INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CHRONIC HIGH-ALTITUDE HYPOXIA leads to pulmonary hypertension and, subsequently, to right ventricular (RV) hypertrophy. The left ventricle (LV) usually does not hypertrophy, unless there is rather severe and prolonged intermittent hypoxia (9, 26). It is well established that the heart of animals adapted to chronic hypoxia exhibits increased tolerance to an acute ischemic injury (reviewed in Ref. 8) and impaired chronotropic and inotropic responsiveness to -adrenergic stimulation (13, 24). These responses, as well as the agonist-stimulated Ca²⁺ transients in cardiac myocytes, appear to be similar in the hypertrophic RV and nonhypertrophic LV (1, 25), suggesting that they relate to hypoxia rather than to hypertrophy.

Reduced adrenergic responsiveness of the chronically hypoxic heart is due to alterations of the -adrenoceptor (-AR)-adenylyl cyclase (AC) signaling system. A number of in vivo studies demonstrated a downregulation of -ARs and a desensitization of AC in the hypoxic heart (10, 12, 14, 36). These changes appear to reflect increased sympathetic activity, as evidenced by elevated plasma and urine concentration of catecholamines (2, 16) and impaired neuronal reuptake of norepinephrine, which leads to its increased synaptic levels (14). However, direct effects of chronic hypoxia may differ, as illustrated by increased -ARs expression and unchanged AC activity in isolated cardiac myocytes (11).

-ARs, as well as some other types of receptors, are coupled to AC through G proteins. It has been proposed that either a decrease in the stimulatory G (G_s) proteins or an increase in the inhibitory G (G_i) proteins may participate in the impairment of -AR responsiveness of chronically hypoxic hearts under in vivo conditions (7). The available information is, however, relatively scarce and rather inconclusive. Taken together, total G_s protein levels have been reported to be either decreased or unchanged, and those of G_i protein either increased or unchanged (6, 10, 14, 24). Whereas in certain studies changes in G proteins induced by chronic hypoxia were similar in the two ventricles (24), others point to significant right-to-left differences (6, 10). Moreover, a discord among gene expression, protein levels, and their functional activity was observed in both RV and LV of chronically hypoxic animals (6).

Many of the characteristic cardiopulmonary changes induced by chronic hypoxia are fully reversible after recovery of animals under normoxic atmosphere for a sufficient period of time, usually from 2 to 5 wk. However, some changes, such as increased cardiac ischemic tolerance, myocardial fibrosis, and muscularization of distal pulmonary arterioles may persist for a considerably longer period (reviewed in Ref. 23). To our knowledge, no information is available concerning possible reversibility of hypoxia-induced alterations of myocardial -AR-AC signaling system and cardiac -adrenergic responsiveness.

In the present study, we examined the effects of severe, chronic, intermittent high-altitude (IHA) hypoxia corresponding to 7,000 m above sea level on the AC signaling system in the RV and LV myocardium of adult rats. We also studied reversibility of hypoxia-induced changes in animals recovering from IHA hypoxia for 5 wk under normoxic conditions.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Materials. [-³²P]ATP, [³H]cAMP, and [³H]CGP-12177 were purchased from Amersham Biosciences (Buckinghamshire, UK), and scintillation cocktail CytoScint from ICN Biomedicals (Irvine, CA). Acrylamide and bis-acrylamide were from SERVA (Heidelberg, Germany), and aluminum oxide 90 (neutral, activity I) was from Merck (Darmstadt, Germany). All other chemicals were from Sigma Chemical (St. Louis, MO), and they were of the highest purity available.

Experimental model. Adult male Wistar rats were exposed to IHA hypoxia in a hypobaric chamber for 8 h/day, 5 consecutive days/wk. Barometric pressure (PB) was lowered stepwise, starting at the level equivalent to an altitude of 2,000 m during the first exposure. The level equivalent to the highest altitude of 7,000 m (PB = 308 mmHg, 41.1 kPa; PO₂ = 65 Torr, 8.7 kPa) was reached after 13 exposures; the total number of exposures was 25 (i.e., 5-wk adaptation period). For the remaining period of 1 day and 2 whole days/wk, the animals were kept at PB and PO₂ equivalent to an altitude of 200 m (742 mmHg, 98.9 kPa and 155 Torr, 20.7 kPa, respectively). It has been shown earlier that all major hypoxia-induced cardiopulmonary effects are already complete after 25 exposures, and no appreciable changes occur thereafter (28). One-half of the group of adapted animals was employed 24 h after the last hypoxic (H) exposure, whereas the remaining animals were kept at these normoxic conditions for a further 35 days [hypoxic-recovery (H_R)]. The age-matched control groups of animals were kept for the corresponding period of time under normoxic conditions [normoxic (N) and normoxic-recovery (N_R)]. All animals had free access to water and a standard laboratory diet. The study was conducted in accordance with the "Guide for the Care and Use of Laboratory Animals," published by the US National Institutes of Health (NIH Publication no. 85-23, revised 1996).

Assessment of cardiac inotropic response to isoproterenol. We decided to examine responses of the RV, which exhibited more pronounced hypoxia-induced changes in AC signaling than the LV. Animals were killed by decapitation, and the heart was quickly excised, washed in cold (5°C) saline, and perfused in the Langendorff mode under constant pressure (9.8 kPa) and nonrecirculating conditions. A Krebs-Henseleit perfusion solution contained the following (in mmol/l): 118.0 NaCl, 4.7 KCl, 1.25 CaCl₂, 1.2 MgSO₄, 25.0 NaHCO₃, 1.2 KH₂PO₄, 7.0 glucose, and 2.0 sodium pyruvate. The solution was saturated by a mixture of 95% O₂ and 5% CO₂ (pH 7.4), and its temperature was maintained at 37°C. The hearts were electrically stimulated by pulses of 1-ms duration at 350 beats/min. The contractile function of the RV was measured under isovolumetric conditions by using a nonelastic balloon inserted into the ventricular cavity, filled with water, and connected to a pressure transducer. The transducer output was amplified (HGM, Experimetria, Hungary), and the analog pressure signal was analyzed on-line by using our custom-designed software. The following parameters were derived: developed pressure, maximal rate of pressure development, and maximal rate of pressure fall. Coronary flow was measured by a timed collection of coronary effluent and subsequently normalized to ventricular weight. After a stabilizing period (20 min), the volume of the intraventricular balloon was gradually increased to reach diastolic pressure of 3-4 mmHg. The concentration-response curve of isoproterenol (3 × 10¹⁰ to 1 × 10⁷ mol/l) was measured in a cumulative manner in the presence of 1 × 10³ mol/l ascorbic acid. The geometric EC₅₀ value was estimated as the concentration of isoproterenol at which the maximal rate of pressure development reached 50% of maximal response.

Tissue procurement and homogenization. The animals assigned to biochemical experiments were killed by decapitation, and the hearts were excised, rinsed in cold (5°C) saline, trimmed of atria and large vessels, and dissected into the LV and RV and the septum. Each part was weighted, and the ventricles were frozen in liquid nitrogen and stored at 70°C until use.

-AR binding. Myocardial -ARs were determined by radioligand-binding assay with the -antagonist [³H]CGP-12177, as described previously (20). Briefly, samples of myocardial membranes (200-µg protein) were incubated in a buffer B (50 mM Tris · Cl, 10 mM MgCl₂, and 1 mM ascorbic acid; pH 7.4) containing 4 nM [³H]CGP-12177 at 37°C for 2 h (total volume of 0.5 ml). The binding reaction was terminated by adding 3 ml of ice-cold buffer C (50 mM Tris · Cl and 10 mM MgCl₂; pH 7.4) and subsequent filtration through GF/C filters presoaked for 1 h with polyethylenimine. The filters were then washed two times with 3 ml of ice-cold buffer C. After addition of 4-ml scintillation cocktail CytoScint, radioactivity retained on the filters was measured by counting for 5 min. Nonspecific binding was defined as that not displaceable by 10 µM L-propranolol, and it represented <25% of total binding.

Electrophoresis, immunoblotting, and densitometric analysis. All types of myocardial preparations were solubilized (1:1) in Laemmli buffer and loaded (50 µg/lane) on standard (10% acrylamide-0.26% bis-acrylamide) or urea (12.5% acrylamide-0.0625% bis-acrylamide containing 6 M urea) polyacrylamide gels. SDS-PAGE was carried out at 200 V for 60 min on a Mini-Protean II apparatus (Bio-Rad, Hercules, CA). After electrophoresis, the resolved proteins were transferred to nitrocellulose membrane (Schleicher and Schuell), blocked with 4% BSA in TBS buffer (10 mM Tris, 150 mM NaCl; pH 8.0) for 1 h, and then incubated with relevant G protein-specific primary antisera for 2 h at room temperature. Anti-G protein -subunit (G_s) antibody was purchased from Sigma Chemical, and preparation and characterization of primary rabbit anti-G_i1,2 antiserum ("SG1") was described previously (17). After three 10-min washes in TBS containing 0.3% Tween 20, the secondary goat anti-rabbit IgG labeled with horseradish peroxidase was applied for 1 h. After another three 10-min washes in TBS-Tween, the blots were visualized by enhanced chemiluminescence technique, according to the manufacturer's instructions (Pierce Biotechnology, Rockford, IL). For reliable identification of the individual G protein subunits, guinea pig myocardial membranes rich in G_s (33) or rat brain microsomes highly abundant with G_i proteins (3) were included as standards on each blot. The immunoblots were scanned (Astra 610P, UMAX) and quantitatively analyzed by the ImageQuant computer program.

Determination of AC activity. Activity of AC was determined according to the method of White (37). The reaction mixture (in a total volume of 0.1 ml) contained 100 µg of protein, 50 mM Tris · HCl buffer (pH 7.4), 5 mM MgCl₂, 1 mM EDTA, 50 U/ml pyruvate kinase, 10 mM potassium phosphoenolpyruvate, 160 µg/ml BSA, 0.2 mM 3-isobutyl-1-methylxanthine, 10 µM GTP, 0.1 mM cAMP, 10,000 counts · min¹ · sample¹ of [³H]cAMP, and 0.4 mM ATP with [-³²P]ATP (~1 × 10⁶ counts · min¹ · sample¹). The assay was run for 30 min at 32°C in the presence of 100 µM GTP, and the following different stimulators were used in separate experiments: forskolin (50 µM), AlF (10 mM NaF and 1 mM AlCl₃), or (-)-isoproterenol (50 µM). The reaction was terminated by adding 0.2 ml of 0.5 M HCl and heating for 5 min at 100°C. The cAMP formed was separated by alumina columns, and the detected amount of [³²P]cAMP was corrected for recovery with [³H]cAMP.

Reconstitution assay of G_s. This assay consists of reconstituting stimulatible AC activity in membranes from S49 cyc lymphoma cells, which lack G_s protein. Samples of membranes or cytosol derived from RV and LV were mixed with 1% sodium cholate and extracted for 60 min at 4°C. After centrifugation (200,000 g, 30 min, 4°C) and heat inactivation (10 min at 30°C), supernatants (extracts) were used in a cyc-reconstitution assay (34). Extracts (corresponding to 20 µg of proteins) were mixed with S49 cyc membranes (50 µg protein) in a buffer solution of the same composition as that used for determination of AC activity (see above). In some cases, cyc membranes were reconstituted directly with cytosolic preparations (without cholate extraction). The reaction mixture was supplemented with activators [100 µM GTP or AlF (10 mM NaF and 1 mM AlCl₃)] and preincubated at 32°C for 15 min. Reaction was then started by the addition of 0.4 mM ATP with [-³²P]ATP (~1 × 10⁶ counts · min¹ · sample¹), and activity of AC was determined as above.

Data analysis. All values are expressed as means ± SE. Biochemical data were determined in at least three independent preparations. Effect of chronic hypoxia on all tested parameters was analyzed by one-way ANOVA, and group-to-group comparisons were done by using unpaired Student's t-test. Significance of effects of different activators on AC activity was determined by Student's t-test.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Body and heart weight. Body (BW) and heart weight parameters of animals adapted to IHA hypoxia and those kept under normoxia for a further 35 days after the last hypoxic exposure are summarized in Table 1. Compared with the normoxic control group, IHA hypoxia led to a significant retardation of body growth accompanied by an increase of the heart weight due to hypertrophy of both ventricles. The RV weight increased much more than the LV weight, resulting in a rise in the RV-to-LV ratio by 31%. Five weeks after the termination of hypoxia, the BWs of animals were still significantly lower than those of age-matched controls. Whereas a complete regression of LV hypertrophy occurred during this period, the RV weight remained slightly but significantly elevated (the RV weight-to-BW ratio by 18%).

RV function and inotropic response to isoproterenol. Adaptation to IHA hypoxia markedly increased all baseline indexes of the RV contractile function and the coronary flow. Five weeks after the hypoxic period, these indexes recovered only partially and remained still significantly higher compared with those of age-matched normoxic controls, except for the coronary flow, which was close to the control value (Table 2). Addition of isoproterenol to the perfusion medium caused concentration-dependent positive inotropic and positive lusitropic effects in all groups of hearts. In the chronically hypoxic group, the maximal response to isoproterenol decreased by 30%, and the EC₅₀ value increased from 3.7 ± 0.3 (controls) to 4.6 ± 0.3 nmol/l, indicating a moderate decrease in sensitivity. Hearts of rats recovering from hypoxia did not exhibit any significant difference in -adrenergic responsiveness compared with those of corresponding normoxic controls (Fig. 1, Table 2).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Effect of increasing concentration of isoproterenol (IPRO) on maximal rate of right ventricular (RV) pressure development [(+dP/dt)max] expressed as percent increase above the baseline value in isolated perfused hearts of rats adapted to intermittent high-altitude hypoxia (H; A) and of animals recovering from hypoxia (HR; B) compared with corresponding age-matched controls [N (A) and NR (B), respectively]. Values are means ± SE from 7-8 hearts in each group. See Table 2 for EC50 values.

-ARs. Our preliminary saturation binding experiments performed on myocardial membranes indicated that 4 nM concentration of [³H]CGP-12177 was sufficient to achieve the maximal binding [dissociation constant (K_D)  350 pM]. The levels of -ARs determined by [³H]CGP-12177 binding were similar in preparations from both RV and LV (~25 fmol/mg protein), and they were not significantly altered by adaptation of rats to IHA hypoxia (Fig. 2).

View larger version (28K): [in this window] [in a new window]   Fig. 2.   -Adrenoceptor density in RV (A) and left ventricular (LV; B) preparations from H, HR, N, and NR rats. The total number of -adrenoceptors was assessed by [3H]CGP-12177 binding as described in MATERIALS AND METHODS. Values are means ± SE of 3 separate determinations performed in duplicates.

G proteins. The G protein -subunits in myocardial membrane and cytosolic fractions were identified on Western blots by specific antisera (Fig. 3A). The long (G_s-L) and short (G_s-S) variants of the G_s were resolved by standard SDS-PAGE (10% polyacrylamide gels). To discern G_s-L and G_s-S in myocardial membranes containing relatively high amounts of these proteins, it was necessary to use a lower load of these preparations (Fig. 3B). Application of urea-SDS-PAGE (13% polyacrylamide gels containing 6 M urea) showed up advantageously for a sufficient resolution of G_i1 and G_i2 proteins (Fig. 3A).

View larger version (75K): [in this window] [in a new window]   Fig. 3.   A: representative immunoblots showing the distribution of stimulatory (Gs) and inhibitory G protein -subunit (Gi) proteins in RV and LV preparations from H, HR, N, and NR rats. After resolution by SDS-PAGE (Gs) or urea-SDS-PAGE (Gi), myocardial membrane or cytosolic fractions (50 µg/lane) were immunoblotted with specific G protein antisera, as described in MATERIALS AND METHODS. Guinea pig myocardial membranes (50 µg/lane) and rat brain microsomes (20 µg/lane) were used as standards (St) for reliable identification of Gs and Gi1/2, respectively. B: the long (Gs-L) and short Gs isoforms (Gs-S) in myocardial membranes were resolved by SDS-PAGE by using a lower load of these preparations (15 µg/lane), and immunoblot detection was conducted as above.

View larger version (36K): [in this window] [in a new window]   Fig. 4.   Effect of H and subsequent HR on the levels of myocardial Gs and Gi proteins. The membrane (A and B) and cytosolic fractions (C and D) derived from rat RV (A and C) and LV (B and D) were analyzed by using immunoblotting, and the relative changes in Gs levels were assessed by densitometric scanning. Values are means ± SE in arbitrary units. Statistically significant differences H vs. N and HR vs. NR rats: * P < 0.05 and ** P < 0.001.

AC activity. To evaluate the influence of adaptation to hypoxia on functional status of the myocardial AC signaling system, activity of AC was determined. AC activity measured under different stimulatory conditions (GTP, AlF, forskolin, or isoprotertenol) was lower by ~50% in RV preparations from hypoxic rats than in those from control animals (Fig. 5). Similarly, but not so profound, depression of AC activity was found in LV preparations. RV preparations from rats recovering from hypoxia still exhibited significantly decreased activity of AC compared with that from corresponding controls. In contrast, AC activity in samples of LV from H_R rats did not significantly differ from that determined in age-matched controls (N_R).

View larger version (55K): [in this window] [in a new window]   Fig. 5.   Effect of chronic hypoxia on myocardial adenylyl cyclase (AC) activity. GTP-, fluoride-, forskolin-, and isoproterenol-stimulated AC activities were measured in RV (A) and LV (B) preparations from H, HR, N, and NR rats. Values are means ± SE in pmol cAMP · min1 · mg protein1 and represent the average of 3 separate determinations performed in duplicate. Statistically significant differences H vs. N and HR vs. NR: * P < 0.05, ** P < 0.01, *** P < 0.001; H vs. HR:  P < 0.05.

G_s protein activity. Functioning of membrane-bound as well as of cytosolic G_s protein in both RV and LV preparations was assessed by a cyc-reconstitution assay. Whereas membrane-bound G_s was always extracted in a standard way with sodium cholate before being mixed with cyc membranes, cytosolic G_s was analyzed both directly and after extraction with sodium cholate. GTP or AlF were used as activators for measurements of cyc-reconstitutive AC activity. Reconstitution analysis indicated a significant decrease (by ~20-35%) in functional activity of membrane-bound G_s in both RV and LV after adaptation to hypoxia (Figs. 6 and 7, respectively). Interestingly, no such clear changes in G_s functional activity were observed in myocardial membrane preparations from H_R rats; a significant reduction (by ~22%) in AlF-stimulated reconstitutive AC activity was found only in RV but not in LV samples. A bit more variable data were obtained from the analysis of cytosolic preparations. A marked decrease was determined in cyc-reconstituted activity of cytosolic G_s in both RV and LV of chronically hypoxic animals. This decrease was clearly detectable, especially in the case when sodium cholate extraction was omitted (Figs. 6 and 7). In addition, the presence of sodium cholate in the assay substantially (~3 times) reduced AlF-stimulated AC activity. Functional activity of G_s in the presence of sodium cholate was rather increased in cytosolic RV, but not in LV, preparations from H_R rats. By contrast, a mild decrease in G_s functional activity was observed in cytosolic LV samples prepared in cholate-free conditions.

View larger version (46K): [in this window] [in a new window]   Fig. 6.   Effect of chronic hypoxia on Gs functional activity in the membrane and cytosolic fractions prepared from RV. GTP and fluoride were used as activators for measurements of a cyc-reconstituted AC activity. Activity of AC was determined after complementation of cyc membranes with sodium cholate (NaCH) extracts of membranes or cytosol. Besides that, the cytosolic fractions were analyzed also without NaCH extraction. Values are means ± SE in pmol cAMP · min1 · mg protein1 and represent the average of 3 separate determinations performed in duplicate. Statistically significant differences H vs. N and HR vs. NR: * P < 0.05, ** P < 0.01.

View larger version (46K): [in this window] [in a new window]   Fig. 7.   Effect of chronic hypoxia on Gs functional activity in the membrane and cytosolic fractions prepared from LV. GTP and fluoride were used as activators for measurements of a cyc-reconstituted AC activity. Activity of AC was determined after complementation of cyc membranes with NaCH extracts of membranes or cytosol. Besides that, the cytosolic fractions were analyzed also without NaCH extraction. Values are means ± SE in pmol cAMP · min1 · mg protein1 and represent the average of 3 separate determinations performed in duplicate. Statistically significant differences H vs. N and HR vs. NR: * P < 0.05, ** P < 0.01; H vs. HR:  P < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The observed effects of long-term exposure of rats to severe IHA hypoxia on heart weight parameters and indexes of baseline RV performance correspond to data reported in previous studies using the same experimental model (9, 26). The RV weight increased by 48%, and the RV developed pressure of the isolated perfused heart was 65% higher than that of normoxic controls. The LV weight increased as well, but only by 13%. Whereas the complete regression of LV hypertrophy occurred after 5 wk of recovery under normoxic atmosphere, the relative RV weight remained still significantly higher (by 18%) compared with the age-matched control value. Similarly, the RV contractile parameters did not fully normalize in animals recovering from hypoxia. It is unclear how long a period of time is needed for a complete regression of these hypoxia-induced changes. In our previous study, significant RV hypertrophy persisted even 60 days after the termination of hypoxic exposure (26). This might be related to a remodeling of ventricular myocardium with marked and persisting accumulation of fibrillar collagenous proteins (26).

In subsequent parts of the present study, we aimed to examine in complexity the G protein-mediated AC signaling in myocardium of rats adapted to IHA hypoxia. Our measurements of -ARs indicated that hypoxia did not significantly influence the total number of these receptors, either in RV or LV, and the receptor levels remained unchanged, even after a 5-wk recovery period under normoxic conditions. A number of controversial data have been published concerning the effects of hypoxia on myocardial -ARs. Voekel and coworkers (36) observed a decrease in -AR density in both RV and LV from Wistar rats after 5 wk of hypoxia, and similar results have been reported by others (14). By contrast, Kacimi and coworkers (7) did not find any change in myocardial -ARs after 2 wk and a mild decline in the receptor levels only in LV after 3 wk of hypoxia. Similar discordant results have been obtained in studies focused on -AR distribution in cultured cells exposed to hypoxia (11, 15, 29, 35). These discrepancies might be, in part, explicable by differences in binding techniques. The major role, however, may be attributed to differences in experimental models, time course, and severity of exposure to hypoxia and other specific conditions.

The G_s and G_i proteins are another crucial regulatory components of the AC signaling system, which could be influenced by adaptation to chronic hypoxia. The content of predominantly membrane-associated forms of myocardial G_s protein appeared not to be affected by chronic IHA hypoxia, but the amount of G_s-L increased compared with that of normoxic controls. In parallel, there were no significant changes in the distribution of G_i proteins in any ventricular preparations. These results are basically in line with some previously published data. Relative levels of immunodetectable membrane-bound G_s and G_i proteins remained unchanged in RV as well as in LV samples from rats exposed to a 30-day hypobaric hypoxia (6). Partially similar observations have been reported by Pei and coworkers (24) in myocytes derived from chronically hypoxic rats. Interestingly, these authors also did not find any changes in G_i protein, but they concurrently detected increased G_s-L and decreased G_s-S levels in the particulate fractions from both RV and LV, compared with the corresponding samples from normoxic animals. These data (increase in G_s-L) fit, at least partly, with our present findings of the raised G_s-L content (and increased G_s-L/G_s-S) in the soluble fraction of myocardial samples derived from chronically hypoxic rats. Differential behavior of G_s isoforms attesting for their inequality have been observed under various physiological and pathophysiological conditions (reviewed in Ref. 22). The physiological significance of alerations in the G_s-L/G_s-S is not clear.

The possible dissociation between the levels of myocardial G_s protein and its functional activity was assessed by reconstitution experiments. These measurements revealed a lower functional activity of membrane-bound G_s in both RV and LV from chronically hypoxic rats. Attenuated functional activity of myocardial G_s due to adaptation to chronic hypoxia was previously described by Kacimi and coworkers (6). Our measurements of G_s functional activity in cytosolic preparations also provided some interesting novel information. Chronic hypoxia apparently reduced G_s activity in both RV and LV, but this effect was clearly detectable only when cyc membranes were complemented with samples of cytosol without sodium cholate. Involvement of cholate extraction blunted the differences and strongly suppressed fluoride-stimulated reconstitutive AC activity. Sodium cholate negatively affects AC activity (18, 34), and it is conceivable that the presence of this substance in reaction mixture also impairs the interaction between fluoride and GDP-bound G_s subunit. The existence of soluble forms of some trimeric G proteins has been noticed in several other earlier studies (4, 5, 19, 21, 32), but their role in cellular signaling still remains obscure.

It has been shown that adaptation to chronic hypoxia is often associated with diminution of AC activity (7, 24, 36). Our present results are in line with these observations. Chronic IHA hypoxia clearly reduced activity of AC in both RV and LV, irrespective of what kind of stimulators were used in the assay (GTP, fluoride, forskolin, or isoproterenol). These data suggest that, besides reduced bioactivity of G_s protein, it is also impaired catalytic function of AC that accounts for attenuated -adrenergic responsiveness observed in RV preparations from hypoxic rats.

Adaptation of rats to IHA hypoxia increases the tolerance of their hearts to a subsequent acute ischemia-reperfusion injury. This cardioprotection manifests itself as limitation of myocardial infarct size, improved recovery of cardiac contractile function after ischemia, and reduced incidence and severity of ischemic and reperfusion ventricular arrhythmias (reviewed in Refs. 8, 23). It seems likely that desensitization of the AC signaling system may contribute to these protective effects of chronic hypoxia by blunting adrenergic stimulation of the heart during ischemia and decreasing its oxygen demand. Moreover, this phenomenon may be responsible for protection of the chronically hypoxic heart against necrotic lesions induced by toxic doses of isoproterenol (27).

The new observation of this study is that altered distribution of cytosolic G_s protein, decreased functional activity of membrane-bound G_s, and derangement of the AC signaling in RV, induced by chronic hypoxia, were not fully amended after a 5-wk recovery of rats under normoxic atmosphere. In fact, the AC activity remained at the same level as in the chronically hypoxic group, and even more elevated levels of cytosolic G_s-L were detectable in RV preparations from these animals. The reason for these persisting alterations is not clear; they do not appear to be related to myocardial hypertrophy, as the RV mass remained only slightly higher in H_R rats than in age-matched controls. Despite deranged AC signaling, the ventricular contractile responsiveness to -adrenergic stimulation returned to normal in H_R animals. It remains to be determined at which level or component of the signaling system the compensation may occur. It can be speculated that the diminished function of the AC signaling might be, at least partially, compensated by recruitment of the ancillary -adrenergic signaling pathway.

In conclusion, adaptation of rats to IHA hypoxia led to increased localization of G_s-L in the cytosolic fractions, and it was associated with reduced bioactivity of membrane-bound G_s protein and lower AC activity, especially in RV. These changes may explain decreased sensitivity to -adrenergic stimulation and may be considered as cardioprotective under conditions of acute myocardial ischemia. Alterations in the AC signaling system do not appear to be readily reversible, even after a 5-wk recovery of animals at normoxia.