Genetics and genomics of exposure to high altitude




Introduction


During the past several decades, efforts to link differences in the genetic code with an individual’s physiology have yielded important insights for human health and disease. Within the past decade, studies of high altitude genetics have sought to determine how variations in DNA underlie physiological differences between and within individuals who encounter the same environmental stress of high altitude hypoxia. This has become increasingly accessible through the advent of high-throughput genotyping technologies, which have revealed important insights regarding genetic adaptations in three continental highland populations (Moore 2017; Simonson 2015), as outlined in Chapter 4. These populations’ unique evolutionary histories provide an exceptional opportunity to identify important genetic factors that have been selected for over hundreds of generations at altitude. Genetic factors underlying lowlanders’ acute responses to high altitude have been examined in various contexts yet, in contrast to the extreme genetic signatures identified in native highland populations, these studies rely on extensive case-control datasets and remain to be more extensively explored. Recent analyses of -omic-level signals (e.g., the evaluation of gene expression through transcriptomics and/or epigenomics; proteomics and metabolomics) in both chronic and acute high altitude exposure provide an additional layer of information regarding the genetic pathways involved in responses to hypoxia and promising insights into generational and short-term responses at high altitude.


History


It has long been recognized that highlanders have a different biological make-up compared with lowlanders. As far back as 1878, Bert (1878) in his typically urbane fashion wrote:


Kellas was one of the first people to recognize the superior performance of Sherpas at high altitude. He used them as partners on some of his high climbs after his first expedition in 1907 when he took two Swiss guides, but both were badly affected by mountain sickness (Kellas 1912). Barcroft in his account of the International High Altitude Expedition to Cerro de Pasco in 1921–22 remarked both on the extraordinary physical ability of the indigenous people and on their remarkable phenotype such as their very large chests (Barcroft et al. 1923).


Monge, in his classical book Acclimatization in the Andes, devoted a great deal of space to the special characteristics of the Peruvian Quechua highlanders (Monge 1948). He was adamant that these people should not be judged by the standards of lowlanders, and he was incensed by Barcroft’s comments that “all dwellers at high altitudes are persons of impaired physical and mental powers.” Monge’s response was that “Andean man being different from sea-level man, his biological personality must be measured with a scale distinct from that applied to the men of the lower valleys and plains.” There were, thus, many references to the fact that highlanders were different from lowlanders in the early literature but, of course, there was no way of studying the genetic basis if indeed one existed.


Variation at the Molecular Level


The human genome, comprising more than 3.2 billion nucleotides of deoxyribonucleic acid (DNA) in the form of adenine (A), thymine (T), guanine (G), and cytosine (C), is organized in double-stranded bases within 23 pairs of chromosomes in the nucleus and ≈16,569 nucleotides in the circular mitochondrial genome. Only a small portion (∼2%) of the human genome encodes genes that are transcribed and translated into a sequence of amino acids that make up proteins to carry out various functions within the cell. The cellular roles of regulatory sequence outside coding portions of the genome (ENCODE Consortium 2012) and nonprotein-coding ribonucleic acids (RNAs) that are transcribed from DNA (Bartoszewski and Sikorski 2018) are active areas of research.


Recent estimates indicate that two individuals differ at approximately 0.6% sites across the genome (∼20 million base pairs) (1000 Genomes Project Consortium 2015). These variable sites are referred to as polymorphisms and may exist as a single base pair difference, such as a single nucleotide polymorphism (SNP) or variant (SNV), or in the form of inserted or deleted portions of genome sequence (insertions/deletions commonly referred to as indels) or copy number variants. Most polymorphisms are considered neutral as they have no effects on the individual. In other cases, they may alter the expression of a gene or impact the protein product with notable effects.


Omics analysis


In addition to examining genetic and genomic contributions to phenotypes, various other large-scale “-omic” approaches are employed to characterize global patterns of gene expression (e.g., transcriptomics) as well as protein and metabolic profiles (proteomics and metabolomics, respectively) in both tissue- and development-specific contexts. Epigenetic modifications, alterations in gene expression above (“epi”) the DNA nucleotide sequence, change in response to environmental cues and are therefore highly relevant in the context of the high altitude environment. Specific markers on the histone proteins that package DNA or direct methylation of specific DNA nucleotides result in heterochromatin and euchromatin states, which preclude and promote transcription of DNA, respectively. Methylome and genome-wide DNA hypersensitivity assays are additional -omics approaches that provide complementary insights into genome-wide patterns of cellular and molecular function.


Candidate genes


Some of the first genetic analyses in highlanders examined polymorphisms in “candidate” genes—single genes that were hypothesized to harbor specific polymorphisms that contribute to the variation in traits observed within and between populations at altitude. One of the first examples of such studies involved examination of the ACE gene, which encodes an angiotensin converting enzyme, and reported an association between the I allele and performance at extreme altitude (Montgomery et al. 1998). The ACE insertion/deletion (I/D) polymorphism has been associated with improvements in performance and exercise duration in a variety of populations (Puthucheary et al. 2011). It has been concluded that while the I allele may relate to improved endurance performance and enhanced oxygen utilization in general, the alternate D allele is associated with gains in strength with training and the associated acquisition of increased muscle bulk in response to muscle strength training (Puthucheary et al. 2011). Despite these reports and numerous attempts to discover genetic variants associated with elite athletic performance, injury predisposition, and elite/world-class athletic status, there has been limited progress to date (Pitsiladis et al. 2016). Past reliance on candidate gene studies, predominantly focusing on genotyping a limited number of SNPs or the I/D variants in small, often heterogeneous cohorts (i.e., made up of athletes of quite different sport specialties), has not generated the kind of results that could offer solid opportunities to bridge the gap between basic research and deliverables in altitude medicine. Following identification of the hypoxia-inducible factor (HIF) pathway(s), many candidate gene studies were performed to test for associations between these genes and/or downstream targets.


The advent of microarray technologies provided an opportunity to simultaneously capture information from thousands to millions of polymorphisms (i.e., SNPs or SNVs) across the genome. This propelled the field of genetics research beyond the single candidate gene approach to examination of genome-wide patterns of variation. The genomics approaches include whole genome and exome sequence analyses, based on entire genome and protein-coding portions only. Rather than providing important but somewhat limited information about thousands to millions of linked genetic markers (signposts scattered across millions of nucleotides) on genotyping arrays, these analyses capture fine-scale information about consecutive DNA sites with potential functional roles.


Hypoxia-Inducible Factors


Description of hypoxia-inducible factors


HIFs are transcription factors that respond to changes in available oxygen in the cellular environment. A transcription factor is a protein that binds to a specific DNA sequence and regulates the transcription of genetic information from DNA to messenger RNA (mRNA). Increasing the rate of gene transcription is referred to as upregulation, while decreasing the rate of gene transcription is called downregulation. There are additional proteins such as coactivators that also play a role in transcription of DNA but typically do not contain DNA-binding domains.


Most oxygen-breathing species express the well conserved transcription factor HIF-1, a heterodimer comprised of an alpha and beta subunit known as HIF-1α and HIF-1β. HIF was originally discovered as a protein that bound to the hypoxia response element (HRE) of the EPO gene under hypoxic conditions (Semenza and Wang 1992). The EPO gene encodes erythropoietin, the hormone controlling red cell production. HIF-1 plays a critical role in many responses of the cell to hypoxia (Samanta and Semenza 2017). In cellular hypoxia, the transcription of several hundred mRNAs is increased and or decreased (Semenza 2011).


HIF responses to hypoxia


The HIF family of transcriptional complexes control the expression of various genes involved in erythropoiesis, angiogenesis, metabolic pathways, vasomotor tone, and cell proliferation and survival. An overview of HIF-1 responses under conditions of normoxia and hypoxia is shown in Figure 6.1. The HIF-α subunit is found at very low levels under normoxic conditions in contrast to the constitutively expressed HIF-β. Under normoxic conditions, HIF-α is hydroxylated by prolyl hydroxylase domain (PHD) proteins (PHD-1, PHD-2, PHD-3) and, through interaction with the von Hippel-Lindau (VHL) protein, ubiquitin (Ub) is added and targets the HIF-α subunit for degradation (Figure 6.2).

Figure 6.1

Figure 6.1Some of the functions of HIF-1α. The large number of genes that are upregulated by the mechanism shown in Figure 6.1 results in extensive physiological effects. (From Wilson et al. 2005, with permission.)

Figure 6.2

Figure 6.2Regulation of HIF-1 by hypoxia. In normoxia (left panel), oxygen-dependent hydroxylation of HIF-1α by prolyl hydroxylase domain proteins (PHD) leads to interaction with von Hippel-Lindau (VHL) and addition of ubiquitin (Ub) resulting in proteasomal degradation. In hypoxia (right panel), HIF-1α is not targeted for degradation and moves to the nucleus where it binds with HIF-1β and combines with coactivators at the hypoxia response element (HRE) to effect gene transcription. (Reprinted with permission of the American Thoracic Society. Copyright 2012 American Thoracic Society. Shimoda LA and Semenza GL. (2011). HIF and the lung: role of hypoxia-inducible factors in pulmonary development and disease. Am J Respir Crit Care Med. 183:152–156.)


Under hypoxic conditions, oxygen-dependent PHD activity is decreased and this results in stabilization of HIF-1α, which then translocates to the nucleus, binds HIF-1β, and recruits coactivator proteins to the HIF binding site: the HIF response element, HRE (a specialized short sequence of DNA of 50 or less base pairs). The result is activation of transcription of various target genes.


The processing of HIF-2α is similar to that of HIF-1α, as shown in Figure 6.2, yet HIF-1α and HIF-2α isoforms exhibit different patterns of expression with tissue-specific effects (Semenza 2011) and impact unique suites of genes (Hu et al. 2003). Different phenotypes are observed in Hif1α and Hif2α knockout mice. The transcriptional profiles of HIF-1α and HIF-2α and functions are better characterized than HIF-3α, which appears to exhibit various functions based on differences in promoter and transcription initiating sites, alternative splice variants that influence mRNA products, and developmental- or tissue-specific expression patterns (Duan 2016).


Congenital Chuvash polycythemia


This unusual condition is of interest because of the light it sheds on the HIF pathway shown in Figure 6.2. The Chuvash Republic is a small area in the east of Russia, and some of its residents have a congenital form of polycythemia associated with genetic abnormalities. High hemoglobin concentrations of 22.6 ± 1.4 g dL−1 have been reported while platelet and white blood cell counts are normal (Sergeyeva et al. 1997). The affected people are homozygous for a missense mutation, that is a change in a single nucleotide that results in a codon that codes for a different amino acid. The consequence is partial impaired binding of the VHL protein to hydroxylated HIF-1α subunits. This results in HIF activity that is inappropriately elevated for a given PO2. In addition to the very high red cell concentrations, these people have elevated ventilation and pulmonary vascular tone. Hypoxia-induced changes in respiration and pulmonary vascular pressures occur at a higher PO2 than in normal individuals indicating a generalized defect in oxygen sensing (Smith et al. 2006; Smith et al. 2008a).


In separate studies of HIF pathway mutations in humans, HIF2α gain-of-function was associated with elevated basal ventilation and pulmonary artery pressure, a moderately enhanced pulmonary vascular sensitivity to hypoxia, but a normal ventilatory sensitivity to hypoxia (Formenti et al. 2011). A PHD2 mutation was associated with elevated basal ventilation and pulmonary artery pressure, increased ventilatory sensitivity to hypoxia, but only a relatively modest increase in the pulmonary vascular sensitivity to hypoxia (Talbot et al. 2017).


Regulation of cellular metabolism by HIF


It has been found that HIF-1 is expressed in very primitive animals such as the worm Caenorhabditis elegans that lacks specialized respiratory or circulatory systems. This suggests that HIF was initially developed to allow individual cells to survive in low oxygen environments. Consistent with this, HIF-1α assists cell survival under hypoxic conditions by switching metabolism from oxidative to glycolytic. This is done by upregulating genes that increase the flux of glucose to pyruvate, such as pyruvate dehydrogenase kinase, genes that inactivate pyruvate dehydrogenase, which converts pyruvate to acetyl-CoA, and genes that increase lactate dehydrogenase, which converts pyruvate to lactate (Kim et al. 2006). Glycolysis is a relatively inefficient path for production of adenosine triphosphate (ATP) compared with oxidative metabolism. The latter produces 18 times as much ATP per mole of glucose compared with glycolysis.


This discovery of the role of HIF-1 in cellular metabolism has some interesting implications. Surprisingly, HIF-1α null fibroblasts (that is with both copies of the gene missing) produce higher ATP levels at 1% oxygen than wild-type cells at 20% oxygen. This demonstrates that under these severely hypoxic conditions, oxygen is not limiting ATP production. However, under these conditions, the HIF-1α null cells die because of the buildup of reactive oxygen species (ROS). It has therefore been suggested that the advantage of switching to glycolysis is to prevent excess mitochondrial generation of ROS (Semenza 2012), which would interfere with mitochondrial electron transport under hypoxic conditions. This work raises the question of whether cells switch to glycolysis because insufficient oxygen is available, or because oxidative metabolism under hypoxic conditions leads to a lethal concentration of ROS. Recent evidence supports the notion that HIF switches cells from oxidative to glycolytic metabolism to reduce mitochondrial superoxide generation and increase the synthesis of NADPH and glutathione in order to maintain redox homeostasis under hypoxic conditions (Samanta and Semenza 2017).


HIF and hypoxic pulmonary hypertension


Alveolar hypoxia (for example, as occurs at high altitude) causes hypoxic pulmonary vasoconstriction, which, if present for sufficient duration, leads to pulmonary vascular remodeling, including thickening of the vessel wall due to the proliferation of smooth muscle and fibroblasts and extension of smooth muscle cells into previously nonmuscular precapillary arterioles. Recently, a role for HIF-1α and HIF-2α in the development of hypoxic pulmonary hypertension has been investigated. It is known that increases in intracellular calcium ion concentration and intracellular pH contribute to the growth and contraction of pulmonary artery smooth muscle cells under hypoxic conditions. There is evidence that the increase in intracellular pH is related to an increased activity and expression of the sodium/hydrogen ion exchanger isoform 1 (NHE1) (Rios et al. 2005). In addition, the increase in intracellular calcium ion concentration is due to the increased expression of canonical transient receptor potential (TRPC) proteins and enhanced calcium ion entry through nonselective cation channels (Wang et al. 2006). Both the increase in intracellular pH and calcium ion concentration are mediated by HIF-1 (Shimoda et al. 2001).


Iron-dependent processes play an important role in pulmonary vasoconstriction, and iron is an obligate cofactor in the regulation of the HIF system. Infusion of iron or iron chelators blunt or aggravate pulmonary vasoconstriction (Smith et al. 2008b) and, given that iron decreases the sensitivity of the HIF system to hypoxia, this could underlie the inverse relationship between iron status and pulmonary vasoconstriction (Frise and Robbins 2015; Smith et al. 2008a).


Tibetans studied at sea level exhibited blunted pulmonary vasoconstriction, relative to Han Chinese, during both 10 minutes of acute and an eight-hour sustained hypoxic exposure. This difference was attributed to hyporesponsive HIF transcriptional response to hypoxia in Tibetans (Petousi et al. 2014).


It is now known that digoxin, a drug that has been used to treat heart failure for a very long time, inhibits HIF-1. To test the possible therapeutic value of this drug in hypoxic pulmonary hypertension, mice were injected with digoxin or saline and exposed to ambient air or hypoxia for three weeks. It was shown that the digoxin-treated animals had attenuation of the development of right ventricular hypertrophy, pulmonary vascular remodeling, increases in intracellular calcium and pH in the pulmonary artery smooth muscle cells, and a fall in right ventricular pressures (Abud et al. 2012). These exciting results suggest a possible role for digoxin in the treatment of high altitude pulmonary hypertension (HAPH).


HIF and intermittent hypoxia


Intermittent hypoxia occurs in several forms among high altitude travelers. For example, many people who travel to high altitude experience periodic breathing at night during which they have a crescendo-decrescendo pattern in their breathing movements punctuated by periods of no airflow (apnea) lasting anywhere from 5 to 20 seconds (Chapter 17). Another pattern of intermittent hypoxia is seen in mine workers or telescope operators who commute to high altitude. For example, the workers of the Collahuasi mine in north Chile live in Iquique at sea level but are bused up to the mine where they work at an altitude as high as 4500 m, although they sleep at the lower altitude of 3800 m. A common pattern is to spend a week at the mine followed by a week at sea level with the pattern being repeated for many months or years, which likely impacts HIF signaling in terms of chronic and intermittent hypoxia (Prabhakar and Semenza 2012).


It is therefore interesting that the patterns of HIF production are affected by intermittent hypoxia. Studies have been carried out on animals exposed to patterns of intermittent hypoxia that are similar to those that occur in many people who suffer from obstructive sleep apnea. It is known that these people who cycle between hypoxia and reoxygenation dozens of times during the night have increased sympathetic activation and increased levels of catecholamines that lead to systemic hypertension and other adverse cardiovascular outcomes related to endovascular dysfunction. Rats exposed to intermittent hypoxia for 35 days (lowest oxygen concentration 3–5%) developed systemic hypertension and left ventricular hypertrophy (Fletcher et al. 1992).


The roles of HIF-1 and HIF-2 are key to regulating various physiological adaptations to chronic hypoxia yet, in cases of intermittent hypoxia, such as experienced during sleep disordered breathing, an imbalance causes oxidative stress and negative cardiorespiratory effects (Prabhakar and Semenza 2012). As already discussed, HIF-1α is expressed at low levels under normoxic conditions, and it is also induced by chronic intermittent hypoxia. However, it is remarkable that the results of intermittent hypoxia on HIF-1α and HIF-2α are diametrically opposed. Experiments conducted on rats showed that intermittent hypoxia upregulated HIF-1α but downregulated HIF-2α both in tissue from rats exposed to intermittent hypoxia and in cells in culture (Nanduri et al. 2009). The results are all the more surprising because continuous hypoxia upregulates both HIF-1α and HIF-2α in the cell cultures. Thus, intermittent hypoxia has a differential effect upon these two structurally related HIF transcriptional activators. These studies also showed that the downregulation of HIF-2α in rats exposed to intermittent hypoxia could be inhibited by calpain proteases (calpain is a protein belonging to a family of calcium-dependent proteases). This may have implications for the prevention of the pathology caused by intermittent hypoxia (Nanduri et al. 2009). Recent work suggests epigenetic modifications, such as changes in DNA methylation that influence gene expression, underlie respiratory and cardiovascular pathologies due to chronic intermittent hypoxia via the HIF pathway (Nanduri et al. 2017).


HIF and lung injury


Acute lung injury, such as that caused by the acute respiratory distress syndrome, results in severe hypoxemia due to damage to the alveolar epithelium and pulmonary capillaries and subsequent development of pulmonary edema. The possible role of HIF in these changes has been briefly studied although a clear pattern has not emerged yet. In a hypoxic ischemia/reperfusion model using isolated ferret lungs, levels of HIF-1α mRNA and protein were increased during hypoxic ischemia. These were associated with an increase in vascular endothelial growth factor (VEGF), which was thought to be a mediator of the increased pulmonary vascular permeability in this model (Becker et al. 2000). By contrast, other studies using both endothelial cell cultures and mice lacking one HIF-1α allele found downregulation of HIF-1α and adenosine kinase in this model of lung injury (Morote-Garcia et al. 2008). Further research is needed to define roles for each of the HIF isoforms and the development of targeted pharmacological approaches of specific PHDs or HIFs (Eltzschig et al. 2014).


HIF and cancer


Most solid cancerous tumors outrun their blood supply with the result that the center of the lesion is hypoxic and drives cancer progression (Schito and Semenza 2016). It is therefore not surprising that HIF-1α is upregulated in these tumors. In addition, there may be loss of function of the von Hippel-Lindau (VHL) protein, which results in increased expression of both HIF-1α and HIF-2α. There is consequently considerable interest in the use of chemotherapeutic agents against cancer that function as HIF-1 inhibitors, which is a very active area of research.


HIF and ischemia


Variation in polymorphisms in the HIF-1 gene may influence its activity and affect the presentation of coronary artery disease. Patients with coronary artery disease often develop collateral vessels in response to narrowing of a coronary artery, and patients with collaterals are likely to have smaller infarcts if the main vessel is occluded (Resar et al. 2005). It has been shown that patients with coronary artery disease who have collateral vessels have an increased frequency of an SNP affecting HIF-1α expression. This change in HIF-1α was associated with stable exertional angina rather than a serious myocardial infarction (Hlatky et al. 2007).


HIF-1α +/− mice, with one wild-type copy of HIF-1α and one nonfunctional copy, exhibit greater susceptibility and impaired recovery to perfusion and greater damage following arterial ligation (Bosch-Marce et al. 2007). Ischemic preconditioning in mice exposed to short-term, several-minute exposures to ischemia and reperfusion are protected against longer (e.g., half hour) episodes of ischemia (Cai et al. 2008; Eckle et al. 2008), while HIF-1 and HIF-2 activity provides protection in the absence of preconditioning (Semenza 2011). Therefore, HIF plays important roles to ensure the heart survives periods deprived of oxygen.


Generational Adaptation to High Altitude in Highland Populations


Various genomic studies conducted within the past decade provide compelling evidence to support the long-standing hypothesis that genetic adaptations to high altitude occurred over hundreds of generations in populations from three continental regions: the Tibetan Plateau, Andean Altiplano, and Ethiopian highlands (Azad et al. 2017; Moore 2017; Simonson 2015). While studies of Tibetan adaptation have been most extensively studied, various reports in Andean and Ethiopian highlanders have emerged and provide important cross-population insights into distinct adaptations and lack of adaptation at high altitude.


Evidence for genetic adaptation in Tibetans


In 2010, several publications based on genome-wide microarray analyses of DNA from Tibetan populations provided unprecedented evidence for genetic adaptation to high altitude (Beall et al. 2010; Simonson et al. 2010; Yi et al. 2010). These studies aimed to identify which genes, among the three billion DNA bases that make up the human genome, are associated with adaptations to high altitude.


The ability to identify these important adaptive genetic factors is unique based on Tibetans’ population history. Ancestors of present-day Tibetans had genetic variants that provided the ability to survive and reproduce despite unavoidable challenges at high altitude. These beneficial variants were passed down each generation and increased rapidly in the population over time, leaving a striking pattern of variation in the genome (distinct from neutrally evolving regions; Figure 6.3) that can be detected in as few as 30 individuals (Pickrell et al. 2009). Genes within these adaptive regions are considered strong candidates for adaptation (Simonson 2015).

Figure 6.3

Figure 6.3A signature of adaptation in a portion of the human genome across evolutionary time. Each panel represents a collection of the same chromosomal region sampled at successive generations. A beneficial genetic variant will increase in frequency over many generations, leaving a pattern of decreased genetic variation (extended haplotype homozygosity). The first panel represents a neutrally evolving region; the second and third panels represent signals detected from an incomplete and complete selective sweep, respectively, that occurred over many generations of selective pressure. In the latter case, the region harboring the adaptive genetic variant(s) is fixed (or nearly fixed) in the population. (Copyright 2015. Simonson TS. (2015). Altitude adaptation: a glimpse through various lenses. High Alt Med Biol. 16(2):125–137.)


As illustrated in Figure 6.1, HIF activity regulates VEGF genes resulting in angiogenesis, EPO, which controls erythropoiesis, and genes for tyrosine hydroxylase, which alter the sensitivity of the carotid body to hypoxia (Prabhakar and Semenza 2012). HIF is therefore a master switch in the general response of the body to hypoxia.


In addition to searching for signatures of adaptation, it is also possible to identify extreme or even subtle significant differences in the frequency of specific genetic variants, i.e., those that are more common in highlanders compared to other populations. These approaches, and/or a combination thereof, have been useful for identifying targets of evolutionary adaptation to high altitude.


The first three genomic studies to examine genotype-phenotype relationships between adaptive genomic variants and hemoglobin concentration in Tibetans applied these various techniques. Simonson et al. (2010) performed two tests of selection across nearly one million SNPs and identified top genes for adaptation as those that were both contained within a region exhibiting a selective sweep and likely involved in hypoxia sensing and response pathways (Simonson et al. 2010). Ten genes were identified in this overlap and six were related to the HIF system, including EPAS1 and EGLN1, which encode HIF-2α and the oxygen-sensing prolyl hydroxylase PHD2, respectively. EGLN1 and PPARA genes were further associated with relatively lower hemoglobin concentration in Tibetans, and subsequent studies identified associations with metabolic parameters suggesting reduced fatty acid oxidation (Ge et al. 2012) and decreased fatty acid oxidation, greater oxygen utilization, and protection from oxidative stress in skeletal muscle (Horscroft et al. 2017).


Yi et al. (2010) analyzed sequence data from the exomes, the protein-coding portion, of 50 Tibetan genomes. Using this strategy, the investigators captured the coding sequences of 92% of all genes. Again, the EPAS1 gene among others was identified as a strong candidate for natural selection. In fact, the frequency of one variant of the EPAS1 gene differed between Tibetans and Han Chinese by 78% (87% versus 9%). This variant was associated with hemoglobin concentration and erythrocyte count, consistent with other reports mentioned below. An interesting conjecture raised by Yi et al. (2010) was that the Tibetan and Han Chinese populations diverged less than 3000 years ago, although more recent estimates suggest greater divergence times based on anthropological evidence for the date of settlements in Tibet (Aldenderfer 2011).


The approach taken by Beall et al. (2010) was to scan the entire human genome looking for dramatic differences in allele frequencies in Tibetans compared with Han Chinese. They examined more than 500,000 SNPs and identified eight variants that were significantly increased in Tibetans. All eight of the overrepresented variants were on chromosome 2 close to the EPAS1 gene. These variants of EPAS1 were correlated with lower hemoglobin concentrations in two additional groups of Tibetans. The authors argued that the change in EPAS1 could be critical in high altitude adaptation in Tibetans because the resulting reduced erythropoietic response helped to avoid the development of chronic mountain sickness. There are also other deleterious effects of a high hematocrit, such as flow abnormalities in the microcirculation including rouleaux formation that can interfere with oxygen delivery. In addition, the increase in viscosity raises vascular resistance, including that in the lung. As illustrated in Figure 6.1, HIFs are involved in the regulation of numerous genes and pathways so the reduced erythropoietin response, although apparently clearly advantageous, may be only the tip of the iceberg for high altitude adaptation.


Numerous studies provide additional support for some of these same genetic adaptations across various locations in Tibet (Bigham et al. 2010; Peng et al. 2011; Xu et al. 2011; Wuren et al. 2014) and among Sherpa (Jeong et al. 2014), with links between another adaptive gene identified in this population (HYOU1) and hemoglobin concentration (Jeong et al. 2014). In addition to EPAS1 and EGLN1 HIF-related genes, several non-HIF pathway targets of selection have been reported in more than one independent study, including CYP17A1, HBB/HBG2, HFE, PKLR, and HMOX2 (Simonson et al. 2010; Yi et al. 2010; Wuren et al. 2014; Simonson 2015). Regulatory variants in HMOX2, which is involved in heme catabolism and oxygen sensing in carotid body chemoreceptors, are further associated with increased gene expression and decreased hemoglobin concentration in 1250 Tibetan men (Yang et al. 2016). Hundreds of distinct putatively adaptive genes have been identified in individual studies to date. These inconsistencies may reflect differences in analytical approaches, population locations/histories, and/or the stage and degree of adaptation.


Despite tremendous progress on the genomics front, the precise functional variants that provide benefits in Tibetans remain largely unknown. One exception is the EGLN1 gene, whereby variants in the first exon found at high frequency in Tibetans (Asp4Glu; Cys127Ser) exhibit a gain of PHD2 function (a lower Km for oxygen), which leads to increased HIF degradation under hypoxic conditions and a potential disruption in erythroid progenitor proliferation (Lorenzo et al. 2014). Other reports suggest these variants underlie loss of PHD2 function via defective binding of cochaperone p23 that would lead to increased HIF activity (Song et al. 2014). A study examining gene expression in a population from India indicates genetic variants within the first intron of EGLN1 lead to increased expression and high altitude pulmonary edema (HAPE) (Aggarwal et al. 2010). More recent analyses of whole genome sequences (Hu et al. 2017) indicate that nonprotein coding variants, including those in heterochromatic or DNA methylated portions of the genome, are crucial for adaptation. In such cases, the effects of increased or decreased gene expression could vary across tissues and/or stages of development (in contrast to protein-coding variants that result in uniform alterations across all cells). An understanding of these changes could provide much needed insight into molecular mechanisms of adaptation.


Whole genome sequencing of DNA from Neanderthals and Denisovans has provided evidence that genetic material from these archaic human populations is present in modern humans today. Genetic variants in the EPAS1 gene region in Tibetans, noted as one of the strongest adaptive signatures in Tibetans, is most similar to Denisovan DNA compared to DNA of other human populations (Huerta-Sánchez et al. 2014; Hu et al. 2017). Therefore, archaic genetic admixture provided variation that helped Tibetans adapt to the high altitude environment. This finding highlights the importance of understanding distinct population histories, and unique genetic backgrounds, in studies of genetic adaptation to high altitude.


Genetic adaptations in Andean populations


Bigham et al. (2009) was the first to provide an analysis of more than 500,000 SNPs in the genomes of two Andean populations, the Quechua and Aymara, and, in a subsequent report, compared these findings with those based on similar analyses in Tibetans. They reported more chromosomal regions showing evidence of positive selection in Andeans than in Tibetans (37 versus 14) and concluded the genetic basis for altitude adaptation was dissimilar in the two populations. One exception was in the EGLN1 gene region, which exhibited signatures of adaptation in both Andeans and Tibetans. Of the reported Tibetan variants in the first exon of EGLN1, c.12C>G (Asp4Glu) is absent and 380G>C (Cys127Ser) is found at low frequency in Andeans from Cerro de Pasco, Peru (Heinrich et al. 2019), further supporting different adaptive mechanisms despite an overlap in the adaptive genetic signal. Preliminary epigenetic investigation at specific sites in the EGLN1 region in Andeans suggests distinct levels of methylation between Andeans with and without excessive erythrocytosis (Julian 2017). More than 30 other genes were reported by Bigham et al. (2010), including EDNRA and PRKAA1, which were both associated with birthweight and the latter with metabolic homeostasis in subsequent genotype-phenotype analysis (Bigham et al. 2014). Several additional genes reported in these studies have since been reported as adaptive in other Andean populations or identified in Tibetans (e.g., the B hemoglobin locus and EDNRA) (Simonson 2015; Table 6.1).









































































































Table 6.1 Signals of altitude adaptation reported in more than one study of human highlanders and/or (1) other species exposed to generations of high altitude or hypoxia, (2) associated with relevant phenotypes (updated from Simonson 2015)

Candidate gene region


Tibetan/Sherpa, Andean, Ethiopian


Other intermediate/highland populations


Phenotype association in human highland population


Hypoxia-adaptive significance in nonhuman species


EPAS1


Tibetan1, 2, 3, 5, 6, 7, 9; Sherpa10; Andean31


Deedu Mongolian15


Hemoglobin concentration (Tibetan2, 3; Amhara Ethiopian23); lactate, free fatty acids (Tibetan27)


Tibetan dog17, 18, 33, Tibetan horse32, Tibetan sheep34, yak35, Tibetan pig3638, Tibetan goats39, 40, plateau zokors42, 43, naked mole rat42, Tibetan saker falcon44, Tibetan chicken45, Tibetan hot-spring snake46, yellow-billed pintail, cinnamon teal, and speckled teal 47


EGLN1


Tibetan1, 3, 5, 6, 7, 8, 9, 12, 13; Andean 4, 11; Sherpa10


Daghestani16


Hemoglobin concentration (Tibetan1); gain29 and loss30 of function in exon 1 Asp4Glu/Cys127Ser


Yellow-billed pintail, cinnamon teal, and speckled teal 47


PPARA


Tibetan1; Amhara and Omotic Ethiopian23



Hemoglobin concentration (Tibetan1, Amhara and Omotic Ethiopian23); lactate, free fatty acids (Tibetan27)


HMOX2/NMRAL1


Tibetan1, 5, 9




Hemoglobin beta gene region


Tibetan1,3; Andean4




Deer mice19, 20, hummingbird21, dog17, 18


PKLR


Tibetan1,3


Deedu Mongolian15



Tibetan pika41


CYP17A1


Tibetan1,9




HFE


Tibetan1,3,9




EDNRA


Tibetan1; Andean4



Birth weight (Andeans26)


CYP2E1


Tibetan1


Deedu Mongolian15



PPARG


Tibetan1


Deedu Mongolian15



HYOU1/HMBS


Sherpa10



Hemoglobin concentration (Sherpa10)


SENP1/ANP32D


Andean14; Andean28




Drosophila14


ADAM17


Tibetan1




Yak22, toad-headed sand lizards48, Tibetan pig36


ARNT2, CBARA1, THRB, VAV3


Amhara Ethiopian23



THRB and hemoglobin concentration (Amhara Ethiopian23)


SNP rs10803083 (chromosome 1)


Amhara Ethiopian24



Hemoglobin concentration in Amhara Ethiopian24


BHLHE41


Amhara, Oromo, and Tigray Ethiopian25




PRKAA1


Andean11



Birth weight (Andeans26); maternal genotypes associated with uterine artery diameter and metabolic homeostasis (Andeans26)


EDNRB


Andean4, Amhara Ethiopian27




CIC, LIPE, PAFAH1B3


Amhara/Oromos Ethiopian27




Involved in hypoxia tolerance in Drosophila27


1  Simonson et al. (2010).


2  Beall et al. (2010).


3  Yi et al. (2010).


4  Bigham et al. (2010).


5  Peng et al. (2011).


6  Xu et al. (2011).


7  Wang et al. (2011).


8  Ge et al. (2012).


9  Wuren et al. (2014).


10 Jeong et al. (2014).


11 Bigham et al. (2009).


12 Aggarwal et al. (2010).


13 Xiang et al. (2013).


14 Zhou et al. (2013).


15 Xing et al. (2013).


16 Pagani et al. (2012).


17 Li et al. (2014).


18 Gou et al. (2014).


19 Storz et al. (2009).


20 Natarajan et al. (2013).


21 Projecto-Garcia et al. (2013).


22 Qiu et al. (2012).


23 Schienfeldt et al. (2012).


24 Alkorta-Aranburu et al. (2012).


25 Huerta-Sánchez et al. (2013).


26 Bigham et al. (2014).


27 Udpa et al. (2014).


28 Cole et al. (2014).


29 Lorenzo et al. (2014).


30 Song et al. (2014).


31 Eichstaedt et al. (2017).


32 Hendrickson (2013).


33 VonHoldt et al. (2017).


34 Wei et al. (2016).


35 Qi et al. (2018).


36 Ai et al. (2013).


37 Li et al. (2013).


38 Ai et al. (2014).


39 Wang et al. (2011).


40 Song et al. (2016).


41 Ge et al. (2013).


42 Shaoet al. (2015).


43 Cai et al. (2018).


44 Pan et al. (2017).


45 Li et al. (2017).


46 Li et al. (2018).


47 Graham and McCracken (2019).


48 Yang et al. (2014).

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 25, 2021 | Posted by in RESPIRATORY | Comments Off on Genetics and genomics of exposure to high altitude

Full access? Get Clinical Tree

Get Clinical Tree app for offline access