Materials and methods

Sample preparation

This retrospective study used de-identified residual blood samples from routine health examinations. Thirty eligible male individuals aged 18–35 years were equally divided into three groups (n = 10): Tianjin, Ningxia Yinchuan, and Xizang Linzhi cohorts. The inclusion criteria were as follows: healthy individuals with normal physical examination results and at least 3 consecutive years of local residence. The exclusion criteria were as follows: chronic or hereditary diseases, long-term medication use, irregular lifestyle, and short-term residence. Blood samples were collected from the antecubital vein between 6 and 8 a.m. (22 °C ± 2 °C) and irreversibly anonymized. To address the insufficient RNA yield for individual microarray hybridizations, we randomly pooled equal volumes (400 μL) of RNA from participants within each group to create three non-overlapping composite pools. Each pool was then analyzed as a distinct biological replicate without technical duplication.

Total RNA extraction

Whole blood samples were collected using PAXgene Blood RNA tubes (PreAnalytiX, Switzerland; QIAGEN). Total RNA was extracted and purified using the QIAGEN serum/Plasma Kit (Cat 217184, QIAGEN) following the manufacturer’s standard instructions. RNA concentration, purity, and integrity were verified using the NanoDrop ND-2000 spectrophotometer and Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, United States).

Oligonucleotide microarray hybridization

Total RNA was amplified and labeled using the Agilent Low Input Quick Amp Labeling Kit, One-Color (Cat# 5190–2305). Labeled cRNA was purified using the RNeasy Mini Kit (Cat#74106, QIAGEN). Hybridization was performed in an Agilent hybridization oven at 65 °C with rotation at 10 rpm for 17 h with 1.65 μg of Cy3-labeled cRNA per array. Slides were washed using the Gene Expression Wash Buffer Kit (Cat#5188–5327, Agilent). Microarrays were scanned using an Agilent Microarray Scanner (G2565CA) at a resolution of 3 μm, with PMT set at 100%, 20-bit depth.

Gene expression profile

Raw expression data were extracted using Agilent Feature Extraction software (version 12.0) and normalized by quantile normalization using the limma package in the R/Bioconductor platform. Quality control analyses were conducted, including principal component analysis (PCA), sample clustering analysis, and sample correlation analysis. All samples were processed in a single batch to avoid batch effects, and three independent pooled RNA biological replicates were included for each group (sea level (0 m), 1,000 m, and 3,000 m). All raw array signals were log-transformed to satisfy normality assumptions for a Student’s t-test and ANOVA. The Benjamini–Hochberg (BH) procedure was applied to calculate the false discovery rate (FDR) for multiple-testing correction. Differentially expressed circRNAs were defined as transcripts with |log2 (fold change)| > 1 and FDR <0.05.

Gene ontology analysis

GO enrichment was performed using Fisher’s exact test using the clusterProfiler R package (v4.20.0; https://bioconductor.org/packages/release/bioc/html/clusterProfiler.html) from the R/Bioconductor platform. Genes were assigned to three major GO categories: biological process (BP), molecular function (MF), and cellular component (CC). The thresholds for significant enrichment were set to a minimum of two differentially expressed host genes mapping to a given term and a raw p < 0.05. For visualization, the top 30 GO terms, ranked in descending order by enrichment factor, were displayed in dot plots. Dot size corresponds to the count of enriched DEGs, and the color gradient represents the FDR-adjusted statistical significance.

KEGG enrichment analysis

KEGG pathway enrichment was performed using the host genes of the differentially expressed circRNAs. Fisher’s exact test was used for enrichment analysis via the clusterProfiler package in R/Bioconductor. The screening criteria for significant enrichment included a minimum of two enriched DEGs per pathway and a raw p < 0.05. The enrichment factor was calculated exactly as described for the GO analysis. The top 30 pathways, ranked in descending order by enrichment factor, were visualized using dot plots. KEGG pathways were classified into six major categories: cellular processes, environmental information processing, genetic information processing, human diseases, metabolism, and organismal systems. Dot size represents the number of enriched DEGs, and the color gradient reflects the FDR-adjusted statistical significance.

Cell culture: Human umbilical vein endothelial cells (HUVECs) were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Sigma, MO, United States) supplemented with 10% fetal bovine serum (FBS; Invitrogen, CA, United States), 100 U/mL penicillin, and 100 μg/mL streptomycin (Invitrogen, CA, United States). Cells were maintained in a humidified incubator with 5% CO2 at 37 °C.

Cell transfection

HIF-1α vector control vectors were designed by RiboBio (Shanghai, China). Transfections were performed using Lipofectamine 2000 (Invitrogen, CA, United States) following the manufacturer’s protocols.

qRT-PCR validation

qRT-PCR was performed using the QuantiNova™ SYBR Green PCR Kit (QIAGEN, Hilden, Germany) according to the manufacturer’s instructions. The Applied Biosystems 7900 Real-Time PCR System was used to quantify the relative expression of circRNAs. qRT-PCR primers were custom-synthesized by GenePharma (Shanghai, China). The sequences are not publicly available but can be requested from the corresponding author.

Statistics analysis

All experimental data were analyzed using GraphPad Prism software (version 8.0) and are presented as the mean ± standard deviation (SD). Prior to statistical analysis, data were assessed for normality and homogeneity of variance. Pairwise comparisons between groups were conducted using Student’s t-test. For all transcriptome-wide comparisons, multiple-testing correction was applied using the Benjamini–Hochberg (BH) method to control the false discovery rate (FDR). Statistical significance was defined as an FDR <0.05. Statistical significance was set at p < 0.05. Asterisks indicate significance levels: p < 0.05 (), p < 0.01 (), p < 0.001 (), and p <0.0001 (****); ns indicates no significant difference (p > 0.05).

Informed consent

This study was performed using de-identified residual health examination samples. Since no personal information could be identified, the requirement for written informed consent was waived by the Institutional Review Board.