundancy effects. The closer the loci are, the stronger the interactions are.

3.6 Functional association of EPAS1 and ATP6V1E2 in HAPC

To further explore the potential mechanisms of EPAS1 and ATP6V1E2 in regulating HAPC, we first constructed EPAS1-ATP6V1E2 PPI to analyze the direct and indirect molecular associations between the two, and then revealed the biological pathways in which they are jointly involved by KEGG pathway analysis, and Figure 6 shows the final results obtained. The PPI plot (Figure 6A) shows that there is an interaction between EPAS1 and ATP6V1E2. They are both directly linked and indirectly functionally associated through some intermediary molecules (e.g., EGLN1, HIF3α). KEGG pathway analysis (Figures 6A,B) also showed that both EPAS1 and ATP6V1E2 and their related proteins were involved in the HIF-1 signaling pathway, in addition to EPAS1 being involved in signaling pathways related to tumors such as renal, prostate and colorectal cancers, whereas ATP6V1E2 was involved in the mTOR signaling pathway, oxidative phosphorylation, and collector tubular acid secretion pathways.

Figure 6

Figure 6. Biological functions of EPAS1 and ATP6V1E2. (A) Protein interaction map of EPAS1. (B) KEGG passage results of proteins related to EPAS1. (C) KEGG passage results of proteins related to ATP6V1E2.

4 Discussion

At present, HAPC remains a serious threat to the health of highlanders. In China, its main treatments include phlebotomy and therapeutic erythrocyte dialysis, and there is still a lack of effective means of curing the disease (23). Several studies have shown that Chinese Tibetans have a lower incidence of HAPC compared to other highlanders and lowlanders living at the same altitude, a phenomenon closely related to genetic factors (8, 17, 24). Therefore, an in-depth study of HAPC from a genetic perspective is important for the prediction, diagnosis, treatment, and prevention of this disease.

Low-pressure hypoxia due to altitude is the main cause of HAPC. Although the pathogenesis of HAPC remains unclear, the widely accepted hypothesis is that HIF-1 upregulates EPO secretion during exposure to hypoxia at high altitude, leading to increased erythrocyte (25). HIF-1 belongs to the PAS family of hypoxia-regulated transcription factors and consists of an oxygen-sensitive α-subunit (HIF-α) and a constitutively expressed β-subunit (HIF-β, also known as ARNT) (26). Under normoxic conditions, prolyl hydroxylases (PHDs) hydroxylate key proline residues of HIF-1α and HIF-2α using oxygen molecules and α-ketoglutarate as substrates (27). Hydroxylated HIF-α is recognized by von Hippel–Lindau to form the E3 ubiquitin ligase complex, which is then degraded via the proteasome pathway (28, 29). In contrast, under hypoxic conditions, this oxygen-dependent pathway ceases, resulting in the intracellular stabilization and accumulation of HIF-1α and HIF-2α and the formation of a heterodimer with ARNT, which then binds to the hypoxia-responsive element, locating within the regulatory element of the HIF target gene (30, 31). HRE activates the expression of hypoxia-related genes, such as target genes of EPO, vascular endothelial growth factor, and lactate dehydrogenase A (18). In particular, upregulation of EPO production promotes increased erythropoiesis, a molecular mechanism that links tissue hypoxia to a compensatory erythropoietic response. It has been found that treatment of HAPC rats with the HIF-2α inhibitor PT2385 resulted in significant reductions in the levels of EPO, HGB, RBC, and HCT compared with the untreated group (32). This result suggests that PT2385 may directly regulate red lineage hematopoiesis by inhibiting the HIF-2α-EPO pathway, thereby reducing the abnormal elevation of HGB and RBCs (32).

Oxygen-dependent regulation of HIF-2α is a central mechanism for cellular adaptation to chronic hypoxia, and it plays a key role in the transcriptional regulation of EPO (33). HIF-2α is encoded by EPAS1 gene, whose function is critical for hypoxic adaptation. Our functional studies of EPAS1 gene also indicate that EPAS1, as a core transcriptional regulator of the hypoxic response, plays a key role in hypoxic adaptation by interacting with proteins such as ARNT, HIF-1α, and TP53, and by integrating regulatory mechanisms such as hypoxic signaling (HIF-1) and other mediators. Specific mutations in EPAS1 (e.g., Tibetan-adapted mutations) reduce the incidence of HAPC, whereas certain mutations may increase HAPC susceptibility. Using population genetic analysis, we found that polymorphisms at EPAS1 rs1868092, rs4953396 and rs4953354 were significantly associated with HAPC in the Chinese Tibetan population. Among them, rs1868092-G, rs4953396-C and rs4953354-A were risk alleles for HAPC, and genetic modeling analysis showed that the mutant genotypes of these three SNPs significantly increased the susceptibility to HAPC in codominant, dominant and recessive models. In addition, the rs4953354 locus showed a super dominant effect. This finding is consistent with previous conclusions, both of which indicate that there is a significant correlation between EPAS1 gene polymorphisms and susceptibility to HAPC (34, 35).

This study identified a significant association between the ATP6V1E2 rs896210 polymorphism and the risk of HAPC. This gene encodes a key subunit of the V-ATPase, which is essential for maintaining lysosomal acidification. An unexpected study on Hif-2a conditional knockout mice provided key insights into its function: in this mouse model, downregulation of ATP6V1E2 expression directly led to impaired lysosomal acidification (36). This suggests that one potential mechanism of ATP6V1E2 in high-altitude adaptation may be: its variants influence lysosomal acidification and function (e.g., protein degradation, autophagy) by regulating V-ATPase activity, thereby indirectly modulating the stability and turnover of hypoxia-responsive proteins like HIF-α. Ultimately, this forms a synergistic regulatory network with EPAS1. Concurrently, this study confirmed through KEGG pathway analysis that both EPAS1 and ATP6V1E2 are enriched in the HIF-1 signaling pathway, indicating that they synergistically participate in the development of HAPC via this critical hypoxia response pathway. Although direct functional studies of ATP6V1E2 remain limited, its significance has been validated through multiple lines of evidence: not only is it located adjacent to the EPAS1 gene and shares a selective signal (9), but this study further directly confirmed their presence within a tightly integrated functional network through PPI network analysis. Collectively, these findings establish ATP6V1E2 as a key player in the hypoxic adaptation regulatory network.

This study also revealed the synergistic contribution of ATP6V1E2 and EPAS1 loci to Tibetan high-altitude adaptation through haplotype and linkage disequilibrium analysis. We identified two functionally distinct haplotypes: AGCA and GGCA as risk haplotypes for HAPC, while GAAG exhibits protective effects. Further linkage analysis confirmed strong linkage disequilibrium between rs896210 and rs4953396. Crucially, multiple interaction model analysis demonstrated robust statistical interactions among the four SNP loci, collectively determining HAPC risk. This suggests polygenic interactions are more critical than single-gene variants. This finding aligns with genomic evidence implicating polygenic regions (Genes such as ATP6V1E2, TMEM247, RHOQ, PIGF, and CRIPT) in high-altitude adaptation (9, 37), offering new insights into the functional role of this gene cluster. Consequently, future research may explore genetic testing for screening high-risk populations and investigate potential drug targets based on these genetic variations, providing novel strategies for HAPC prevention and precision treatment.

However, this study has several limitations. First, the subjects were exclusively Tibetan individuals from the Lhasa region of Tibet, potentially limiting the generalizability of the findings. Future studies should include populations from different geographic regions and ethnic groups to validate broader applicability. Second, the limited sample size, but post-hoc power analysis indicates sufficient detection capability for strong genetic effects (OR > 4.0), with some loci exhibiting >90% power, ensuring the reliability of major positive findings. However, the power for detecting weak effects (OR < 1.5) was low (most < 20%), requiring cautious interpretation of negative results. Future studies should expand sample sizes to enhance statistical power for detecting subtle genetic effects. Finally, this study focused solely on limited sites within the ATP6V1E2 and EPAS1 genes, which to some extent limited the comprehensive assessment of polygenic interactions. Future research should conduct deep saturation mutation screening of these two genes and subsequently extend the analysis to other members of the EPAS1 gene cluster to elucidate the polygenic inheritance architecture of HAPC at the systems level.

5 Conclusion

This study found that specific SNPs at the ATP6V1E2 and EPAS1 loci are not only independently associated with HAPC susceptibility but also exhibit significant genetic interactions, thereby providing new evidence for the polygenic interaction hypothesis. More importantly, establishing risk assessment models based on these SNPs holds promise for providing a theoretical basis for early screening and personalized prevention and treatment of HAPC among the Tibetan population in this region.

Data availability statement

The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding authors.

Ethics statement

The studies involving humans were approved by the Research Ethics Committee of the First Affiliated Hospital of Dali University (Approval No. DFY20171210002, Date: December 10, 2017). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

LR: Conceptualization, Data curation, Investigation, Methodology, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing. YL: Methodology, Software, Visualization, Writing – original draft. DL: Data curation, Methodology, Validation, Writing – original draft. ZC: Methodology, Validation, Visualization, Writing – review & editing. GW: Conceptualization, Methodology, Project administration, Supervision, Visualization, Writing – review & editing. YZ: Methodology, Project administration, Resources, Supervision, Writing – review & editing.

Funding

The author(s) declared that financial support was not received for this work and/or its publication.

Acknowledgments

The authors would like to express their sincere gratitude to The First Affiliated Hospital of Dali University.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

1 ^http://analysis.bio-x.cn/myAnalysis.php

2 ^https://cn.string-db.org/

References

1. López, O, Arce, S, Chávez, R, Alaniz, A, Lancellotti, D, Chiapella, M, et al. Spirometry reference values for an Andean high-altitude population. Respir Physiol Neurobiol. (2018) 247:133–9. doi: 10.1016/j.resp.2017.09.016,

PubMed Abstract | Crossref Full Text | Google Scholar

2. Wang, Z, Liu, F, Ye, S, Jiang, P, Yu, X, Xu, J, et al. Plasma proteome profiling of high-altitude polycythemia using TMT-based quantitative proteomics approach. J Proteome. (2019) 194:60–9. doi: 10.1016/j.jprot.2018.12.031,

PubMed Abstract | Crossref Full Text | Google Scholar

3. Jiang, C, Liu, F, Luo, Y, Li, P, Chen, J, Xu, G, et al. Gene expression profiling of high altitude polycythemia in Han Chinese migrating to the Qinghai-Tibetan plateau. Mol Med Rep. (2012) 5:287–93. doi: 10.3892/mmr.2011.632,

PubMed Abstract | Crossref Full Text | Google Scholar

4. Guan, W, Ga, Q, Li, R, Bai, ZZ, Wuren, T, Wang, J, et al. Sleep disturbances in long-term immigrants with chronic mountain sickness: a comparison with healthy immigrants at high altitude. Respir Physiol Neurobiol. (2015) 206:4–10. doi: 10.1016/j.resp.2014.11.007,

PubMed Abstract | Crossref Full Text | Google Scholar

5. Rankin, E, Biju, M, Liu, Q, Unger, T, Rha, J, Johnson, R, et al. Hypoxia-inducible factor-2 (Hif-2) regulates hepatic erythropoietin in vivo. J Clin Invest. (2007) 117:1068–77. doi: 10.1172/JCI30117,

PubMed Abstract | Crossref Full Text | Google Scholar

6. Chen, S. Investigation on the incidence of high altitude polycythemia and its Hemoglobin characteristics in A Tibetan population. Zhongguo Yi Xue Ke Xue Yuan Xue Bao Acta Acad Med Sin. (1992) 14:237–43.

Google Scholar

7. Hancco, I, Bailly, S, Baillieul, S, Doutreleau, S, Germain, M, Pépin, JL, et al. Excessive erythrocytosis and chronic mountain sickness in dwellers of the highest city in the world. Front Physiol. (2020) 11. doi: 10.3389/fphys.2020.00773,

PubMed Abstract | Crossref Full Text | Google Scholar

8. de, A, Miranda, JJ, Gilman, R, Dávila-Román, V, León-Velarde, F, Rivera-Ch, M, et al. Prevalence, clinical profile, iron status, and subject-specific traits for excessive erythrocytosis in Andean adults living permanently at 3,825 meters above sea level. Chest. (2014) 146:1327–36. doi: 10.1378/chest.14-0298,

PubMed Abstract | Crossref Full Text | Google Scholar

9. Deng, L, Zhang, C, Yuan, K, Gao, Y, Pan, Y, Ge, X, et al. Prioritizing natural-selection signals from the deep-sequencing genomic data suggests multi-variant adaptation in Tibetan highlanders. Natl Sci Rev. (2019) 6:1201–22. doi: 10.1093/nsr/nwz108,

PubMed Abstract | Crossref Full Text | Google Scholar

10. Buroker, NE, Ning, XH, Zhou, ZN, Li, K, Cen, WJ, Wu, XF, et al. Epas1 and Egln1 associations with high altitude sickness in Han and Tibetan Chinese at the Qinghai-Tibetan plateau. Blood Cells Mol Dis. (2012) 49:67–73. doi: 10.1016/j.bcmd.2012.04.004,

PubMed Abstract | Crossref Full Text | Google Scholar

11. Zhao, Y, Zhang, Z, Liu, L, Zhang, Y, Fan, X, Ma, L, et al. Associations of high altitude polycythemia with polymorphisms in Epas1, Itga6 and Erbb4 in Chinese Han and Tibetan populations. Oncotarget. (2017) 8:86736. doi: 10.18632/oncotarget.21420,

PubMed Abstract | Crossref Full Text | Google Scholar

12. Yi, X, Liang, Y, Huerta-Sanchez, E, Jin, X, Cuo, Z, Pool, J, et al. Sequencing of fifty human exomes reveals adaptation to high altitude. Science (New York, NY). (2010) 329:75–8. doi: 10.1126/science.1190371,

PubMed Abstract | Crossref Full Text | Google Scholar

13. Karaghiannis, V, Maric, D, Garrec, C, Maaziz, N, Buffet, A, Schmitt, L, et al. Comprehensive in silico and functional studies for classification of Epas1/Hif2a genetic variants identified in patients with erythrocytosis. Haematologica. (2023) 108:1652–66. doi: 10.3324/haematol.2022.281698,

PubMed Abstract | Crossref Full Text | Google Scholar

14. Lee, F, and Percy, M. The Hif pathway and erythrocytosis. Annu Rev Pathol Mech Dis. (2011) 6:165–92. doi: 10.1146/annurev-pathol-011110-130321,

PubMed Abstract | Crossref Full Text | Google Scholar

15. Ja, S, Wang, L, and Heasley Le, HC-J. Hypoxia regulates alternative splicing of Hif and non-Hif target genes. Molecul Cancer Res. (2014) 12:1233–43. doi: 10.1158/1541-7786.MCR-14-0149,

PubMed Abstract | Crossref Full Text | Google Scholar

16. Li, C, Li, X, Xiao, J, Liu, J, Fan, X, Fan, F, et al. Genetic changes in the Epas1 gene between Tibetan and Han ethnic groups and adaptation to the plateau hypoxic environment. Peerj. (2019) 7:E7943. doi: 10.7717/peerj.7943,

PubMed Abstract | Crossref Full Text | Google Scholar

17. Beall, C, Cavalleri, G, Deng, L, Elston, R, Gao, Y, Knight, J, et al. Natural selection on Epas1 (Hif2α) associated with low hemoglobin concentration in Tibetan highlanders. Proc Natl Acad Sci USA. (2010) 107:11459. doi: 10.1073/pnas.1002443107,

PubMed Abstract | Crossref Full Text | Google Scholar

18. Imai-Senga, Y, Sun-Wada, G-H, Wada, Y, and Futai, M. A human gene, Atp6e1, encoding a testis-specific isoform of H+-Atpase subunit E. Gene. (2002) 289:7–12. doi: 10.1016/S0378-1119(02)00542-5,

PubMed Abstract | Crossref Full Text | Google Scholar

19. León-Velarde, F, Maggiorini, M, Reeves, JT, Aldashev, A, Asmus, I, Bernardi, L, et al. Consensus statement on chronic and subacute high altitude diseases. High Alt Med Biol. (2005) 6:147–57. doi: 10.1089/ham.2005.6.147,

PubMed Abstract | Crossref Full Text | Google Scholar

20. Mallet, R, Burtscher, J, Pialoux, V, Pasha, Q, Ahmad, Y, Millet, G, et al. Molecular mechanisms of high-altitude acclimatization. Int J Mol Sci. (2023) 24:1698. doi: 10.3390/ijms24021698,

PubMed Abstract | Crossref Full Text | Google Scholar

21. Guo, Z, Fan, C, Li, T, Gesang, L, Yin, W, Wang, N, et al. Neural network correlates of high-altitude adaptive genetic variants in Tibetans: a pilot, exploratory study. Hum Brain Mapp. (2020) 41:2406–30. doi: 10.1002/hbm.24954,

PubMed Abstract | Crossref Full Text | Google Scholar

22. Shi, Yong Yong, and He, Lin. Publisher correction: Shesis, a powerful software platform for analyses of linkage disequilibrium, haplotype construction, and genetic association at polymorphism loci. Cell Res 2023, doi: 10.1038/s41422-023-00805-3, [Epub ahead of print.].

PubMed Abstract | Crossref Full Text | Google Scholar

23. Dong, Y, Dun, B, Wang, DP, Zhuo, ML, Fang, J, Yang, ZC, et al. Therapeutic erythrocytapheresis is effective in treating high altitude polycythemia on the Qinghai-Tibet plateau. Wilderness Environ Med. (2020) 31:426–30. doi: 10.1016/j.wem.2020.07.006,

PubMed Abstract | Crossref Full Text | Google Scholar

24. Beall, C. Andean, Tibetan, and Ethiopian patterns of adaptation to high-altitude hypoxia. Integr Comp Biol. (2006) 46:18–24. doi: 10.1093/icb/icj004,

PubMed Abstract | Crossref Full Text | Google Scholar

25. Richmond, T, Chohan, M, and Barber, D. Turning cells red: signal transduction mediated by erythropoietin. Trends Cell Biol. (2005) 15:146–55. doi: 10.1016/j.tcb.2005.01.007,

PubMed Abstract | Crossref Full Text | Google Scholar

26. Me, P, Je, H, Tanabe, Y, and Ts, S. Cross-species insights into genomic adaptations to hypoxia. Front Genet. (2020) 11. doi: 10.3389/fgene.2020.00743,

PubMed Abstract | Crossref Full Text | Google Scholar

27. Rj, A, Tian, Y-M, Rr, R, Turley, H, Al, H, Cw, P, et al. Differential function of the prolyl hydroxylases Phd1, Phd2, and Phd3 in the regulation of hypoxia-inducible factor. J Biol Chem. (2004) 279:38458–65. doi: 10.1074/jbc.M406026200,

PubMed Abstract | Crossref Full Text | Google Scholar

28. Ma, M, Qi, H, Chung, J, Ehl, L, Kondo, Y, Hara, S, et al. Multiple splice variants of the human Hif-3 alpha locus are targets of the Von Hippel-Lindau E3 ubiquitin ligase complex. J Biol Chem. (2003) 278:11032–40. doi: 10.1074/jbc.M208681200,

PubMed Abstract | Crossref Full Text | Google Scholar

29. Ph, M, Ms, W, Gw, C, Sc, C, Ec, V, Me, C, et al. The tumour suppressor protein Vhl targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature. (1999) 399:271–5. doi: 10.1038/20459,

PubMed Abstract | Crossref Full Text | Google Scholar

30. Se, T, Bagnall, J, Mason, D, Levy, R, Dg, F, and See, V. Differential sub-nuclear distribution of hypoxia-inducible factors (Hif)-1 and −2 alpha impacts on their stability and mobility. Open Biol. (2016) 6:160195. doi: 10.1098/rsob.160195,

PubMed Abstract | Crossref Full Text | Google Scholar

31. Wu, D, Potluri, N, Lu, J, Kim, Y, and Rastinejad, F. Structural integration in hypoxia-inducible factors. Nature. (2015) 524:303–8. doi: 10.1038/nature14883,

PubMed Abstract | Crossref Full Text | Google Scholar

32. Li, K, Luobu, G, Wu, P, Ciren, D, Xiao, X, Li, K, et al. Ameliorating effects of the Hif-2α inhibitor Pt2385 on high-altitude polycythemia. Europ J Pharm Sci. (2024) 201:106857. doi: 10.1016/j.ejps.2024.106857,

PubMed Abstract | Crossref Full Text | Google Scholar

33. Haase, V. Regulation of erythropoiesis by hypoxia-inducible factors. Blood Rev. (2013) 27:41–53. doi: 10.1016/j.blre.2012.12.003,

PubMed Abstract | Crossref Full Text | Google Scholar

34. Xu, J, Yang, YZ, Tang, F, Ga, Q, Tana, W, and Ge, RL. EPAS1Gene polymorphisms Are associated with high altitude polycythemia in Tibetans at the Qinghai-Tibetan plateau. Wilderness Environ Med. (2015) 26:288–94. doi: 10.1016/j.wem.2015.01.002,

PubMed Abstract | Crossref Full Text | Google Scholar

35. Ge, RL, Simonson, T, Gordeuk, V, Prchal, J, and McClain, D. Metabolic aspects of high-altitude adaptation in Tibetans. Exp Physiol. (2015) 100:1247–55. doi: 10.1113/EP085292,

PubMed Abstract | Crossref Full Text | Google Scholar

36. Schreiber, T, Koll, N, Padberg, C, Reyes, B, Quinting, T, Malyshkina, A, et al. Reduced vacuolar ATPase protects mice from friend virus infection – an unintended but instructive effect in Hif-2afl mice. J Cell Sci. (2024) 137:Jcs261893. doi: 10.1242/jcs.261893,

PubMed Abstract | Crossref Full Text | Google Scholar

37. Arciero, E, Almarri, M, Mezzavilla, M, Xue, Y, Hallast, P, Hammoud, C, et al. Whole-genome sequences provide insights into the formation and adaptation of human populations in the Himalayas. Curr Biol. (2025) 35:3624–3637.e4. doi: 10.1016/j.cub.2025.06.048,

PubMed Abstract | Crossref Full Text | Google Scholar