A novel mechanism for high-altitude adaptation in hemoglobin of black-spotted frog (Pelophylax nigromaculatus)

Understanding how animals living in highland adapt to extreme conditions is critical to evolutionary biology. In contrast to birds and mammals, little information was available on the adaptation mechanisms for O2 transport in high-altitude ectothermic vertebrates. Here we report for the first time on hematological parameters, amino acid sequences of α and β chains of hemoglobin (Hb), O2 affinity of purified hemoglobins (Hbs) and their sensitivities to anion allosteric effector (H+, Cl−, ATP) and temperature in the high-altitude (2,292 m) black-spotted frogs (Pelophylax nigromaculatus) from the Qinghai-Tibet plateau (QTP) compared with the low-altitude (135 m) population. Our results showed that high-altitude black-spotted frogs exhibit significantly increased relative lung mass, hematocrit, and hemoglobin concentration, but significantly decreased body mass and erythrocyte volume, which could improve the blood O2 carrying capability. Compared with the low-altitude population, the purified Hbs of high-altitude black-spotted frogs possessed significantly higher intrinsic Hb-O2 affinity, similar low anion allosteric effector sensitivities, Bohr effects and temperature sensitivities. The elevated Hb-O2 affinity of highland frogs could maximize the O2 extraction from the lungs. Molecular dynamics simulations showed that the Gln123Glu substitution on α2 chain in highland frogs could form a hydrogen bond with 127Lys on α2 chain, resulting in the elimination of a hydrogen bond between 127Lys on α2 chain and 141Arg on α1 chain. This could weaken the interaction between two semirigid dimers (α1β1 and α2β2) and then lead to the high intrinsic Hb-O2 affinity in high-altitude black-spotted frogs.


Introduction
High-altitude environment was characterized by hypobaric hypoxia, high ultraviolet radiation and cold. Previous studies indicated that both ambient oxygen partial pressure ( P O 2 ) and temperature decreased with the increase in elevation (McClelland and Scott, 2019). Thus, a primary concern for high-altitude animals is how to keep life events under these great stresses. Extensive researches have shown that animals have adapted to high altitude in many aspects including gene (mutation, evolution and expression), protein (structural and function),

OPEN ACCESS EDITED BY
Liming Chang, Chengdu Institute of Biology (CAS), China hematological parameters, and organs (lung, eardrum, vomerine tooth among others) (Simonson et al., 2010;Storz et al., 2010;McClelland and Scott, 2019). Adaptive changes to enhance the ability of O 2 transport in hypobaric hypoxia for high-altitude animals may occur in hematological parameters and oxygenation properties of hemoglobin.
Hemoglobin, the major component of red blood cells, plays an indispensable role in oxygen transport in most vertebrates. The structure of vertebrate hemoglobin is a heterotetramer composed of two α chains (α 1 and α 2 ) and two β chains (β 1 and β 2 ). α 1 and α 2 chains have the same amino acid sequence, and β 1 and β 2 chains have the same amino acid sequence. They were numbered to facilitate the exploration of spatial structural properties of Hb. Heme is located in a hydrophobic pocket on each chain (α 1 , α 2 , β 1 , β 2 ). During O 2 binding, the two semi-rigid dimers (α 1 β 1 and α 2 β 2 ) glide to each other, resulting in the transition of the spatial conformation of Hb from the tense state (T state, low O 2 affinity) to the relaxed state (R state, high O 2 affinity). Many ectothermic vertebrates express multiple αand β-type globin genes, resulting in multiple or heterogeneous Hbs (isoHbs). Hb heterogeneity and 'polymorphism' (intraspecific differences in Hbs) provide a basis for a division of labor among component Hbs under different conditions (Weber and Jensen, 1988).
Animals adapted to high altitudes have successfully overcome the harsh environment, providing an important opportunity to look for insight into the molecular and physiological adaptation of hemoglobin. Many studies found that high-altitude natives evolved elevated Hb-O 2 affinity (Simonson et al., 2010;Qiu et al., 2012;Qu et al., 2013;Lorenzo et al., 2014). For example, the significantly increased intrinsic Hb-O 2 affinity of the bar-headed goose (Anser indicus) is largely caused by the α-119Pro → Ala mutation which weakens the interaction within the semi-rigid dimer (Jessen et al., 1991;Weber et al., 1993;Jendroszek et al., 2018;Natarajan et al., 2018). Weber et al. (2002) found that Andean frog (Telmatobius peruvianus) at 3,800 m exhibited higher O 2 affinity than low-altitude population which could be attributed to the acetylation modification of NH 2 -terminal Val at α-chain and the amino acid substitution of α131Thr → Ala. Pu et al. (2019) found that plateau zokor (Eospalax baileyi) Hbs possessed high intrinsic Hb-O 2 affinity compared to mice which might be related to the substitution of Ser by Asn at position 131 on the α 2 -chain. The adaptive adjustments of hematological traits and increased Hb-O 2 affinity were also found in high-altitude Asiatic toads (Bufo gargarizans) which could improve O 2 carrying capacity under hypoxia environment (Pu et al., 2021).
The evolved elevated Hb-O 2 affinity of high-altitude species could enhance the O 2 extraction capacity from the lungs, on the other hand, it could compromise O 2 unloading to the tissues in the systemic capillaries. The allosteric effects of organic phosphate, chloride, protons, and CO 2 on Hb-O 2 affinity could compensate for the side effect, which binds to specific sites and stabilize the low-affinity Hbs (T state). The main organic phosphates were 2,3-diphos-phoglycerate (DPG) for mammals, inositol pentaphosphate (IP5) for birds, and adenosine triphosphate (ATP) for reptiles. Due to the exothermic nature of hemoglobin oxygenation, an increase in temperature could decrease the Hb-O 2 affinity and favor O 2 unloading to warm exercising tissues and organs. However, this temperature effect would hinder the O 2 loading in the cold limbs for animal living in polar or alpine environments. As we know ectotherms lack the ability of temperature regulation, whether the low temperature sensitivities of Hbs founded in arctic or alpine mammals is true in alpine reptiles remains to be investigated (Weber and Campbell, 2011). However, the low temperature sensitivity of Hbs was reported in Asiatic toad (Pu et al., 2021).
Black-spotted frog (Pelophylax nigromaculatus) belonging to the genus Pelophylax in the family of Ranidae has a wide distribution in East Asia (Gong et al., 2013). Due to their widespread, rapid propagation and easily collected, black-spotted frogs have extensive value in academic, edible and medical aspects. Current research on the black-spotted frog mainly focus on toxicology, conservation biology, and immunology (Yu et al., 2020). Few studies were available on the high-altitude adaptation of this specie despite they can expand to the Qinghai-Tibetan Plateau (QTP).
This study aimed to explore whether blood O 2 carrying capability of high-altitude black-spotted frog undergoes adaptive adjustments compared with low-altitude population. We compared their hematological traits and oxygenation properties of purified Hbs, sequenced the subunit gene of hemoglobin and analyzed the structure properties of Hb tetramers. We hypothesized that there are obvious changes in hematological traits between high-and low-altitude population. Moreover, Hbs of high-altitude black-spotted frog may exhibit the genetic-based higher Hb-O 2 affinity to obtaining adequate O 2 under hypoxia environment compared with low-altitude population.

Animal and samples
High-altitude black-spotted frogs (N = 6) were captured at Guide County, Qinghai Province, China (36°10′87″N, 101°53′43″E, 2292 m). Low-altitude black-spotted frogs (N = 12) were captured at Ningyang County, Shandong Province, China (35°46′8″N, 116°48′12″E, 135 m). High-and low-altitude black-spotted frogs were collected in April and June 2019, respectively. Our experiment received ethics approval from the Ethics Committee of Animal Experiments at Lanzhou University and following guidelines from the China Council on Animal Care. The climate data of sampling site is available in the supplementary attachment (Supplementary Table S1). The climate data of sampling site comes from the national data center of the meteorological discipline in China. 1 Ether was used for anesthetizing all frogs. Blood was drawn from the ventricle by heparinized syringe after exposing heart. A portion of fresh blood was used to measure hematological parameters, another portion was washed three times with a threefold excess of 0.67% NaCl. Then the packed washed cells were dispensed (100 μL per tube) and immediately frozen in liquid nitrogen and stored at −80°C for subsequently purifying.

Relative lung mass and hematological parameters
Relative lung mass was calculated as lung weights/body mass in each sample. The hemiglobincyanide (HiCN) method was used for assessing hemoglobin concentration ([Hb]), which measured the absorbance of a mixture of 10 μL fresh blood and 2.5 mL Van Kampen-Zijlstra solution at the wavelength of 540 nm (Moharram et al., 2006). Red blood cell count (RBC) was measured using a hemocytometer with the mixture of 10 μL blood and 1.99 mL RBC diluent under microscope (Motic, China). Hematocrit (Hct) was measured as the proportion of red blood cells in whole blood by centrifuging fresh blood at 3,000 g for 10 min in microhematocrit capillaries. Mean corpuscular hemoglobin concentration (MCHC) was calculated as ([Hb]/Hct)*100, mean corpuscular volume (MCV) was calculated as Hct/RBC, and mean corpuscular hemoglobin (MCH) was calculated as [Hb]/RBC. The long and short diameters (a and b) of erythrocytes fixed on the dried blood smears were determined by the ocular micrometer on a microscope (SOPTOP, China). At least 20 erythrocytes were measured for each sample. Then, erythrocyte volume was calculated using formula of (4/3)πab 2 .

Hb purification and O 2 equilibrium curves
The frozen red blood cells were redissolved with the threefold volume of ice-cold 10 mmol/L Hepes (pH 7.8) and lysed on ice for 20 min. Then the obtained hemolysate was centrifuged at 4°C (9,000 g, 10 min) to remove membranes and cellular debris, and supernatants were injected into an IexCap Q 6FF 5-mL column (Smartlifesciences, China) on the Äkta Pure Chromatography System (GE Healthcare Life Sciences). After equilibrated with 20 mmol/L Tris⋅HCl (pH 8.4, 0.5 mmol/L EDTA), the supernatant was eluted with a linear gradient of 0-400 mmol/L NaCl at a flow rate of 1 mL/min to get the mixture of hemoglobin isoforms (Hbs) and remove residual endogenous phosphates (Rollema and Bauer, 1979;Bonaventura et al., 1999). The absorbance of the eluate was monitored at the wavelength of 280 nm. The collected Hbs were further desalted by dialysis against three changes of 200-fold volume of 10 mmol/L Hepes buffer (pH 7.6, 0.5 mmol/L EDTA) at 4°C. The final purified Hbs were concentrated to the concentration of ~1.3 mmol/L heme by ultrafiltration using Amicon ® Ultra-4 Centrifugal Filter Unit fitted with a 10-kDa cut-off filter (Millipore, Ireland) and stored at −80°C for subsequent experiments.
We measured the O 2 equilibrium curves of purified Hbs from high-and low-altitude Black-spotted frogs using a home-made modified diffusion chamber based on the method as described by Weber (1981Weber ( , 1992, Weber et al. (1993), and Damsgaard et al. (2013). The purified Hbs samples were dissolved in 0.2 mol/L Hepes buffers with pH of 7.0 and 7.5 in the absence (stripped) and presence of Cl − (adding to 0.1 mol/L KCl) and/or ATP (100-fold molar excess tetrameric Hbs). Then the absorbances of the prepared assay solution (3-5 μL) were measured at 436 nm to calculate the fractional O 2 saturations under corresponding O 2 tension ( P O 2 , mmHg) which was stepwise increased by mixing ultrapure N 2 and atmospheric air. The fractional saturation of O 2 ( S O 2 ) was calculated by S O 2 = (A−A 0 )/ (A 100 −A 0 ), where A 0 and A 100 were the absorbances of 0 and 100% O 2 saturation that was obtained by equilibrating with ultrapure N 2 and atmospheric air, respectively. SevenCompact pH/Ion Meter S220 (Mettler Toledo, Switzerland) equipped with an InLab micro pH electrode (Mettler Toledo, Switzerland) was used for measuring pH values of the prepared assay solution. Before each measurement, the electrode was calibrated by standard pH buffers and the temperature was measured using an ATC temperature probe (Mettler Toledo, Switzerland). Each condition was repeated six times at least.
Since saturation range≈0.1-0.9) were used to fit this equation by nonlinear sigmoidal Hill fitting (r 2 > 0.99) on the GraphPad Prism 9 (GraphPad Software, United States) and yielded P 50 and n 50 (means±SE) for each condition. P 50 is the P O 2 when S O 2 of hemoglobin is 50%, and n 50 is the Hill cooperativity coefficient. The Bohr effect (Riggs, 1988) describes the change in Hb-O 2 affinity that results from a change in H+ concentration and is often expressed numerically as ΔlogP 50 /ΔpH (Bohr factor Φ), where pH refers to that of the Hb solution. The Bohr factors were measured in conditions of 0.1 mol/L Hepes buffers in pH 7.0 and pH 7.5 and equal to the slope of linear plots of logP 50 as a function of pH. Globin sequencing, Hb homology modeling, and molecular dynamic simulation Considering close phylogenetic relationship, the gene sequences of α and β globin of the common frog [Rana temporaria, LOCUS: α (XM_040357009.1), β (NC_053494.1)] were used for designing primers to amplified the coding sequence of α and β goblins of highand low-altitude black-spotted frogs. For each population, samples (N = 5) were collected from liver and used for total DNA extraction (Rapid Animal Genomic DNA Isolation kit, Sangon Biotech, China) and RNA extraction [Steadypure Quick RNA Extraction Kit, Accurate Biotechnology (Hunan) Co., Ltd., Changsha, China], followed by cDNA reverse transcription [Evo M-MLV Kit, Accurate Biotechnology (Hunan) Co., Ltd., Changsha, China]. Using PCR protocol (Taq polymerase, Fermentas, Burlington, ON, Canada) with one pair primer, we had successfully amplified the sequence of mRNA α and β, but the mRNA sequence of β gene was not complete. Since it is difficult to design suitable primer, we amplified β gene of black spotted frog from total DNA with two pair segmented primers (Table 1) and then β gene sequence assembly and translation was performed with software (DNAMAN 6.0.3.40, United States). PCR protocol: 1 cycle of 5 min at 94°C, 35 cycles of 30 s at 94°C, 30 s at annealing temperature, and 60 s at 72°C, and one final cycle of 10 min at 72°C. The obtained sequence of β gene identical matches both the partial CDS sequence that we obtained from cDNA and the mRNA sequence identified in the published liver transcriptome data of black spotted frogs (GSE132282_unigene: TRINITY_DN76909_c0_g1). However, another type of β gene identified in the common frog cannot be cloned Frontiers in Ecology and Evolution 04 frontiersin.org from mRNA (cDNA) in the liver of black-spotted frogs, and it was also not identified in the liver transcriptome data of black-spotted frogs. So, we had successfully amplified the CDS sequence of one type of α and β gene, respectively, and speculated that there might be only one type of α and β gene expressed in the black spotted frog. The sequencing data have been uploaded to GenBank (highaltitude α: OQ079690; high-altitude β: OQ079692; low-altitude α: OQ079691; low-altitude β: OQ079693). We obtained the total length of these genes. Since the methionine (Met), encoded by the start codon, is not involved in the formation of the final polypeptide, the amino acid sequences of α and β chains are numbered from the first amino acid after Met. Then, SWISS-MODLE online software 2 was used to construct the tetramer structures of deoxy Hb for high-and low-altitude frogs using the Side-necked turtle Hb structure (PDB ID: 3at5) in the T state as the template. The amino acid sequence of the template was 55.24% similar to the Hb of both high-and low-altitude black-spotted frogs. Molecular dynamics simulations (MD) were further performed to explore the quaternary structural properties of deoxy Hb under physiological conditions using the NAMD and VMD software (Humphrey et al., 1996;Phillips et al., 2005). The psfgen plugins of VMD were used for adding missing hydrogen atoms to homology models, then, which were immersed in equilibrated TIP3P water boxes with physiological salt concentrations (0.12 mol/L NaCl). NAMD was performed for energy minimization and molecular dynamics simulations using CHARMM version c35b2 with the all-atom 27 protein force field. And there were some constraint conditions in each simulation: p = 1 atm, t = 310 K and PME (Particle-Mesh-Ewald) method for calculating the electrostatic interactions. After 50 ns MD simulation, when the root mean square deviation (RMSD) of the structure showed good convergence, we analyzed the structural features of tetrameric Hb using VMD and its plugin.

Statistical analyses
Data were tested for normality and homogeneity of the variance using SPSS v16.0 (SPSS Corp., IL, United States). The significance of each result between high-and low-altitude frogs was analyzed by ANOVA. Data were presented as means ± standard error (SE) and statistical significance was accepted at p < 0.05.

Relative lung mass and hematological parameters
The results of relative lung mass, body mass and hematological parameters of high-altitude and low-altitude black-spotted frogs are presented in Table 2. High-altitude black-spotted frogs exhibited significantly higher values in M L (relative lung mass), Hct, [Hb], MCH, and MCHC (p < 0.05), but significantly lower value in erythrocyte volume (p < 0.05) and body mass (p < 0.001) compared with low-altitude population.

Oxygenation properties of frog Hbs
Purified Hbs of high-altitude black-spotted frogs exhibited significantly higher overall and intrinsic O 2 affinity compared with the low-altitude population as the P 50 values of the high-altitude frog exhibited significantly lower values than the low-altitude population under every same experimental condition (Table 3). The n 50 values of Hbs in both high-and low-altitude populations were higher than 2.49 in different experimental conditions, which means that O 2 binding was positively cooperative (Table 3).
Sensitivities of ATP and/or Cl − of purified Hbs were indexed by the difference between the logP 50 in the absence (stripped) and the presence of ATP and/or Cl − in both high-and low-altitude blackspotted frogs (Table 4). Compared with the stripped state, O 2 equilibrium curves shifted to right and the Hb-O 2 affinities were significantly reduced in the presence of Cl − and/or ATP for both highand low-altitude populations. Furthermore, the effect of ATP on Hb-O 2 affinities was greater than Cl − in both high-and low-altitude frogs, and the reduced impact reached the maximum when ATP and Cl − exist simultaneously. Sensitivities of ATP and/or Cl − of purified Hbs in high-altitude population were slightly greater than low-altitude populations with one exception (at 20°C, pH 7.0, Table 4).
Hb-O 2 affinity was increased and the O 2 equilibrium curves move to right in both high-and low-altitude frogs when the pH value  increased from 7.0 to 7.5 as the P 50 values significantly decreased ( Figure 1A; Table 3). Bohr factor (Φ = ΔlogP 50 /ΔpH) is equal to the slope of linear curve of logP 50 versus pH in the range of pH 7.0-7.5 (Figures 1C,D; Table 5). Under all conditions, high-and low-altitude black-spotted frog Hbs exhibited roughly similar Bohr factors. Furthermore, when the temperature raised from 10°C to 20°C, the O 2 equilibrium curves move to right and Hb-O 2 affinity was reduced in both high-and low-altitude frogs ( Figure 1B). Compared with low-altitude black-spotted frog Hbs, the high-altitude black-spotted frog Hbs exhibited little difference in temperature sensitivity (enthalpy of oxygenation, ΔH) under the same conditions (Table 5).

Amino acid sequences alignment and molecular dynamic simulation
We had successfully obtained one type of CDS sequence of α and β gene. After translation, the amino acid sequences of α and β chains were presented in Figure 2A. There are two amino acid substitutions of Hb on α chain (Met63Val and Gln123Glu) and one amino acid substitution on β chain (Gln76His) between high-and low-altitude black-spotted frogs (Figure 2A). The backbone RMSD ≤0.1 Å between the template and the constructed Hb model was considered usable.
After evaluating the accuracies of constructed Hb models for highand low-altitude black-spotted frogs, we performed the molecular dynamic simulation to predict the stable conformations of the Hb models. Our results of the molecular dynamic simulation showed that the RMSD of the backbone atom exhibited excellent convergence (about 3.0-4.0 Å) after simulating 30 ns (Supplementary Figure S1). Then we analyzed the interaction between semirigid dimers in the final 20 ns of the trajectories. The mutation was changed from amide group (-CONH 2 ) to carboxyl (-COOH) at the site of 123 on α 2 chain and the properties of amino acid were changed from uncharged polar to negatively charged. It causes an extra hydrogen bond formation between α 2 -123-Glu (oxygen atom of side-chain carboxyl group) and α 2 -127-Lys (hydrogen atom of side-chain amino group) in highaltitude black-spotted frogs Hb ( Figure 2B). This hydrogen bond causes the change of the spatial orientation of α 2 -127-Lys side chain, increase the distance between α 2 -127-Lys and α 1 -141-Arg and decrease the likelihood of hydrogen bond formation. Thus, it found that an additional hydrogen bond between α 1 -141-Arg (oxygen atom of carboxyl-terminus) and α 2 -127-Lys (hydrogen atom of side-chain amino group) presented in low-altitude black-spotted frogs Hb ( Figure 2C), but did not exist in high-altitude black-spotted frogs Hb ( Figure 2B). In addition, the substitutions of Met63Val on α chain and Gln76His on β chain in the high-altitude frogs do not affect hydrogen bond interaction between dimers based on our analysis of MD results.

Discussion
The adaptive adjustments in the O 2 transport system of highaltitude natives occurred in several steps including O 2 utilization in mitochondria, intrinsic Hb-O 2 affinity, cardiorespiratory system and O 2 diffusion and circulation in tissue (McClelland and Scott, 2019). Our research investigated the adaptation mechanisms in the O 2 transport system of high-altitude black-spotted frogs dwelling on the QTP from two aspects: phenotypic plasticity and genetic change.
In terms of phenotypic plasticity, we found that M L (relative lung mass), Hct, [Hb], MCH and MCHC increased, but erythrocyte volume and body mass decreased in high-altitude black-spotted frogs   Frontiers in Ecology and Evolution 06 frontiersin.org (Table 2). The higher [Hb] of high-altitude black-spotted frogs could improve the O 2 carrying capacity of the blood. Similar phenomenon was also found in high-altitude Asiatic toads (Bufo gargarizans) and lake titicaca frogs (Telmatobius culeus) (Hutchison et al., 1976;Pu et al., 2021). The Hct of high-altitude black-spotted frog was higher than low-altitude black-spotted frog. The moderate increase in the Hct could allow the blood to transport more O 2 to aerobically tissues (Guyton and Richardson, 1961). Similar high Hct were also found in Bufo spinulosus limensis (39.53 ± 8.92%), high-altitude Asiatic toads (41.87 ± 0.92%), male Bufo bufo (42.4 ± 1.2%) (Ostojic et al., 2000;Donmez et al., 2009;Pu et al., 2021). On the other hand, high-altitude black-spotted frogs had smaller erythrocyte volume, which was presumed to increase the relative surface area of RBC and minimize the adverse effects of increased blood viscosity (Dunlap, 2006). There was a contradiction between MCV and erythrocyte volume in our results. This may be attributed to the low plasma volume of highaltitude frogs. Lower plasma could lead to the higher Hct, consequently, lead to the higher MCV of the high-altitude frogs although its measured erythrocyte volume was lower than the low-altitude frogs. The speculated low plasma volume of high-altitude frogs may be related to the lower humidity at Guide compared to Ningyang (Supplementary Table S1). However, more experiments about the plasma volume is needed to prove these deductions. Body mass is inversely proportional to surface-area-to-volume ratio, which constrains transport rates for water and respiratory gases across the skin. The smaller frogs have larger contact area with the air and are Representative O 2 equilibrium curves of high-and low-altitude black spotted frogs Hbs in pH7.0 and pH7.5 at 20°C (A), and at 10°C and 20°C in pH 7.0 (B) in the presence of Cl − and ATP. Bohr plots [logP 50 vs. pH] of high-altitude (C) and low-altitude (D) black-spotted frogs Hbs in the absence (stripped) or/and presence of Cl − and ATP at 20°C. The slope of each of these linear plots is Bohr factor (Φ) reported in Table 4. The sample sizes of high-and low-altitude black spotted frogs are 6 and 12, respectively. Each experiment has been carried out six times. Frontiers in Ecology and Evolution 07 frontiersin.org expected to obtain a higher proportion of oxygen through the skin. Hence, the lower body mass of the high-altitude black-spotted frog may be a strategy for high-altitude adaptation (Toledo and Jared, 1993). Furthermore, our results showed that there was no significant difference in RBC between high and low-altitude black-spotted frogs.
This was inconsistent with previous studies. High-altitude Asiatic toads, Lizards (Phrynocephalus erythrurus), and domesticated yaks (Bos grunniens) possess higher RBC than low-altitude populations (Ding et al., 2014;Lu et al., 2015;Pu et al., 2021). So, we caution against the over interpretation of this finding possibly due to our small sample size. Amino acid substitution of Gln123Glu on α 2 chain affects the spatial structure of Hb of black-spotted frog. Amino acid sequence alignments of α-and β-chains in hemoglobin of high-and low-altitude black spotted frog (A). The constructed spatial structure of Hb in high-altitude black-spotted frog (B). The constructed spatial structure of Hb in low-altitude black-spotted frog (C). Substitution of Gln to Glu at position 123 on α 2 chain of high-altitude frog could form an additional hydrogen bond with 127Lys on the same chain, resulting in the change of Hb conformation and the loss of hydrogen bond interaction between α 1 and α 2 chain compared with the low-altitude frog. Red dashed lines indicated the hydrogen bonds formed between 127Lys on α 2 chain and 141Arg on α 1 chain in the Hb of low-altitude frog. Each experiment has been carried out three times.
Frontiers in Ecology and Evolution 09 frontiersin.org environments, and it is a valuable reference for adaptive study in other high-altitude ectothermic amphibians.

Data availability statement
The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

Ethics statement
The animal study was reviewed and approved by the Ethics Committee of Animal Experiments at Lanzhou University.