Author + information
- Received May 6, 2008
- Revision received December 12, 2008
- Accepted December 15, 2008
- Published online April 14, 2009.
- Elise Belaidi, PhD*,
- Marie Joyeux-Faure, PhD*,†,
- Christophe Ribuot, PhD*,†,
- Sandrine H. Launois, MD, PhD*,†,‡,
- Patrick Levy, MD, PhD*,†,‡ and
- Diane Godin-Ribuot, PhD*,†,* ()
- ↵*Reprint requests and correspondence
: Prof. Diane Godin-Ribuot, Laboratoire HP2, INSERM ERI017, Université Joseph Fourier, Institut Jean Roget, BP 170, 38042 Grenoble Cedex 9, France
Objectives Our aim was to investigate the involvement of the endothelin (ET) system in the cardiovascular consequences of intermittent hypoxia (IH).
Background Obstructive sleep apnea (OSA) syndrome is an important risk factor for cardiovascular morbidity. Chronic IH, a major component of OSA, is thought to be responsible for most of the cardiovascular complications occurring during OSA, but the underlying mechanisms remain to be determined.
Methods Chronic IH was applied in rats genetically prone to develop hypertension (spontaneous hypertensive rats [SHR]) and their normotensive controls. The cardiovascular effects were assessed in vivo and in Langendorff perfused hearts. Hypoxia inducible factor (HIF)-1 activity and targeting of the myocardial ET-1 gene and activation of the ET system were investigated using tissue chromatin immunoprecipitation, enzyme-linked immunoadsorbent assay, immunostaining, and Western blotting.
Results Chronic IH enhanced hypertension development and infarct size in SHR compared with that seen in control rats. This was accompanied by an increase in myocardial big ET-1, ET-1, and ET-A receptor expression and by an enhanced coronary vascular reactivity to ET-1 in SHR only. Myocardial HIF-1 activity was increased, and HIF-1 was shown to be linked to the promoter of the myocardial ET-1 gene after chronic IH only. Moreover, administration of bosentan, a mixed ET receptor antagonist, during chronic IH prevented both the increase in blood pressure and in infarct size.
Conclusions In SHR, activation of the ET system, mediated by HIF-1 activity, is responsible for the enhanced susceptibility to chronic IH and for its associated cardiovascular consequences leading to hypertension and ischemic injury. Furthermore, the beneficial effects of bosentan suggest exploring ET antagonists as possible therapeutic tools in OSA.
Obstructive sleep apnea (OSA) syndrome is characterized by recurrent episodes of pharyngeal collapse occurring during sleep, leading to intermittent hypoxia (IH) and hypercapnia and terminated by an arousal. OSA is a growing health concern as it affects up to 5% of middle-aged men and women with a higher prevalence in the elderly (1). Moreover, OSA is recognized as an important and independent risk factor for systemic hypertension (2), arrhythmias (3), stroke (4), and coronary heart disease (5). Patients with OSA have an increased rate of cardiovascular morbidity and mortality (6). Last, OSA is also associated with cardiovascular subclinical conditions such as diastolic hypertension (7), diastolic ventricular dysfunction (8), and early atherosclerosis (9).
Using various biological markers, numerous studies have shown an increased oxidative stress in OSA patients with production of reactive oxygen species correlated with the severity of oxygen desaturations and reversed by continuous positive airway pressure (CPAP) treatment (10,11).
IH appears to be the main causal factor in the pathogenesis of cardiovascular complications in OSA (12). Thus, rodent models of IH have been extensively used to study the mechanisms of OSA-related morbidity. It is now established that systemic hypertension (13), dyslipidemia, and early atherosclerosis (14) as well as vascular remodeling (15) occur after only a few weeks of IH exposure. In addition, chronic IH enhances myocardial sensitivity to ischemia, resulting in larger infarct sizes (16,17). Among the possible mechanisms responsible for some of these changes, increased oxidative stress and activation of the endothelin (ET) system have been proposed (17–19).
Assessing the ET system in response to IH is well founded. Hypoxia is a powerful stimulus for the production of ET-1 by endothelial cells and cardiomyocytes; in particular, its expression is increased under the hypoxia-inducible factor (HIF)-1 transcriptional control (20). Although the vasoconstrictor effects of ET-1 were the first to be characterized, accumulating evidence shows that it is also a powerful proinflammatory cytokine (21) involved in endothelial dysfunction and atherosclerosis (22). Moreover, ET-1 plasma concentrations are elevated in OSA patients (23–25), positively correlated with the severity of nocturnal hypoxia (23,25), and corrected by CPAP treatment (23,24). Rats exposed to chronic IH not only exhibit increased plasma levels of ET-1 (18) but also enhanced vascular reactivity to ET-1 (26,27).
The aim of the present study was to characterize the mechanisms and role of the ET system activation in the development of some of the cardiovascular consequences of IH, namely systemic hypertension and enhanced myocardial infarction. To this end, we used a novel approach by investigating animals with an inbred genetic predisposition to cardiovascular disease, which might render them more sensitive to IH and thus more prone to develop cardiovascular complications. We chose to perform the study on spontaneously hypertensive rats (SHR) because they display, in addition to systemic hypertension, other cardiovascular risk factors also encountered in OSA such as increased oxidative stress (28). Moreover, the fact that hypertension development in SHR is not ET-dependent (29) ensures that ET-related alterations can be mainly attributed to IH.
Thus, we hypothesized that SHR exposed to IH would exhibit a greater response than normotensive control rats, and that this enhanced susceptibility could be related to a specific activation of the ET system. In order to investigate the contribution of HIF-1, we used myocardial chromatin immunoprecipitation, a powerful technique that allows the direct evaluation of transcription factor binding on gene promoters in vivo. Finally, to confirm the role of the ET system in the development of the cardiovascular complications, we evaluated the effects of oral administration of a dual receptor antagonist, bosentan, throughout IH exposure.
See the Online Appendixfor an expanded Methods section.
Nine-week-old male SHR and Wistar Kyoto rats (WKY) were exposed to IH 8 h per day for 14 consecutive days. Control rats were exposed to normoxic cycles (N) only. Bosentan-treated animals were fed bosentan-admixed chow (100 mg/kg/day) throughout the period of exposure.
Systolic arterial blood pressure was measured by plethysmography throughout IH or N exposure and mean arterial blood pressure (MABP) by arterial catheterization at day 15. Vascular and myocardial responses to intracoronary ET-1 and infarct size (after a 30-min global ischemia and a 120-min reperfusion sequence) were assessed in Langendorff-perfused hearts at day 15.
Immunohistochemistry and Immunoblotting
Immunofluorescence staining with fluorescence microscopy was performed on cryosections carefully matched for position within the heart. Heart protein extracts were subjected to Western blot analysis with the same antibodies used for immunostaining.
Enzyme-Linked Immunoadsorbent Assays
Myocardial ET-1 and big ET-1 levels as well as myocardial HIF-1α activation were quantified using enzyme-linked immunoadsorbent assays.
Chromatin Immunoprecipitation Assay
Myocardial chromatin immunoprecipitation was performed as previously described (30), using an anti–HIF-1α antibody and polymerase chain reaction (PCR) amplification of the ET-1 gene promoter.
Experimental data are presented as mean ± standard error of the mean. Levels of myocardial HIF-1 activation in SHR rats exposed to N or IH were compared using a ttest. All other data were analyzed using 2-way analysis of variance (ANOVA) (repeated measures ANOVA in the case of systolic arterial blood pressure) with Bonferroni post-hoc comparisons. A value of p < 0.05 was considered statistically significant.
Effect of Genetic Susceptibility to Hypertension on the Response to IH
The effects of chronic IH on systemic blood pressure and myocardial susceptibility to ischemia were more pronounced in SHR. Thus, chronic IH enhanced hypertension development in SHR, in particular during the second week of exposure (Δ systolic arterial blood pressure: 28 ± 11 mm Hg and 64 ± 8 mm Hg at days 8 and 15 compared with 19 ± 8 mm Hg and 33 ± 7 mm Hg, respectively, in N rats, p < 0.001) while it did not significantly affect blood pressure in WKY (Δ: 16 ± 9 mm Hg at day 15) (Fig. 1A).This was confirmed by catheter recording at day 15 showing a significant effect of IH on MABP in SHR (175 ± 3 mm Hg vs. 147 ± 4 mm Hg in N rats, p < 0.001) but not in WKY (Fig. 1B). Although infarct size values of WKY submitted to chronic IH tended to be higher, a significant increase in infarct size was seen only in SHR (35 ± 4% in IH vs. 21 ± 3% in N rats, p < 0.01) (Fig. 2).
HIF-1 and ET system activation
Coronary and Myocardial Response to ET-1
SHR exposed to IH exhibited a greater increase in coronary perfusion pressure in response to intracoronary ET-1 compared with SHR exposed to N or to WKY (Δ: 115 ± 7 mm Hg vs. 81 ± 7 mm Hg , 78 ± 7 mm Hg, and 77 ± 7 mm Hg, respectively, in response to 3.10−6mol/l, p < 0.01) (Fig. 3A).This was accompanied by a significant increase in left ventricular end-diastolic pressure (Δ: 51 ± 7 mm Hg vs. 2 ± 8 mm Hg, 3 ± 7 mm Hg, and 4 ± 8 mm Hg, respectively, in response to 3.10−6mol/l, p < 0.01) (Fig. 3B) and a significant decrease in left ventricular developed pressure (Δ: −85 ± 20 mm Hg vs. −20 ± 12 mm Hg, −24 ± 10 mm Hg, and −8 ± 10 mm Hg, respectively, in response to 3.10−6mol/l, p < 0.01) (Fig. 3C).
Myocardial and Coronary ET-A and ET-B Receptor Expression
Myocardial expression of both ET-A and ET-B receptors was higher in SHR than in WKY. Moreover, chronic IH induced an up-regulation in myocardial ET-A in SHR only. ET-B receptor expression was not affected by IH (Fig. 4A).In accordance, immunofluorescent staining showed a higher density of ET-1 receptors in coronary arteries of SHR as well as an up-regulation of ET-A receptors with chronic IH in SHR only (Fig. 4B).
Myocardial Big-ET-1 and ET-1 Concentrations
Myocardial big ET-1 (13.4 ± 4.0 pg/g compared with 3.6 ± 0.8 pg/g in N rats, p < 0.01) and ET-1 (85.0 ± 3.5 pg/g vs. 69.3 ± 3.6 pg/g in N rats, p < 0.01) levels were significantly increased after exposure to chronic IH in SHR but not in WKY (Fig. 5).Specific activation of the ET system in SHR exposed to IH resulted in a higher content of big ET-1 after a 30-min global ischemia in Langendorff-perfused hearts (increase of 597.2 ± 192.0% compared with 119.2 ± 37.5% in WKY, p < 0.01).
Myocardial HIF-1 Activation and Targeting of the ET-1 Gene
In SHR, chronic exposure to IH induced a 1.6-fold increase in activated myocardial HIF-1α (linked with HIF-1β to form HIF-1) in nuclear extracts (54.5 ± 6.1 arbitrary units in IH vs. 35.2 ± 3.7 arbitrary units in N rats, p < 0.05) (Fig. 6A).In vivo chromatin immunoprecipitation assay on myocardial extracts from SHR enabled us to cross-link and immunoprecipitate myocardial HIF-1α bound to its target genes. Figures 6B and 6C show PCR amplification of the hypoxia response element (HRE) site proximal to the promoter of the myocardial ET-1 gene. The increase in the PCR signal of the immunoprecipitated fraction from the IH group provided evidence that HIF-1α targeting of the myocardial ET-1 gene is induced by chronic IH in SHR. Lack of amplification in mock conditions confirmed the specificity of the immunoprecipitation technique. PCR signals from deoxyribonucleic acid fragments before immunoprecipitation (input) show that the immunoprecipitation results are not due to differences in sample deoxyribonucleic acid content.
Effects of Bosentan Treatment
In SHR, oral bosentan administration throughout exposure to N or IH completely prevented the IH-induced enhancement of hypertension and increase in infarct size (24 ± 3% and 25 ± 2% in N and IH bosentan-treated rats, respectively). Bosentan treatment did not, however, affect the development of genetically related hypertension (Fig. 7).In WKY, administration of bosentan was without effect (data not shown).
OSA is associated with cardiovascular diseases such as hypertension, ischemic heart disease, and stroke and with increased plasma ET levels. In this study, we found that exposure of SHR to chronic IH in a manner similar to that seen in OSA patients resulted in enhanced hypertension development, increase in infarct size, and activation of the ET system and that bosentan treatment during IH exposure completely prevented the deleterious cardiovascular alterations. In addition, we established a critical role for HIF-1 in the up-regulation of the ET-1 gene in response to chronic IH.
We have previously demonstrated that chronic IH enhances myocardial susceptibility to infarction in the rat (16). This was also recently reported in mice and was shown to be related to increased oxidative stress (17). As for the hypertensive effects of chronic IH, we were unable to observe a significant increase in MABP after 35 days of exposure (16). This is in accordance with some studies (31,32) but in contrast with others showing a moderate but statistically significant increase in MABP in the rat (13,19,33). Overall, the resulting effect of IH on blood pressure is thought to be critically dependent on the duration and severity of the IH exposure. The results of the present study show that the cardiovascular consequences of IH may also be determined by pre-existing constitutional traits. Indeed, SHR were more susceptible to chronic IH, and this was evidenced by significant increases in MABP and in infarct size after 14 days of exposure that were not seen in WKY.
We propose that this increased susceptibility is due to a unique hypoxia-dependent activation of the ET system in SHR, since such activation was not seen in WKY and since bosentan treatment completely prevented the deleterious cardiovascular effects. In contrast to other animal models of hypertension, hypertension development in SHR is not related to the ET system. Thus, SHR do not normally exhibit enhanced ET-1 expression in blood vessels (34), and chronic treatment with ET antagonists does not lower blood pressure in SHR (29,35). Actually, our observation that bosentan treatment prevented the IH-induced but not the genetically related increase in blood pressure is in accordance with these studies.
Nevertheless, SHR displayed an activation of the ET system in response to chronic IH while WKY did not. The basis for this paradoxical effect may lie in the results of our chromatin immunoprecipitation assay analyzing the interaction between HIF-1α and the myocardial ET-1 gene HRE. Hypoxia has been shown to induce ET transcription in several cell types including cardiomyocytes (36) and endothelial cells (37). The HRE of the human and rat ET-1 promoter have been characterized (38), and HIF-1α binding to the ET-1 gene after hypoxia has been demonstrated in endothelial cells using gel shift assays (39). However, our study is the first to provide in vivo evidence of this interaction and, notably, to report targeting of the cardiac ET-1 promoter by HIF-1α in SHR after chronic IH.
Activation of HIF-1 in response to chronic IH has been demonstrated in vitro (40). In vivo, increases in tissue HIF-1α protein content have been reported (41,42), and data from mice with heterozygous HIF-1α deficiency have provided indirect evidence that HIF-1 is involved in chronic IH (43). In the current study, we have measured nuclear HIF-1 activity and have shown that it was significantly increased in myocardial extracts from SHR submitted to IH, thus providing the first demonstration that HIF-1 is activated in vivo in response to chronic IH.
While HIF-1α stabilization in sustained hypoxic conditions is related to inhibition of prolyl hydroxylase activity, there is increasing evidence that production of oxygen-derived radicals, in particular superoxide anions, during IH plays a major role in HIF-1α up-regulation (43,44), and a hallmark of hypertension development in SHR is increased oxidative stress (28,45). Thus, increased superoxide production in SHR could enhance their susceptibility to IH by promoting HIF-1α activity and its subsequent targeting of the ET-1 gene. In support for this hypothesis, Troncoso Brindeiro et al. (19) have shown that nicotinamide adenine dinucleotide phosphate oxidase activity and superoxide production were elevated after IH exposure and that increases in ET-1 synthesis and blood pressure could be prevented by the superoxide dismutase mimetic tempol. In addition to HIF-1, other transcription factors known to modulate ET-1 gene expression, such as nuclear factor kappa B and activator protein-1 (20), are activated by exposure to IH (46) and could also have played a role in the up-regulation observed in this study.
Our observation that HIF-1 targeting of the ET-1 gene is responsible for the deleterious effects of chronic IH is important since HIF-1 activation is known to be cardioprotective (47). Indeed, we have shown that HIF-1 targeting of the inducible nitric oxide synthase gene results in cardioprotection after acuteexposure to IH (30). In accordance, mice exposed to IH present a similar time course in myocardial susceptibility to ischemia (17). It thus appears that with prolonged exposure, the beneficial effects are over-ridden, and this could be due to the repeated activation of the ET-1 gene along with the enhanced expression of ET-1 receptors.
Indeed, SHR had, overall, a higher expression level of both ET-1 receptor subtypes in the myocardium compared with WKY, a feature also observed in the kidney (48). However, in response to IH, SHR, but not WKY, underwent up-regulation of myocardial and, in particular, of coronary ET-A receptors while ET-B receptors remained unaffected. Similar results have been obtained by Allahdadi et al. (26), which have shown up-regulation of ET-A receptor in mesenteric arteries and more recently in lung, renal medulla, and coronary arteries (49) after chronic IH while ET-B receptor expression remained unchanged. Overexpression of pulmonary ET, ET-A, and in some studies ET-B receptors is also seen in response to chronic sustained hypoxia and is thought to participate in the development of pulmonary hypertension (50). Chronic hypoxia also increases the expression of ET-1 and ET-A receptors in glomus cells of the carotid body leading to increased hypoxic sensitivity (51). Thus, hypoxia appears to be a powerful stimulus for ET-1 receptor expression sufficient to overcome homologous down-regulation. Interestingly, a recent study has demonstrated the presence of several HRE in the human ET-B receptor promoter and has shown that HIF-1 can up-regulate ET-B receptors (52). Whether the same applies to ET-A receptors remains to be demonstrated. Also, organ culture studies have shown that protein kinase C and mitogen-activated protein kinase signaling pathways are involved in ET receptor up-regulation in coronary arteries (53). This could provide the basis for the ET receptor up-regulation induced by chronic IH in SHR since among the agents known to activate these pathways are reactive oxygen species and ET-1 itself (54,55).
Coronary ET-A up-regulation in SHR resulted in pronounced vasoconstriction and myocardial dysfunction in response to intracoronary ET-1. In view of the well-known detrimental effects of ET-1 on the ischemic myocardium (56), this, along with the increased activation of the ET system, which persisted in the ischemic myocardium, could explain the enhanced susceptibility to infarction induced by chronic IH. The dramatic reduction in infarct size brought about by bosentan treatment is in agreement with this hypothesis.
Although this is the first report of enhanced ET-1 vasoconstriction in coronary arteries after chronic IH, increased responses have been observed in other vascular territories (26,27). Thus, a generalized vascular hyper-responsiveness to ET-1 could account for the development of hypertension, and this is supported by the preventive effects of ET-1 receptor blockade observed in various studies including the present one (18,49). A role for the ET system in the IH-induced development of hypertension was first proposed by Kanagy et al. (18). In the current study, using the mixed antagonist bosentan, we show that activation of the ET system by IH has far more widespread consequences resulting in hypertension development but also in aggravated infarction.
Our findings in rats bring into light a potential treatment strategy for the cardiovascular complications of OSA syndrome. Nocturnal application of CPAP is an effective and safe treatment for sleep apnea. Although its regular use is associated with a decrease in blood pressure and cardiovascular risk (57), CPAP therapy is cumbersome and up to 30% of OSA patients will discontinue their treatment over time (58). Alternative strategies could consist of targeting OSA consequences rather than eliminating apneas in noncompliant patients. Given that some of the IH-induced cardiovascular changes can be prevented by ET-1 blockade, the effects of ET-1 antagonist administration could be tested in OSA patients.
Another interesting clinical perspective brought about by the present study is that, in OSA patients, susceptibility to IH and cardiovascular injury could be amplified by the presence of comorbidities resulting in, among other things, increased activation of HIF-1 and expression of deleterious genes such as the ET-1 gene.
The authors thank Ms. Nolwenn Miguet, Mrs. Sandrine Cachot, and M. Patrick Méresse for their excellent technical support and Actelion Pharmaceuticals (Basel, Switzerland) for their gracious gift of bosentan.
For the expanded Methods section of this study, please see the online version of this article.
This study was supported, in part, by a grant from AGIRà dom (Grenoble, France). Dr. Launois is the recipient of an AREA award from Actelion Pharmaceuticals for a clinical project on sleep apnea patients.
- Abbreviations and Acronyms
- continuous positive airway pressure
- hypoxia-inducible factor
- hypoxia response element
- intermittent hypoxia
- mean arterial blood pressure
- obstructive sleep apnea syndrome
- polymerase chain reaction
- spontaneous hypertensive rats
- Wistar Kyoto rats
- Received May 6, 2008.
- Revision received December 12, 2008.
- Accepted December 15, 2008.
- American College of Cardiology Foundation
- Peker Y.,
- Kraiczi H.,
- Hedner J.,
- Loth S.,
- Johansson A.,
- Bende M.
- Alchanatis M.,
- Tourkohoriti G.,
- Kosmas E.N.,
- et al.
- Barcelo A.,
- Barbe F.,
- de la Pena M.,
- et al.
- McNicholas W.T.,
- Bonsigore M.R.
- Joyeux-Faure M.,
- Stanke-Labesque F.,
- Lefebvre B.,
- et al.
- Troncoso Brindeiro C.M.,
- da Silva A.Q.,
- Allahdadi K.J.,
- Youngblood V.,
- Kanagy N.L.
- Yamashita K.,
- Discher D.J.,
- Hu J.,
- Bishopric N.H.,
- Webster K.A.
- Lerman A.,
- Holmes D.R. Jr..,
- Bell M.R.,
- Garratt K.N.,
- Nishimura R.A.,
- Burnett J.C. Jr.
- Gjorup P.H.,
- Sadauskiene L.,
- Wessels J.,
- Nyvad O.,
- Strunge B.,
- Pedersen E.B.
- Belaidi E.,
- Beguin P.C.,
- Levy P.,
- Ribuot C.,
- Godin-Ribuot D.
- Kraiczi H.,
- Magga J.,
- Sun X.Y.,
- Ruskoaho H.,
- Zhao X.,
- Hedner J.
- Greenberg H.E.,
- Sica A.,
- Batson D.,
- Scharf S.M.
- Gan X.T.,
- Chakrabarti S.,
- Karmazyn M.
- Yuan G.,
- Nanduri J.,
- Bhasker C.R.,
- Semenza G.L.,
- Prabhakar N.R.
- Natarajan R.,
- Salloum F.N.,
- Fisher B.J.,
- Kukreja R.C.,
- Fowler A.A. 3rd.
- Hocher B.,
- Rohmeiss P.,
- Zart R.,
- et al.
- Allahdadi K.J.,
- Cherng T.W.,
- Pai H.,
- et al.
- Chen J.,
- He L.,
- Dinger B.,
- Stensaas L.,
- Fidone S.
- Patel N.,
- Gonsalves C.S.,
- Malik P.,
- Kalra V.K.
- Pernow J.,
- Wang Q.-D.
- McNicholas W.T.