Discussion

Living at HA is challenging due to the low oxygen pressure. A comprehensive physiological response that protects tissues from hypoxic injury is essential for adaptation and survival (D'Alessandro and Xia, 2020). HA responses are induced by multiple factors, and HA adaptation is critical for the survival of populations. Differences in ethnicity, geographical environment, and evolutionary history may lead to different levels of HA response (Getu et al., 2025; Nieves-Colón et al., 2022; Mairbäurl et al., 2020). Therefore, it is of great significance to clarify the adaptation patterns of different populations to the HA environment. With the decreasing cost of next-generation sequencing (NGS) technology, a variety of whole-genome sequencing studies have been performed to explore the molecular mechanism of HA adaptation (Zaborowska et al., 2021; Rathore et al., 2022; Hofmann et al., 2021). A microarray study revealed that key genes associated with hypoxic stress and endoplasmic reticulum stress, including HIF-1α, STAT3, and EGFR, exhibit significantly lower expression levels in Tibetan placentas than in European placentas (Tenzing et al., 2021). In addition, a study constructed a predictive model including 50 genes, such as LRRC18 and HCAR3, by integrating two public microarray datasets, providing insights into the risk of HAPC in Han Chinese populations after exposure to high altitude (Wang et al., 2022). These studies indicate that the HA environment can extensively regulate human genome expression and that genome-wide expression detection represents an effective strategy for revealing HA adaptation mechanisms.

CircRNAs are a class of non-coding RNAs without 5′terminal caps or 3′terminal poly(A) tails, which form closed-loop structures through covalent bonding (Hansen et al., 2013; Kos et al., 1986; Wang et al., 2014). CircRNAs are mainly derived from gene exons, but other types also exist, including circRNAs originating from introns, intergenic regions, antisense strands, and sense-overlapping regions. Some circRNAs contain miRNA response elements and can act as competitive endogenous RNAs (ceRNAs) by binding miRNAs and serving as miRNA sponges in cells, thereby relieving the inhibition of target genes by miRNAs and upregulating target gene expression (Hansen et al., 2013; Shang et al., 2019). In recent years, circRNAs have been regarded as key biomarkers for many human diseases, including cancer (Conn et al., 2024; Zhang et al., 2023). However, whether circRNAs participate in HA adaptation and regulation remains unclear.

In this study, we selected permanent residents from Tianjin sea level (0 m), Ningxia Yinchuan (1,000 m), and Xizang Linzhi (3,000 m) as study subjects. All participants were healthy male individuals who had resided continuously in the corresponding area for more than 3 consecutive years and had consistent genetic backgrounds and basic lifestyles to minimize confounding factors. We performed a genome-wide expression analysis to characterize circRNA regulation in three populations living at different altitudes. As a major type of genomic structural variation, CNVs drive long-term adaptive evolution in extreme environments. Unlike SNVs with mild effects, CNVs strongly regulate gene transcription by altering gene dosage and chromatin structure. We observed CNVs in the parent genes of circRNAs and identified differentially expressed circRNAs among the three populations. Using GO and KEGG analyses, we explored the potential functions and pathways of the parent genes of differentially expressed circRNAs. It is worth noting that while functional enrichment analyses of host genes offer preliminary predictive insights, the intrinsic biological functions of circRNAs cannot be definitively established without in vitro experimental validation. We identified five common differentially expressed circRNAs in the three populations and predicted their binding proteins. Finally, we verified the differential expression of these five circRNAs among the three populations and confirmed by qRT-PCR that HIF-1α enhanced their relative expression. While these unidirectional transfection assays reveal a positive correlation between HIF-1α and the five circRNAs, they do not verify the reverse regulation of the HIF-1 pathway by circRNAs. In the absence of knockout and rescue experiments, this association must be interpreted as a preliminary exploratory clue rather than a definitive causal link. In summary, we identified five circRNAs differentially expressed in three populations at different altitudes: hsa_circ_0044526, hsa_circ_0022498, hsa_circ_0044534, hsa_circ_0044520, and hsa_circ_0026102. GO analysis showed that biological processes represented the most significantly enriched function of the parent genes. For instance, biological regulation was one of the most enriched terms in 3,000 m vs. 1,000 m; 1,000 m vs. sea level (0 m); and 3,000 m vs. sea level (0 m) comparisons. KEGG analysis indicated that the parent genes were involved in multiple pathways, including the HIF-1 signaling pathway, suggesting that HIF-1 may serve an important function in HA adaptation. AGO2 and EIF4A3 were the common RNA-binding proteins predicted to interact with these five circRNAs, which is consistent with previous studies reporting that AGO2 and EIF4A3 participate in hypoxic regulation (Hale et al., 2014; Li et al., 2020).

HIF-1 is a transcription factor that is essential for responses to low oxygen levels or hypoxia (Yuan et al., 2024). It consists of two subunits: the oxygen-regulated HIF-1α and the constitutively expressed HIF-1β (Wang et al., 1995; Jiang et al., 2025). HIF-1 acts as the master regulator of numerous hypoxia-induced genes under hypoxic conditions (Dengler et al., 2014). Proteins encoded by HIF-1 target genes enhance oxygen supply and mediate adaptive responses to oxygen depletion (Yang et al., 2025; Gao et al., 2025; Jin et al., 2018). Although known as a hypoxia-inducible factor, HIF-1 can induce a series of adaptive responses not only by reducing oxygen utilization but also through other stimuli, such as nitric oxide or various growth factors (Caballano-Infantes et al., 2022; Giaccia et al., 2004; So et al., 2022). In this study, KEGG analysis identified the HIF-1 signaling pathway as a key pathway for the parent genes of differentially expressed circRNAs. Moreover, we confirmed that HIF-1α increased the relative expression of the five circRNAs. This unidirectional regulatory observation provides only preliminary correlative clues linking these circRNAs to HIF-1-related hypoxic signals and altitude-associated molecular phenotypes.

This study enrolled only male subjects with a limited sample size, which may restrict the generalizability of the conclusions. The present findings are observational and correlative rather than mechanistic due to the lack of in-depth functional experiments, such as circRNA overexpression and knockdown. In addition, other vital hypoxia regulators, including EPAS1, EGLN1, and PPARA, may also contribute to high-altitude adaptation and warrant further investigation. Despite these limitations, this study provides a comprehensive overview of circRNA expression profiles in human populations at different altitudes. The identification of differentially expressed circRNAs and their association with the HIF-1 pathway expands our understanding of the genetic basis and molecular regulatory network of HA adaptation. The five circRNAs identified in this study may serve as novel molecular markers for evaluating HA adaptability. Collectively, this study reveals transcriptomic characteristics during HA adaptation and offers new insights into the molecular mechanisms underlying human adaptation to high altitude.