Abstract
Objective
The crucial role of haemoglobin in endurance performance has been well documented. We examined whether polymorphisms in the HBB gene modified aerobic capacity.
Methods
102 recruits were trained by running 5000 m three times per week for 18 weeks. Exercise intensity progressively increased from an initial heart rate corresponding to 95% of the individual baseline ventilatory threshold during the first 10 weeks to 105% during the last 8 weeks. The phenotypes measured were running economy and VO2max. Running economy was determined by measuring submaximal VO2 for 5 min at a constant running speed of 12 km·h−1 and VO2max was obtained during an incremental test to exhaustion. Genomic DNA was extracted from white cells of peripheral blood and the −551C/T, intron2,+16C/G and +340 A/T genotypes were examined relative to the TAA site variants by PCR‐RFLP.
Results
Genotype distributions were in Hardy‐Weinberg equilibrium at three loci. None of the running economy and VO2max‐related traits were associated with the three polymorphisms or haplotypes at baseline, while the training response of running economy was associated with −551C/T and intron2,+16C/G polymorphisms. Subjects homozygous for intron2,+16C/C or −551C/C had decreased oxygen cost of running compared to the other individuals.
Discussion
It was concluded that the −551C/C or intron2,+16C/C genotype might explain part of the individual variation in the cardiorespiratory adaptation to endurance training.
Keywords: haplotype, polymorphism, running economy, training, VO2max
Haemoglobin is a protein‐based component of red blood cells and is primarily responsible for transferring oxygen from the lungs to respiring tissues. The crucial role of haemoglobin in exercise has been well documented.1,2,3,4 An acute reduction in blood haemoglobin concentration, even when the circulating blood volume is maintained, results in lower maximal oxygen uptake (VO2max) and endurance performance due to the reduction in the oxygen carrying capacity of the blood.4 Conversely, an increase in haemoglobin concentration is associated with enhanced VO2max and endurance capacity, which is also proportional to the increase in the oxygen carrying capacity of the blood; the effects on endurance capacity appear more pronounced and prolonged than on VO2max.1 Compared to VO2max, less research has focused on studying the factors that determine running economy, which is one of the main indicators of submaximal aerobic capacity. Recent investigations nevertheless suggest that some strategies originally intended to enhance blood oxygen transport capacity (for example, intermittent hypoxic exposure) might also improve running economy.5 On the other hand, the VO2 cost of leg pedalling exercise decreases with increasing blood haemoglobin concentration, at least in elite endurance cyclists.6 Since haemoglobin is an excellent buffer, an increase in the circulating haemoglobin mass helps improve the blood buffering capacity.7 A higher blood buffering capacity could facilitate lactate8 and proton release from the active muscle and increase the capacity to produce energy via the anaerobic pathways.
The haemoglobin molecule has been proposed as an ideal O2 sensor.9,10,11 The O2 sensing properties of the haemoglobin molecule appear to involve conformational changes resulting from O2 desaturation.12 On desaturation, red blood cells could elicit vasodilatation through four independent mechanisms: release of adenosine 5′‐triphosphate (ATP),13 release of S‐nitrosylated molecules originally bound to β93 cysteine residues of oxyhaemoglobin,14,15 deoxygenated haemoglobin acting as a nitrite reductase and conversion to nitric oxide via the reaction with haemoglobin at the site of ferric (Fe3+) and ferrous (Fe2+) haemoglobin.16,17,18,19
Normal adult haemoglobin (molecular weight, 64 500) consists of two α and two β polypeptide chains, each bound to a haeme group. The human α‐globin gene cluster is located on chromosome 16 (16p13.3), spans about 30 kb and includes seven loci: 5′‐ζ – pseudo ζ – μ – pseudo α‐1 – α‐2 – α‐1 – θ‐3′. The α‐2 (HBA2) and α‐1 (HBA1) coding sequences are identical. The order of the genes in the β‐globin cluster is 5′‐ε – γ‐G – γ‐A – δ – β‐3′, which are located on 11p15.5. These genes are arranged in the order of their expression during development. The α (HBA) and β (HBB) loci determine the structure of the two types of polypeptide chains in adult haemoglobin (Hb). Two α chains plus two β chains constitute HbA, which in the normal adult constitutes about 97% of the total haemoglobin. α Chains combine with δ chains to form HbA‐2, which with HbF (fetal haemoglobin) makes up the remaining 3% of adult haemoglobin. Data from the Heritage Family Study suggest that linkages at p values of 0.01 and better are observed with markers on 4q, 8q, 11p15.1 and 14q for VO2max before training and with markers on 1p, 2p, 4q, 6p and 11p14 for the change in VO2max in response to a 20 week standardised endurance training program.20 In this regard, the HBB gene might be viewed as a functional and positional candidate gene for endurance capacity. To the best of our knowledge, however, no information is available concerning the association between HBB gene polymorphisms and aerobic capacity. Therefore, the principal aim of this study was to examine the genetic association between the HBB gene and VO2max and running economy in an untrained state (baseline) and in response to aerobic training (changes) in a young Chinese Han population.
Methods
Subjects
Written informed consent was obtained from each subject prior to commencement of the training regimen. The study was approved by the local institutional ethics committee and was carried out in accordance with the ethical standards laid down in the 2004 Declaration of Helsinki. A total of 181 biologically unrelated male recruits from a local military police force volunteered to participate in our study and filled in a questionnaire designed by a professor experienced in statistics. The recruits were healthy, untrained and of Chinese Han origin in Northern China with similar lifestyles; untrained status was defined as not having engaged in regular vigorous physical activity in the past 2 years (less than two 20 (or more) minute sessions/week of vigorous exercise (running, cycling, or swimming)). The questionnaire also sought background information (for example, nationality, birth place, lifestyle, physical activity, smoking, blood glucose level, and blood lipids, etc). However, only 102 subjects completed both pre‐ and post‐training tests (see below) and the entire exercise training program. Their physical characteristics were (mean±SD): age 18.8±0.9 years, height 171.7±5.8 cm, and body mass 60.3±6.5 kg. No subjects took any medications before or during the program. During the program, all subjects followed the same type of diet and the same daily schedule in terms of training and rest hours. Apart from this program, the subjects were only engaged in military training without other endurance training.
Endurance training program
Subjects ran 5000 m on a 400 m track three times per week for 18 weeks and were supervised by an exercise specialist. Subjects underwent adaptive training for 2 weeks prior to formal training. The intensity of the exercise progressively increased from an initial heart rate corresponding to 95% of the individual baseline ventilatory threshold during the first 10 weeks to 105% during the last 8 weeks. To maintain target heart rate, subjects wore a heart rate monitor (S810 Polar; Polar, Kempele, Finland) during each running session.
Running economy, VO2max and ventilatory threshold measurements
Subjects undertook a battery of anthropometric and physiological tests prior to and after the 18 week exercise training program. Body composition was determined by anthropometric measurement combined with bioelectrical impedance analysis (BIA‐Tanita; Tanita, Tokyo, Japan). Running economy was determined by measuring submaximal VO2 for a 5 min period while running on a treadmill (Treadmill E6; Jaeger, Hoechberg, Germany) at a constant speed (12 km·h−1).21 Gas exchange data were continuously measured with an automated open‐circuit system (Mijnhault Oxycon Champion; Jaeger). VO2, pulmonary ventilation, respiratory exchange ratio (RER) and heart rate were measured during the running economy test. VO2max was measured following the running economy test and during an incremental test. Starting at 12 km·h−1 (0.5% incline), treadmill speed and incline were increased every minute by 0.5 km·h−1 and 2%, respectively, until subjects' volitional exhaustion. VO2max was reached if the two following criteria were met simultaneously20: (a) RER ⩾1.10; (b) change in VO2 ⩽5% or 100 ml·min−1 in the last three 30 s intervals. Running economy and VO2max were expressed in l·min−1, ml·min−1·total body mass (TBM) −1 in kg, and ml·min−1·lean body mass (LBM) −1 in kg. The ventilatory threshold was concurrently determined by three validated methods22: (a) the ventilatory equivalent method; (b) the excess carbon dioxide method; and (c) the modified V‐slope method using 30 s averaged data. Visual evaluation to determine ventilatory threshold was carried out independently by two experienced investigators using these three methods. Training response was calculated as [Δ = (post‐training minus baseline)/baseline×100%]. The reliability of the techniques used for LBM, VO2max and running economy determination have been previously corroborated by our group, with intra‐class correlation for repeated measurements ranging between 0.93 (LBM) and 0.97 (submaximal VO2) (p<0.001).
Genotype classifications
Briefly, genomic DNA was extracted from white cells of peripheral blood using the Wizard Genomic DNA Purification Kit (Promega, Madison, Wisconsin, USA). −551C/T, intron2,+16C/G and +340 A/T relative to the TAA site were selected. The −551C/T polymorphism is in the region which functions as a silencer in transient expression.23,24 Intron2,+16C/G is in the alternative splicing region of mRNA, which might lead to differential expression. +340 A/T is in the 3′‐untranslated region (3′‐UTR) which is critical for normal β‐globin mRNA stability.25,26,27,28 The primers were designed by Primer 5.0 software based on the sequence NM_000518. The primer sequence and polymerase chain reaction‐restricted fragment length polymorphism (PCR‐RFLP) conditions are shown in table 1.
Table 1 Primers and PCR‐RFLP conditions for polymorphism analysis of the HBB gene.
| RefSNP ID | Primers (5′→3′) | Position | Variation | Tm | MgCl2 | Restriction enzyme |
|---|---|---|---|---|---|---|
| Up: ctt tgg gtt gta agt ga | −551, | C/T | 60°C | 2.5 mM | RsaI | |
| Down: ttg gga tat gta gat gg | ||||||
| Rs10768683 | Up: tta ggc tgc tgg tgg tc | +16, intron2 | C/G | 58°C | 2 mM | AvaII |
| Down: caa tca ttc gtc tgt ttc c | ||||||
| Up: tta ggc tgc tgg tgg tc | +340 relative to the TAA site, 3′‐UTR | A/T | 56°C | 2.5 mM | HinfI | |
| Down: caa tca ttc gtc tgt ttc c |
Statistical analyses
Data were expressed as mean±SD. A χ2 analysis was employed to confirm that the observed genotype frequencies were in Hardy‐Weinberg equilibrium. The associations between endurance performance phenotypes and the three genetic markers were analysed by ANOVA using SPSS version 11.5 (SPSS Inc., Chicago, Illinois, USA). Lewontin's D′ and r2 were estimated using SHEsis software.29 Haplotype associations were estimated using Haplo.Score.30 The post hoc power analysis was computed by G*Power.31
Results
After training, the oxygen cost of running decreased from 2.69±0.26 to 2.48±0.25 l·min−1 (p<0.001) with the per cent change ranging from −25.0% to 8.0%. The VO2max increased from 3.44±0.38 to 3.49±0.40 l·min−1 (p<0.05) with the percent change ranging from −13.6% to 24.1%.
Genotype distributions were in Hardy‐Weinberg equilibrium at three loci (p>0.05). By pair‐wise linkage analysis, the Lewontin's D′ and r2 were 0.737 and 0.513 between −551C/T and intron2,+16C/G, 0.670 and 0.231 between −551C/T and +340A/T, and 0.933 and 0.474 between intron2,+16C/G and +340A/T, respectively. When the three polymorphisms were considered together, four haplotypes were identified (>5%): T‐C551T‐C+C16G‐A+A340T (30%), C‐C551T‐C+C16G‐T+A340T (13%), C‐C551T‐G+C16G‐T+A340T (8%), and T‐C551T‐G+C16G‐T+A340T (41%).
Table 2 displays running economy‐related traits stratified by genotype at baseline and their response to training. None of the three polymorphisms was significantly associated with baseline running economy‐related traits. Nevertheless, the training response of running economy was associated with −551C/T and intron2,+16C/G polymorphisms, and individuals homozygous for intron2,+16C/C or −551C/C exhibited a lower cost of running (when VO2 was expressed as ml·min−1·TBM−1 in kg and ml·min−1·LBM−1 in kg), that is, better running economy, than the other subjects. In addition, there was no significant association with haplotypes at baseline or trainability.
Table 2 Running economy at baseline and response to training according to HBB genotype.
| −551C/T | Intron2,16C/G | +340A/T in 3′‐UTR | |||||||
|---|---|---|---|---|---|---|---|---|---|
| CC (n = 28) | TC (n = 48) | TT (n = 26) | GG (n = 26) | CG (n = 51) | CC (n = 25) | TT (n = 41) | AT (n = 49) | AA (n = 12) | |
| Baseline | |||||||||
| RE(HR) (bpm) | 168±9 | 167±10 | 169±9 | 171±9 | 165±10 | 169±10 | 169±10 | 166±9 | 169±10 |
| RE(VE) (l·min−1) | 75.14±9.61 | 75.69±12.08 | 76.99±7.29 | 77.10±8.47 | 73.22±10.19 | 80.00±11.03 | 78.00±8.69 | 73.85±11.20 | 76.83±10.83 |
| RE(VO2) (l·min−1) | 2.74±2.34 | 2.64±2.98 | 2.71±2.14 | 2.69±0.20 | 2.61±0.26 | 2.82±0.28 | 2.70±2.17 | 2.66±3.05 | 2.74±0.23 |
| RE(VO2·TBM−1) | 45.13±3.91 | 44.41±3.81 | 44.90±2.54 | 45.10±3.15 | 44.22±3.26 | 45.39±4.39 | 44.57±3.07 | 44.82±3.65 | 44.93±4.75 |
| (ml·min−1·TBM−1 in kg) | |||||||||
| RE(VO2·LBM−1) | 72.53±5.74 | 71.26±5.69 | 72.25±4.19 | 72.81±4.67 | 70.92±4.90 | 72.80±6.60 | 71.96±4.55 | 71.76±5.38 | 71.95±7.79 |
| (ml·min−1·LBM−1 in kg) | |||||||||
| Training response (%) | |||||||||
| ΔRE(HR) | −4.38±3.40 | −2.17±6.55 | −2.73±4.21 | −3.66±4.00 | −1.78±6.50 | −4.49±2.91* | −2.86±5.72 | −2.58±5.31 | −4.53±3.80 |
| ΔRE(VE) | −15.27±8.27 | −13.52±12.00 | −14.50±9.21 | −15.76±10.84 | −11.98±10.63 | −17.29±8.30* | −15.60±9.87 | −12.51±11.34 | −16.74±6.20 |
| ΔRE(VO2) | −8.69±5.89 | −6.80±6.91 | −7.72±5.83 | −8.68±6.08 | −6.34±6.59 | −8.86±5.94 | −7.48±6.22 | −7.06±6.42 | −9.83±6.69 |
| ΔRE(VO2·TBM−1) | −11.40±5.58* | −8.38±6.55 | −9.08±5.30 | −9.59±6.53 | −8.20±5.81 | −11.49±5.75* | −8.60±6.35 | −9.24±5.61 | −12.27±6.60 |
| ΔRE(VO2·LBM−1) | −7.14±3.56* | −5.24±4.09 | −5.64±3.28 | −5.94±4.05 | −5.15±3.64 | −7.17±3.67* | −5.35±3.94 | −5.80±3.55 | −7.67±4.14 |
*p<0.05.
LBM, lean body mass; RE, running economy; ΔRE(HR), the training response of heart rate at RE; ΔRE(VO2), the training response of VO2 (l·min−1) at RE; ΔRE(VO2·LBM−1), the training response of VO2 (ml·min−1·LBM−1 in kg) at RE; ΔRE(VO2·TBM−1), the training response of VO2 (ml·min−1·TBM−1 in kg) at RE; RE(HR), heart rate at RE; RE(VE), ventilatory threshold at RE; RE(VO2), VO2 at RE expressed as l·min−1; RE(VO2·LBM−1), VO2 at RE expressed as ml·min−1·LBM−1 in kg; RE(VO2·TBM−1), VO2 at RE expressed as ml·min−1·TBM−1 in kg; TBM, total body mass.
The VO2max data are shown in table 3 at baseline and their response to training; no significant difference was observed among different genotypes or haplotypes.
Table 3 VO2max at baseline according to HBB genotypes.
| −551C/T | Intron2,+16C/G | +340A/T in 3′‐UTR | |||||||
|---|---|---|---|---|---|---|---|---|---|
| CC (n = 28) | TC (n = 48) | TT (n = 26) | GG (n = 26) | CG (n = 51) | CC (n = 25) | TT (n = 41) | AT (n = 49) | AA (n = 12) | |
| Baseline | |||||||||
| VO2max (l·min−1) | 3.53±0.35 | 3.40±0.41 | 3.43±0.36 | 3.39±3.60 | 3.40±3.63 | 3.57±4.27 | 3.44±3.75 | 3.42±3.85 | 3.51±4.16 |
| VO2max·TBM−1 (ml·min−1·TBM−1 in kg) | 57.69±4.26 | 57.10±4.26 | 56.59±3.93 | 56.56±3.82 | 57.44±4.20 | 57.10±4.51 | 56.50±3.82 | 57.58±4.18 | 57.48±5.20 |
| VO2max·LBM−1 (ml·min−1·LBM−1 in kg) | 93.02±7.10 | 91.64±6.70 | 91.45±7.43 | 91.55±6.89 | 92.28±6.68 | 91.78±7.81 | 91.44±6.53 | 92.40±6.78 | 92.01±9.36 |
| Training response (%) | |||||||||
| ΔVO2max (l·min−1) | 2.73±6.91 | 0.91±6.01 | 1.25±9.24 | 0.49±7.96 | 1.55±7.10 | 2.43±6.57 | 1.55±7.94 | 1.19±6.57 | 2.55±7.19 |
| ΔVO2max·TBM−1 (ml·min−1·TBM−1 in kg) | −2.29±6.56 | −2.84±6.46 | −2.98±7.54 | −2.66±7.76 | −3.02±6.19 | −2.18±6.83 | −1.94±8.05 | −3.62±5.14 | −1.71±7.47 |
| ΔVO2max·LBM−1 (ml·min−1·LBM−1 in kg) | 1.04±5.89 | 1.03±6.38 | 1.06±8.57 | 0.87±8.03 | 0.97±6.67 | 1.37±5.97 | 1.65±8.07 | 0.27±5.75 | 2.03±6.38 |
*p<0.05.
LBM, lean body mass; TBM, total body mass; VO2max, maximal oxygen uptake; VO2max·LBM−1, VO2max expressed as ml·min−1·LBM−1 in kg; VO2max·TBM−1, VO2max expresses as ml·min−1·TBM−1 in kg. VO2max (l·min−1), the training response of VO2max (l·min−1); ΔVO2max·LBM−1 (ml·min−1·LBM−1 in kg), the training response of VO2max·LBM−1 (ml·min−1·LBM−1 in kg); ΔVO2max·TBM−1 (ml·min−1·TBM−1 in kg), the training response of VO2max·TBM−1 (ml·min−1·TBM−1 in kg).
Discussion
The noteworthy finding from this study was that the −551C/T and intron2,+16C/G polymorphisms in the HBB gene were associated with aerobic capacity at submaximal intensities, that is, running economy. Cardiorespiratory fitness (which is often quantified through VO2max determination) is as strong a predictor of mortality as other conventional risk factors such as hypercholesterolaemia, cigarette smoking and hypertension.32,33 However, most daily activities are executed at submaximal exercise intensities, and thus submaximal aerobic capacity is an important variable. Running economy is typically defined as the energy demand for a given velocity of submaximal intensity.5 In our study, the −551C/T and intron2,+16C/G polymorphisms in the HBB gene were associated with running economy training response. Only one study has investigated the heritability of running economy at running speed of 5, 6, 7 and 8 km·h−1 by using the monozygotic and dizygotic twin method in an untrained state, but the researchers did not detect any genetic component in running economy (nor in VO2max relative to body mass).34 The heritability of VO2max, on the other hand, has been well documented in previous research.20,35 Therefore, it was postulated that the report of Rodas et al was questionable due, possibly, to the limited sample size.34 Moreover, these alleles might contribute to the trainability of running economy but do not appear to influence running economy in the untrained state. The HBB gene would be a new gene associated with the training response of running economy.
The molecular mechanism underlying the association observed here remains to be elucidated. −551C/T is in strict linkage disequilibrium with (AT)x(T)y between nucleotides −610 and −490 relative to the cap site. The (AT)x(T)y is associated with the binding capacity of the β protein (BP1), which is a negative regulatory transacting factor, so −551C/T may be involved in negative regulation.23,24 The −551T/T motif might bind the putative BPI repressor more strongly than the −551C/T and −551C/C motifs did. Intron2,+16C/G is in the splicing region. The possible significance of intron2,+16C/G was predicted by NNSPLICE and SpliceView computer programs.36,37 Intron2+16G, but not intron2+16C, might contain two possible branch sites, and the new branch site might induce a truncated protein. Hence, individuals who are homozygous for −551C/C or intron2,+16C/C might have a higher haemoglobin concentration than normal. In addition, the HBB gene is located close to the sulfonylurea receptor (SUR) gene and the voltage‐gated potassium channel gene (KCNA11).20 SUR is expressed in pancreatic β‐cells where it forms ATP sensitive potassium channels together with KCNA11, and is thereby involved in the regulation of insulin secretion. KCNA11 is expressed in several tissues, including heart and skeletal muscle, where it plays a role in the coupling of cell metabolism to membrane potential. Thus, the −551C/T and intron2,+16C/G polymorphisms might be in linkage disequilibrium with some other functional polymorphisms and/or might be one of the accumulated genes responsible for aerobic capacity at submaximal intensity in humans. An obvious next step is comprehensive sequencing of this region and functional studies to clarify how the HBB gene polymorphisms contribute to aerobic capacity.
What is already known on this topic
The crucial role of haemoglobin in exercise has been well documented.
The blood haemoglobin concentration is related to maximal oxygen uptake (VO2max) and endurance performance.
New information suggests that the haemoglobin molecule is an ideal O2 sensor.
What this study adds
No information is available concerning the association between HBB gene polymorphisms and endurance performance.
Our data indicate that individuals homozygous for the −551C or intron2,+16C allele might have a greater capacity to improve running economy in response to endurance exercise training.
Our study has some limitations. One potential limitation of our investigation was the method (BIA‐Tanita, Japan) we used to estimate LBM since the algorithms used with this methodology were not designed for a Chinese population. Although the intra‐class correlation coefficient for repeated tests was 0.93 (p<0.001), our method has not been validated against other accepted methods (for example, under water weighing or magnetic resonance imaging). Thus, there may be errors predicting LBM using BIA‐Tanita. Finally, the addition of a group of approximately 100 controls not undergoing any training and being tested before and after the same time period as our study subjects would have strengthened our investigation.
In summary, our data indicate that individuals homozygous for the −551C or intron2,+16C allele might have a greater capacity to improve running economy in response to endurance exercise training. Thus, it is suggested that HBB genotypes might explain some of the individual variation in the cardiorespiratory adaptation induced by aerobic training, at least in a male Chinese Han population. Further research on larger populations and/or of different ethnic origin is necessary to fully determine causality between HBB genotypes and individual differences in physiological adaptations to endurance training.
Acknowledgements
This work is supported by Ministry of Science and Technology of the People's Republic of China (grant code: 2003BA904B04).
Abbreviations
ATP - adenosine 5′‐triphosphate
LBM - lean body mass
RER - respiratory exchange ratio
TBM - total body mass
UTR - untranslated region
Footnotes
Competing interests: None declared.
References
- 1.Calbet J A, Lundby C, Koskolou M.et al Importance of hemoglobin concentration to exercise: acute manipulations. Respir Physiol Neurobiol 2006151132–140. [DOI] [PubMed] [Google Scholar]
- 2.Richardson R S. Oxygen transport and utilization: an integration of the muscle systems. Adv Physiol Educ 200327183–191. [DOI] [PubMed] [Google Scholar]
- 3.Shaskey D J, Green G A. Sports haematology. Sports Med 20002927–38. [DOI] [PubMed] [Google Scholar]
- 4.Schaffartzik W, Barton E D, Poole D C.et al Effect of reduced hemoglobin concentration on leg oxygen uptake during maximal exercise in humans. J Appl Physiol 199375491–498. [DOI] [PubMed] [Google Scholar]
- 5.Saunders P U, Pyne D B, Telford R D.et al Factors affecting running economy in trained distance runners. Sports Med 200434465–485. [DOI] [PubMed] [Google Scholar]
- 6.Lucia A, Hoyos J, Santalla A.et al Curvilinear VO2: power output relationship in a ramp test in professional cyclists: possible association with blood hemoglobin concentration. Jpn J Physiol 20025295–103. [DOI] [PubMed] [Google Scholar]
- 7.Cerretelli P, Samaja M. Acid‐base balance at exercise in normoxia and in chronic hypoxia. Revisiting the “lactate paradox”. Eur J Appl Physiol 200390431–448. [DOI] [PubMed] [Google Scholar]
- 8.Koskolou M D, Roach R C, Calbet J A.et al Cardiovascular responses to dynamic exercise with acute anemia in humans. Am J Physiol 1997273H1787–H1793. [DOI] [PubMed] [Google Scholar]
- 9.Saltin B, Kiens B, Savard G.et al Role of hemoglobin and capillarization for oxygen delivery and extraction in muscular exercise. Acta Physiol Scand Suppl 198655621–32. [PubMed] [Google Scholar]
- 10.Calbet J A. Oxygen tension and content in the regulation of limb blood flow. Acta Physiol Scand 2000168465–472. [DOI] [PubMed] [Google Scholar]
- 11.Singel D J, Stamler J S. Chemical physiology of blood flow regulation by red blood cells: the role of nitric oxide and S‐nitrosohemoglobin. Annu Rev Physiol 20056799–145. [DOI] [PubMed] [Google Scholar]
- 12.Buehler P W, Alayash A I. Oxygen sensing in the circulation: “cross talk” between red blood cells and the vasculature. Antioxid Redox Signal 200461000–1010. [DOI] [PubMed] [Google Scholar]
- 13.Ellsworth M L. Red blood cell‐derived ATP as a regulator of skeletal muscle perfusion. Med Sci Sports Exerc 20043635–41. [DOI] [PubMed] [Google Scholar]
- 14.Jia L, Bonaventura C, Bonaventura J.et al S‐nitrosohaemoglobin: a dynamic activity of blood involved in vascular control. Nature 1996380221–226. [DOI] [PubMed] [Google Scholar]
- 15.Joshi M S, Ferguson T B, Jr, Han T H.et al Nitric oxide is consumed, rather than conserved, by reaction with oxyhemoglobin under physiological conditions. Proc Natl Acad Sci U S A 20029910341–10346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Cosby K, Partovi K S, Crawford J H.et al Nitrite reduction to nitric oxide by deoxyhemoglobin vasodilates the human circulation. Nat Med 200391498–1505. [DOI] [PubMed] [Google Scholar]
- 17.Huang Z, Shiva S, Kim‐Shapiro D B.et al Enzymatic function of hemoglobin as a nitrite reductase that produces NO under allosteric control. J Clin Invest 20051152099–2107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Gladwin M T, Crawford J H, Patel R P. The biochemistry of nitric oxide, nitrite, and hemoglobin: role in blood flow regulation. Free Radic Biol Med 200436707–717. [DOI] [PubMed] [Google Scholar]
- 19.Nagababu E, Ramasamy S, Abernethy D R.et al Active nitric oxide produced in the red cell under hypoxic conditions by deoxyhemoglobin‐mediated nitrite reduction. J Biol Chem 200327846349–46356. [DOI] [PubMed] [Google Scholar]
- 20.Bouchard C, Rankinen T, Chagnon Y C.et al Genomic scan for maximal oxygen uptake and its response to training in the Heritage Family Study. J Appl Physiol 200088551–559. [DOI] [PubMed] [Google Scholar]
- 21.Krahenbuhl G S, Williams T J. Running economy: changes with age during childhood and adolescence. Med Sci Sports Exerc 199224462–466. [PubMed] [Google Scholar]
- 22.Feitosa M F, Gaskill S E, Rice T.et al Major gene effects on exercise ventilatory threshold: the HERITAGE Family Study. J Appl Physiol 2002931000–1006. [DOI] [PubMed] [Google Scholar]
- 23.Berg P E, Williams D M, Qian R L.et al A common protein binds to two silencers 5′ to the human beta‐globin gene. Nucleic Acids Res 1989178833–8852. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Elion J, Berg P E, Lapoumeroulie C.et al DNA sequence variation in a negative control region 5′ to the beta‐globin gene correlates with the phenotypic expression of the beta s mutation. Blood 199279787–792. [PubMed] [Google Scholar]
- 25.Semenza G L, Malladi P, Surrey S.et al Detection of a novel DNA polymorphism in the beta‐globin gene cluster. J Biol Chem 19842596045–6048. [PubMed] [Google Scholar]
- 26.Jiang Y, Xu X S, Russell J E. A nucleolin‐binding 3′ untranslated region element stabilizes beta‐globin mRNA in vivo. Mol Cell Biol 2006262419–2429. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Antoniou M, deBoer E, Habets G.et al The human beta‐globin gene contains multiple regulatory regions: identification of one promoter and two downstream enhancers. EMBO J 19887377–384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Russell J E, Liebhaber S A. The stability of human beta‐globin mRNA is dependent on structural determinants positioned within its 3′ untranslated region. Blood 1996875314–5323. [PubMed] [Google Scholar]
- 29.Shi Y Y, He L. SHEsis, a powerful software platform for analyses of linkage disequilibrium, haplotype construction, and genetic association at polymorphism loci. Cell Res 20051597–98. [DOI] [PubMed] [Google Scholar]
- 30.Schaid D J, Rowland C M, Tines D E.et al Score tests for association between traits and haplotypes when linkage phase is ambiguous. Am J Hum Genet 200270425–434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Erdfelder E, Faul F, Buchner A. GPower: a general power analysis program. Behav Res Methods Instrum Comput 1996281–11. [Google Scholar]
- 32.Blair S N, Kampert J B, Kohl H W., 3rdet al Influences of cardiorespiratory fitness and other precursors on cardiovascular disease and all‐cause mortality in men and women. JAMA 1996276205–210. [PubMed] [Google Scholar]
- 33.Wei M, Kampert J B, Barlow C E.et al Relationship between low cardiorespiratory fitness and mortality in normal‐weight, overweight, and obese men. JAMA 19992821547–1553. [DOI] [PubMed] [Google Scholar]
- 34.Rodas G, Calvo M, Estruch A.et al Heritability of running economy: a study made on twin brothers. Eur J Appl Physiol 199877511–516. [DOI] [PubMed] [Google Scholar]
- 35.Wolfarth B, Bray M S, Hagberg J M.et al The human gene map for performance and health‐related fitness phenotypes: the 2004 update. Med Sci Sports Exerc 200537881–903. [PubMed] [Google Scholar]
- 36.Reese M G, Eeckman F H, Kulp D.et al Improved splice site detection in Genie. J Comput Biol 19974311–323. [DOI] [PubMed] [Google Scholar]
- 37.Rogozin I B, Milanesi L. Analysis of donor splice sites in different eukaryotic organisms. J Mol Evol 19974550–59. [DOI] [PubMed] [Google Scholar]