- Email: [email protected]
Effects of Angiotensin II Receptor Blockade on Hypoxia-induced Right Ventricular Hypertrophy in Rats
J Mol Cell Cardiol 29, 2931–2939 (1997)
Growth Factors and Cardiac Hypertrophy
Effects of Angiotensin II Receptor Blockade on Hypoxia-induced Right Ventricular Hypertrophy in Rats Michael Irlbeck1, Takaaki Iwai2, Tobias Lerner1 and Heinz-Gerd Zimmer3 1
Department of Anaesthesiology, University of Munich, Germany; 2Jikei University School of Medicine, Tokyo, Japan; and 3Carl-Ludwig-Institute of Physiology, University of Leipzig, Germany M. I, T. I, T. L H.-G. Z. Effects of Angiotensin II Receptor Blockade on Hypoxia-induced Right Ventricular Hypertrophy in Rats. Journal of Molecular and Cellular Cardiology (1997) 29, 2931–2939. It was the aim of the present study to characterize the hemodynamic, biochemical and morphologic effect of angiotensin II receptor blockade on hypoxia-induced right ventricular hypertrophy in rats. Isolated right ventricular hypertrophy was induced in female Sprague–Dawley rats by intermittent hypoxia (IH; 10% O2, 8 h/ day, 5 days/week, 20 days of exposition, n=15). After completion of IH, left- (LV) and right-ventricular (RV) hemodynamic parameters were measured under room air conditions in the intact, thiopental-anesthetized animals with special Millar ultraminiature tipcatheter-manometers. Cardiac output was determined using the thermodilution method. Cell volume (CV) of isolated cardiomyocytes was measured with a Coulter Channellyzer after collagenase cell isolation. The specific activities of the myocardial pentose phosphate pathway enzymes glucose-6-phosphate-dehydrogenase (G-6-PD) and 6-phosphogluconate-dehydrogenase (6-PGD) were determined using a spectrophotometric assay. IH caused a rise in right ventricular systolic pressure (RVSP) from 38.1±0.83 to 58.1±1.42 mmHg and an increase in the RV weight/body weight ratio (RVW/BW) from 0.884±0.053 to 1.166±0.049 mg/g. The activities of G-6-PD and 6-PGD were significantly increased after IH in the RV, but not in the LV. CV was increased from 24 248±1193 to 29 541±1765 lm3, myocardial cell length was unchanged. IH had no influence on the LV parameters or cardiac output. Co-infusion of the angiotensin II receptor antagonist losartan (LO; 12 mg/kg/d i.p., n=14) during the IH period reduced the rises in RVSP (49.4±2.06 mmHg), RVW/ BW (0.99±0.072 mg/g), G-6-PD and 6-PGD significantly, but not completely. The increase in CV, however, was prevented (24 524±2370 lm3) entirely. We conclude from these data that the IH-induced RV-hypertrophy was primarily of the concentric type. LO attenuated the hypoxia-induced isolated RV hypertrophy and significantly reduced the metabolic response of the RV. The LO effect was most potent with regard to the increase in cardiomyocyte volume. 1997 Academic Press Limited K W: Angiotensin II; Losartan; Right ventricle; Pulmonary hypertension; Hypoxia; Hypertrophy.
Introduction Various experimental models exist for the evaluation of right ventricular hypertrophy. Biventricular hypertrophy of the left and right rat heart has been observed and studied after experimentally induced chronic myocardial infarction (by ligation of the left anterior descending coronary artery) (Zimmer et al., 1989, 1990), infusion of catecholamines (Irlbeck et al., 1996a) or treatment with triiodo-
thyronine (Zimmer et al., 1988). Isolated right ventricular hypertrophy without growth responses of the left ventricle is seen after experimental pulmonary artery stenosis (Zierhut et al., 1990) or by induction of pulmonary hypertension by lung irradiation (Zimmer et al., 1988) or chronic exposure to hypoxia (Rabinovitch et al., 1979; Ostadal and Widimsky, 1990; Kolar and Ostadal, 1991). In the chronic hypoxia model, alveolar hypoxia leads to pulmonary vasoconstriction, as described
Please address all correspondence to: Dr Michael Irlbeck, Department of Anaesthesiology, University of Munich, Klinikum Grosshadern, 81366 Muenchen, FRG.
0022–2828/97/112931+09 $25.00/0
mc970528
1997 Academic Press Limited
2932
M. Irlbeck et al.
by v. Euler and Liljestrand (1947). Prolonged exposure to inspiratory hypoxia causes pulmonary hypertension and consequent isolated right ventricular hypertrophy without growth responses of the left ventricle. Evidence exists that angiotensin II contributes to the development of cardiac hypertrophy (Weber and Brilla, 1991; Sadoshima and Izumo, 1993; Everett et al., 1994; Lopez et al., 1994). Apart from its blood pressure raising effects, angiotensin II is a potent growth stimulating peptide. The mechanisms of action include induction of protein synthesis, mitosis of vascular smooth muscle cells and increased synthesis of collagen type I and III in fibroblasts, leading to myocardial fibrosis. It was therefore interesting to study the role of angiotensin II in hypoxia-induced right ventricular hypertrophy by blocking its action during the evolvement of the hypertrophy. All known physiologic effects of angiotensin II are mediated through the type 1 angiotensin II receptor (AT1). A newly developed class of angiotensin II receptor antagonists specific for the AT1 receptor has proven to be a valuable addition to the tools available for research on the renin– angiotensin system and cardiac hypertrophy. Losartan (LO), the most prominent substance in this class, has just recently been introduced to clinical use and is a promising new drug in the control of essential hypertension, left ventricular hypertrophy and heart failure. Angiotensin infusion has been shown to prevent both the acute and chronic hypoxia-induced increase in pulmonary artery pressure and the chronic hypoxia-induced right ventricular hypertrophy, presumably by pulmonary vasodilatation mediated by angiotensin-induced release of prostacyclin (Rabinovitch et al., 1988). However, both in an isolated perfused lung model (Nossaman et al., 1994, 1995) and in a study with intact rats (Morrell et al., 1995), no effect of the angiotensin receptor antagonist LO on the acute hypoxia-induced increase in pulmonary artery pressure was observed. Conversely, LO, as well as the ACE inhibitor captopril, was able to reduce the amount of right ventricular hypertrophy during chronic hypoxia treatment (Morrell et al., 1995). In view of these contradictory findings, the exact role of angiotensin in the development and modulation of hypoxiainduced right ventricular hypertrophy remains unclear. In order to more fully understand the role of angiotensin II in the development of hypoxia-induced right ventricular hypertrophy, we examined the preventative effects of the specific AT1 receptor antagonist LO on the resulting mor-
phologic, hemodynamic and biochemical changes. In addition to hemodynamic measurements, we examined the changes in cardiomyocyte size and in the activities of the oxidative pentose phosphate pathway enzymes glucose-6-phosphate-dehydrogenase (G-6-PD) and 6-phosphogluconate-dehydrogenase (6-PGD).
Materials and Methods Experimental protocol Right ventricular hypertrophy was induced in 29 female Sprague–Dawley rats (Charles River Wiga GmbH, Sulzfeld, Germany) by intermittent hypoxia for 8 h per day, 5 days per week, for a total of 20 days of exposition. The rats were placed in a plexiglas chamber (45×30×18 cm3) at atmospheric pressure. The animals were able to move around freely in their cages or chamber with free access to tap water and control rat chow diet (Altromin C 100, Altromin GmbH, Lage, Germany). No more than six animals were present in the chamber at one time. The desired O2 concentration was achieved by mixing pressurized air with N2. Gas flows were controlled with Rota⊆ flowmeters and the O2 concentration was mon¨ itored at the chamber outflow with a Drager Oxydig ¨ ¨ O2 Sensor (Dragerwerk AG, Lubeck, Germany). Gas flow was adjusted to approximately 150 l/h. FiO2 was 0.16 (16 vol.%) at the start of hypoxia treatment and then was reduced stepwise to 0.10 over a period of 8 days. For continuous application of LO, miniature infusion pumps (alza, Palo Alto, CA, USA) were implanted intraperitoneally the day before the start of the hypoxia treatment. The chosen dose of LO (12 mg/kg/d) was found to have little effect on systemic blood pressure in normotensive Sprague– Dawley rats after 2 weeks of continuous administration (Irlbeck et al., 1996b) and is well within the effective dose range necessary to reduce blood pressure in SHR animals (Timmermans et al., 1993). In those animals who did not receive LO, a piece of silicone tubing was implanted instead of the minipumps. The infusion rate of the minipumps was only 0.5 l/h, hence we did not feel that the use of expensive minipumps was necessary in the control animals. Since a filling of the pumps only lasts for 14 days, the pumps and silicone tubing were changed after 2 weeks. The animals were anesthetized with ether for the pump implantation and change procedure. LO was donated ¨ by MSD Sharp & Dome (Munchen, Germany).
AT1 Blockade and Right Ventricular Hypertrophy
Hemodynamic measurements All hemodynamic measurements were done under room air conditions and after the rats had been out of the hypoxia chamber for at least 1 h. For measurement of the hemodynamic parameters, all animals were anesthetized by intraperitoneal injection of 80 mg/kg thiopental sodium (Trapanal, Byk Gulden, Konstanz, Germany). Depth of anesthesia was evaluated by pinching the animal’s paw with forceps, and additional thiopental was injected if necessary. The animals breathed spontaneously through a cannula which was placed in the trachea by tracheotomy for better airway control and to allow mechanical ventilation if necessary. For measurement of left and right ventricular hemodynamics, Millar ultraminiature pressure tipcatheters (model SPR-249, 3 French and model SPR-503, 2 French; Millar Instruments, INC., Houston, TX, USA) were used. The catheters were calibrated with a mercury pressure manometer (Hugo Sachs Elektronik, March-Hugstetten, Germany) prior to every measurement. The left heart catheter was inserted into the right carotid artery and advanced into the left ventricle. Systolic and diastolic aortic pressures (SAP, DAP) were measured after withdrawal of the left heart catheter from the ventricle. Mean arterial pressure (MAP) was calculated as the mean of SAP and DAP. After collection of the left heart data, the right heart catheter was introduced via the right internal jugular vein into the right ventricle. Heart rate (HR), dP/dtmax and pressure were recorded continuously for 15 min for each ventricle. Thereafter, the left and right tip catheters were replaced by a thermosensitive 1.5 F microprobe (Columbus Instruments, Columbus, OH, USA) and a polyethylene tube (inner diameter 0.58 mm), respectively, for the determination of cardiac output with the thermodilution method. For this procedure, 0.1 ml of cold saline (18°C) was injected into the right atrium. The temperature curve was recorded with a Cardiomax II computer (Columbus Instruments, Columbus, OH, USA). Each cardiac output value was calculated as the mean of five consecutive measurements. Total peripheral resistance (TPR) was estimated by dividing MAP by cardiac output and stroke work (SW) was calculated by multiplication of stroke volume with systolic ventricular pressures. After completion of the hemodynamic measurements, the hearts were excised and the coronary arteries of those hearts which were subjected to biochemical analysis were slowly flushed with ice cold saline. The atria were cut off and the right
2933
ventricular free wall was trimmed away. Both ventricles were weighed after freezing in liquid nitrogen. The hearts in which cell isolation was done were gently flushed with warmed, oxygenated Tyrode solution in the Langendorff mode.
Biochemical measurements The activities of G-6-PD (EC 1.1.1.49) and 6-PGD (EC 1.1.1.44) were measured according to the methods of Glock and McLean (1953, 1954). After homogenization of the hearts in the perfusion medium, pH control (7.0) and centrifugation (Beckman ultracentrifuge model L5-65 at 35 000×g for 30 min), the supernatants were dialysed overnight. The enzyme activity was then measured spectrophotometrically. Protein content of the ventricular homogenates was determined according to the method of Lowry et al. (1951). A ¨ commercial kit from Sigma Chemical (Munchen, Germany) was used. The activity of the enzymes was expressed as units/g protein.
Cell isolation and measurements After excision of the hearts and flushing with 20 ml of Tyrode solution, the hearts were perfused on a modified Langendorff apparatus for 15 min with calcium free Tyrode solution to wash out any remaining blood and calcium-containing Tyrode solution. Thereafter, isolated cardiomyocytes were prepared by perfusion with collagenase (Type II in calcium free Tyrode, Sigma Chemical) for approximately 30 min. Collagenase concentration and perfusion duration was optimized for cell isolation of right ventricular free wall cardiomyocytes. The ventricles were then separated, minced in calcium free Tyrode and poured through nylon mesh. The cardiomyocytes were fixed in glutaraldehyde and viewed under a microscope for quality control of the isolation procedure. Cell volume measurements were done with a Coulter Counter Channelyzer (Coulter Electronics, Krefeld, Germany) (Nash et al., 1979). Since cardiomyocytes are primarily rod-shaped cells, a shape factor of 1.05 was used when calculating cell volume (Hurley, 1970; Gerdes et al., 1987). Cell length measurements were carried out under light micro¨ scopy using a Zeiss Axioskop (Carl Zeiss, Munchen, Germany). For each ventricle, the average length of 30 measured cardiomyocytes was used. Cardiomyocyte cross-sectional area was calculated by dividing cell volume by cell length.
2934
M. Irlbeck et al.
Table 1 Heart rate, cardiac output, stroke volume and total peripheral resistance in all groups of the study Heart rate (HR; beats/min) Controls LO Hypoxia Hypoxia+LO
Cardiac output (CO; ml/kg/min)
391.6±5.2 (14) 367.0±6.8∗ (14) 360.3±7.1∗ (15) 343.2±10.0∗† (14)
412.0±11.3 443.1±29.3 464.9±13.0 412.6±15.6
(13) (14) (15) (11)
Stroke volume (SV; ml/kg)
Total peripheral resistance (TPR; mmHg×kg×min/ml)
1.05±0.03 (13) 1.21±0.08 (14) 1.30±0.05∗ (15) 1.18±0.06 (11)
0.327±0.018 (13) 0.260±0.014∗ (14) 0.274±0.014∗ (15) 0.257±0.021∗ (11)
LO, losartan (12 mg/kg/d). All data as mean±... Number of experiments in parenthesis. ∗P<0.05 v controls; †P<0.05 v losartan.
Table 2 Aortic pressures and left ventricular parameters LVSP (mmHg)
DAP (mmHg)
MAP (mmHg)
LV dP/dtmax (mmHg/s)
LV SW (mmHg×ml/kg)
Controls 149.4±3.3 (14) 114.1±3.9 (14) 132.0±3.6 (14) 11 936±544 (14) LO 126.6±2.7∗ (14) 94.8±3.4∗ (14) 110.9±3.0∗ (14) 10 957±356 (14) Hypoxia 147.7±5.0† (15) 104.7±5.7 (15) 126.3±5.1† (15) 11 626±425 (15) Hypoxia+LO 116.1±3.8∗‡ (14) 80.4±4.6∗†‡ (14) 98.5±4.1∗†‡ (14) 8971±444∗†‡ (14)
156.5±4.0 (13) 153.7±11.5 (14) 190.8±7.8∗† (15) 140.1±4.4‡ (11)
Left ventricular systolic pressure (LVSP), diastolic aortic pressure (DAP), mean aortic pressure (MAP), left ventricular maximal rise in pressure (LVdP/dtmax), left ventricular stroke work (LV SW). LO, losartan (12 mg/kg/d). All data as mean±... Number of experiments in parenthesis. ∗P<0.05 v controls; †P<0.05 v losartan; ‡P<0.05 v hypoxia.
Table 3 Right ventricular parameters
Controls LO Hypoxia Hypoxia+LO
RVSP (mmHg)
RV dP/dtmax (mmHg/s)
RVEDP (mmHg)
RVSW (mmHg×ml/kg)
38.1±0.83 (16) 36.8±0.97 (15) 58.1±1.42∗† (15) 49.5±2.06∗†‡ (13)
2704±102 (16) 2640±83 (15) 3782±191∗† (15) 3046±263‡ (13)
2.50±0.38 (11) 3.06±0.35 (8) 3.87±0.35∗ (13) 3.23±0.51 (11)
39.9±1.5 (13) 45.2±3.6 (14) 75.4±3.1∗,† (15) 59.5±3.1∗†‡ (11)
Right ventricular systolic pressure (RVSP), right ventricular maximal rise in pressure (RV dP/dtmax), right ventricular end-diastolic pressure (RVEDP) and right ventricular stroke work (RVSW). LO, losartan (12 mg/kg/d). All data as mean±... Number of experiments in parenthesis. ∗P<0.05 v controls; †P<0.05 v losartan; ‡P<0·05 v hypoxia.
Statistical analysis The data were analysed and mean±... For evaluation of nificance, one way ANOVA was multiple comparison procedure Duncan. A value of P<0.05 was nificant.
expressed statistical used with according considered
as sigthe to sig-
Results Left ventricular and systemic circulatory parameters As shown in Tables 1 and 2, 4 weeks of intermittent hypoxia treatment caused a small decrease in heart rate (HR). Left ventricular maximal rise in pressure
(LV dP/dtmax) as well as LV and aortic pressures (LVSP, MAP, DAP) were unchanged from control values, and cardiac output was slightly, but not significantly elevated. Total peripheral resistance (TPR) was significantly decreased after hypoxia exposure and LV stroke volume (SV) as well as stroke work (SW) were elevated compared to control values. Blockade of the AT1 receptors with LO treatment reduced HR, LVSP and aortic pressures in the control animals. In the hypoxic animals, LO reduced LVSP, aortic pressures and LV dP/dtmax. Cardiac output and SV were unaffected by LO. LO significantly reduced TPR only in the control animals, if had no effect on the already low TPR in the hypoxia-treated rats. LO did, however, reduce LV SW in the hypoxic animals.
2935
AT1 Blockade and Right Ventricular Hypertrophy Table 4 Left and right ventricular morphologic parameters Day 1 (Bodyweight; g) Controls LO Hypoxia Hypoxia+LO
194.9±4.4 207.9±5.0 206.1±5.9 200.4±7.6
(16) (16) (15) (14)
LVW/BW (mg/g) Controls LO Hypoxia Hypoxia+LO
2.925±0.189 2.515±0.173 2.925±0.200 2.549±0.135
(8) (8) (8) (7)
Day 20 243.3±5.6 256.6±5.9 255.7±4.8 255.2±9.9
LV weight (mg) (16) (16) (15) (14)
706.1±40.4 632.0±27.9 756.5±51.8 662.1±51.8
(8) (8) (8) (7)
RV weight (mg)
RVW/BW (mg/g)
204.1±12.1 (8) 200.0±10.4 (8) 302.5±16.0∗† (8) 254.9±18.4∗†‡ (7)
0.844±0.053 (8) 0.784±0.029 (8) 1.166±0.049∗† (8) 0.990±0.072†‡ (7)
Left and right ventricular weight/body weight ratio (LVW/BW; RVW/BW). LO, losartan (12 mg/kg/ d). All data as mean±... Number of experiments in parenthesis. ∗P<0.05 v controls; †P<0.05 v losartan; ‡P<0.05 v hypoxia.
15
(a)
[Units/mg protein]
Right ventricular functional parameters *† 10
5
0
[Units/mg protein]
15
Control Control + LO Hypoxic Hypoxic + LO 8 7 9 7
Hypoxia induced a marked elevation of right ventricular systolic pressure (RVSP; 52%; Table 3), RV maximal rise in pressure (dP/dtmax; 49%) and RV stroke work (RVSW; 89%). LO had no effect on these parameters under control conditions, but attenuated the hypoxia-induced increase. RVSP and RVSW were still elevated from control values, however. As shown in Table 3, RV end diastolic pressure was elevated after hypoxia exposure and not significantly influenced by LO infusion.
(b)
Heart and cell sizes 10
5
0
Control Control + LO Hypoxic Hypoxic + LO 8 8 9 7
Figure 1 (a) Right and (b) left ventricular glucose-6phosphate-dehydrogenase (G-6-PD) in all groups of the study. LO, losartan (12 mg/kg/d). All data as mean± ... Number of experiments shown below group name. ∗P<0.05 v controls; †P<0.05 v losartan.
Neither LV weight nor LV myocyte size were significantly affected by intermittent hypoxia or longterm treatment with LO (Tables 4 and 5). In contrast, the ventricular weight/body weight ratio (RVW/BW) and myocyte cell volume (CV) of the RV were significantly elevated after 4 weeks of intermittent hypoxia (Tables 4 and 5). The increase in cell volume was exclusively due to an increase in cell cross-sectional area, since the cell lengths were very similar in all groups. LO slightly decreased RVW/BW and RVCV under normoxic conditions, but in the hypoxia-treated rats, the gain in RV mass was significantly attenuated in parallel with RVSP. The increase in RVCV was completely prevented by LO.
2936
M. Irlbeck et al.
Table 5 Left and right cell parameters LV cell volume (lm3) Controls LO Hypoxia Hypoxia+LO
27 048±1924 26 703±3234 25 342±2250 23 729±1740
(4) (5) (4) (6)
RV cell volume (lm3)
RV cell length (lm)
24 248±1193 (8) 22 896±855 (8) 29 541±1765∗† (7) 24 524±2370‡ (7)
132.3±3.1 129.5±1.7 130.9±1.6 127.6±1.4
(8) (8) (7) (7)
RV CSA (lm2) 183.2±7.9 (8) 176.6±4.9 (8) 225.1±11.4∗† (7) 191.7±17.1‡ (7)
Right ventricular cell cross sectional area (RV CSA). LO, losartan (12 mg/kg/d). All data as mean±... Number of experiments in parenthesis. ∗P<0.05 v controls; †P<0.05 v losartan; ‡P<0.05 v hypoxia.
25
The activities of two enzymes of the oxidative pentose phosphate pathway, G-6-PD and 6-PGD, were unaffected by hypoxia treatment or LO in the LV, but both enzyme activities were significantly enhanced by intermittent hypoxia in the RV. LO had no significant effect on either enzyme activity in the RV under control conditions, but it attenuated the hypoxia-induced increase (Figs 1 and 2).
20
[Units/mg protein]
Biochemical parameters
10
[Units/mg protein]
25
Control Control + LO Hypoxic Hypoxic + LO 8 7 9 7 (b)
20 15 10 5 0
Although total peripheral resistance was down by about 15%, the left ventricular pressures and contractility (as measured by dP/dtmax) were not markedly affected by intermittent hypoxia treatment. In contrast, right ventricular systolic pressure and contactility remained elevated even after restoring room air conditions. The observed decrease in HR of the hypoxiatreated animals is in line with previous reports from the literature. In humans and other mammals, heart rate tends to be reduced after prolonged hypoxia (Maher et al., 1975, 1978). The angiotensin II receptor antagonist LO had only slight effects on the hemodynamic values under normoxic conditions. This effect of LO consisted mainly of a decrease in left ventricular and
‡
5
Discussion
Left ventricular and systemic circulatory parameters
*†
15
0
The results of our experiments showed that intermittent hypoxia was sufficient to induce right ventricular hypertrophy. This hypertrophy was confined to the right ventricle, neither the left ventricular weight/body weight ratio nor the left ventricular myocyte cell volume was increased in the hypoxia-treated rats.
(a)
Control Control + LO Hypoxic Hypoxic + LO 8 8 9 7
Figure 2 (a) Right and (b) left ventricular 6-phosphogluconate-dehydrogenase (6-PGD) in all groups of the study. LO, losartan (12 mg/kg/d). All data as mean±... Number of experiments shown below group name. ∗P<0.05 v controls; †P<0.05 v losartan; ‡P<0.05 v hypoxia.
aortic pressures. TPR was also reduced, hence cardiac output was slightly elevated. This pressure reduction in the systemic circulation was not observed in a previous experiment, where LO was infused intravenously for 2 weeks (Irlbeck et al., 1996b). It is also different from observations during acute infusion of LO in conscious, normotensive animals (Wong et al., 1990), where LO did not lower blood pressure. We interpret our finding as
AT1 Blockade and Right Ventricular Hypertrophy
a mild vasodilating effect of LO. This effect could be explained by inhibition of some basal angiotensin action on systemic resistance vessels. It did not seem to affect left ventricular contractility since LV dP/dtmax was not depressed by LO.
2937
in RVSP was most likely due to the blockade of angiotensin action on the pulmonary vessels.
Heart and cell sizes Right ventricular functional parameters Intermittent hypoxia induced a marked and sustained increase in RVSP. In pre-series experiments, the elevation of RVSP after 4 weeks of intermittent hypoxia was observed even after the rats had been under room air conditions for more than 15 h, suggestive of fixed pulmonary hypertension. We were unable to catheterize the pulmonary artery, and therefore do not know the pulmonary artery resistance, but it is most likely that the elevation of RVSP was secondary to a high right ventricular afterload induced by hypoxic pulmonary vasoconstriction (Rabinovitch et al., 1979). LO had no effect on the right ventricular parameters under control conditions, but reduced the development of hypoxia-induced RV hypertension. Control values, however, were not reached. It could be speculated that this pressure reduction was due to a vasodilatory effect of LO in the pulmonary circulation or due to direct inhibition of hypoxic pulmonary vasoconstriction. However, LO had no measurable effect on RVSP in the controls and, as shown by Morrell et al. (1995), does not seem to interfere with the vasoconstrictive response of the rat pulmonary circulation to acute hypoxia. On the other hand, human experiments done by Kiely et al. (1995) showed significantly less acute hypoxic pulmonary vasoconstriction in healthy volunteers pretreated with LO, suggestive of some species difference in this respect. Other results from Morrell et al. (1995) show that LO is able to reduce the development of medial thickening and peripheral muscularization of small pulmonary arteries in a comparable hypoxic rat model. It is likely that blockade of the AT1 receptors prevents pulmonary artery hypertrophy and thus ameliorates the development of pulmonary hypertension. The other alternative, a negative inotropic effect on the RV itself is not as likely despite the lower RV dP/dtmax. With a reduction in RV contractility and a constant (high) pulmonary artery pressure, an increase in RVEDP should have been observed, specially since cardiac output was unchanged. However, RVEDP was reduced in the hypoxic animals treated with LO and, therefore, the reduction
As expected, intermittent hypoxia caused isolated RV hypertrophy as judged by the increase in the ventricular weight/body weight ratio and by the increase in myocyte cell volume. Since myocyte cell length in the hypoxia-treated animals was not different from the controls, the RV hypertrophy in this model was of the concentric type. This compares to morphological changes seen after 4 weeks of large LV infarction (Zimmer et al., 1990). In that model, a RVSP of 79±2 mmHg was measured in rats which had developed large LV infarction. RVW/ BW and myocyte volume were doubled and RV myocytes increased in length under those conditions. It should be noted that the LVs of those rats were in severe failure with a LVEDP of 32±2 mmHg as compared to 3.4±0.8 mmHg in the controls. In our present experiments, LO reduced the weight gain (VW/BW) of the RV to approximately the same extent as RVSP, but not completely to control values. Interestingly, the increase in cell volume and myocyte cross-sectional area was completely prevented by LO. More experiments are therefore needed to find out what makes up the remaining increase in mass of the RV. [A qualitative and quantitative examination of RV collagen is currently being done and should provide valuable additional information.]
Biochemical parameters Glucose-6-phosphate dehydrogenase (G-6-PD), the first and rate-limiting enzymes of the oxidative pentose phosphate cycle, has been shown in previous experiments to be associated with myocardial hypertrophy. The activity of G-6-PD was elevated with catecholamine-induced myocardial hypertrophy, with thyroxin-induced hypertrophy in spontaneously hypertensive rats and with hypertrophy after banding of the aortic arch. In a model of chronic myocardial infarction, the activity of G-6PD was elevated as well (Zimmer, 1992; Zimmer et al., 1992). In a different model of right ventricular hypertrophy, where the pulmonary artery was banded for 2 weeks, G-6-PD activity was not elevated. In our experiments, chronic intermittent hypoxia
2938
M. Irlbeck et al.
treatment induced an increase in the activity of G6-PD in the right ventricle, which was comparable to the increase seen in the left ventricle with other models of cardiac hypertrophy. The absence of an increase in G-6-PD activity in the left ventricle shows that the change was not induced by systemic hypoxia. It is also clear now that the increase in G-6-PD activity seen in other models of cardiac hypertrophy was not due to lack of oxygen, but was indeed induced by the additional cardiac work load. Another enzyme in the sequence of the pentose phosphate cycle is 6-phosphogluconate dehydrogenase (6-PGD), which, just like G-6-PD, acts as one of the major producers of NADPH in the heart muscle cell. None of the previous experiments had shown any correlation between cardiac hypertrophy and activity of 6-PGD. However, in the present model of 4-weeks duration, a significant hypoxia-induced elevation of 6-PGD activity was observed in the right ventricle, although less pronounced than the increase of 6-PGD. This difference could be explained by the longer duration of our current experiment. Since the previous experiments had a maximal duration of only 2 weeks, it is possible that a longer time is needed to induce an increase in 6-PGD activity. LO reduced both enzyme activities after 4 weeks of intermittent hypoxia to values near the control range. It is not clear from our experiments whether this effect of LO, as well as the effect on myocyte cell size and ventricular mass was a direct result of AT1 receptor blockade or an indirect consequence brought about by the reduction in pressure. Since LO completely prevented the hypoxia-induced myocyte growth despite the still elevated RVSP, some direct effects are likely.
Acknowledgments This study was supported by the Deutsche Forschungsgemeinschaft DFG (Zi 199/8-4). The excellent technical assistance of Heike Kartmann and Eveline Musiol is gratefully acknowledged. In addition, we would like to thank MSD Sharp & Dome for their partial support.
References E AD, T-MR A, F A, G RA, 1994. Angiotensin receptor regulates cardiac hypertrophy and transforming growth factor-b1 expression. Hypertension 23: 587–592.
G AM, M JA, H JM, 1987. Regional changes in myocyte size and number in propranolol-treated hyperthyroid rats. Lab Invest 57: 708–713. G GE, ML P, 1953. Further studies on the properties and assay of glucose 6-phosphate dehydrogenase and 6-phosphogluconate dehydrogenase of rat liver. Biochem J 55: 400–408. G GE, ML P, 1954. Levels of enzymes of the direct oxidative pathway of carbohydrate metabolism in mammalian tissues and tumours. Biochem J 56: 171–175. H J, 1970. Sizing particles with a Coulter counter. Biophys J 10: 74–79. ¨ I M, M O, I T, Z H-G, 1996a. Different response of the rat left and right heart to norepinephrine. Cardiovasc Res 31: 157–162. ¨ I M, M O, I T, Z H-G, 1966. Influence of angiotensin II receptor blockade on chronic noradrenaline stimulation of the rat left and right heart. Exp Clin Cardiol 1: 49–55. K DG, C RI, L BJ, 1995. Acute hypoxic pulmonary vasoconstriction in man is attenuated by type I angiotensin II receptor blockade. Cardiovasc Res 30: 875–880. K F, O B, 1991. Right ventricular function ¨ in rats with hypoxic pulmonary hypertension. Pflugers Arch 419: 121–126. L JJ, L BH, I JR, W EO, S H, et al., 1994. Distribution and function of cardiac angiotensin AT1- and AT2-receptor subtypes in hypertrophied rat hearts. Am J Physiol 267: H844– 852. L OH, R NJ, F AL, R RJ, 1951. Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265–275. M JT, M SC, C A, W DL, H LH, 1975. Cardiovascular responsiveness to b-adrenergic stimulation and blockade in chronic hypoxia. Am J Physiol 228: 477–481. M JT, D JC, W DL, C A, 1978. Mechanism of the attenuated cardiac response to badrenergic stimulation in chronic hypoxia. J Appl Physiol 44: 647–651. M NW, M KG, S KR, 1995. Role of angiotensin-converting enzyme and angiotensin II in development of hypoxic pulmonary hypertension. Am J Physiol 269: H1186–1194. N GB, T PER, P T, T VW, S RD, et al., 1979. Size measurements on isolated rat heart cells using analysis and light scatter flow cytometry. BBA 587: 99–111. N BD, F CJ, W J, K PJ, 1994. Analysis of angiotensins I, II, and III in pulmonary vascular bed of the rat. Am J Physiol 266: L389–396. N BD, F CJ, K AD, K PJ, 1995. Anaysis of responses to ANG IV: effects of PD-123319 and DuP-753 in the pulmonary circulation of the rat. Am J Physiol 268: L302–308. O B, W J, 1990. Chronic intermittent hypoxia and cardiopulmonary system. Prog Resp Res 26: 1–11. R M, G W, N AS, M OS, R L, 1979. Rat pulmonary circulation after chronic hypoxia: hemodynamic and structural features. Am J Physiol 236: H818–827. R M, M M, R HC, M
AT1 Blockade and Right Ventricular Hypertrophy K, O’B H, et al., 1988. Angiotensin II prevents hypoxic pulmonary hypertension and vascular changes in rat. Am J Physiol 254: H500–508. S J, I S, 1993. Molecular characterization of angiotensin II-induced hypertrophy of cardiac myocytes and hyperplasia of cardiac fibroblasts. Critical role of the AT1 receptor subtype. Circ Res 73: 413–423. T P, W PC, C AT, H WF, B P, et al., 1993. Angiotensin II receptors and angiotensin II receptor antagonists. Pharm Rev 45: 205–251. . E US, L G, 1947. Observations on the pulmonary arterial blood pressure in the cat. Acta Physiol Scand 12: 301–320. W KT, B CG, 1991. Pathological hypertrophy and cardiac interstitium. Fibrosis and renin-angiotensin-adlosterone system. Circulation 83: 1849– 1865. W PC, P WA, C AT, D JV, C DJ, et al., 1990. Nonpeptide angiotensin II receptor antagonists. VIII. Characterization of functional antagonism displayed by DuP 753, an orally active antihypertensive agent. J Pharmacol Exp Therap 252: 719–725. Z W, Z H-G, G AM, 1990. Influence of
2939
ramipril on right ventricular hypertrophy induced by pulmonary artery stenosis in rats. J Cardiovasc Pharmacol 16: 480–468. Z H-G, 1992. The oxidative pentose phosphate pathway in the heart: regulation, physiological significance, and clinical implications. Basic Res Cardiol 87: 303–316. Z H-G, Z W, S RC, V AE, 1988. Right heart catheterisation in rats with pulmonary hypertension and right ventricular hypertrophy. Basic Res Cardiol 83: 48–57. Z H-G, M PA, M G, 1989. Myocardial infarction in rats: effects of metabolic and pharmacologic intervention. Basic Res Cardiol 84: 332– 343. Z H-G, G AM, L S, M G, 1990. Changes in heart function and cardiac cell size in rats with chronic myocardial infarction. J Mol Cell Cardiol 22: 1231–1243. ¨ Z H-G, L-B B, K-R C, N T, Z W, 1992. Effects of norphenephrine on the oxidative pentose phosphate pathway in the rat heart. Circ Res 71: 451–459.
Growth Factors and Cardiac Hypertrophy
Effects of Angiotensin II Receptor Blockade on Hypoxia-induced Right Ventricular Hypertrophy in Rats Michael Irlbeck1, Takaaki Iwai2, Tobias Lerner1 and Heinz-Gerd Zimmer3 1
Department of Anaesthesiology, University of Munich, Germany; 2Jikei University School of Medicine, Tokyo, Japan; and 3Carl-Ludwig-Institute of Physiology, University of Leipzig, Germany M. I, T. I, T. L H.-G. Z. Effects of Angiotensin II Receptor Blockade on Hypoxia-induced Right Ventricular Hypertrophy in Rats. Journal of Molecular and Cellular Cardiology (1997) 29, 2931–2939. It was the aim of the present study to characterize the hemodynamic, biochemical and morphologic effect of angiotensin II receptor blockade on hypoxia-induced right ventricular hypertrophy in rats. Isolated right ventricular hypertrophy was induced in female Sprague–Dawley rats by intermittent hypoxia (IH; 10% O2, 8 h/ day, 5 days/week, 20 days of exposition, n=15). After completion of IH, left- (LV) and right-ventricular (RV) hemodynamic parameters were measured under room air conditions in the intact, thiopental-anesthetized animals with special Millar ultraminiature tipcatheter-manometers. Cardiac output was determined using the thermodilution method. Cell volume (CV) of isolated cardiomyocytes was measured with a Coulter Channellyzer after collagenase cell isolation. The specific activities of the myocardial pentose phosphate pathway enzymes glucose-6-phosphate-dehydrogenase (G-6-PD) and 6-phosphogluconate-dehydrogenase (6-PGD) were determined using a spectrophotometric assay. IH caused a rise in right ventricular systolic pressure (RVSP) from 38.1±0.83 to 58.1±1.42 mmHg and an increase in the RV weight/body weight ratio (RVW/BW) from 0.884±0.053 to 1.166±0.049 mg/g. The activities of G-6-PD and 6-PGD were significantly increased after IH in the RV, but not in the LV. CV was increased from 24 248±1193 to 29 541±1765 lm3, myocardial cell length was unchanged. IH had no influence on the LV parameters or cardiac output. Co-infusion of the angiotensin II receptor antagonist losartan (LO; 12 mg/kg/d i.p., n=14) during the IH period reduced the rises in RVSP (49.4±2.06 mmHg), RVW/ BW (0.99±0.072 mg/g), G-6-PD and 6-PGD significantly, but not completely. The increase in CV, however, was prevented (24 524±2370 lm3) entirely. We conclude from these data that the IH-induced RV-hypertrophy was primarily of the concentric type. LO attenuated the hypoxia-induced isolated RV hypertrophy and significantly reduced the metabolic response of the RV. The LO effect was most potent with regard to the increase in cardiomyocyte volume. 1997 Academic Press Limited K W: Angiotensin II; Losartan; Right ventricle; Pulmonary hypertension; Hypoxia; Hypertrophy.
Introduction Various experimental models exist for the evaluation of right ventricular hypertrophy. Biventricular hypertrophy of the left and right rat heart has been observed and studied after experimentally induced chronic myocardial infarction (by ligation of the left anterior descending coronary artery) (Zimmer et al., 1989, 1990), infusion of catecholamines (Irlbeck et al., 1996a) or treatment with triiodo-
thyronine (Zimmer et al., 1988). Isolated right ventricular hypertrophy without growth responses of the left ventricle is seen after experimental pulmonary artery stenosis (Zierhut et al., 1990) or by induction of pulmonary hypertension by lung irradiation (Zimmer et al., 1988) or chronic exposure to hypoxia (Rabinovitch et al., 1979; Ostadal and Widimsky, 1990; Kolar and Ostadal, 1991). In the chronic hypoxia model, alveolar hypoxia leads to pulmonary vasoconstriction, as described
Please address all correspondence to: Dr Michael Irlbeck, Department of Anaesthesiology, University of Munich, Klinikum Grosshadern, 81366 Muenchen, FRG.
0022–2828/97/112931+09 $25.00/0
mc970528
1997 Academic Press Limited
2932
M. Irlbeck et al.
by v. Euler and Liljestrand (1947). Prolonged exposure to inspiratory hypoxia causes pulmonary hypertension and consequent isolated right ventricular hypertrophy without growth responses of the left ventricle. Evidence exists that angiotensin II contributes to the development of cardiac hypertrophy (Weber and Brilla, 1991; Sadoshima and Izumo, 1993; Everett et al., 1994; Lopez et al., 1994). Apart from its blood pressure raising effects, angiotensin II is a potent growth stimulating peptide. The mechanisms of action include induction of protein synthesis, mitosis of vascular smooth muscle cells and increased synthesis of collagen type I and III in fibroblasts, leading to myocardial fibrosis. It was therefore interesting to study the role of angiotensin II in hypoxia-induced right ventricular hypertrophy by blocking its action during the evolvement of the hypertrophy. All known physiologic effects of angiotensin II are mediated through the type 1 angiotensin II receptor (AT1). A newly developed class of angiotensin II receptor antagonists specific for the AT1 receptor has proven to be a valuable addition to the tools available for research on the renin– angiotensin system and cardiac hypertrophy. Losartan (LO), the most prominent substance in this class, has just recently been introduced to clinical use and is a promising new drug in the control of essential hypertension, left ventricular hypertrophy and heart failure. Angiotensin infusion has been shown to prevent both the acute and chronic hypoxia-induced increase in pulmonary artery pressure and the chronic hypoxia-induced right ventricular hypertrophy, presumably by pulmonary vasodilatation mediated by angiotensin-induced release of prostacyclin (Rabinovitch et al., 1988). However, both in an isolated perfused lung model (Nossaman et al., 1994, 1995) and in a study with intact rats (Morrell et al., 1995), no effect of the angiotensin receptor antagonist LO on the acute hypoxia-induced increase in pulmonary artery pressure was observed. Conversely, LO, as well as the ACE inhibitor captopril, was able to reduce the amount of right ventricular hypertrophy during chronic hypoxia treatment (Morrell et al., 1995). In view of these contradictory findings, the exact role of angiotensin in the development and modulation of hypoxiainduced right ventricular hypertrophy remains unclear. In order to more fully understand the role of angiotensin II in the development of hypoxia-induced right ventricular hypertrophy, we examined the preventative effects of the specific AT1 receptor antagonist LO on the resulting mor-
phologic, hemodynamic and biochemical changes. In addition to hemodynamic measurements, we examined the changes in cardiomyocyte size and in the activities of the oxidative pentose phosphate pathway enzymes glucose-6-phosphate-dehydrogenase (G-6-PD) and 6-phosphogluconate-dehydrogenase (6-PGD).
Materials and Methods Experimental protocol Right ventricular hypertrophy was induced in 29 female Sprague–Dawley rats (Charles River Wiga GmbH, Sulzfeld, Germany) by intermittent hypoxia for 8 h per day, 5 days per week, for a total of 20 days of exposition. The rats were placed in a plexiglas chamber (45×30×18 cm3) at atmospheric pressure. The animals were able to move around freely in their cages or chamber with free access to tap water and control rat chow diet (Altromin C 100, Altromin GmbH, Lage, Germany). No more than six animals were present in the chamber at one time. The desired O2 concentration was achieved by mixing pressurized air with N2. Gas flows were controlled with Rota⊆ flowmeters and the O2 concentration was mon¨ itored at the chamber outflow with a Drager Oxydig ¨ ¨ O2 Sensor (Dragerwerk AG, Lubeck, Germany). Gas flow was adjusted to approximately 150 l/h. FiO2 was 0.16 (16 vol.%) at the start of hypoxia treatment and then was reduced stepwise to 0.10 over a period of 8 days. For continuous application of LO, miniature infusion pumps (alza, Palo Alto, CA, USA) were implanted intraperitoneally the day before the start of the hypoxia treatment. The chosen dose of LO (12 mg/kg/d) was found to have little effect on systemic blood pressure in normotensive Sprague– Dawley rats after 2 weeks of continuous administration (Irlbeck et al., 1996b) and is well within the effective dose range necessary to reduce blood pressure in SHR animals (Timmermans et al., 1993). In those animals who did not receive LO, a piece of silicone tubing was implanted instead of the minipumps. The infusion rate of the minipumps was only 0.5 l/h, hence we did not feel that the use of expensive minipumps was necessary in the control animals. Since a filling of the pumps only lasts for 14 days, the pumps and silicone tubing were changed after 2 weeks. The animals were anesthetized with ether for the pump implantation and change procedure. LO was donated ¨ by MSD Sharp & Dome (Munchen, Germany).
AT1 Blockade and Right Ventricular Hypertrophy
Hemodynamic measurements All hemodynamic measurements were done under room air conditions and after the rats had been out of the hypoxia chamber for at least 1 h. For measurement of the hemodynamic parameters, all animals were anesthetized by intraperitoneal injection of 80 mg/kg thiopental sodium (Trapanal, Byk Gulden, Konstanz, Germany). Depth of anesthesia was evaluated by pinching the animal’s paw with forceps, and additional thiopental was injected if necessary. The animals breathed spontaneously through a cannula which was placed in the trachea by tracheotomy for better airway control and to allow mechanical ventilation if necessary. For measurement of left and right ventricular hemodynamics, Millar ultraminiature pressure tipcatheters (model SPR-249, 3 French and model SPR-503, 2 French; Millar Instruments, INC., Houston, TX, USA) were used. The catheters were calibrated with a mercury pressure manometer (Hugo Sachs Elektronik, March-Hugstetten, Germany) prior to every measurement. The left heart catheter was inserted into the right carotid artery and advanced into the left ventricle. Systolic and diastolic aortic pressures (SAP, DAP) were measured after withdrawal of the left heart catheter from the ventricle. Mean arterial pressure (MAP) was calculated as the mean of SAP and DAP. After collection of the left heart data, the right heart catheter was introduced via the right internal jugular vein into the right ventricle. Heart rate (HR), dP/dtmax and pressure were recorded continuously for 15 min for each ventricle. Thereafter, the left and right tip catheters were replaced by a thermosensitive 1.5 F microprobe (Columbus Instruments, Columbus, OH, USA) and a polyethylene tube (inner diameter 0.58 mm), respectively, for the determination of cardiac output with the thermodilution method. For this procedure, 0.1 ml of cold saline (18°C) was injected into the right atrium. The temperature curve was recorded with a Cardiomax II computer (Columbus Instruments, Columbus, OH, USA). Each cardiac output value was calculated as the mean of five consecutive measurements. Total peripheral resistance (TPR) was estimated by dividing MAP by cardiac output and stroke work (SW) was calculated by multiplication of stroke volume with systolic ventricular pressures. After completion of the hemodynamic measurements, the hearts were excised and the coronary arteries of those hearts which were subjected to biochemical analysis were slowly flushed with ice cold saline. The atria were cut off and the right
2933
ventricular free wall was trimmed away. Both ventricles were weighed after freezing in liquid nitrogen. The hearts in which cell isolation was done were gently flushed with warmed, oxygenated Tyrode solution in the Langendorff mode.
Biochemical measurements The activities of G-6-PD (EC 1.1.1.49) and 6-PGD (EC 1.1.1.44) were measured according to the methods of Glock and McLean (1953, 1954). After homogenization of the hearts in the perfusion medium, pH control (7.0) and centrifugation (Beckman ultracentrifuge model L5-65 at 35 000×g for 30 min), the supernatants were dialysed overnight. The enzyme activity was then measured spectrophotometrically. Protein content of the ventricular homogenates was determined according to the method of Lowry et al. (1951). A ¨ commercial kit from Sigma Chemical (Munchen, Germany) was used. The activity of the enzymes was expressed as units/g protein.
Cell isolation and measurements After excision of the hearts and flushing with 20 ml of Tyrode solution, the hearts were perfused on a modified Langendorff apparatus for 15 min with calcium free Tyrode solution to wash out any remaining blood and calcium-containing Tyrode solution. Thereafter, isolated cardiomyocytes were prepared by perfusion with collagenase (Type II in calcium free Tyrode, Sigma Chemical) for approximately 30 min. Collagenase concentration and perfusion duration was optimized for cell isolation of right ventricular free wall cardiomyocytes. The ventricles were then separated, minced in calcium free Tyrode and poured through nylon mesh. The cardiomyocytes were fixed in glutaraldehyde and viewed under a microscope for quality control of the isolation procedure. Cell volume measurements were done with a Coulter Counter Channelyzer (Coulter Electronics, Krefeld, Germany) (Nash et al., 1979). Since cardiomyocytes are primarily rod-shaped cells, a shape factor of 1.05 was used when calculating cell volume (Hurley, 1970; Gerdes et al., 1987). Cell length measurements were carried out under light micro¨ scopy using a Zeiss Axioskop (Carl Zeiss, Munchen, Germany). For each ventricle, the average length of 30 measured cardiomyocytes was used. Cardiomyocyte cross-sectional area was calculated by dividing cell volume by cell length.
2934
M. Irlbeck et al.
Table 1 Heart rate, cardiac output, stroke volume and total peripheral resistance in all groups of the study Heart rate (HR; beats/min) Controls LO Hypoxia Hypoxia+LO
Cardiac output (CO; ml/kg/min)
391.6±5.2 (14) 367.0±6.8∗ (14) 360.3±7.1∗ (15) 343.2±10.0∗† (14)
412.0±11.3 443.1±29.3 464.9±13.0 412.6±15.6
(13) (14) (15) (11)
Stroke volume (SV; ml/kg)
Total peripheral resistance (TPR; mmHg×kg×min/ml)
1.05±0.03 (13) 1.21±0.08 (14) 1.30±0.05∗ (15) 1.18±0.06 (11)
0.327±0.018 (13) 0.260±0.014∗ (14) 0.274±0.014∗ (15) 0.257±0.021∗ (11)
LO, losartan (12 mg/kg/d). All data as mean±... Number of experiments in parenthesis. ∗P<0.05 v controls; †P<0.05 v losartan.
Table 2 Aortic pressures and left ventricular parameters LVSP (mmHg)
DAP (mmHg)
MAP (mmHg)
LV dP/dtmax (mmHg/s)
LV SW (mmHg×ml/kg)
Controls 149.4±3.3 (14) 114.1±3.9 (14) 132.0±3.6 (14) 11 936±544 (14) LO 126.6±2.7∗ (14) 94.8±3.4∗ (14) 110.9±3.0∗ (14) 10 957±356 (14) Hypoxia 147.7±5.0† (15) 104.7±5.7 (15) 126.3±5.1† (15) 11 626±425 (15) Hypoxia+LO 116.1±3.8∗‡ (14) 80.4±4.6∗†‡ (14) 98.5±4.1∗†‡ (14) 8971±444∗†‡ (14)
156.5±4.0 (13) 153.7±11.5 (14) 190.8±7.8∗† (15) 140.1±4.4‡ (11)
Left ventricular systolic pressure (LVSP), diastolic aortic pressure (DAP), mean aortic pressure (MAP), left ventricular maximal rise in pressure (LVdP/dtmax), left ventricular stroke work (LV SW). LO, losartan (12 mg/kg/d). All data as mean±... Number of experiments in parenthesis. ∗P<0.05 v controls; †P<0.05 v losartan; ‡P<0.05 v hypoxia.
Table 3 Right ventricular parameters
Controls LO Hypoxia Hypoxia+LO
RVSP (mmHg)
RV dP/dtmax (mmHg/s)
RVEDP (mmHg)
RVSW (mmHg×ml/kg)
38.1±0.83 (16) 36.8±0.97 (15) 58.1±1.42∗† (15) 49.5±2.06∗†‡ (13)
2704±102 (16) 2640±83 (15) 3782±191∗† (15) 3046±263‡ (13)
2.50±0.38 (11) 3.06±0.35 (8) 3.87±0.35∗ (13) 3.23±0.51 (11)
39.9±1.5 (13) 45.2±3.6 (14) 75.4±3.1∗,† (15) 59.5±3.1∗†‡ (11)
Right ventricular systolic pressure (RVSP), right ventricular maximal rise in pressure (RV dP/dtmax), right ventricular end-diastolic pressure (RVEDP) and right ventricular stroke work (RVSW). LO, losartan (12 mg/kg/d). All data as mean±... Number of experiments in parenthesis. ∗P<0.05 v controls; †P<0.05 v losartan; ‡P<0·05 v hypoxia.
Statistical analysis The data were analysed and mean±... For evaluation of nificance, one way ANOVA was multiple comparison procedure Duncan. A value of P<0.05 was nificant.
expressed statistical used with according considered
as sigthe to sig-
Results Left ventricular and systemic circulatory parameters As shown in Tables 1 and 2, 4 weeks of intermittent hypoxia treatment caused a small decrease in heart rate (HR). Left ventricular maximal rise in pressure
(LV dP/dtmax) as well as LV and aortic pressures (LVSP, MAP, DAP) were unchanged from control values, and cardiac output was slightly, but not significantly elevated. Total peripheral resistance (TPR) was significantly decreased after hypoxia exposure and LV stroke volume (SV) as well as stroke work (SW) were elevated compared to control values. Blockade of the AT1 receptors with LO treatment reduced HR, LVSP and aortic pressures in the control animals. In the hypoxic animals, LO reduced LVSP, aortic pressures and LV dP/dtmax. Cardiac output and SV were unaffected by LO. LO significantly reduced TPR only in the control animals, if had no effect on the already low TPR in the hypoxia-treated rats. LO did, however, reduce LV SW in the hypoxic animals.
2935
AT1 Blockade and Right Ventricular Hypertrophy Table 4 Left and right ventricular morphologic parameters Day 1 (Bodyweight; g) Controls LO Hypoxia Hypoxia+LO
194.9±4.4 207.9±5.0 206.1±5.9 200.4±7.6
(16) (16) (15) (14)
LVW/BW (mg/g) Controls LO Hypoxia Hypoxia+LO
2.925±0.189 2.515±0.173 2.925±0.200 2.549±0.135
(8) (8) (8) (7)
Day 20 243.3±5.6 256.6±5.9 255.7±4.8 255.2±9.9
LV weight (mg) (16) (16) (15) (14)
706.1±40.4 632.0±27.9 756.5±51.8 662.1±51.8
(8) (8) (8) (7)
RV weight (mg)
RVW/BW (mg/g)
204.1±12.1 (8) 200.0±10.4 (8) 302.5±16.0∗† (8) 254.9±18.4∗†‡ (7)
0.844±0.053 (8) 0.784±0.029 (8) 1.166±0.049∗† (8) 0.990±0.072†‡ (7)
Left and right ventricular weight/body weight ratio (LVW/BW; RVW/BW). LO, losartan (12 mg/kg/ d). All data as mean±... Number of experiments in parenthesis. ∗P<0.05 v controls; †P<0.05 v losartan; ‡P<0.05 v hypoxia.
15
(a)
[Units/mg protein]
Right ventricular functional parameters *† 10
5
0
[Units/mg protein]
15
Control Control + LO Hypoxic Hypoxic + LO 8 7 9 7
Hypoxia induced a marked elevation of right ventricular systolic pressure (RVSP; 52%; Table 3), RV maximal rise in pressure (dP/dtmax; 49%) and RV stroke work (RVSW; 89%). LO had no effect on these parameters under control conditions, but attenuated the hypoxia-induced increase. RVSP and RVSW were still elevated from control values, however. As shown in Table 3, RV end diastolic pressure was elevated after hypoxia exposure and not significantly influenced by LO infusion.
(b)
Heart and cell sizes 10
5
0
Control Control + LO Hypoxic Hypoxic + LO 8 8 9 7
Figure 1 (a) Right and (b) left ventricular glucose-6phosphate-dehydrogenase (G-6-PD) in all groups of the study. LO, losartan (12 mg/kg/d). All data as mean± ... Number of experiments shown below group name. ∗P<0.05 v controls; †P<0.05 v losartan.
Neither LV weight nor LV myocyte size were significantly affected by intermittent hypoxia or longterm treatment with LO (Tables 4 and 5). In contrast, the ventricular weight/body weight ratio (RVW/BW) and myocyte cell volume (CV) of the RV were significantly elevated after 4 weeks of intermittent hypoxia (Tables 4 and 5). The increase in cell volume was exclusively due to an increase in cell cross-sectional area, since the cell lengths were very similar in all groups. LO slightly decreased RVW/BW and RVCV under normoxic conditions, but in the hypoxia-treated rats, the gain in RV mass was significantly attenuated in parallel with RVSP. The increase in RVCV was completely prevented by LO.
2936
M. Irlbeck et al.
Table 5 Left and right cell parameters LV cell volume (lm3) Controls LO Hypoxia Hypoxia+LO
27 048±1924 26 703±3234 25 342±2250 23 729±1740
(4) (5) (4) (6)
RV cell volume (lm3)
RV cell length (lm)
24 248±1193 (8) 22 896±855 (8) 29 541±1765∗† (7) 24 524±2370‡ (7)
132.3±3.1 129.5±1.7 130.9±1.6 127.6±1.4
(8) (8) (7) (7)
RV CSA (lm2) 183.2±7.9 (8) 176.6±4.9 (8) 225.1±11.4∗† (7) 191.7±17.1‡ (7)
Right ventricular cell cross sectional area (RV CSA). LO, losartan (12 mg/kg/d). All data as mean±... Number of experiments in parenthesis. ∗P<0.05 v controls; †P<0.05 v losartan; ‡P<0.05 v hypoxia.
25
The activities of two enzymes of the oxidative pentose phosphate pathway, G-6-PD and 6-PGD, were unaffected by hypoxia treatment or LO in the LV, but both enzyme activities were significantly enhanced by intermittent hypoxia in the RV. LO had no significant effect on either enzyme activity in the RV under control conditions, but it attenuated the hypoxia-induced increase (Figs 1 and 2).
20
[Units/mg protein]
Biochemical parameters
10
[Units/mg protein]
25
Control Control + LO Hypoxic Hypoxic + LO 8 7 9 7 (b)
20 15 10 5 0
Although total peripheral resistance was down by about 15%, the left ventricular pressures and contractility (as measured by dP/dtmax) were not markedly affected by intermittent hypoxia treatment. In contrast, right ventricular systolic pressure and contactility remained elevated even after restoring room air conditions. The observed decrease in HR of the hypoxiatreated animals is in line with previous reports from the literature. In humans and other mammals, heart rate tends to be reduced after prolonged hypoxia (Maher et al., 1975, 1978). The angiotensin II receptor antagonist LO had only slight effects on the hemodynamic values under normoxic conditions. This effect of LO consisted mainly of a decrease in left ventricular and
‡
5
Discussion
Left ventricular and systemic circulatory parameters
*†
15
0
The results of our experiments showed that intermittent hypoxia was sufficient to induce right ventricular hypertrophy. This hypertrophy was confined to the right ventricle, neither the left ventricular weight/body weight ratio nor the left ventricular myocyte cell volume was increased in the hypoxia-treated rats.
(a)
Control Control + LO Hypoxic Hypoxic + LO 8 8 9 7
Figure 2 (a) Right and (b) left ventricular 6-phosphogluconate-dehydrogenase (6-PGD) in all groups of the study. LO, losartan (12 mg/kg/d). All data as mean±... Number of experiments shown below group name. ∗P<0.05 v controls; †P<0.05 v losartan; ‡P<0.05 v hypoxia.
aortic pressures. TPR was also reduced, hence cardiac output was slightly elevated. This pressure reduction in the systemic circulation was not observed in a previous experiment, where LO was infused intravenously for 2 weeks (Irlbeck et al., 1996b). It is also different from observations during acute infusion of LO in conscious, normotensive animals (Wong et al., 1990), where LO did not lower blood pressure. We interpret our finding as
AT1 Blockade and Right Ventricular Hypertrophy
a mild vasodilating effect of LO. This effect could be explained by inhibition of some basal angiotensin action on systemic resistance vessels. It did not seem to affect left ventricular contractility since LV dP/dtmax was not depressed by LO.
2937
in RVSP was most likely due to the blockade of angiotensin action on the pulmonary vessels.
Heart and cell sizes Right ventricular functional parameters Intermittent hypoxia induced a marked and sustained increase in RVSP. In pre-series experiments, the elevation of RVSP after 4 weeks of intermittent hypoxia was observed even after the rats had been under room air conditions for more than 15 h, suggestive of fixed pulmonary hypertension. We were unable to catheterize the pulmonary artery, and therefore do not know the pulmonary artery resistance, but it is most likely that the elevation of RVSP was secondary to a high right ventricular afterload induced by hypoxic pulmonary vasoconstriction (Rabinovitch et al., 1979). LO had no effect on the right ventricular parameters under control conditions, but reduced the development of hypoxia-induced RV hypertension. Control values, however, were not reached. It could be speculated that this pressure reduction was due to a vasodilatory effect of LO in the pulmonary circulation or due to direct inhibition of hypoxic pulmonary vasoconstriction. However, LO had no measurable effect on RVSP in the controls and, as shown by Morrell et al. (1995), does not seem to interfere with the vasoconstrictive response of the rat pulmonary circulation to acute hypoxia. On the other hand, human experiments done by Kiely et al. (1995) showed significantly less acute hypoxic pulmonary vasoconstriction in healthy volunteers pretreated with LO, suggestive of some species difference in this respect. Other results from Morrell et al. (1995) show that LO is able to reduce the development of medial thickening and peripheral muscularization of small pulmonary arteries in a comparable hypoxic rat model. It is likely that blockade of the AT1 receptors prevents pulmonary artery hypertrophy and thus ameliorates the development of pulmonary hypertension. The other alternative, a negative inotropic effect on the RV itself is not as likely despite the lower RV dP/dtmax. With a reduction in RV contractility and a constant (high) pulmonary artery pressure, an increase in RVEDP should have been observed, specially since cardiac output was unchanged. However, RVEDP was reduced in the hypoxic animals treated with LO and, therefore, the reduction
As expected, intermittent hypoxia caused isolated RV hypertrophy as judged by the increase in the ventricular weight/body weight ratio and by the increase in myocyte cell volume. Since myocyte cell length in the hypoxia-treated animals was not different from the controls, the RV hypertrophy in this model was of the concentric type. This compares to morphological changes seen after 4 weeks of large LV infarction (Zimmer et al., 1990). In that model, a RVSP of 79±2 mmHg was measured in rats which had developed large LV infarction. RVW/ BW and myocyte volume were doubled and RV myocytes increased in length under those conditions. It should be noted that the LVs of those rats were in severe failure with a LVEDP of 32±2 mmHg as compared to 3.4±0.8 mmHg in the controls. In our present experiments, LO reduced the weight gain (VW/BW) of the RV to approximately the same extent as RVSP, but not completely to control values. Interestingly, the increase in cell volume and myocyte cross-sectional area was completely prevented by LO. More experiments are therefore needed to find out what makes up the remaining increase in mass of the RV. [A qualitative and quantitative examination of RV collagen is currently being done and should provide valuable additional information.]
Biochemical parameters Glucose-6-phosphate dehydrogenase (G-6-PD), the first and rate-limiting enzymes of the oxidative pentose phosphate cycle, has been shown in previous experiments to be associated with myocardial hypertrophy. The activity of G-6-PD was elevated with catecholamine-induced myocardial hypertrophy, with thyroxin-induced hypertrophy in spontaneously hypertensive rats and with hypertrophy after banding of the aortic arch. In a model of chronic myocardial infarction, the activity of G-6PD was elevated as well (Zimmer, 1992; Zimmer et al., 1992). In a different model of right ventricular hypertrophy, where the pulmonary artery was banded for 2 weeks, G-6-PD activity was not elevated. In our experiments, chronic intermittent hypoxia
2938
M. Irlbeck et al.
treatment induced an increase in the activity of G6-PD in the right ventricle, which was comparable to the increase seen in the left ventricle with other models of cardiac hypertrophy. The absence of an increase in G-6-PD activity in the left ventricle shows that the change was not induced by systemic hypoxia. It is also clear now that the increase in G-6-PD activity seen in other models of cardiac hypertrophy was not due to lack of oxygen, but was indeed induced by the additional cardiac work load. Another enzyme in the sequence of the pentose phosphate cycle is 6-phosphogluconate dehydrogenase (6-PGD), which, just like G-6-PD, acts as one of the major producers of NADPH in the heart muscle cell. None of the previous experiments had shown any correlation between cardiac hypertrophy and activity of 6-PGD. However, in the present model of 4-weeks duration, a significant hypoxia-induced elevation of 6-PGD activity was observed in the right ventricle, although less pronounced than the increase of 6-PGD. This difference could be explained by the longer duration of our current experiment. Since the previous experiments had a maximal duration of only 2 weeks, it is possible that a longer time is needed to induce an increase in 6-PGD activity. LO reduced both enzyme activities after 4 weeks of intermittent hypoxia to values near the control range. It is not clear from our experiments whether this effect of LO, as well as the effect on myocyte cell size and ventricular mass was a direct result of AT1 receptor blockade or an indirect consequence brought about by the reduction in pressure. Since LO completely prevented the hypoxia-induced myocyte growth despite the still elevated RVSP, some direct effects are likely.
Acknowledgments This study was supported by the Deutsche Forschungsgemeinschaft DFG (Zi 199/8-4). The excellent technical assistance of Heike Kartmann and Eveline Musiol is gratefully acknowledged. In addition, we would like to thank MSD Sharp & Dome for their partial support.
References E AD, T-MR A, F A, G RA, 1994. Angiotensin receptor regulates cardiac hypertrophy and transforming growth factor-b1 expression. Hypertension 23: 587–592.
G AM, M JA, H JM, 1987. Regional changes in myocyte size and number in propranolol-treated hyperthyroid rats. Lab Invest 57: 708–713. G GE, ML P, 1953. Further studies on the properties and assay of glucose 6-phosphate dehydrogenase and 6-phosphogluconate dehydrogenase of rat liver. Biochem J 55: 400–408. G GE, ML P, 1954. Levels of enzymes of the direct oxidative pathway of carbohydrate metabolism in mammalian tissues and tumours. Biochem J 56: 171–175. H J, 1970. Sizing particles with a Coulter counter. Biophys J 10: 74–79. ¨ I M, M O, I T, Z H-G, 1996a. Different response of the rat left and right heart to norepinephrine. Cardiovasc Res 31: 157–162. ¨ I M, M O, I T, Z H-G, 1966. Influence of angiotensin II receptor blockade on chronic noradrenaline stimulation of the rat left and right heart. Exp Clin Cardiol 1: 49–55. K DG, C RI, L BJ, 1995. Acute hypoxic pulmonary vasoconstriction in man is attenuated by type I angiotensin II receptor blockade. Cardiovasc Res 30: 875–880. K F, O B, 1991. Right ventricular function ¨ in rats with hypoxic pulmonary hypertension. Pflugers Arch 419: 121–126. L JJ, L BH, I JR, W EO, S H, et al., 1994. Distribution and function of cardiac angiotensin AT1- and AT2-receptor subtypes in hypertrophied rat hearts. Am J Physiol 267: H844– 852. L OH, R NJ, F AL, R RJ, 1951. Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265–275. M JT, M SC, C A, W DL, H LH, 1975. Cardiovascular responsiveness to b-adrenergic stimulation and blockade in chronic hypoxia. Am J Physiol 228: 477–481. M JT, D JC, W DL, C A, 1978. Mechanism of the attenuated cardiac response to badrenergic stimulation in chronic hypoxia. J Appl Physiol 44: 647–651. M NW, M KG, S KR, 1995. Role of angiotensin-converting enzyme and angiotensin II in development of hypoxic pulmonary hypertension. Am J Physiol 269: H1186–1194. N GB, T PER, P T, T VW, S RD, et al., 1979. Size measurements on isolated rat heart cells using analysis and light scatter flow cytometry. BBA 587: 99–111. N BD, F CJ, W J, K PJ, 1994. Analysis of angiotensins I, II, and III in pulmonary vascular bed of the rat. Am J Physiol 266: L389–396. N BD, F CJ, K AD, K PJ, 1995. Anaysis of responses to ANG IV: effects of PD-123319 and DuP-753 in the pulmonary circulation of the rat. Am J Physiol 268: L302–308. O B, W J, 1990. Chronic intermittent hypoxia and cardiopulmonary system. Prog Resp Res 26: 1–11. R M, G W, N AS, M OS, R L, 1979. Rat pulmonary circulation after chronic hypoxia: hemodynamic and structural features. Am J Physiol 236: H818–827. R M, M M, R HC, M
AT1 Blockade and Right Ventricular Hypertrophy K, O’B H, et al., 1988. Angiotensin II prevents hypoxic pulmonary hypertension and vascular changes in rat. Am J Physiol 254: H500–508. S J, I S, 1993. Molecular characterization of angiotensin II-induced hypertrophy of cardiac myocytes and hyperplasia of cardiac fibroblasts. Critical role of the AT1 receptor subtype. Circ Res 73: 413–423. T P, W PC, C AT, H WF, B P, et al., 1993. Angiotensin II receptors and angiotensin II receptor antagonists. Pharm Rev 45: 205–251. . E US, L G, 1947. Observations on the pulmonary arterial blood pressure in the cat. Acta Physiol Scand 12: 301–320. W KT, B CG, 1991. Pathological hypertrophy and cardiac interstitium. Fibrosis and renin-angiotensin-adlosterone system. Circulation 83: 1849– 1865. W PC, P WA, C AT, D JV, C DJ, et al., 1990. Nonpeptide angiotensin II receptor antagonists. VIII. Characterization of functional antagonism displayed by DuP 753, an orally active antihypertensive agent. J Pharmacol Exp Therap 252: 719–725. Z W, Z H-G, G AM, 1990. Influence of
2939
ramipril on right ventricular hypertrophy induced by pulmonary artery stenosis in rats. J Cardiovasc Pharmacol 16: 480–468. Z H-G, 1992. The oxidative pentose phosphate pathway in the heart: regulation, physiological significance, and clinical implications. Basic Res Cardiol 87: 303–316. Z H-G, Z W, S RC, V AE, 1988. Right heart catheterisation in rats with pulmonary hypertension and right ventricular hypertrophy. Basic Res Cardiol 83: 48–57. Z H-G, M PA, M G, 1989. Myocardial infarction in rats: effects of metabolic and pharmacologic intervention. Basic Res Cardiol 84: 332– 343. Z H-G, G AM, L S, M G, 1990. Changes in heart function and cardiac cell size in rats with chronic myocardial infarction. J Mol Cell Cardiol 22: 1231–1243. ¨ Z H-G, L-B B, K-R C, N T, Z W, 1992. Effects of norphenephrine on the oxidative pentose phosphate pathway in the rat heart. Circ Res 71: 451–459.