Clinical Autonomic Research

, 19:335

A role for succinate dehydrogenase genes in low chemoresponsiveness to hypoxia?

Authors

    • Université Paris 13, UFR SMBH EA2363
    • AP-HP, Hôpital Avicenne, Service de Physiologie, Explorations Fonctionnelles et Médecine du Sport
  • Anne-Paule Gimenez-Roqueplo
    • AP-HP, Hôpital Européen Georges Pompidou, Département de Génétique
    • INSERM
    • Collège de France
    • Université Paris Descartes
  • Séverine Peyrard
    • AP-HP, Hôpital Européen Georges Pompidou, Centre d’Investigations Cliniques 9201
    • INSERM, Centre d’Investigations Cliniques 9201
    • Université Paris Descartes
  • Annabelle Vénisse
    • AP-HP, Hôpital Européen Georges Pompidou, Département de Génétique
    • INSERM
    • Collège de France
    • Université Paris Descartes
  • Laure Marelle
    • AP-HP, Hôpital Européen Georges Pompidou, Centre d’Investigations Cliniques 9201
  • Nelly Burnichon
    • AP-HP, Hôpital Européen Georges Pompidou, Département de Génétique
    • INSERM
    • Collège de France
    • AP-HP, Hôpital Européen Georges Pompidou, Centre d’Investigations Cliniques 9201
  • Anissa Bouzamondo
    • Société Française de Cardiologie
  • Xavier Jeunemaitre
    • AP-HP, Hôpital Européen Georges Pompidou, Département de Génétique
    • INSERM
    • Collège de France
    • Université Paris Descartes
  • Michel Azizi
    • AP-HP, Hôpital Européen Georges Pompidou, Centre d’Investigations Cliniques 9201
    • INSERM, Centre d’Investigations Cliniques 9201
    • Université Paris Descartes
  • Jean-Luc Elghozi
    • AP-HP, Hôpital Necker, UF de Pharmacologie Clinique
    • Université Paris Descartes
Research Article

DOI: 10.1007/s10286-009-0028-z

Cite this article as:
Richalet, J., Gimenez-Roqueplo, A., Peyrard, S. et al. Clin Auton Res (2009) 19: 335. doi:10.1007/s10286-009-0028-z

Abstract

The detection of hypoxia by the carotid bodies elicits a ventilatory response of utmost importance for tolerance to high altitude. Germline mutations in three genes encoding subunit B, C and D of succinate dehydrogenase (SDHB, SDHC and SDHD) have been associated with paragangliomas of the carotid body. We hypothesized that SDH dysfunction within the carotid body could result in low chemoresponsiveness and intolerance to high altitude. The frequency of polymorphisms of SDHs, hypoxia-inducible factor type 1 (HIF1α) and angiotensin converting enzyme (ACE) genes was compared between 40 subjects with intolerance to high altitude and a low hypoxic ventilatory response at exercise (HVRe ≤ 0.5 ml min−1 kg−1; HVR− group) and 41 subjects without intolerance to high altitude and a high HVRe (≥0.80 ml min−1 kg−1; HVR+). We found no significant association between low or high HVRe and (1) the allele frequencies for nine single nucleotide polymorphisms (SNPs) in the SDHD and SDHB genes, (2) the ACE insertion/deletion polymorphism and (3) four SNPs in the HIF1α gene. However, a marginal significant association was found between the synonymous polymorphism c.18A>C of the SDHB gene and chemoresponsiveness: 8/40 (20%) in the HVR− group and 3/41 (7%) in the HVR+ group (p = 0.12). A principal component analysis showed that no subject carrying the 18C allele had both high ventilatory and cardiac response to hypoxia. In conclusion, no clear association was found between gene variants involved in oxygen sensing and chemoresponsiveness, although some mutations in the SDHB and SDHD genes deserve further investigations in a larger population.

Keywords

Altitude toleranceHypoxiaCarotid bodiesVentilation

Introduction

Rapid access to high altitude in non-acclimatized sea level residents may be associated with the occurrence of acute mountain sickness (AMS), high altitude pulmonary edema (HAPE) or high altitude cerebral edema (HACE) [5, 14, 25]. Some physiological characteristics have been associated with intolerance to high altitude. Low sensitivity of peripheral chemoreceptors to hypoxia is responsible for reduced alveolar ventilation at rest or exercise, resulting in an exaggerated hypoxemia and a higher risk of developing AMS or HAPE [9, 16, 24, 26]. Concerning HACE, physiological risks are less clear, although it shares with AMS some symptoms and probably pathophysiological mechanisms [14]. Some subjects seem particularly intolerant to high altitude with recurrent severe forms of AMS, HAPE or HACE when they travel to high altitude, suggesting that there is an individual susceptibility to altitude tolerance. Similarly, in diseases associated with chronic exposure to high altitude such as chronic mountain sickness, alterations in chemosensitivity have been put forward as responsible for decreased ventilation, arterial hypoxemia and consequently severe polycythemia [27]. Therefore, a common pathophysiological mechanism, involving the ventilatory response to hypoxia, could be shared by acute and chronic intolerance to hypoxia.

Numerous studies have explored the association of the susceptibility to high altitude-related diseases and polymorphisms or expression of genes involved in oxygen sensing or physiological responses to hypoxia [21, 29]. These include the angiotensin converting enzyme (ACE) [1, 10, 20, 21, 23, 33], AT1 receptor [17], hypoxia inducible factor type 1 (HIF1α) [3, 12, 21, 22], VEGF [3, 12], eNOS [11], iNOS [12], erythropoietin [19, 22], surfactant proteins [30] and hsp70 [35] genes. However, no clear association has been found up to now between clinical tolerance to high altitude and the polymorphism of one single gene because (1) tolerance to hypoxia is probably depending on the expression of multiple factors, and (2) the sample size of the case–control studies was probably inadequate to evidence significant differences in genotypes [29].

No study has investigated the possible responsibility of the genes coding for succinate dehydrogenase in the tolerance to high altitude. Indeed, an increased incidence of carotid paraganglioma, a tumor located in the carotid body, has been observed at high altitude [28]. Paragangliomas are highly vascularized tumors that can be inherited through germline mutations in three genes encoding subunit B, C or D of the succinate dehydrogenase enzyme (SDHB, SDHC and SDHD). A direct molecular link was recently found between SDH inactivation and the regulation of the hypoxia-angiogenesis pathway [13]. Succinate accumulated in SDH-related tumors inhibits the prolylhydroxylase activity and leads to an abnormal stabilization of HIF-1α. Moreover, carotid bodies are the main organ to detect changes in arterial blood oxygen tension and elicit a ventilatory response to hypoxia, and it has been suggested that altitude could be a phenotypic modifier in hereditary paraganglioma [4]. Therefore, any alteration in the hypoxia sensing pathways due to SDH dysfunction within the carotid body could result in low chemoresponsiveness. Consequently, we hypothesized that high altitude intolerance and blunted ventilatory response to hypoxia could be associated with genetic variants of the SDH genes.

Materials and methods

Subjects

We included two groups of subjects either with both low chemoresponsiveness to hypoxia and clinical intolerance to high altitude (HRV− group, n = 40) or with both high chemoresponsiveness to hypoxia and good clinical tolerance to high altitude (HRV+ group, n = 41) according to the procedure described below. Subjects had to satisfy both physiological and clinical criteria to be included in either group. Subjects complying with the two criteria were retrospectively selected from a cohort of 2,044 subjects investigated at Hôpital Avicenne, Bobigny, before going for trekking or expedition to high altitude (>4,500 m). Respiratory and cardiac response to acute hypoxia (FIO2 = 0.115, equivalent altitude ~4,800 m) were measured at rest and exercise (~30% maximal O2 consumption) following a standardized protocol modified from Richalet [8, 24, 26]. Briefly, the test is composed of four successive phases of 4 min each: rest in normoxia, rest in hypoxia, exercise in hypoxia and exercise in normoxia. Minute ventilation, heart rate and arterial O2 saturation (SaO2) were measured continuously during the test (respectively with Oxycon Jeager metabograph, EKG and transcutaneous oximetry) and mean values in the last minute of each phase were taken for the three variables (see Table 2). Calculated variables were the difference in SaO2 between normoxia and hypoxia at rest and at exercise, the cardiac response to hypoxia (HCR) at rest and exercise, as the ratio difference in heart rate/difference in SaO2 between normoxia and hypoxia, the ventilatory response to hypoxia (HVR) as the ratio difference in ventilation/difference in SaO2/body weight between normoxia and hypoxia. The main variable used to discriminate subjects with low (HVR−) and high ventilatory response to hypoxia (HVR+) was exercise HVR (HVRe) which has been shown to be the best predictor of low tolerance to high altitude [8, 24, 26]. Low chemoresponsiveness was defined when HVRe ≤ 0.5 l min−1 kg−1 and high chemoresponsiveness when HVRe ≥ 0.8 l min−1 kg−1. During their stay at high altitude, a self-assessment questionnaire [32] was used to identify the occurrence of severe AMS (Lake Louise score > 3), HAPE or HACE. Then, subjects were classified as tolerant (Lake Louise score ≤ 3) or intolerant (occurrence of severe AMS, HAPE or HACE) to high altitude. All subjects were included in the study in the Hôpital Européen Georges Pompidou Clinical Investigation Center after given written informed consent. After interview and clinical examination, 20 ml of blood sampled from an antecubital vein was collected and sent to the Department of Genetics.

The protocol was approved by the Comité Consultatif de Protection des Personnes dans la Recherche Biomédicale (Paris-Necker, France). All procedures were in accordance with the Declaration of Helsinki principles and institutional guidelines.

Genotyping

Leukocyte DNA was extracted by a standard method. Six different single nucleotide polymorphisms (SNPs) were genotyped on the coding sequence of SDHB, three on SDHD and one on SDHC gene by direct sequencing of exon 1 (c.18A>C; p.Ala6Ala–rs2746462), intron 2 (c.73-36T>G; c.200+33G>A–rs2647196; c.200+35G>A), exon 4 (c.300T>C; p.Ser100Ser) and exon 5 (c.487T>C; p.Ser163Pro–rs 33927012) of the SDHB gene, of exon 1 (c.34G>A; p.Gly12Ser–rs34677591), exon 2 (c.152A>G; p.His50Arg–rs11214077) and exon 3 (c.204C>T; p.Ser68Ser–rs9919552) of the SDHD gene and of intron 1 (c.21-97C>T–rs4255403) of the SDHC gene as previously described [2]. Four of them were non-synonymous. The ACE insertion/deletion genotype was determined by two PCR using an insertion-specific primer pair (hace5a, hace5c) to avoid the classical mistyping of ID subjects. Four single nucleotide polymorphisms (P582S, A588T, T600T and Q979E) of HIF1α gene were screened by direct sequencing of the exon 12 of the HIF1a using the following primers 5′-TTAGTCTGAAGTGACTTTGAGT-3′ (forward) and 5′-ATCATGTACTATCACATACATT-3′ (reverse). The annealing temperature used was 58°C for the amplification.

Additionally, we analyzed one SNP on the SDHC sequence, the insertion/deletion (I/D) polymorphism of the angiotensin converting enzyme gene and four SNP in the HIF1α gene.

Statistical analysis

A sample size of 40 subjects by group was initially calculated as the minimum required to detect a 25% difference in the allele frequency (heterozygotes + mutated homozygotes) of different SNPs of SDHB, SDHD and SDHC genes between the control group (HVR+) and the HVR− participants, with an alpha risk of 5% and a power of 80%.

For descriptive purposes, we calculated genotype frequencies for each individual polymorphism and evaluated Hardy–Weinberg equilibrium using a one-degree of freedom goodness-of-fit test among controls. If not specified otherwise, the genotype frequencies between HVR+ and HVR− participants were compared using Fisher’s exact tests and a dominant mode (homozygote-common versus heterozygote and homozygote-rare) because of the low frequency of the polymorphisms tested. Polymorphisms with no mutated alleles and those in complete linkage disequilibrium with other polymorphism(s) were not tested (indicated by n.t. in tables). Clinical and physiological characteristics were compared between groups (HVR+ vs. HVR− participants or wild-type homozygotes vs. heterozygotes and mutated homozygotes) using unpaired Student or Wilcoxon tests for continuous variables and the Fisher-exact test for binary variables. To further investigate the association between the c.18A>C polymorphism on SDHB gene and the various physiological parameters measured, a principal component analysis was conducted to determine a physiological profile of AC heterozygotes and CC homozygotes using linear combinations of ventilation, heart rate and SaO2 measured on the four separate conditions (normoxia, rest in hypoxia, exercise in hypoxia and exercise in normoxia). The objective of this principal component analysis is to reduce the dimensionality (number of physiological parameters) combining the original variables in only two principal components which retain most of the original variability in the data. The horizontal axis represents a linear combination of the four measures of ventilatory response and the four measures of SaO2. The vertical axis indicates the cardiac response (linear combination of the four measures of heart rate in the different conditions). Fifty-four percent of the variance was explained by the two principal axes: the horizontal axis represented 30% of the total variance and the vertical axis, 24% of the total variance.

Statistical analyses were performed with SAS statistical software (Cary, NC 27513, USA) and a p value less than 0.05 was considered as significant.

Results

The clinical and physiological characteristics of the 81 subjects are given in Tables 1 and 2. The two groups were similar in gender distribution, age, height and body weight. By selection, 68, 8 and 10% of HVR− patients had a history of severe AMS, HAPE or HACE, respectively, whereas HVR+ patients had no history of intolerance to high altitude (Table 1). SaO2 and ventilation in hypoxia (rest and exercise), as well as cardiac and ventilatory response to hypoxia (rest and exercise), were significantly lower in the HVR− than in the HVR+ group (Table 2; Fig. 1). The hypoxia-induced decrease in SaO2 (ΔSaO2) was also significantly higher in the HVR− than in the HVR+ group at rest and exercise (Table 2; Fig. 1).
Table 1

Clinical characteristics of the cases and control subjects

 

HVR+

n = 41

HVR–

n = 40

Total

n = 81

Male (%)

21 (51)

12 (30)

33 (41)

Age (years)

46.4 ± 11.5

47.5 ± 10.1

47.0 ± 10.8

Height (cm)

170.6 ± 10.5

168.2 ± 8.5

169.4 ± 9.6

Weight (kg)

67.5 ± 14.9

63.8 ± 10.6

65.7 ± 13.0

History of severe AMS (%)

0

27 (68)*

27 (33)

History of HAPE (%)

0

3 (8)

3 (4)

History of HACE (%)

0

4 (10)

4 (5)

Mean ± SD. * p < 0.05, HVR− versus HVR+

HVR+, HVR− subjects with respectively high and low hypoxic ventilatory response at exercise; AMS acute mountain sickness; HAPE high altitude pulmonary edema; HACE high altitude cerebral edema

Table 2

Physiological characteristics of the cases and control subjects

 

HVR+

n = 41

HVR–

n = 40

Total

n = 81

SaO2 (%)

 Rest (normoxia)

97.7 ± 0.9

97.6 ± 1.2

97.7 ± 1.1

 Rest (hypoxia)

87.9 ± 2.5

85.6 ± 4.5*

86.8 ± 3.8

 Exercise (normoxia)

96.6 ± 1.3

96.6 ± 1.6

96.6 ± 1.4

 Exercise (hypoxia)

76.5 ± 3.1

65.0 ± 5.9*

70.8 ± 7.4

Heart rate (min−1)

 Rest (normoxia)

77.2 ± 11.9

76.6 ± 11.2

76.9 ± 11.5

 Rest (hypoxia)

89.1 ± 12.7

86.3 ± 12.5*

87.8 ± 12.6

 Exercise (normoxia)

114.8 ± 11.5

116.6 ± 12.3

115.7 ± 11.9

 Exercise (hypoxia)

132.2 ± 11.2

134.1 ± 11.5*

133.1 ± 11.3

Ventilation (l min−1)

 Rest (normoxia)

9.9 ± 3.1

9.1 ± 3.4

9.5 ± 3.2

 Rest (hypoxia)

12.8 ± 3.7

10.9 ± 3.2*

11.9 ± 3.6

 Exercise (normoxia)

29.0 ± 7.0

29.7 ± 7.1

29.4 ± 7.0

 Exercise (hypoxia)

43.0 ± 9.5

34.5 ± 6.9*

38.8 ± 9.3

Power output at exercise (W)

60 [30;120]

60 [30;90]

60 [30;120]

ΔSaO2 at rest (%)

9.9 ± 2.6

12.0 ± 4.4*

10.9 ± 3.7

ΔSaO2 at exercise (%)

20.1 ± 2.5

31.6 ± 5.7*

25.8 ± 7.2

Resting HCR (min−1 %−1)

1.2 [0.3;3.0]

0.7 [0.1;2.0]*

1.0 [0.1;3.0]

Exercise HCR (min−1 %−1)

0.8 [0.3;1.7]

0.6 [0.1;1.0]*

0.7 [0.1;1.7]

Resting HVR (l min−1 kg−1)

0.4 [0;1.4]

0.3 [0;1.8]*

0.3 [0;1.8]

Exercise HVR (l min−1 kg−1)

1.0 [0.8;1.7]

0.3 [0;0.5]*

0.8 [0;1.7]

Data are mean ± SD or median [minimum; maximum]

HVR+, HVR− subjects with respectively high and low hypoxic ventilatory response at exercise; SaO2 arterial oxygen saturation; ΔSaO2 decrease in SaO2 from normoxia to hypoxia; HCR hypoxic cardiac response; HVR hypoxic ventilatory response

p < 0.05, HVR− versus HVR+

https://static-content.springer.com/image/art%3A10.1007%2Fs10286-009-0028-z/MediaObjects/10286_2009_28_Fig1_HTML.gif
Fig. 1

Individual values of ventilatory and cardiac response to hypoxia, at rest and at exercise. Groups (HVR+ and HVR−) were separated using the value of exercise HVR: HVR+: exercise HVR ≥ 0.8; HVR−: exercise HVR ≤ 0.5 l min−1 kg−1. The line represents the median

The allele frequencies for the nine different single SNPs in the SDHD or SDHB genes are given in Table 3. All allele frequencies were in Hardy–Weinberg equilibrium. There was no significant association between the distribution of any of these SNPs and the HVR+ and HVR− phenotype (Table 3). No association was found either, between the six complementary genetic variants tested (SDHC, ACE and HIF1α) and the hypoxic ventilatory response (Table 4).
Table 3

Allele frequencies of nine SNPs of the SDHB and SDHD genes according to the ventilatory response to hypoxia at exercise

 

HVR+

n = 41 (%)

HVR−

n = 40 (%)

All

n = 81 (%)

p value

SDHB, c.18A>C, p.A6A

 AA

38 (93)

32 (80)

70 (86)

0.12

 AC

3 (7)

7 (18)

10 (12)

 

 CC

0

1 (2)

1 (1)

 

SDHB, c.73-36T>G

 TT

41 (100)

40 (100)

81 (100)

n.t.

 TG

0

0

0

 

 GG

0

0

0

 

SDHB, c.200+33G>A

 GG

33 (80)

36 (90)

69 (85)

0.35

 GA

6 (15)

4 (10)

10 (12)

 

 AA

2 (5)

0

2 (2)

 

SDHB, c.200+35G>A

 GG

38 (93)

38 (95)

76 (94)

1.00

 GA

3 (7)

2 (5)

5 (6)

 

 AA

0

0

0

 

SDHB, c.300C>T, p.S100S

 CC

41 (100)

39 (98)

80 (99)

0.49

 CT

0

1 (2)

1 (1)

 

 TT

0

0

0

 

SDHB, c.487T>C, p.S163P

 TT

39 (95)

39 (98)

78 (96)

1.00

 TC

2 (5)

1 (2)

3 (4)

 

 CC

0

0

0

 

SDHD, c.34G>A, p.G12S

 GG

41 (100)

38 (95)

79 (98)

0.24

 GA

0

2 (5)

2 (2)

 

 AA

0

0

0

 

SDHD, c.152A>G, p.H50R

 AA

37 (90)

40 (100)

77 (95)

0.12

 AG

4 (10)

0

4 (5)

 

 GG

0

0

0

 

SDHD, c.204C>T, p.S68S

 CC

41 (100)

38 (95)

79 (98)

n.t.

 CT

0

2 (5)

2 (2)

 

 TT

0

0

0

 

HVR+, HVR− subjects with respectively high and low hypoxic ventilatory response at exercise; n.t. not tested

Table 4

SDHC, ACE and HIF1α genotypes according to the ventilatory response to hypoxia at exercise

 

HVR+

n = 39 (%)

HVR−

n = 40 (%)

All

n = 79 (%)

p value

SDHC, c.21-97C>T

 CC

32 (82)

35 (88)

67 (85)

0.55

 CT

7 (18)

5 (12)

12 (15)

 

 TT

0

0

0

 

ACE, I/D*

 DD

14 (36)

18 (45)

32 (41)

0.38

 ID

20 (51)

14 (35)

34 (43)

 

 II

5 (13)

8 (20)

13 (16)

 

HIF1α, c.1772C>T, p.P582S

 CC

30 (77)

35 (88)

65 (82)

0.25

 CT

9 (23)

5 (12)

14 (18)

 

 TT

0

0

0

 

HIF1α, c.1790G>A, p.A588T

 GG

38 (97)

40 (100)

78 (99)

0.49

 GA

1 (3)

0

1 (1)

 

 AA

0

0

0

 

HIF1α, c.1800A>T, p.T600T

 AA

37 (95)

36 (90)

73 (92)

0.68

 AT

1 (3)

3 (7)

4 (5)

 

 TT

1 (3)

1 (3)

2 (3)

 

HIF1α, c.2035C>G, p.Q679E

 CC

38 (97)

40 (100)

78 (99)

n.t.

 CG

1 (3)

0

1 (1)

 

 GG

0

0

0

 

HVR+, HVR− subjects with respectively high and low hypoxic ventilatory response at exercise

* For ACE, analysis was performed using an additive mode (homozygote-common versus heterozygote vs. homozygote-rare)

Among the 15 SNPs analyzed, only the synonymous polymorphism c.18A>C on SDHB gene tended to be associated with a low chemoresponsiveness: eight among the 40 HVR− subjects (20%) whereas only three among the 41 HVR+ subjects (7%) were AC heterozygotes + CC homozygotes (p = 0.12).

Finally, we analyzed the distribution of the subjects harboring the 18C allele for SDHB (AC heterozygotes + CC homozygotes) according to cardiac and ventilatory responses to hypoxia after combining the HVR+ and HVR− groups (Fig. 2). The presence of the 18C allele was significantly associated with a lower cardiac response to hypoxia at rest (median [interquartile range]: 0.47 [0.33;1.13] vs. 1.00 [0.57;1.46]; p = 0.0199). Nine out of the 11 AC heterozygotes + CC homozygotes had a cardiac response to hypoxia at rest lower than the median value for the 70 AA homozygotes (Fig. 2d). The other physiological variables did not discriminate significantly the subjects harboring the 18C allele from the AA homozygotes (Fig. 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs10286-009-0028-z/MediaObjects/10286_2009_28_Fig2_HTML.gif
Fig. 2

Individual values of ventilation (a) and heart rate (b) in hypoxia, and ventilatory (c) and cardiac (d) response to hypoxia, at rest and at exercise. Filled circles heterozygotes + mutated homozygotes subjects; open circles wild-type homozygote subjects for 18C SDHB allele. The lines represent the median (plain line heterozygotes + mutated homozygotes subjects, dotted line wild-type homozygote subjects)

The results of the principal component analysis, which allows taking into account the various physiological parameters to define each subject, showed that the AC heterozygotes and CC homozygotes subjects were largely scattered (Fig. 3). However, none of the 11 subjects carrying the 18C allele of SDHB had coordinates in the upper right quadrant of the scatter plot which corresponds to subjects with both elevated cardiac and ventilatory chemoreflex.
https://static-content.springer.com/image/art%3A10.1007%2Fs10286-009-0028-z/MediaObjects/10286_2009_28_Fig3_HTML.gif
Fig. 3

Principal component analysis. The horizontal axis is a linear combination of ventilatory and SaO2 measurements in the four experimental conditions. The vertical axis is a linear combination of heart rate measurements in the four experimental conditions. No subject carrying the 18C allele of SDH (SDHB, c.18A>C; p.A6A) is present in the upper right quadrant corresponding to both elevated ventilatory and cardiac response to hypoxia. Filled circles heterozygotes + mutated homozygotes subjects; open circles wild-type homozygotes subjects

Discussion

This is the first study that tries to correlate, in a relatively large group, the presence of variants of genes involved in oxygen sensing and ventilatory response to hypoxia associated with tolerance to high altitude. A special attention was paid to polymorphisms of SDHB and SDHD genes since these genes were associated with the dysfunction of hypoxia-angiogenesis pathways.

The hypothesis of a crucial role of mitochondria in oxygen sensing has recently resurged with the discovery of mutations in genes encoding for the mitochondrial complex II in patients affected by tumors arising either from the carotid body, the organ responsible for controlling oxygen homeostasis in the whole organism, or in organs particularly sensitive to hypoxia during development (adrenal medulla). However, the role of mitochondria and their dysfunction in oxygen sensing is still largely controversial. The incidence of paraganglioma is increased in populations living at high altitude and exposed to chronic hypoxia, suggesting that succinate dehydrogenase activity may play a physiological role as an oxygen sensor [6]. The germline mutations of SDHx genes induce an inactivation of succinate dehydrogenase enzyme, an accumulation of its substrate, the succinate, in high quantities in tumors that inhibit the HIF prolyl hydroxylase and increase the stability of HIF1α in targeted tissue. Consequently, HIF1α could activate its target genes involved in a variety of physiological responses to hypoxia such as erythropoiesis, glucose metabolism, pH regulation, vascular tone regulation and angiogenesis. In another way, recent data suggest that SDH deficiency induces ROS overproduction that could also play a role in hypoxia tolerance [18]. Therefore, we hypothesized that an impairment of SDH enzymatic activity due to one or several SDHx polymorphisms could lead to SDH dysfunction and affect the tolerance to acute and or chronic hypoxia resulting from an abnormal HIF-signaling pathway.

Although no significant association was found with any of the SNPs studied, a tendency was found with the c.18A>C polymorphism on SDHB gene which was more frequent in the low tolerant group (20 vs. 7%, p = 0.12). Moreover, a principal component analysis evidenced that no subject with this mutation had both high cardiac and ventilatory responses to hypoxia at exercise.

We also tested other candidate genes and found no association between four polymorphisms of the HIF1α gene and susceptibility to AMS, HAPE or HACE. Few studies have explored the association of polymorphisms of the HIF1α gene with tolerance to high altitude, such as in groups of different ethnic origin [31]. Another single study did not find any association between susceptibility to HAPE and blunted HVR and polymorphisms of the tyrosine hydroxylase (TH) gene, even though TH is a rate-limiting enzyme in the carotid body response to hypoxia [15].

Finally, in accordance with previous studies [10, 20, 23] we found no association with intolerance to high altitude and the ACE I/D polymorphism. Similarly, in a Quechua population from Peru, no association was found between ACE polymorphism and HVR, although I/I individuals showed a slightly higher SaO2, compared to I/D or D/D [7]. ACE polymorphisms had been rather associated with physical performance at high altitude [33, 34], which is not similar to the susceptibility to high altitude diseases (AMS, HAPE and HACE).

Study limitations

First, since the allele frequencies of the SNPs studied were relatively low, one limitation of this study is the sample size of our population. The sample size was calculated for an absolute difference in AC + CC frequency of 24% between HRV+ and HRV− subjects (6 vs. 30%). However, with a maximal difference observed among the nine SNPs tested in the SDHD or SDHB genes between the two groups of 13% (7 vs. 20%), the power of the study was 40% instead of the 80% planned.

Second, the phenotype was defined on physiological variables measured during a standardized exercise and on the basis of a history of intolerance to high altitude based on a questionnaire (clinical evaluation). The physiological variables used for the discrimination of the two groups appeared robust enough to give a clear cut difference in phenotypes between tolerant and intolerant subjects. A large significant difference was found between the two groups for all calculated variables (ΔSaO2, HCR, HVR both at rest and exercise) even though only exercise HVR was used to separate the two groups (see Table 2). The clinical evaluation was added to the discrimination process only to ascertain the fact that the physiological differences were correlated with the clinical outcome at high altitude. However, the clinical assessment was subjective since based on an auto-evaluation.

In conclusion, although no significant association was clearly found between polymorphisms of SDH genes and chemoresponsiveness to hypoxia or tolerance to high altitude, a tendency was found for the implication of c.18A>C mutation of SDHB gene, which would confirm the importance of the mitochondrial pathways of oxygen transfer in the chemoresponsiveness of carotid bodies. The evaluation of larger groups of subjects may allow giving a conclusive answer to the question of genes involved in tolerance to high altitude.

Acknowledgments

We thank the nurses of the Centre d’Investigations Cliniques and the technicians of the molecular genetic laboratory at the Hôpital Européen Georges Pompidou for their technical support. This study was supported by a grant delivered by the Société Française d’Hypertension Artérielle and promoted by the Société Française de Cardiologie.

Copyright information

© Springer-Verlag 2009