Experimental Physiology

Transcription

1 Experimental Physiology Effect of intermittent hypoxia on cardiovascular function, adrenoceptors and muscarinic receptors in Wistar rats R. Germack *t, F. Leon-Velarde *$, R. Valdes De La Barra 5, J. Farias 5, G. Soto 5 and 7. P. Richalet * * Laboratoire Rdponses cellulaires et fonctionnelles a l hypoxie, EA 2363, Universitd Paris XIII, Bobigny, France, 4 Laboratorio de Fisiotogia de Altura, Universidad Arturo Prat, Iquique, Chile and * Laboratorio de Transporte de OxigenolIIA, Universidad Cayetano Heredia, Lima, Peru (Revised manuscript received 21 February 2002; accepted 9 April 2002) The usual model of intermittent hypoxia (sleep apnoea) corresponds to repeated episodes of hypoxia from a few seconds to a few hours interspersed with episodes of normoxia. The aim of this study was to evaluate in rats the effect of two periods of intermittent exposure for 2 months to hypoxia (IHX1,24 h in hypoxia (428 Torr), 24 h in normoxia; IHX2,48 h in hypoxia (428 Torr), 24 h in normoxia) as a new model of hypoxia simulating intermittent exposure to high altitude experienced by Andean miners. We assessed the haematological parameters, time course of resting heart rate and systolic blood pressure. We also evaluated the expression of adrenergic and muscarinic receptors. IHXl and IHX2 produced an increase in haematocrit, haemoglobin concentration and mean corpuscular volume as previously seen in most hypoxic models. IHXl and IHX2 induced a simiiar sustained elevation of systolic blood pressure (132 k 2 and 135 f 3 mmhg, respectively, vs. the control level of 121 & 16 mmhg) after I0 days of exposure without change in heart rate. Right ventricular (RV) hypertrophy (225 f 13 and 268 k 15 mg g-i, vs. 178 k 7 mg g- ) and downregulation of a,-adrenoceptor (RV 127 k 21 and 94 k 16 fmol mg- vs. 157 f 8 fmol mg- ; left ventricle (LV): 141 f 5 and 126 f 9 fmol mg- vs. 152 f 5 fmol mg- ) have been found in both groups, with right ventricular hypertrophy being greater and a,-adrenoceptor density being lower in IHX2 than in HX1 groups. These data indicate that both parameters are related to the time of exposure to hypoxia. IHXl and IHXZ produced the same magnitude of upregulation of muscarinic receptors (LV, 60%; RV, 40%), and no change in /3-adrenoceptors. In conclusion, exposure to intermittent hypoxia led to polycythaemia and RV hypertrophy as observed in other types of hypoxia. A specific cardiovascular response was seen, that is an increase in blood pressure without change in heart rate, which was different from the one observed in episodic and chronic hypoxia. Furthermore, this model involved specific modifications of a,-adrenergic and muscarinic expression. Expen mental Physiologv (2002) 87.4, Intermittent hypoxia, which is experienced by a great number of Andean workers (e.g. miners and observatory workers), is a new model of exposure to hypoxia whereby periods of stay at higher altitude are interspersed with periods of stay at sea level. These periods may be as short as one day to several days (Richalet et al. 2002). This model is different from acute (sport and tourism) episodic (sleep apnoea) or chronic (permanent residence) exposure. Episodic and chronic hypoxia are characterized by marked cardiovascular effects to offset a global decrease in tissue oxygen supply, including polycythaemia and an overall sympathetic stimulation (Moore-Gillon & Cameron, 1985; Richalet, 1990; Rostrup, 1998; Greenberg et al. 1999; Neubauer, 2001). In parallel, hypoxic pulmonary vaso- constriction leads to pulmonary hypertension and a right ventricular hypertrophy (Rumsey et al. 1999; Neubauer, 2001). Acute and episodic hypoxia have been found to be associated with an increase in systemic blood pressure and resting heart rate (Fletcher, 2001). In chronic hypoxia, these parameters were unchanged or decreased in rats (Kuwahira et al. 1993; Gonzalez et al. 1998; Favret et al. 2001) or slightly increased in humans (Richalet, 1990; Antezana et al. 1992). In acute (days) and chronic hypoxia in vivo, the progressive blunting of the chronotropic responsiveness has been attributed to a decreased cardiac sensitivity to adrenergic stimulation (Richalet 1990; Antezana et al. 1992) andlor an increased cholinergic activation of the heart (Hartley et al. 1974; Selvamurthy et Publication of The Physiological Society t Corresponding author: richalet@smbh.univ-parisi3.fr 2375

2 454 R. Germak and others Exp. Physiol al. 1981; Zhuang et al. 1993). However, Bernardi et al. (2001) showed that cardiac vagal activity was unchanged in intermittent hypoxia. These modifications can be related in part to alterations in the expression of adrenergic and cholinergic receptors. Indeed, a downregulation of j3-adrenergic receptors (P-ARs) and upregulation of muscarinic receptors have been reported (Kacimi et al. 1993; Lebn-Velarde et al. 1996; Richalet, 1997; Favret et al. 2001). Favret et al. (2001) showed that a-ar density increased after hypoxic exposure for 1-3 days, but was downregulated after this period in chronic hypoxia. A similar downregulation of a,-ars has been observed in vitro in cardiomyocytes exposed to chronic hypoxia (Li et al. 1995). Furthermore, a,-ars are one of the determinants of the cardiac hypertrophic response to pulmonary and/or systemic pressure changes (Benfey, 1990). The hypertrophy induced by a,-ar stimulation is characterized by increased protein synthesis, myofibrillar re-organization and the re-expression of fetal genes (Ikeda et al. 1991; Zimmer et al. 1995). Thus, the a- and /?-adrenergic and cholinergic systems seem to be involved together in the regulation of the responses of the heart to hypoxic exposure. The aim of this study was to establish a new model of intermittent hypoxia with longer hypoxic exposure than the usual models, and to determine the cardiovascular pattern and expression of adrenergic and muscarinic receptors involved in the control of cardiac function, which can be different from those observed in episodic and chronic hypoxia. For this purpose, we evaluated the effects of two different periods of intermittent hypoxia for 2 months on the haernatological parameters, resting heart rate and systolic blood pressure. The density of adrenergic and muscarinic cardiac receptors was also determined. METHODS Animals and hypoxia procedure Male Wistar rats (351? 6 g, n = 25) were separated into three groups: NX, normoxia group (control); IHX1, intermittent hypoxia group 1; IHX2, intermittent hypoxia group 2. For the hypoxic exposure of IHXl, the hypobaric chamber was brought to normobaria for 24 h every other day (24 h hypoxial24 h normoxia), and for the hypoxic exposure of IHX2, the chamber was brought to normobaria for 24 h after 2 days in hypoxia (48 h hypoxia/24 h normoxia). Both groups were exposed to a 4600 m simulated altitude (428 Torr) for a period of 60 days. The control group was housed in the same room and in comparable conditions to the hypoxic groups. In this model we took into account the metabolic rate of a rat which is 4 times greater (0.88 ml g-' h-i) than the metabolic rate of a man (0.23 ml g-' h-'1 (Prosser, 1973). This information was used to calculate the number of days per rat (1 and 2 days) that corresponded approximately to the number of days per man (4 or 8 days) in terms of the effect of exposure to intermittent hypoxia as experienced by Andean workers. In these groups we measured the effect of intermittent hypoxia on the haematological parameters, body weight (BW, g), the time course of changes in heart rate (HR, beats min-i), systolic blood pressure (SBP, mmhg), and the expression of adrenergic and cholinergic receptors. All procedures were performed in agreement with the regulations of the French 'Ministhe de 1'Agriculture' for animal care. Heart rate and blood pressure measurement, and haematological parameters Resting HR and SBP were measured in conscious rats each day under normobaric conditions after each hypoxia exposure, using an inflatable tail-cuff and pressure sensor (RTBP , Kent Scientific, Torrington, CT, USA). The signal was sent through a pre-amplifier to a Workbench data acquisition system. The average of 7-10 measurements was taken for HR and SBP. Data were collected and analysed using the RTBP-001 -DS worksheet (Kent Scientific). On the days of measurement, the animals were acclimated to a movement-limiting Plexiglas chamber for min. After vasodilatation of the tail by warming it, repetitive measurements of systolic pressure were made. The average of 2 days of measurements before beginning hypoxia was used as baseline HR and SBP. Body weight (BW) was also recorded before each BP measurement. After the last in vivo measurements were performed, the animals were killed by stunning and cervical dislocation. The blood sample ( pl) was withdrawn from the cava vein through a 3 cm incision in the peritoneum immediately after the animal was killed, when the blood was still under free-flowing conditions, placed in heparinized microtubules, and spun in a microcentrifuge for the measurement of the haematocrit (Hct). The other haematological parameters were measured using a Coulter Electronics counter (Hialeah, FL, USA). After the blood samples had been obtained, the hearts were quickly removed and dissected free of fat and large vessels. The ventricles were separated from the atria and rapidly put into liquid nitrogen. They were then immediately placed at -80 "C until use. Crude membrane preparation The wet weight of the combined LV plus septum and of the RV were determined rapidly before membrane preparation. The ventricles were minced and homogenized in 10 ml hypotonic buffer (30 mm Tris HC1, 100mM NaCl, 5 m MgC12, ~ 1 mm EGTA, 10 pg ml-' leupeptin and 300 j l phenylmethylsulphonyl ~ fluoride; ph 7.5) with a polytron tissue homogenizer (6 x 5 s bursts). The suspension was centrifuged at 1000 g for 10 min. Soon after, the supernatant was centrifuged at 4500Og for 45 min at 4 C. The pellet was resuspended in 10 ml binding buffer (50 mm Tris HCl, 5 mm MgC1,; ph 7.5) and centrifuged at g for 45 min at 4 "C. The final pellet was resuspended in binding buffer to a final concentration of 2-3 mg protein ml-l and stored at -80 C for subsequent binding studies. Protein concentration was determined by dye-binding assay using a commercial kit (Bio-Rad, Munchen, Germany) and bovine serum albumin was used as standard. All the products were purchased from Sigma-Aldrich Chimie (Lyon, France). Binding studies Saturation studies were performed with 30-7Opg of crude membrane protein using [3H]pra~~~in (a,-antagonist) at concentrations ranging from to 1.5 nm for a,-ars, ['HICGP (@,/@,-antagonist) at concentrations ranging from 0.06 to 3 nm for P-ARs or [3H]quinuclidinyl benzilate (QNB; muscarinic antagonist) at concentrations ranging from 0.01 to 2 nm for muscarinic receptors in a 96-well microplate. To

3 Exp. Physiol Efect of intermittent hypoxia on cardiovascular function 455 Table I. Effect of intermittent hypoxia on haematological parameters Hct Hb RBC CHCM VCM ("/.I (g dl-i) (x lo6 cells pl-') (a dl-') (fl) NX (n = 6) 40k k f k 2.5 IHXl (n=9) 50+_5.8* * 5.2k * ** IHX2(n=9) 51k5.9" 16.4k2.3' 5.1k f1.2* *** The values are means f S.D. of the following parameters: haematocrit (Hct), haemoglobin concentration (Hb), red blood cell count (RBC), mean corpuscular haemoglobin concentration (CHCM) and mean corpuscular volume (VCM). * P < 0.05, **P < 0.01 and ***P < vs. normoxic control (NX). the binding buffer, 50 pi radioligand and 50 pl crude membrane were added to obtain a final volume of 200 ~ 1 The. incubation was carried out in a shaking water bath at 25 C for 60 min for a-ars and muscarinic receptors, and at 37 C for 60 min for P-ARs and then stopped by rapid filtration through glass fibre filters (Filtermats-Receptor binding, Skatron, Lier, Norway) prewetted in binding buffer. The radioactivity trapped by filters was measured using a liquid scintillation p-counter (LS6OOOSC, Beckman, USA). Non-specific binding was determined in the presence of 1 FM prazosin, 10 PM propranolol and 10 p~ atropine to a- and P-ARs and muscarinic receptors, respectively. The radiochemicals were obtained from Amersham Pharmacia Biotech GmBh (Les Ulis, France). Data and statistical analysis The binding parameters (dissociation constant, Kd and receptor density, BmU) were determined using a LIGAND non-linear curve-fitting program (Munson & Rodbard, 1980). The binding data are expressed as mean f S.E.M. For the other data, the values are expressed as mean f S.D. Differences across all testing conditions as well as differences between the groups IHXl and IHX2 were determined by a repeated measures two-way analysis of variance for physiological measurements and one-way analysis of variance ANOVA for receptor expression. Tukey's post hoc comparisons were used to identify significant differences among means. P < 0.05 was considered as statistically significant. RESULTS Haematological parameters, heart rate and blood pressure Intermittent hypoxia produced polycythaemia, as has been observed previously in our models and in other models of hypoxic exposure (Moore-Gillon & Cameron, 1985; Neubauer, 2001). Table 1 shows all the haematological parameters for the NX, IHXl and IHX2 groups. IHXl and IHX2 groups showed higher values of Hct, haemoglobin (Hb) and mean corpuscular volume than the NX group. Blood pressure increased after 10 days of hypoxia (IHXl, 132 & 2 mmhg; IHX2, 135 k 3 mmhg) when compared with NX (121 f 16 mmhg; P < 0.05), and when compared with the blood pressure values obtained in the same animals in normoxia (IHX1, 121 +_ 3 mmhg; IHX2, 121 k 7 mmhg). Blood pressure remained elevated until the end of the experiment (Fig. 1A). Heart rate was not modified by exposure to hypoxia in either group (IHX1, 342 k 17 beats min-'; IHXZ, 345 f 7 beats rnin-') when e v) v, 170- a rr m a c1i ' s.* *. g, 120- R w 110 I 8 1 I I I Time exposure (days) f e I - I Time exposure (days) Figure 1 Systolic arterial blood pressure (A) and mean heart rate (B) in rats exposed to 60 days of intermittent hypoxia. NX, normoxia; IHXl, 24 h hypoxia/24 h normoxia; IHX2,48 h hypoxial24 h normoxia. Measurements were made each normobaric day after each period of exposure to hypoxia. Heart rate values were not significantly different at the end of the exposure. Blood pressure values from days 15 to 60 in hypoxia were significantly higher than controls for both IHXl and IHX2 values; ** P < 0.01.

4 ~ 456 R. Germak and others Exp. Physiol Table 2. Effect of intermittent hypoxia on ventricle weight and weight ratio Left ventricle Right ventricle LV ratio RV ratio (mg) (mg) (rng g-'1 (mg g-y NX (n = 7) 635? k f f 0.02 IHXl (n = 9) 609 f f 13"* 1.64 f f 0.04a* IHX2 (n = 9) 592 f a**,b* 1.73 f f 0.04a**,b* The values are means f S.E.M. Left ventricle (LV) or right ventricle (RV) ratio: left or right ventricle weightslbody weight. * P < 0.05, ** P < vs. control (NX), vs. HX1. Table 3. Effect of intermittent hypoxia on adrenoceptor expression in left and right ventricles Left ventricle Right ventricle Kd Bmax Kd Bmax (nm) (fmol mg-') (nm) (fmol mg-i) a,-adrenoceptor (n = 6) (n = 6) NX 0.46 k f f k 8 IHXl 0.47 k k k f 21 IHXZ 0.41 f ' 0.43 f f 16" P- Adrenoceptor (n = 6) (n = 4) NX 0.28 f f f k 7 IHXl 0.28 f k f f 7 IHXZ 0.31 k k k 8 The values are means k S.E.M. of separate experiments performed in duplicate. The dissociation constants (Kd) are expressed in nm and the receptor densities (Bmax) in fmol of radioligand bound per mg of membrane proteins: ['Hlprazosin for a,-adrenoceptors and ['HICGP for P-adrenoceptors. * P < 0.05 vs. norrnoxic control (NX). ~~~ ~~ Table 4. Effect of intermittent hypoxia on muscarinic receptor expression in left and right ventricles Left ventricle Right ventricle Kd Bmax Kd Bmax (nm) (fmol rng-i) (nm) (fmol mg-') (n = 5) (n = 6) NX 0.18? * f f 81 IHXl 0.22f f 110* 0.27 f f 83 * IHX2 0.19f * 0.40 f f 107 * The values are means f S.E.M. of separate experiments performed in duplicate. The dissociation constants (Kd) are expressed in nm and the receptor densities (BmaX) in fmol of ['HIQNB bound per mg of membrane proteins. * P < 0.05 vs. normoxic control (NX). compared with NX (335 k 10 beats min-i), and with their own heart rate values in normoxia (IHX1, 334 k 10 beats min-'; IHX2, 342 * 23 beats min-') (Fig. 1B). There was no statistical difference between the two groups exposed to intermittent hypoxia, either for blood pressure or heart rate. Body and ventricular weights Intermittent hypoxic exposure for 60 days induced in both animal groups a significant reduction in body weight gain compared to NX (IHX1, 378 f 34 g; IHX2, 348 +_ 22 g versus NX, 398 f 29 g; P < 0.05). Intermittent hypoxia resulted in right ventricular hypertrophy as assessed by the ratio of RV weight to body weight (Table 2). The degree of hypertrophy was greater in the IHXZ group (22%). The wet weight of the LV and the ratio of LV weight to body weight of either group were not changed significantly compared to the control group. a,- and /3-adrenoceptor expression The a, - and P-adrenoceptor binding characteristics are shown in Table 3. Figure 2 is an illustration of the binding experiments. The a,-ar density was lower in hypoxia without change in affinity whatever the model of hypoxia as compared to the normoxic control although the difference between IHXl group and NX was not significant. The longest continuous exposure to hypoxia (IHX2) was associated with a significant decrease in the a,-ar density. This decrease was greater in RV than in LV when compared to the control group. The diminution of the a,-ar density for the IHXl and IHXZ groups was 7% and 17% for the LV, and 19 Yo and 40 Yo for the RV, respectively.

5 Exp. Physiol Effect of intermittent hypoxia on cardiovascular function 457 In contrast to the a,-ar expression, intermittent hypoxia produced no change in the P-AR number in RV as well as in LV (Table 3). The affinity ofp-ars and a,-ars was not modified by intermittent hypoxia. Muscarinic receptor expression Muscarinic receptors, implicated in the P-antiadrenergic effect, were increased by about 60 Yo in LV and 40 5% in RV in intermittent hypoxia independent of the time of exposure (Table 4). Unlike the a,-adrenoceptor, the modification of the muscarinic receptor density was greater in LV. The affinity was not altered by hypoxic exposure in any group. DISCUSSION Different types of hypoxic exposure have been described in the literature as alternating periods of hypoxia and reoxygenation as in episodic hypoxia. In these models of intermittent hypoxia, the cycle length of exposure to hypoxia ranges from a few seconds to a few hours per day (Neubauer, 2001). In our model, the period of hypoxia was longer, from 1-2 days. We estimated the number of days per rat that would correspond to the number of days per man in terms of the effect of exposure to intermittent hypoxia (Prosser, 1973) in order to simulate the intermittent exposure to high altitude experienced by Andean miners. To establish a new model and characterize the mechanism of adaptation to this type of intermittent hypoxia, we evaluated the changes in haematological parameters, heart rate and systolic blood pressure, as well as the expression of adrenergic and muscarinic receptors. Our results showed that adaptation to intermittent hypoxia led to an increase in systolic blood pressure without change in heart rate. This pattern is different from the one observed in episodic and chronic models of hypoxia, and involved specific modifications in a,-adrenergic and muscarinic receptor density without changes in P-adrenoceptor expression. Intermittent hypoxia (IHX) in our model produced polycythaemia as observed in chronic hypoxia and different models of intermittent hypoxia (Moore-Gillon & Cameron, 1985; Kacimi et al. 1992). Hb concentration and Hct increased by around 25 %, which was less than in rats exposed to chronic hypoxia, with increases of 40 and 70 %, respectively (Kacimi et al. 1992). No change in Hct level was found in episodic hypoxia associated with very short repetitive exposure (30 s of hypoxia; Fletcher et al. 1992). These data suggest that the degree of changes in blood parameters depends on the duration of exposure to hypoxia. Increased Hct, which enhances the vascular resistance by increasing blood viscosity, contributes to pulmonary and systemic hypertension. Systemic blood pressure increases significantly in episodic and progressive hypoxia (Fletcher, ) in contrast to chronic exposure where no changes, or only small variations were observed (Richalet, 1990; An tezana et al. 1992; Kuwahira et al, 1993; Gonzalez et al. 1998; Favret et al. 2001). Our results are in good agreement with those obtained using these models of intermittent hypoxia. Tail-cuff systolic blood pressure, as measured here, and invasively measured mean arterial pressure, were shown to be similar in rats exposed to intermittent hypoxia (Fletcher et al. 1992). Although we did not measure cardiac output and mean arterial pressure, the observed 30 Yo increase in systolic blood pressure in both groups may reflect systemic hypertension and an increase in systemic vascular resistance. The polycythaemia alone is not sufficient to explain this hypertension since it has not been described in chronic hypoxia. Indeed, the chronically intermittent stimulation of peripheral chemoreceptors has been shown to be originally implicated in the increase in systemic blood pressure in hypoxia (Fletcher et al. 1992, LeBke et al. 1997) whereas the desensitization of peripheral chemoreceptors observed in chronic hypoxia may prevent the development of systemic hypertension. It appears also that the increased B ~~]~razosine], n~ OM( AlHXl O I W ~[~~~~razosinel, n~ Figure 2 Expression of a,-adrenoceptors. Specific ['Hlprasozin binding to left (A) and right (B) ventricles from rats exposed to 60 days of intermittent hypoxia. NX, normoxia; IHX1,24 h hypoxial24 h normoxia; ihx2,48 h hypoxiat24 h normoxia. Each saturation curve is representative of an experiment that was performed in duplicate. S.D. values not shown are within the size of the symbol.

6 458 R. Gerrnak and others Exp. Physiol activity of adrenergic and renin-angiotensin systems contributes to the early chronic elevated blood pressure in intermittent hypoxia (Flecher, 2001). Finally, erythropoietin may have a direct growth effect on blood vessels and exacerbate systemic hypertension in already hypertensive animals (Gogusev et al. 1994). In our model, IHX did not modify resting heart rate whereas in episodic hypoxia (sleep apnoea) this variable is increased (Fletcher, 2001) and it is unchanged or decreased in chronic hypoxia (Kuwahira et al. 1993; Gonzalez et al. 1998; Favret et al. 2001). This result indicates that a longer exposure time to hypoxia, even with alternance in normoxia produced a cardiovascular pattern specific to our model of IHX switching towards a chronic pattern with respect to resting heart rate. In the present model of intermittent hypoxia, hypertrophy of the right ventricle (RV) was greater in the IHX2 group (51 Yo) than in the HX1 group (26%) indicating that RV hypertrophy is also related to the duration of exposure to hypoxia. No significant change was found in the left ventricle (LV) in either group. Adaptation to hypoxia leads to a cardiac asymmetry since hypoxia-induced atrophy of the LV results in reduced oxidative capacity while the RV is hypertrophied due to pressure overload produced by pulmonary hypertension (Rumsey et al. 1999) Favret et al. 2001). The lack of hypertrophy of the LV in spite of systemic hypertension may have been influenced by an enhanced catabolic or a decreased anabolic state as demonstrated by a lower weight gain in both hypoxic groups. We can also postulate that the hypertrophic stimulus produced by the increase in afterload was not large enough to produce hypertrophy of the LV. It has been shown that the activation of a,-ars stimulates the hypertrophic response in cardiomyocytes, both in vivo (Zimmer et al. 1995) and in vitro (Ikeda et al. 1991). The downregulation of a,-ars was greater in the hypertrophied RV (40%) than in the LV (17%). These data suggest that the downregulation of a,-ars may be more related to hypertrophy than to hypoxia. A prolonged exposure to the agonist results in inactivation and desensitization of G-protein-coupled receptors, such as adrenoceptors and muscarinic receptors, by a downregulation process (Bunemann et al. 1999). Noradrenaline (NA, norepinephrine) levels are increased in episodic hypoxia as a result of sympathetic stimulation (Bao et al. 1997; Greenberg et al. 1999). Thus, an elevated circulating NA level can explain the downregulation of a,-ars. Similar results were obtained in vitro by combining hypoxia and NA (Li et al. 1995). However, the a,-ar density was decreased without change in P-AR number, in contrast to chronic hypoxia where both a,- and P-ARs were downregulated (Favret et al. 2001). Favret et al. (2001) showed that the decrease in resting heart rate is consistent with the downregulation of P-ARs in chronic hypoxia. In our model, resting heart rate was unchanged, which is in accordance with the unaltered P-AR density. Furthermore, it is well known that NA stimulates a-ars preferentially to P-ARs (Liggett & Raymond, 1993), which can explain the downregulation observed only in a,-ars in IHX. We can also suggest that the adrenergic stimulation is not sufficient in IHX (1 or 2 days in hypoxiall day in normoxia) to induce a decrease in P-AR density, particularly when no downregulation was reported in chronic hypoxia before 3 days of exposure (Favret et al. 2001). Muscarinic receptors are involved in the regulation of cardiac function by mediating negative chronotropic and inotropic effects. These receptors have been shown to be upregulated in chronic hypoxia (Kacimi et al. 1993; Leon- Velarde et al. 1996; Favret et al. 2001). Furthermore, this regulation was rapid and sustained indicating that the cholinergic system is the first modulated by hypoxia. Our results are in agreement with these studies. The same magnitude of upregulation of muscarinic receptors was found in both ventricles in IHXl and IHX2 (LV, 60%; RV, 40 YO). It has been shown that chronic heart failure, leading to an increase in plasma concentrations of catecholamines, results in a decreased response to isoprenaline and an increased response to carbachol (Borst et al. 1999). Furthermore, P-AR stimulation leads to an increase in muscarinic receptor density. These results are supported by the downregulation of muscarinic receptors induced by a chronic P-AR blockade in the rat heart (Marquetant et al. 1992). These findings suggest cross-talk between P-ARs and muscarinic receptors, which can take place in chronic hypoxia and partly in intermittent hypoxia since no P-AR downregulation was observed. However, muscarinic receptors are downregulated after stimulation by the agonist like most G-protein-coupled receptors (Haddad & Roussel, 1998). Thus the upregulation observed in intermittent hypoxia cannot be related to an increase in vagal activity but rather to a decrease or a lack of change in cardiac vagal activity, as shown by Bernardi et al. (2001) in intermittent hypoxia. In conclusion, intermittent hypoxia led to changes that were in some ways similar to other models of hypoxia and also specific to this model: a moderate increase in haematological constants, a sustained elevation of systolic blood pressure without modification of resting heart rate, and the development of right ventricular hypertrophy. The regulation of cardiac receptors involved an upregulation of muscarinic receptors and a downregulation of a,-ars without change in P-ARs. The density of a,-ars and the hypertrophic process were related to the total duration of exposure to hypoxia, in contrast to muscarinic receptor expression which was independent of exposure time. Studies need to be carried out to further our understanding of the mechanisms of regulation implicated in this new model of intermittent hypoxia, and especially to elucidate the physiological responses produced by the changes in receptor expression.

7 Exp. Physiol Efiect of intermittent hypoxia on cardiovascular function 459 ANTEZANA, A. M., RICHALET, 1. P., ANTEZANA, G., SPIELVOGEL, H. & KACIMI, R. (1992). Adrenergic system in high altitude residents. International Journal of Sports Medicine 13, S96-S100. BAO, G., METREVELI, N., LI, R., TAYLOR, A. & FLETCHER, E. (1997). Blood pressure response to chronic episodic hypoxia: role of sympathetic nervous system. Journal of Applied Physiology 83, BENFEY, B. G. (1990). Function of myocardial alpha-adrenoceptors. Life Sciences 46, BERNARDI, L., PASSINO, C., SEREBROVSKAYA, Z., SEREBROVSKAYA, T. & APPENZELLER, 0. (2001). Respiratory and cardiovascular adaptations to progressive hypoxia; effect of interval hypoxic training. European Heart journal 22, BORST, M. M., SZALAI, P., HERZOG, N., KOBLER, W. & STRASSER, R. H. (1999). Transregulation of adenylyl-cyclase-coupled inhibitory receptors in heart failure enhances anti-adrenergic effects on adult rat cardiomyocytes. Cardiovascular Research 44, BONEMA", M., LEE, K. B., PALS-RYLAARSDAM, R., ROSBERRY, A. G. & HOSEY, M. M. (1999). Desensitization of G protein-coupled receptors in the cardiovascular system. Annual Review of Physiology 61, FAVRET, F., RICHALET, J. P., HENDERSON, K. K., GERMACK, R. & GONZALEZ, N. C. (2001). Myocardial adrenergic and cholinergic receptor function in hypoxia: correlation with 0, transport in exercise. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology 280, R FLETCHER, E. C. (2001). Physiological consequences of intermittent hypoxia: systemic blood pressure. journal of Applied Physiology 90, FLETCHER, E. C., LEBKE, J., CULMAN, C., MILLER, C. C. 111 & UNGER, T. (1992). Repetitive, episodic hypoxia causes diurnal elevation of blood pressure in rats. Hypertension 19, GREENBERG, H. E., SICA, A., BATSON, D. & SCHARF, S. M. ( 1999). Chronic intermittent hypoxia increases sympathetic responsiveness to hypoxia and hypercapnia. Journal of Applied Physiology 86, GOGUSEV, J., ZHU, D. L., HEREMBERT, T., MMARGUELLAT, F., MARCHE, P. & DRUEKE, T. (1994). Effect of erythropoietin on DNA synthesis, proto-oncogene expression and phospholipase C activity in rat vascular smooth muscle cells. Biochemical and Biophysical Research Communications 199, GONZALEZ, N. C., CLANCY, R. L., MOUE, Y. & RICHALET, J. P. (1998). Increasing maximal heart rate increases maximal 0, uptake in rats acclimatized to simulated altitude. journal ofapplied Physiology 84, HADDAD E. B. & ROUSSELL, J. (1998). Regulation of expression and function of the M, muscarinic receptor. Trends in Pharmacological Sciences 19, HARTLEY, L. H., VOGEL, J. A. & CRUZ, J. C. (1974). Reduction of maximal exercise heart rate at altitude and its reversal with atropine. Journal of Applied Physiology 36, IKEDA, U., TSURUYA, Y. & YAGINUMA, T. (1991). a,-adrenergic stimulation is coupled to cardiac myocyte hypertrophy. American Journal OfPhysiology 260, H KACIMI, R., RICHALET, J. P., CORSIN, A., ABOUSAHL, I. & CROZATIER, B. ( 1992). Hypoxia-induced down-regulation of 8-adrenergic receptors in rat heart. Journal of Applied Physiology 73, KACIMI, R., RICHALET, J. P. & CROZATIER, B. (1993). Hypoxia-induced differential modulation of adenosinergic and muscarinic receptors in rat heart. journal ofapplied Physiology 75, KUWAHIRA, I., HEELER, N., PIIPER, J. & GONZALEZ, N. C. (1993). Effect of chronic hypoxia on hemodynamics, organ blood flow and 0, supply in rats. Respiration Physiology 92, L~oN-VELARDE, F., RICHALET, J. P., CHAVEZ, J. C., KACIMI, R., RIVERA- CHIRA, M., PALACIOS, 1. A. & CLARK, D. (1996). Hypoxia and normoxia-induced reversibility of autonomic control in Andean guinea pig heart. Journal of Applied Physiology 81, LEBKE, J., FLETCHER, E. C., BAO, G. & UNGER, T. (1997). Hypertension caused by chronic intermittent hypoxia - influence of chemoreceptors and sympathetic nervous system. Journal of Hypertension 15, LI, H. T., LONG, C. S., ROKOSH, D. G., HONBO, N. Y. & ORLINER, J. S. (1995). Chronic hypoxia differentially regulates a 1-adrenergic receptor subtype mrnas and inhibits a1 -adrenergic receptorstimulated cardiac hypertrophy and signaling. Circulation 92, LIGGETT, S. B. & RAYMOND, J. R. (1993). Pharmacology and molecular biology of adrenergic receptors. Baillieres Clinical Endocrinology and Metabolism 7, MARQUETANT, R., BREHM, B. & STRASSER, R. H. (1992). Chronic p- blockade transregulates inhibitory A, adenosine and muscarinic M1 receptors of the adenylyl cyclase system. Journal of Molecular and Cellular Cardiology 24, MOORE-GILLON, J. C. & CAMERON, I. R. (1985). Right ventricular hypertrophy and polycythaemia in rats after intermittent exposure to hypoxia. Clinical Sciences 69, MUNSON, P. J. & RODBARD, D. (1980). Ligand: a versatile computerized approach for the characterization of ligand binding systems. Annals of Biochemistry 107, NEUBAUER, 7. A. (2001). Physiological and pathophysiological responses to intermittent hypoxia. Journal ofapplied Physiology 90, PROSSER, C. L. (1973). Energy metabolism. In Comparative Animal Physiology, 3rd edn. W. B. Saunders Company, Philadelphia. RICHALET, J. P. (1990). The heart and adrenergic system in hypoxia. In Hypoxia: The Adaptations, ed. SUTTON, J. R., COATES, G. J. E. & REMMERS, J. E., pp B. C. Decker Inc., Toronto, Philadelphia. RICHALET, J. P. (1997). Oxygen sensors in the organism: examples of regulation under altitude hypoxia in mammals. Comparative Biochemishy and Physiology 118A, RICHALET, J. P., VARGAS, M., JIMENEZ, D., ANTEZANA, A. M., HUDSON, C., CORTES, G., OSORIO, 1. & LEON, A. (2002). Chilean miners commuting from sea level to 4500 m: a prospective study. High Altitude Medicine and Biology (in the Press). ROSTRUP, M. (1998). Catecholamines, hypoxia and high altitude. Acta Physiologica Scandinavica 162, RUMSEY, W. L., ABBOTT, B., BERTELSEN, D., MALLAMACI, M., HAGAN, K., NELSON, D. & ERECINSKA, M. (1999). Adaptation to hypoxia alters energy metabolism in rat heart. American Journal of Physiology 276, H SELVAMURTHY, W., SAXENA, R. K., KRISHNAMURTHY, N. & NAYAR, H. S. (1981). Autonomic responses of high-altitude natives during sojourn at plains and on return to altitude. Aviation Space and Environmental Medicine 52,

8 460 R. Germak and others Exp. Physiol ZIMMER, H. G., KOLRECK-RUHMKORFF, C. & ZIERHUT, W. (1995). Cardiac hypertrophy induced by alpha- and beta-adrenergic receptor stimulation. Curdioscience 6, ZHUANC, J., DROMA, T., SUTTON, J. R., MCCULLOUGH, R. E., MCCULLOUGH, R. G., GROVES, B. M., RAPMUND, G., JANES, C., SUN, S. 8r MOORE, L. C. (1993). Autonomic regulation of heart rate response to exercise in Tibetan and Han residents of Lhasa (3.658 rn). Journal ofapplied Physiology 75, Acknowledgements This study was supported by FONDEF (Chile) D97-I 1050, and from INSERM/CONICYT /1999.