High altitude genetic adaptation in Tibetans: no role of increased hemoglobin-oxygen affinity

Abstract

High altitude exerts selective evolutionary pressure primarily due to its hypoxic environment, resulting in multiple adaptive responses. High hemoglobin-oxygen affinity is postulated to be one such adaptive change, which has been reported in Sherpas of the Himalayas. Tibetans have lived on the Qinghai-Tibetan plateau for thousands of years and have developed unique phenotypes, such as protection from polycythemia which has been linked to PDH2 mutation, resulting in downregulation of HIF pathway. In order to see if Tibetans also developed high hemoglobin-oxygen affinity as a part of their genetic adaptation, we conducted this study assessing hemoglobin-oxygen affinity and their fetal hemoglobin levels in Tibetan subjects from 3 different altitudes. We found normal hemoglobin-oxygen affinity in all subjects, fetal hemoglobin levels were normal in all except one and no hemoglobin variants in any of the subjects. We conclude that increased hemoglobin-oxygen affinity or increased fetal hemoglobin are not adaptive phenotypes of the Tibetan highlanders.

Introduction

High-altitude hypoxia poses a great challenge to maintain adequate tissue oxygenation, and multiple physiologic responses occur to adapt to such an environment. Random mutations altering these responses undergo natural selection when they are beneficial, then the frequency of these mutations increases in subsequent generations. Thus, the Amerindians in Andes show an increased Bohr effect compared to non-Amerindians at the same altitude¹. In Sherpas of the Himalayas, a left-shifted oxygen dissociation curve has been reported which enables better pulmonary uptake of oxygen in hypoxia². These adaptive changes to hypoxia ensure adequate oxygen delivery at the cellular levels, but the genetic basis of these adaptations is yet to be elucidated.

Native Tibetans have lived at an average of 3000–5000 meters on the Tibetan Plateau for about 25,000 years³, and have acquired genetic adaptations that have enabled them to thrive in this reduced oxygen environment. Most Tibetans thus have evolved distinctive phenotypes, such as reduced low birth weights, low prevalence of pulmonary hypertension and protection from polycythemia and other features of chronic mountain sickness^4,5,6,7,19. All of these phenotypic traits are components of underlying genetic adaptation and some, such as protection from polycythemia, have been linked to a gain-of-function mutation of PDH2 (encoded by EGLN1 gene) resulting in down-regulation of the HIF pathway ^6,7.

Increased hemoglobin-oxygen affinity has been described as an adaptive response to hypoxia in high-altitude Camelidae⁸ similar to what has been reported in the Sherpas². Although no such studies have been done in Tibetans, but it has been reported that Tibetans have higher resting arterial oxygen saturations compared to non-Tibetans⁹. Higher arterial oxygen saturation was reported in Tibetan infants during the first four months of life, compared to Han Chinese counterparts ¹⁰. A genetic basis for enhanced oxygen transport in Tibetans as a part of their high-altitude adaptation has been suggested, using a complex statistical analysis¹¹. Further, recent genomic studies have reported that β-globin haplotypes for adult hemoglobin (β subunit encoded by HBB gene) and γ2 subunit of fetal hemoglobin (HbF, γ2 subunit encoded by HBG2 gene) have undergone genetic selection in Tibetans¹², suggesting that a hemoglobin variant and/or modified transition from HbF to adult hemoglobin may be beneficial factors of Tibetan high altitude adaptation.

The role of hemoglobin-oxygen affinity in oxygen transport at high altitude is complex and cannot be based solely on arterial oxygen saturation. Arterial oxygen saturation is largely a reflection of alveolar function and does not indicate increased oxygen affinity, which is correctly measured by deriving its P50 value – the partial pressure of oxygen at which hemoglobin is 50% saturated with oxygen. In addition to high affinity hemoglobin mutants, several physiologic variables, including 2, 3-diphosphoglycerate (2, 3 DPG), temperature, pH, can also increase hemoglobin-oxygen affinity and thus decrease P50 value. Further, an increased proportion of HbF, that has reduced interaction with 2, 3 DPG, could also increase hemoglobin-oxygen affinity.

To test whether altered hemoglobin-oxygen affinity is a constituent of Tibetan high-altitude adaptation, we conducted a study of direct and indirect oxygen-hemoglobin affinity among Tibetans living at different altitudes.

Materials and methods

A cohort case study on 15 ethnic Tibetans living at 3 different locations in the Tibetan plateau in China, India and the USA at altitudes ranging from 1300 meters, 1730–2300 meters and 4320 meters was conducted with the approval of local IRBs in USA, China, and India. Written informed consents (in Tibetan and English) were obtained from each volunteer after detailed explanation in Tibetan or English as preferred by each volunteer. All studied subjects were healthy ethnic Tibetans ranging from 30 – 75 years (see Table 1). Four non-Tibetan volunteers constituted the control arm. Additional 29 Tibetans, 25 Aymaras, and 5 Caucasian Bolivian high altitude residents participated in evaluation of fetal hemoglobin. (HbF).

Table 1.

Demographics

		Salt Lake City, UT, USA		Srinagar, Jammu & Kashmir, India		Huashixia, Qinghai Province, P. R. China
Altitude		1320 m		1730 m		4320 m
Subjects		5		5		9
Tibetans/non-Tibetans		5/0		5/0		5*/4
Male		2		2		8
Female		3		3		1
Age range		40–77		37–70		23–66

^*Additional 29 Tibetans from Huashixia, China and 25 Aymaras and 5 European Caucasians from La Paz and Tiwanaku (4100m) were also quantitated for fetal hemoglobin (HbF).

Blood sample collection

A 5ml ACD tube of venous blood was obtained from the antecubital vein in each subject, and a small aliquot was used for venous blood gas and complete blood count analysis. High pressure liquid chromatography (HPLC) was done for evaluation of hemoglobin variants and HbF.

Measurement of P50

The hemoglobin-oxygen dissociation and P50 are optimally derived by hemoximeter measurements of the percent saturation of hemoglobin at various partial pressures of oxygen. The resultant curve has a sigmoid shape due to the cooperative binding of oxygen to the four globins in the hemoglobin tetramer, enumerated as Hill coefficient “n”. If a hemoximeter is not available, the P50 can be estimated from the venous blood gas using the measured PO₂, oxygen percent saturation and pH using a formula described by Lichtman and colleagues¹³; however, the Hill coefficient “n” cannot be derived by this method.

For the subjects from Salt Lake City, USA the peripheral blood was evaluated by Hemox Analyzer (TCS Scientific Corporation, New Hope, PA) for obtaining P50 values and “n” Hill coefficients for hemoglobin oxygen binding. The normal range for P50 by Hemox Analyzer is 22–28 mmHg.

On subjects from Huashixia in Qinghai, China, venous blood gases were done on Nova pHOx (Nova Biomedical, Waltham, MA), and in Srinagar, India, on GEM Premier 3000 (Instrumentation Laboratory, Lexington, MA). P50 was derived using the formula below¹³. The normal range for P50 was 22.6 – 29.4 mmHg.

$$P_{50}std=\mathit{\text{antilog}}\frac{\mathit{\text{log}}(\frac{1}{k})}{n};\text{where}\frac{1}{\mathrm{k}}=[\text{antilog}(\mathrm{n}\text{log}{\mathrm{P}\mathrm{O}}_{2(7.4)}]•\frac{100-{\mathrm{S}\mathrm{O}}_2}{{\mathrm{S}\mathrm{O}}_2}$$

A Hill constant “n” for hemoglobin A of 2.7 was used. The PO₂ in venous blood at 37° C was converted to PO₂ at pH 7.4 with the formula:

$$\text{log}{\mathrm{P}\mathrm{O}}_{2(7.4)}=\text{log}{\mathrm{P}\mathrm{O}}_2-[0.5(7.40-\mathrm{p}\mathrm{H})]$$

where pH is measured from the antecubital venous blood¹³.

Results and Discussion

The P50 measured by Hemox Analyzer on Tibetan subjects from Salt Lake City, USA, and calculated by the formula described by Lichtman, et al¹³ on all subjects from differing altitudes were all within normal range, regardless of their arterial oxygenation measured by finger pulse oximeter (see Table 2). No hemoglobin variants were detected by HPLC in any subjects. HbF were less than than 1% in all 19 subjects, but one subject's HbF was 2.1%. HPLC was also done on additional 29 Tibetan (Huashixia 4320 m), 25 Aymara and 5 Caucasian Bolivian residents of La Paz (3800 m) and Tiwanaku (4100 m) and also revealed less than 1% HbF (mean HbF 0.45%, SD 0.14).

Table 2.

P50 and Hemoglobin results.

Location (Altitude)	Subject ID	P50 (mmHg)	Hill Co- off (n)	HbF %	HbA2 %	Hb (g/dl)
Salt Lake City, USA (1320m)	SL-1	25.16	2.83	2.1	2.3	13.5
	SL-2	22.50	2.89	0.2	2.6	13.7
	SL-3	24.06	2.87	0.2	2.5	13.7
	SL-4	24.28	2.83	0.3	2.8	12.0
	SL-5	22.35	2.82	0.3	2.4	13.3
Srinagar, India (1730m)	SR-1	26.38	-	0.5	2.3	15.3
	SR-2	25.95	-	0.6	2.1	12.5
	SR-3	26.55	-	0.6	2.2	11.8
	SR-4	23.68	-	0.7	1.8	14.7
	SR-5	22.72	-	0.8	2.1	11.9
Huashixia, China (4320m)	HX-1	25.99	-	0.6	2.3	19.3
	HX-2	25.75	-	0.5	2.7	16.3
	HX-3	25.75	-	0.3	2.5	13.8
	HX-4	25.74	-	0.7	2.4	15.2
	HX-5	25.80	-	0.3	2.4	15.2
Non- Tibetan Controls at Huashixia, China (4320m)	CNTRL-1	26.01	-	0.7	1.5	16.3
	CNTRL-2	26.00	-	0.5	2.2	17.8
	CNTRL-3	25.74	-	0.4	2.6	16.7
	CNTRL-4	25.42	-	0.3	2.6	15.3

SL: Salt Lake City; SR: Srinagar; HX: Huashixia; CNTRL: Controls; ND: Not Determined

The evolutionary selection of β-globin haplotype (HBB) has been reported in Tibetans¹² and also in high altitude deer mice¹⁴. However, when we analyzed the P50 of the 15 native Tibetan highlanders in differing altitudes, they were all within normal limits, and no mutant hemoglobins were found by HPLC in any of them including the additional 29 Tibetan subjects from high altitude. Thus, our data rule out that this selected HBB Tibetan haplotype is associated with the presence of a β-globin mutant.

Based on the published data of increase in hemoglobin oxygen affinity in high altitude animals as well as in Sherpas^1,2, an increase of HbF could account for potential benefits to Tibetan high-altitude adaptation and, indeed, a selection of HBG2 haplotype (encoding for a γ-globin subunit of HbF) was also reported in Tibetans¹². Further, promoter variants of the HBG2 locus have been reported in native Chilean Andes highlanders, causing delayed transition from HbF to adult hemoglobin¹⁵; however, we are not aware of any previous studies of dysregulated expression of HbF in Tibetans. As there are reports of increased HbF in individuals exposed to hypoxia¹⁶, we also quantitated HbF in 29 additional Tibetan subjects living at 4320 m, 25 Aymara and 5 Caucasian Andean residents and all of them had normal HbF. Our data are consistent with a recent report of normal HbF in subjects with Chuvash polycythemia who have aberrant hypoxia sensing due to an underlying von Hippel-Lindau gene mutation that leads to augmented hypoxia sensing at low altitudes¹⁷.

Thus, our study demonstrates that neither high hemoglobin-oxygen affinity, nor significant changes of 2, 3 DPG levels (as assessed by normal P50), nor elevated HbF are features of the adaptive phenotype of Tibetan highlanders. However we could not rule out that promoter variants of HBB or HBG2 haplotypes may cause delayed transition to adult hemoglobin in Tibetan children.

Small number of sample size is the major limitation in our study, mainly due to the logistic and technical issues related to the access to the remote areas in India and China.

Adaptation to a hypoxic environment involves a series of complex but integrated physiologic responses that are all aimed at ensuring a goal of adequate oxygenation at the cellular level. Although high hemoglobin-oxygen affinity is postulated to be at an advantage at high altitudes¹⁸, it is plausible that other physiologic adaptive mechanisms on the cellular level ensuring adequate oxygenation may offset the need for high hemoglobin-oxygen affinity. Clearly, further in-depth studies of the complex, unique phenotype of Tibetans’ high-attitude adaptation are needed.