Advances in Understanding Adaptive Hemoglobin Concentration at High Altitude

Abstract

Today, more than 81 million people globally live at altitudes ≥ 2,500 m (Tremblay and Ainslie 2021), which corresponds to less than 73% of the oxygen present at sea level, dropping exponentially downwards with increasing elevation. This reduced atmospheric oxygen content, known as high-altitude hypoxia, presents a pronounced physiological challenge to human health, well-being, and reproduction. Nevertheless, there are three global regions where humans have lived in the hypoxic conditions of high altitude for millennia. They include the Andean Altiplano of South America, the Himalayan Plateau of East/Central Asia, and the Semien Plateau of Ethiopia. For decades, biological anthropologists, physiologists, and others have studied human adaptation to hypoxia among the high-altitude populations from these regions (Beall 1982; Frisancho 1969). This work has highlighted that each of these groups has developed unique physiological, genetic, and potentially epigenetic adaptations to life in low oxygen conditions (Alkorta-Aranburu et al. 2012; Beall et al. 2010; Bigham et al. 2010; Childebayeva et al. 2021). One phenotype that has been of particular interest in high-altitude evolutionary studies is hemoglobin.

Hemoglobin (Hb) is the iron-containing protein found in red blood cells (RBC) responsible for oxygen transport. It carries oxygen from the lungs to the various tissues in the body. Hb concentration is a measure of the amount of hemoglobin protein in red blood cells. At high elevation, atmospheric oxygen is limited, thus reducing arterial oxygen content. High-altitude sojourners overcome this reduction by increasing the amount of circulating Hb, initially through reductions in plasma volume and Hb-O2 affinity, and later through increases in red cell volume (Siebenmann et al. 2015). Among high-altitude-adapted populations, we see distinct hematological adaptations to hypoxia both between and in comparison to high-altitude sojourners. Tibetans display a relatively low erythropoietic response and attendant low Hb concentration (Adams and Strang 1975; Beall and Goldstein 1987; Beall and Reichsman 1984). Andeans exhibit elevated concentrations with some individuals presenting with polycythemia, or the increase of hematocrit and/or Hb (Beall et al. 1990, 1998). Hematocrit is related to Hb concentration and measures the percentage of whole blood composed of red blood cells. High-altitude Ethiopians of mainly Amharic ancestry show similar Hb concentrations compared to low-altitude US residents (Beall et al. 2002), but high-altitude Amhara and Oromo exhibit elevated Hb levels compared to their low-altitude (< 1500 masl) counterparts (Scheinfeldt et al. 2012; Alkorta-Aranburu et al. 2012), with Oromo displaying twice the elevation in Hb level compared to Amhara (Alkorta-Aranburu et al. 2012). Collectively, this physiological evidence demonstrates that modifications to Hb concentration play an important role in the biological response to hypoxia.

A cohort of genome-wide investigations published in 2010 demonstrated that Andeans and Tibetans have adapted by natural selection to high-altitude hypoxia (Beall et al. 2010; Bigham et al. 2010; Simonson et al. 2010; Yi et al. 2010), and, among Tibetans, that adaptive genetic variation contributed to their blunted Hb phenotype. In particular, two genes central to the hypoxia inducible factor (HIF) pathway, an evolutionarily ancient transcriptional regulatory pathway for the cellular response to hypoxia (Semenza 2012), showed signatures of positive selection and possessed variation that was associated with Hb concentration: EPAS1 (also known as HIF2A) and EGLN1 (also known as PHD2). These findings provided strong support for the hypothesis that hypoxia has acted as a selective agent on HIF genes to influence Hb concentration. Genome-wide data from Andeans also showed evidence of natural selection at EGLN1 (Bigham et al. 2010) and EPAS1 (Foll et al. 2014) but associations with phenotype were not explored at that time.

In 2013, we and others published “Andean and Tibetan patterns of adaptation to high-altitude” in the American Journal of Human Biology. In this article, we tested for EGLN1 and EPAS1 SNP associations with Hb concentration in Andeans, identifying no significant genotype–phenotype relationships. These findings suggested that the two genes with variation associated with Hb concentration among Tibetans did not contribute to the Andean Hb phenotype. We were careful to emphasize that Hb concentration could indeed have a genetic basis among Andeans and importantly, yet to be identified genetic variants in EGLN1 and/or EPAS1 could contribute. We recommended continued investigation into the genetic contributions to Andean Hb concentration. Since our 2013 publication, much has been learned about the genetic basis of Hb concentration among Andeans, Tibetans, and, to a lesser degree, highland Ethiopians. In this commentary, we discuss the advances in our understanding of the genetic basis of altitude-adaptive Hb concentration, the strides made in characterizing the functional consequence of altitude-adaptive variation on Hb phenotypes, and the integration of epigenetics into Hb studies.

Hb plays a central role in acclimatization, as it is one of the main mechanisms by which the body physiologically adjusts to lower oxygen partial pressure (Jourdanet 1875; Monge and León-Velarde 1991; Viault 1890). At high elevation, reduced environmental oxygen diminishes available oxygen along the oxygen transport chain, physiologically translating to reduced blood oxygen content and diminished tissue oxygenation. The reduction in oxygen upon acute exposure to hypoxia (Figure 1) sets in motion a series of physiological changes, including a hematopoietic response wherein Hb concentration rises over a period of several weeks. This occurs through a cascade of events rooted in oxygen sensing and signaling by the kidney. Upon immediate exposure to high elevation, the kidney detects lowered blood oxygen content. The renal glycoprotein hormone erythropoietin (EPO) then stimulates erythropoiesis in the bone marrow, resulting in increased red blood cell production that compensates for the lowered arterial oxygen content (Knaupp et al. 1992). Activated by the HIF pathway, plasma EPO levels rise rapidly and peak 1–2 days after initial hypoxic exposure (Haase 2010), eventually decreasing as hematocrit increases over a period of a few weeks. For lowland sojourners to high elevation, this leads to an increase in Hb concentration over a period of 1–2 weeks of sustained hypoxia exposure (Childebayeva, Harman, et al. 2019; D'Alessandro et al. 2016). This increased production of erythrocytes promotes greater oxygen-carrying capability that helps overcome the lower ambient oxygen tension and is part of the acclimatization process to high elevation.

Hb concentration is a highly polygenic phenotype. Although sometimes elusive in its genomic underpinnings, it has emerged as an extremely useful example of how natural selection can act across diverse, population-specific variation to shape high-altitude-adaptive phenotypes. Investigations seeking to understand the genetic basis of Hb concentration among the three long-term, high-altitude populations have highlighted distinct genetic adaptations within each population and have revealed a breadth of potentially Hb-associated genes. These include HIF-pathway genes such as EPAS1 and EGLN1, as well as genes that participate in other biological pathways including HMOX2, PDE1B, NOS2, and NFKB1 (Amaru et al. 2022; Beall et al. 2010; Scheinfeldt et al. 2012; Yang et al. 2017; Yi et al. 2010).

EPAS1 and EGLN1 are among the few genes that show signals of selection across populations, and for which there exists evidence that naturally selected variation underlies Hb-adaptive phenotypes. EPAS1 encodes for the HIF-2α protein and, together with its paralogue HIF-1α, comprises the α-subunit (HIF-α) of the constitutively expressed heterodimeric HIF transcription factor responsible for oxygen sensing. EGLN1 plays a key role in the regulation of HIF-α. It encodes for the PHD2 enzymatic protein, which together with PHD1 and PHD3, regulates HIF-α. Under normoxia, HIF-α is hydroxylated by PHD, targeting it for proteasomal degradation by the ubiquitin-proteasome pathway (Kaelin Jr. and Ratcliffe 2008). Under hypoxia, HIF-α is stabilized through arrested posttranslational modification by PHD, leading to the upregulation of target genes involved in oxygen homeostasis, including the erythropoietin-encoding gene EPO that controls Hb concentration (Lappin and Lee 2019). The evidence for Hb association for EPAS1 is compelling and has been replicated across studies, whereas the EGLN1 evidence of association with Hb concentration is weaker, underlying the complex polygenic nature of Hb response that is different by population.

Epigenetics studies mitotically and, in some cases, meiotically heritable changes to gene expression that do not involve changes to the DNA sequence (Feil and Fraga 2012; Wolffe and Guschin 2000). Epigenetic modifications, such as DNA methylation (DNAm) and histone tail modifications, undergo reprogramming during early development when the epigenome is the most susceptible to environmental cues (Reik et al. 2001). As such, epigenetic processes long have been hypothesized to play a role in adaptation, especially in the context of early life developmental plasticity that may influence later-in-life health and disease outcomes (Kuzawa and Thayer 2011). In the context of high altitude, hypoxia exposure in utero and during the first years of life may serve to prime the body for life in low oxygen conditions. Although epigenetic modifications have been hypothesized to contribute to high-altitude adaptation (Julian 2017), it is mainly within the last decade that epigenetic change has been studied in this environmental context (Childebayeva, Jones, et al. 2019).

Some of the most compelling epigenetic findings to date derive from studies among Andeans. Differences in DNAm have been identified in Andeans with both lifetime and developmental-only exposures to high altitude compared to Andeans born and raised at low elevation (Childebayeva et al. 2021; Childebayeva, Jones, et al. 2019). This suggests long-term consequences of early-life exposure to hypoxia that persist throughout an individual's lifetime. Early work with Ethiopians identified no significant DNAm differences when comparing high- and low-altitude Oromo and Amhara, but this was likely due to low sample sizes (Alkorta-Aranburu et al. 2012).

Research among high-altitude sojourners, Tibetans, and Andeans indicates a role for epigenetic modification in shaping hemopoietic responses to high-altitude hypoxia. In the context of non-adapted individuals exposed to high-altitude hypoxia (high-altitude sojourners), an increase in DNAm of RXRA has been linked to increased Hb concentration levels (Childebayeva, Harman, et al. 2019). RXRA is a retinoic acid receptor family gene that is essential for normal hematopoiesis during development and in adulthood (Cañete et al. 2017; Oren et al. 2003). No significant association between EPAS1 DNAm and Hb concentration was identified in the same study (Childebayeva, Harman, et al. 2019). In Tibetans from very high altitude (> 4500 m), significantly higher DNAm levels in the promoter regions of TGF-β and BMPR2 have been identified in high-altitude polycythemia patients compared to the healthy controls (Zhaxi et al. 2024). Among Andeans, lower DNAm levels at CpG sites in the CpG island overlapping with the EPAS1 promoter have been found in individuals living at high altitude, with the length of time spent at high elevation being negatively associated with methylation levels (Childebayeva, Jones, et al. 2019). Based on these same data, we identified an association between EPAS1 promoter DNAm and Hb levels that approaches significance (β = −0.04, p = 0.056, n = 454 Andean Quechua from both high [Cerro de Pasco, Peru 4300 m] and low altitude [Lima, Peru 150 m]), model corrected for city of recruitment, sex, age, and BMI (Figure 2). When DNAm occurs within promoter regions, methyl groups interfere with RNA polymerases during transcription. As such, DNAm acts as a repressive mechanism with higher DNAm associated with lower transcription. In this case, lower DNAm of the loci in the CpG island overlapping the EPAS1 promoter is associated with higher Hb levels, suggesting a phenotypic outcome of DNAm. This finding shows the effect of hypoxia on EPAS1 methylation, which in turn mediates Hb levels.

Beyond our discovery of an association between lower EPAS1 promoter DNAm and the altitude of residence in the Peruvian Quechua (Childebayeva, Jones, et al. 2019), we also identified a significant association between intronic EPAS1 SNPs (rs7579899 and rs2044456) and EPAS1 DNAm (Childebayeva et al. 2021). Similar to the finding that gene-specific genetic variation plays a role in the DNAm variation for EPAS1 (Childebayeva, Jones, et al. 2019), SNPs in EGLN1 have been shown to mediate the DNAm levels of the nearby CpG sites (Sharma et al. 2022). Together, these findings indicate a complex interplay between environmental exposures, genetic variation, epigenetic variation, and the resulting phenotype (Figure 3).

Maintaining an optimal balance of Hb concentration is crucial for the body in hypoxia, as this phenotype has the potential for supporting survival under hypoxic conditions but also contributing towards high-altitude disease. Moderately elevated Hb concentrations improve blood oxygen capacity and tissue oxygenation, but excessively elevated Hb concentrations may limit the passive diffusion of oxygen across the alveolar-capillary membrane. The increased blood viscosity caused by an elevated Hb concentration can result in reduced cardiac output and microcirculatory blood flow (Storz 2021). Excessive erythrocytosis, reflected by exceedingly high Hb concentrations, can result in cyanosis, hypoxemia, myocardial infarction, and stroke, largely due to pathologically increased blood viscosity (Corante et al. 2018). Furthermore, excessive erythrocytosis is a key diagnostic feature of CMS, a condition that can develop after an extended period of time living at high altitude and that is characterized by polycythemia, hypoxemia, pulmonary hypertension, and blood hyperviscosity. Among high-altitude adapted populations, Peruvian Andeans display the highest incidence of CMS at rates up to 18% among males, Tibetans very modest incidence rates of roughly 1% among males, and no evidence to date exists for CMS among Ethiopians (Monge et al. 1989; Wu 2005). Candidate genes such as SENP1, NFKB1, ANP32D, SGK3, and COPS5 have been associated with the exaggerated erythropoietic response that characterizes CMS and high-altitude polycythemia among Andean populations (Azad et al. 2016; Hsieh et al. 2016; Song et al. 2025; Stobdan et al. 2017; Zhou et al. 2013). Further, while CMS primarily impacts fitness during post-reproductive years, some investigations have identified excessive erythrocytosis and CMS-associated genes that show signals of selection, including SENP1 and ANP32D (Bermudez et al. 2020; Gazal et al. 2019; Zhou et al. 2013). Although these genes have been identified in association with polycythemia, it is likely that they also may help regulate Hb concentrations outside of this disease context. Furthermore, in Andeans who exhibit increased CMS levels with age, epigenetic modifications could explain, at least to a degree, the likelihood of the development of CMS and should be explored in future avenues of research. Looking beyond advancements in our understanding of the genetic basis for altitude-adaptive Hb phenotypes, research in the last decade has identified genetic variation associated with additional altitude-adaptive traits with implications for human health. These genes and phenotypes include, but are not limited to, EGLN1 variation on exercise or work capacity among Andeans (Brutsaert et al. 2019), PTPRD, CPT2, and TBX5 variation on placental metabolism (O'Brien et al. 2024), and CCDC141 variation on reproductive success measured as number of pregnancies and number of live births (Jeong et al. 2018).

In the more than 10 years since the publishing of “Andean and Tibetan patterns of adaptation to high-altitude” in the American Journal of Human Biology, a broader understanding of the underlying mechanics of high-altitude Hb adaptation has emerged. Research among Tibetans has been particularly fruitful, with several studies functionally characterizing adaptive Tibetan EPAS1 and EGLN1 alleles, demonstrating their impact on protein function. This work has revealed in part how genetic changes translate into altitude-adaptive phenotypes. Among Andeans, recent strides in understanding the genetic basis of adaptive Hb variation have suggested that polygenic natural selection may be ongoing to reduce Hb concentration in this population, and that the HIF genes EGLN1 and EPAS1 may indeed contribute to the Andean Hb phenotype. For Ethiopian highlanders, several genes, including EPAS1, but not EGLN1, have emerged as potential targets of natural selection that may be involved in shaping Hb in this region of the world. In addition to developments in understanding the genetic basis of Hb concentration, epigenetics has emerged as a potential mechanism for mediating high-altitude adaptation via an effect on Hb, including evidence suggesting that EPAS1 DNAm helps shape Hb concentration. The degree to which epigenetic modifications are involved remains an open question and warrants further scientific inquiry. In the next decade, the continued integration of evolutionary and functional approaches in the study of human adaptation to high elevation will continue to provide critical insights into the molecular mechanisms governing adaptive hemopoietic change. As such, it will endure as a model for understanding the functional basis of naturally selected genetic variation and convergent evolution.

Ainash Childebayeva: conceptualization, writing – original draft, writing – review and editing. Kimberly Zhu: writing – original draft, writing – review and editing. Abigail W. Bigham: conceptualization; writing – original draft; writing – review and editing.

The authors have nothing to report.

The authors declare no conflicts of interest.