Regression of ventricular hypertrophy abolishes cardiocyte vulnerability to acute hypoxia.код для вставкиСкачать
THE ANATOMICAL RECORD 226:198-206 (1990) Regression of Ventricular Hypertrophy Abolishes Cardiocyte Vulnerability to Acute Hypoxia CRAIG A. CANBY AND ROBERT J. TOMANEK Department of Anatomy and The Cardiovascular Center, The University of Iowa, Iowa City, Iowa 52242 ABSTRACT Left ventricular hypertrophy (LVH) secondary to a pressure overload commonly leads to perfusion abnormalities that may limit oxygen delivery to the myocardium and, therefore, result in cardiocyte intracellular damage. We initiated this study to test the hypothesis that the increased vulnerability of the hypertrophied left ventricle to acute hypoxia is minimized when LVH regresses and maximal coronary flow returns to normal. Six-month-old spontaneously hypertensive (SHR) and normotensive Wistar-Kyoto (WKY) rats were divided into control or one of two antihypertensive treatment groups. A 3-month treatment consisted of captopril (75-100 mg/kg) or hydralazine (80-160 mg/L) with hydrochlorothiazide (500 mg/L) added to each therapy. At the conclusion of the treatment period, the rats were administered a 7% 02-93% Nz gas mixture for 20 minutes to induce acute hypoxic stress during which time hemodynamics, blood gases, and pH were monitored. The heart was then rapidly fixed by vascular perfusion and prepared for electron microscopy. Captopril and hydralazine were equally effective in lowering arterial pressure in both strains, but only captopril was efficacious in reducing heart mass. Hypoxia-induced changes in hemodynamics, blood gases, and pH were similar in all of the groups; POz was decreased by about 70%. The electron micrographs revealed that the hypertrophied left ventricle consistently showed morphologic evidence of hypoxic damage (as indicated by T-tubular swelling, intracellular edema, and mitochondrial alterations); in contrast hypoxia had little effect on the non-hypertrophied ventricle. Captopril treatment resulted in a disappearance of the lability to hypoxic damage while hydralazine caused a small reduction in the frequency of hypoxic damage. In conclusion, reversal of hypertension alone had little effect on cardiocyte vulnerability to acute hypoxia, but reversal of hypertension in conjunction with regression of LVH prevented the intracellular damage characteristic of hypoxia. Thus, LVH and its associated maximal perfusion decrement, rather than hypertension per se, underlie cardiocyte vulnerability to hypoxic stress. An increase in minimal coronary vascular resistance (MCVR) (Bache and Vrobel, 1979; Tomanek et al., 1985) and a decrease in capillary density (Henquell e t al., 1977; Tomanek e t al., 1979) are associated with left ventricular hypertrophy (LVH) secondary to a pressure overload. Additionally, perfusion abnormalities have been shown to be more pronounced in the endomyocardium than in the epimyocardium (Breisch et al., 1986; Bache et al., 1981). Since myocardial oxygen extraction is near maximum, coronary blood flow must increase to match any increase in myocardial oxygen consumption. Also a n elevated MCVR and a decreased capillary density would limit the ability of the coronary circulation to increase blood flow and would limit oxygen diffusion; thus, it seems likely that the hypertrophied myocardium would be more vulnerable to hypoxia. Indeed, studies have shown that performance in the hypertrophied left ventricle is altered when exposed to hypoxia, compared to the non-hypertrophied ventricle (Lore11 et al., 19861, and that the hypertrophied ventricle, particularly the 0 1990 WILEY-LISS, INC. endomyocardium, is more vulnerable to hypoxic cell damage (Wangler e t al., 1980). We recently demonstrated that the decrement in maximal coronary perfusion, characteristic of pressure-overload hypertrophy, returns to normal in spontaneously hypertensive rats (SHR) following anti-hypertensive treatment which produces regression of cardiac hypertrophy (Canby and Tomanek, 1988). Based on this finding we proposed the hypothesis that the increased vulnerability of the cardiocyte to acute hypoxia is minimized or abolished with regression of LVH and the normalization of maximal coronary flow. Captopril treatment was selected because our previous experience showed that maximal coronary perfusion is normalized in conjunction with the drug’s ability to regress LVH. We also subjected a second group of rats to hydralazine treatment, because this agent normal- Received October 5, 1988; accepted May 16, 1989. HYPERTROPHY AND HYPOXIC DAMAGE TO CARDIOCYTE izes blood pressure without regression of LVH. Thus, the two treatment protocols enabled us to separate the anti-hypertensive effects of a n agent from the effects of a reduction in LV mass. MATERIALS AND METHODS General Experimental Design We assigned male spontaneously hypertensive (SHR) and normotensive Wistar-Kyoto (WKY) rats, obtained from Taconic Farms (Germantown, NY), to control (WKY and SHR), captopril-treated (C-WKY and C-SHR), and hydralazine-treated (H-WKY and HSHR) groups. Captopril (75 mg/kg) or hydralazine (80 mg/L) were dissolved in the rats’ drinking water. Based on the average body weight and water consumption of the treated WKY and SHR groups, 540-730 mg of captopril was dissolved in 1 liter of drinking water. Hydrochlorothiazide (500 mg/L), a diuretic, was also added in order to potentiate the antihypertensive effect of each of the treatment protocols. A 3-month treatment was initiated to coincide with the time that hypertension and LVH were well established (6 months of age) in order to focus on their reversibility. In order to maintain arterial systolic pressure below 140 mmHg, we increased the drug dosage of captopril to 100 mg/kg at week 2, while hydralazine was increased to 120 and 160 mg/L at weeks 4 and 8, respectively. Arterial systolic pressure measurements (tail-cuff method) were determined weekly in the four treated groups and at the beginning and end of the experimental period in the two control groups. Forty-eight hours prior to hypoxic exposure the drug water was replaced with tap water in order to eliminate the acute hemodynamic effects of the drugs during the time of the study. Each animal was given atropine (1.2 mg, i.p.1 and anesthetized with a mixture of ten parts ketamine (100 mg/ml) to one part acepromazine (10 mg/ml). The dosage was 0.10 m1/100 g of body weight. Following anesthesia, the left femoral and right common carotid arteries were cannulated with PE-10 tubing telescoped into PE-50 tubing. Subsequently, the right common carotid catheter was introduced into the left ventricle. The catheters were filled with heparinized saline (70 U/ml) and connected to a n Ailtech M520 pressure transducer, and heart rate as well a s arterial and ventricular pressures were recorded on a Gilson five-channel recorder. Administration of Acute Hypoxia Following measurement of arterial and ventricular pressures, and heart rate, a tracheotomy was performed and the animal was artificially ventilated with a Harvard small animal respirator. Respiration was maintained at 30 strokes per minute and at a tidal volume adjusted for body weight (5 ml for a 300-400 g rat). Following a 5-minute stabilization period, after performing a midsternal thoracotomy, arterial and ventricular pressures and heart rate were recorded and a 0.3 ml blood sample was withdrawn from the femoral artery for measurement of blood gases, pH (Radiometer BMS 3Mk2 pH/blood gas analyzer), and hematocrit. This blood sample was immediately replaced with a n equal volume of strain-matched donor blood. Subsequently, the ventilator was switched to a 7% 02-93% N2 gas mixture. The animal was administered this gas 199 mixture for 20 minutes in order to produce a n acute hypoxic environment. Pressures and heart rate were then recorded and another blood sample was withdrawn for measurement of blood gases, pH, and hematocrit. Electron Microscopy At the conclusion of the 20-minute hypoxic period, heparin (5,000 U) was injected into the inferior vena cava. The heart was arrested in diastole with procaine and removed, and the aorta was attached to a 20-gauge animal feeding needle (Harvard) for perfuse-fixation. A ligature was placed around the aorta so that the ball of the needle was just above the coronary ostia. Next, the right atrium was cut to provide a n outflow channel and the coronary vasculature was perfused with 20 ml of Locke’s solution (which also contained 5,000 U of heparin and 0.5 ml of procaine) a t 37°C via the feeding needle, and then with 100 ml of a modified Karnovsky’s solution (Tomanek et al., 1979, 1981). Subsequently, the organ was blotted dry and total heart weight and left ventricular weight (including the septum) were obtained. The septum was dissected out and the left ventricle was placed in fresh fixative for a n additional 20-24 hours, and then specimens were dissected from epimyocardium and endomyocardium. Both longitudinal and cross sections were obtained. These tissue samples were placed in a 0.1 M sodium cacodylate buffer with 3% sucrose for approximately 24 hours, post-fixed in 1% osmium tetroxide for 2 hours, washed with 0.1 M sodium cacodylate buffer, dehydrated through a graded series of ethanol and propylene oxide, and embedded in Spurr’s plastic. Thin sections were cut both perpendicular and parallel to the long axis of myocytes with a diamond knife on a Reichert ultratome. The sections were then mounted on copper grids and stained with uranyl acetate and lead citrate solutions, and subsequently viewed with a Hitachi 7000 electron microscope. Evaluation of hypoxic damage was done systematically by scanning all cells in the field. In the rare case that capillaries were not clear, indicating inadequate perfusion, the specimen was excluded. We evaluated 3045 cell profiles, cut in cross section, and tabulated the morphological changes characteristic of hypoxia. Thin sections cut parallel to the long axis of cardiocytes were also photographed and used to evaluate sarcomere changes and dilatations of the intercalated discs. Statistical Methods Data regarding most variables were analyzed by analysis of variance and the Bonferroni multiple comparison procedure. Ultrastructural alterations resulting from acute hypoxia were analyzed with a Fisher’s Exact Test for Frequencies. A P 5 0.05 was selected to denote statistical significance for all comparisons. The values presented in this paper are group means SEM. * RESULTS Arterial Systolic Pressure, and Body and Heart Mass Prior to antihypertensive therapy, arterial systolic pressure (tail-cuff) in SHR was approximately 55% higher than in WKY (Fig. 1). At this time all SHR had systolic pressures greater than 180 mmHg and were, C.A. CANBY AND R.J. TOMANEK 200 220 3.4 200 3.2 180 3.0 - B E 0 .- 2 2.8 v f n 0) v 3 2.6 E? m & 2.4 3 140 : U n 120 .-0 0 L &/At U) c% = P+ I 100 0 WKY Control ( n = l l ) 0 WKY Captopril Treated (n=9) 80 A A , l l 0 1 2 l r 2.0 WKY Hydralame Treated ( n = l i ) SHR Control (n=l1) SHR Caplopril Treated (n=10) SHR Hydralame Treated (n=11) l r 3 4 5 6 Week of Treatment 32.2 1.8 l l l l 7 8 9 10 Fig. 1. The effect of captopril and hydralazine treatment on arterial systolic pressure (tail-cuff method in conscious rats) during the first 10 weeks of treatment. *, P <0.0001 vs. SHR;t, P <0.0001 vs. respective control group. therefore, unequivocally hypertensive. Ten weeks of anithypertensive treatment lowered arterial systolic pressure in the C-WKY and H-WKY by 21% and 14%, respectively. In SHR, captopril lowered pressure by 40% while hydralazine reduced pressure by 34%.These reductions in arterial systolic pressure are significant in both rat strains ( P < 0.0001). Figure 2 illustrates the left ventricular weightlbody weight (LVW/BW) ratios. SHR have significantly greater LVWIBW ratio values than WKY (P < 0.0001). Captopril treatment in SHR and WKY was associated with LVWIBW ratios that were 22% and 14% lower compared to their respective controls. While hydralazine did not alter the ratio in WKY, i t did result in a 6% decrease in the ratio of SHR. However, H-SHR still had a ratio that was markedly greater than C-SHR and the control WKY. The data provided in Table 1indicated that captopril treatment was associated with significantly lower values for BW, heart weight (HW), and LVW in both r a t strains, and that hydralazine treatment resulted in lower values for BW, HW, and LVW in SHR, but not in WKY. Hemodynamics Arterial systolic pressure, measured directly via a femoral catheter, in closed-chest, anesthetized SHR was 78% greater than WKY (Table 1).C-SHR and H- WKY UControl SHR a C a p t o p r i l Treated D H y d r a l a z i n e Treated Fig. 2. The effect of captopril and hydralazine treatment on left ventricular weight (LVWIibody weight (BW) ratio. SHR had arterial systolic pressures that were 26% and 12% lower than SHR. Arterial systolic pressures in CWKY and H-WKY are not significantly different from WKY. Table 2 lists the various hernodynamic indices before and after hypoxia in anesthetized, open-chest rats. SHR exhibited significantly greater values for arterial systolic, arterial diastolic and mean pressure, and left ventricular systolic pressure than did WKY. Captopril treatment in SHR reduced the values for all of these variables (P < 0.01). Captopril treatment in WKY and hydralazine treatment in WKY and SHR did not sig- Figs. 3-10. Transmission electron micrographs of left ventricular cardiocytes. Fig. 3. Cross section of cardiocyte from SHR not subjected to acute hypoxia. Note the close packing of myofibrils, mitochondria, and tubular system. Transverse (T)-tubule is indicated by arrow. x 14,400. Fig. 4. Cross section of cardiocyte from SHR following 20 minutes of hypoxia. Note that T-tubules (arrows) are markedly dilated. In comparing to Figure 3, note that the magnification is less in Figure 4, which underestimates the magnitude of the dilations. x 9,900, Figs. 5 and 6. Longitudinal sections of cardiocytes from SHR following 20 minutes of hypoxia. These micrographs show the more severe, but less common alterations. Edema is evidenced by the exaggerated sarcoplasmic areas (Fig. 5).Translucence of mitochondrial matrix and disruption of cristae (arrows) are evident in Figure 6. Figure 5, x 10,200. Figure 6, x 13,600. HYPERTROPHY AND HYPOXIC DAMAGE TO CARDIOCYTE Figs. 3-6 201 202 C.A. CANBY AND R.J. TOMANEK Figs. 7-9. 203 HYPERTROPHY AND HYPOXIC DAMAGE TO CARDIOCYTE TABLE 1. Body weight, heart weight, and arterial systolic pressure' BW ( g ) HW (g) LVW (g) ASP (mmHg) WKY (n = 9) 495 t 11 1.515 f 0.045 1.084 t 0.032 116 t 5 C-WKY (n = 7) 436 k 2 1.196 t 0.052** 0.820 f 0.036** 116 t 6 'BW = body weight, HW = heart weight, LVW artery). * P < 0.01 vs. WKY. ** P < 0.01 vs. respective control group. = H-WKY (n= 11) 464 ? 10 1.476 f 041 1.046 t 0.029 125 t 4 SHR ( n = 11) 432 t 10* 1.671 f .041* 1.345 k 0.029* 207 t 4* C-SHR (n= 9) 376 t 11* 1.298 t 0.045** 0.918 t 0.032** 153 t 5** H-SHR (n = 11) 387 k 10** 1.471 f 0.041** 1.132 f 0.029** 182 f 4** left ventricular weight, and ASP = arterial systolic pressure (direct measurement in femoral TABLE 2. Hemodynamics prior to and following hypoxia' Prior to hypoxia ASP ADP MAP LVSP LVEDP HR After hypoxia ASP ADP MAP LVSP LVEDP HR WKY (n = 9) C-WKY (n=7) H-WKY (n= 11) ( n = 11) 100 f 6 73 4 5 82 f 6 104 t 7 842 345 f 12 91 f 7 68 t 6 76 f 6 90 t 8 9k3 328 t 14 101 t 6 79 f 5 86 + 5 102 t 6 12 t 1 350 f 11 151 k 6* 108 t 5* 122 f 5* 150f6* 9 t 1 341 t 11 120 t 6** 90 f 5** 100 t 6** 120 f 7** 10 t 2 372 t 2 138 f 6 105 f 5 116 ? 5 137 t 6 2+1 356 t 11 66 t 4*** 41 t 3*** 49 f 3*** 75 t 5*** 852 322 17 61 k 4*** 39 t 3*** 46 + 4*** 67 k 6*** 9?2 305 f 19 60 t 4*** 39 t 3*** 46 f 4*** 72 ? 4*** 13 f 1 323 f 15 72 2 4*** 41 f 3*** 51 t 3*** 74 4*** 12 t 1 314 2 15 79 2 4 48 t 3*** 58 f 3*** go f 5*** 922 361 t 17 68 t 4*** 39 t 3*** 48 t 3*** 72 t 4*** 9+1 359 ? 15 * SHR * C-SHR (n = 9) H-SHR (n = 11) 'ASP and ADP = arterial systolic and diastolic pressure (mmHg), respectively; MAP = mean arterial pressure (mmHg); LVSP and LVEDP = left ventricular systolic and end diastolic pressure (mmHg), respectively; HR = heart rate (beatdmin.). * P < 0.01 vs. WKY. ** P < 0.01 vs. respective control. ***P < 0.001 vs. pre-hypoxic value. TABLE 3. Arterial blood gases and pH prior to and following hypoxia Prior PC62 PH After hypoxia PO2 PCOZ PH C-WKY (n = 7) H-WKY (n = 11) 115 t 6 37 + 2 7.36 f 0.02 123 f 7 36 t 2 7.33 + 0.02 101 f 5 38 _t 1 7.42 f 0.02 1.6 t 5 33 t 1 7.34 t 0.02 110 t 6 35 t 2 7.38 + 0.02 108 t 5 36 t 1 7.39 t 0.02 34 + 2* 30 ? 2 * 7.32 t 0.02 37 t 2* 29 k 2* 7.26 f 0.03* 32 t 1* 26 f 2* 7.36 t 0.02 37 f 1" 25 f 2" 7.27 t 0.02* 37 t 2* 23 t 2* 7.33 _t 0.02" 33 t 1* 28 f 2* 7.32 t 0.02* to hypoxia PO, H-SHR (n = 11) WKY (n = 9) SHR (n= 11) C-SHR (n = 9) * P < 0.001vs. value prior to hypoxia. Fig. 7. Cross section from WKY subjected to 20 minutes of hypoxia. Note that the ultrastructural appearance is characteristic of myocardium from rats maintained under normoxic conditions. T-tubules (T) are not dilated. x 14,400. Fig. 8. An intercalated disc from a captopril-treated SHR subjected to 20 minutes of hypoxia is seen in longitudinal section. There is no evidence of dilated T-tubules or expansions of the non-specialized regions (between arrows) of the disc. x 32,000. Figs. 9 and 10. Cross and longitudinal fields from hydralazine treated SHR after 20 minutes of hypoxia. This pharmacological treatment did not minimize the hypoxic alterations characteristic of SHR. Large dilatations in the region of intercalated discs are seen in Figure 9, while Figure 10 illustrates dilated T-tubules (T).Two type of dilatations are evident in Figure 9: the non-specialized regions of the dlsc (NS) and adjacent T-tubules (T). Figure 9, ~ 2 9 , 0 0 0 ,Figure 10, x 8,500. nificantly alter these indices. Left ventricular end diastolic pressure and heart rate are not significantly different for any of the comparisons. All of these indices, except for LV end diastolic pressure and heart rate, were reduced by 20 minutes of hypoxia (P < 0.001). Arterial Blood Gases and pH While there are no significant intergroup differences in arterial blood gases and pH (Table 3) it is evident that the 7% Oz-93% N2 gas mixture markedly reduced POz in all six groups (P < 0.001). Hypoxemia developed in all groups as indicated by an approximate 70% reduction in P02. PCOz fell by 19-32%. Three possible factors may explain the decrease in PCOz. First, the 204 C.A. CANBY AND R.J. TOMANEK TABLE 4. Percentage of rats exhibiting various characteristics of hypoxic damage' % T-tubule dilation only Mitochondrial swelling Mitochondrial disruption Edema A t least one of the above WKY C-WKY H-WKY SHR C-SHR H-SHR (n = 8 ) 251 0 0 0 25 (n=7) 29t 0 0 0 29 (n = 7) 29t 0 0 0 29 (n = 9) 56$; l l t (n = 8 ) 12t* O* 0 O* 12* (n = 8 ) 37$; 25t 12$; 12t 123; 25t 12$; 25t 88 44$ 22* 44$ 100 'Number of rats per group is indicated in parentheses. t , mild-moderate change; $, marked change. For each rat, 30-45 cell profiles were evaluated. * Statistically significant difference (P5 0.05), between drug-treated group and its strain control, determined by Fisher's Exact Test for Frequencies. All differences between SHR and WKY groups, except mitochondrial disruption, are statistically significant. gas mixture did not contain COz. Second, i t is possible that a lower Oz cellular utilization contributed to the decreased PCOz. Finally, a decrease in bicarbonate concentration may have occurred. In addition, hypoxia lowered pH in the C-WKY, SHR, C-SHR, and H-SHR (P < 0.001). Hypoxia-Induced Tissue Damage Figures 3-10 are representative micrographs obtained from the various groups subjected to acute hypoxia. The characteristics of hypoxic damage are summarized in Table 4. When SHR were subjected to hypoxia they predictably exhibited dilated T-tubules (Fig. 4). In four of the nine hearts from SHR cardiocytes showed evidence of edema as demonstrated by large sarcoplasmic areas (Fig. 5), dilatations of the intercalated disc, and mitochondrial alterations (Fig. 6). The latter consisted of a decreased matrix density, swelling, and disruption of cristae. In contrast, hypoxia caused few and infrequent alterations in WKY (Fig. 7). While most cells lacked any alteration, some occasionally displayed dilated T-tubules. Most cells in captopril-treated SHR resembled those of WKY; i.e., most showed no alterations; the only change seen was mild-moderate dilations of T-tubules, but this occurred in only 12% of the animals. Figure 8 demonstrates a normal intercalated disc which contrasts with those commonly observed in untreated SHR. Thus, captopril treatment in SHR completely abolished the structural alterations common to hypoxic exposure. Hydralazine treatment in SHR, which did not affect LVH, had little effect in reducing subcellular damage. Dilatations in the region of intercalated discs (Fig. 9) and expanded T-tubules (Fig. 10) persisted, and the incidence of edema was also unaltered. Thirtyseven percent of the H-SHR group presented mitochondrial changes and cellular edema. As summarized in Table 4, the hypoxic changes consistently observed in SHR were prevented by captopril but not hydralazine treatment. The latter treatment did cause a slight decrease in the severity of the edema and mitochondrial alterations. By using the Fisher's Exact Test for Frequencies (Table 41, the conclusion is reached that captopril significantly lowered the incidences of all characteristic subcellular changes in SHR. DISCUSSION The major findings of this study are: 1)cardiocytes of SHR are more vulnerable to acute hypoxic damage than cells from WKY; and 2) lowering arterial pressure and reversing LVH with captopril prevents the vulnerability to hypoxic damage in SHR. Ischemia or hypoxia, if adequate, causes rather predictable ultrastructural changes. These alterations may be irreversible; irreversibility is related to the length of time the cell is deprived of Oz. In our study the described alterations were reversible. Reversible damage is morphologically characterized by a decrease in glycogen (Jennings and Ganote, 19741, slightly swollen mitochondria, some margination of nuclear chromatin, intracellular edema (Hearse et al., 19751, swollen T-tubules, swollen sarcoplasmic reticulum, and swollen intercalated disks (Denker et al., 1969; McCallister et al., 1979). Irreversible damage is identified by dense intramitochondrial granules, almost complete depletion of glycogen, disorganization of the sarcoplasmic reticulum, disruption of the sarcolemma, pronounced peripheral aggregation of nuclear chromatin, and mitochondrial alterations, such as disruption of the cristae and a decrease in matrix density (Jennings et al., 1965; Jennings and Ganote, 1974). Studies have been conducted to discern the time course of reversible vs. irreversible damage (Schaper et al., 1979; Jennings et al., 1965; Ganote et al., 1975; Edoute et al., 1983). Dog hearts subjected to 15 minutes or less of ischemia in vivo exhibited anatomical characteristics of reversible damage, while 30 minutes or more resulted in irreversible damage (Jennings et al., 1965). A study employing a n isolated, beating, rat heart working against a n afterload of 100 mmHg reported signs of infrequent, irreversible damage in the endomyocardium following just 15 minutes of anoxic perfusion. The frequency of this damage progressively increased after 20 and 25 minutes of anoxia (Edoute et al., 1983). Although previous studies have examined the time sequence of ultrastructural alterations in ischemic myocardium, the effects of various magnitudes of ischemia on myocardial cellular structures were not studied until 1981 when four stages of flow reductions, expressed a s percent of control flow, i.e.: 75-80 (Stage 1); 50-74 (Stage 2); 25-49 (Stage 3); and 0-24 (Stage 41, were examined in our laboratory (Tomanek et al., 1981). These flow reductions were maintained for a period of 12 hours. Stage 1 was characterized by normal cardiac structure except for rare instances of focal mitochondrial disruption. Stage 2 was associated with more pronounced changes in sarcoplasm and mitochondria. Mitochondria exhibited focal swelling and disrup- HYPERTROPHY AND HYPOXIC DAMAGE TO CARDIOCYTE tion in all cells and there was enhancement of the sarcoplasmic compartment in most cells. More severe ischemia (Stage 3) resulted in severe swelling of mitochondria, amorphous densities in many of the mitochondria, glycogen loss, marked enhancement of the sarcoplasm, Z-line irregularity, focal I-band disruption, presence of contraction bands, and clumping and margination of chromatin in many nuclei. Stage 4 was associated with mitochondrial vesiculation and occasional destruction of cristae, vesiculation and marked expansion of the sarcoplasm, decrease in the amount of sarcoplasmic reticulum, disruption of myofibrils, loss of Z-lines, presence of contraction bands, and clumping and margination of chromatin in all nuclei. A number of studies have documented hemodynamic and biochemical consequences of hypoxia. These include a decline in myocardial performance a s suggested by a drop in dP/dt (Carey et al., 1976), and a decrease in myocardial ATP and creatine phosphate (Edoute et al., 1983; Schaper et al., 1979; Hearse et al., 1976). Lore11 et al. (19861, using a n isolated, beating heart, noted t h a t hypoxia alters diastolic properties of both the hypertrophied and non-hypertrophied ventricles. However, the changes are more pronounced in the hypertrophied left ventricle, which undergoes a greater increase in end diastolic pressure and a greater upward shift of the asymptote of left ventricular pressure decay. Thus, hypoxia seemed to be associated with a greater loss of left ventricular diastolic distensibility and a lesser extent of relaxation in hypertrophied hearts than in non-hypertrophied controls. Although hypoxia led to a depression of hemodynamics in our study, we did not discern any differences between hypertrophied and non-hypertrophied left ventricles. Nevertheless, a decrement in more sensitive measures of ventricular performance and contractility may have occurred in the hypertrophied hearts. The effects of hypoxia on the cell structure of the hypertrophied myocardium were previously studied in our laboratory (Wangler et al., 1980). SHR and normotensive, non-hypertrophied controls (WKY) were subjected to 15 minutes of 7% 0?93% Nz. This acute hypoxic episode resulted in a n arterial POz of 43-55 mmHg. When the hypoxic cardiac tissue was examined with the electron microscope, slight and infrequent mitochondrial and T-tubular swelling was observed in the endomyocardium of WKY. However, SHR were associated with focal cellular edema, frequent T-tubular swelling, and a decrease in mitochondrial matrix density. That SHR cardiocytes are more vulnerable to hypoxia than are WKY cardiocytes and that the damage is morphologically characteristic of reversible damage are in agreement with the present results, which are based on a greater degree of hypoxia. Thus, i t is not surprising that we observed characteristics of irreversible damage, albeit infrequent, in the hypertrophied myocardium. This damage was characterized by decreased mitochrondrial matrix density and disruption of mitochondrial cristae. We submit that increased vulnerability to hypoxia is most likely due to perfusion abnormalities associated with LVH secondary to pressure overload, such as increased MCVR (O’Keefe et al., 1978; Wangler et al., 1982) and decreased capillary density (Tomanek et al., 1979; Rakusan et al., 1980; Anversa et al., 1979). Since oxygen extraction in the 205 myocardium is near maximum the cardiocyte depends on a n increase in blood flow to meet any increase in oxygen requirement (Feigl, 1983). An elevated MCVR would tend to limit the ability of the coronary circulation to deliver oxygenated blood to the cardiocytes during times of metabolic stress (e.g., hypoxia). Moreover, decreased capillary density would be expected to result in increased oxygen diffusion distance and a corresponding decrease in myocardial oxygen pressure. Thus, elevated MCVR and reduced capillary density could act in concert to hinder the ability of the coronary circulation to deliver oxygen to the hypertrophied myocardium during physiological stress resulting in ischemic or hypoxic damage. Based on the foregoing discussion i t is concluded that myocardial hypertrophy in response to pressure overload renders the myocardium more vulnerable to hypoxia. Such vulnerability is characterized by ultrastructural alterations as well as biochemical and physiological abnormalities. Our findings indicate that cardiac hypertrophy (which is associated with a decrease in capillary density), rather than elevated blood pressure, is the major contributor to this increased vulnerability. ACKNOWLEDGMENTS We thank The Squibb Institute for Medical Research for their generous supply of captopril. This work was supported by National Heart, Lung, and Blood Institute Grants HL 18629, HL 14388, and HL 32295. LITERATURE CITED Anversa, P., G. Olivetti, M. Melissari, and A.V. Loud 1979 Morphometric study of myocardial hypertrophy induced by abdominal aortic stenosis. Lab. Invest., 40t341-349. Bache, R.J., T.R. Vrobel, C.E. Arentzen, and W.S. Ring 1981 Effect of maximal coronary vasodilation on transmural myocardial perfusion during tachycardia in dogs with left ventricular hypertrophy. Circ. Res. 49t742-750. Bache, R.J., and T.R. Vrobel 1979 Effects of exercise on blood flow in the hypertrophied heart. Am. J. Cardiol., 44t1029-1033. Breisch, E.A., F.C. White, L E . Nimmo, and C.M. Bloor 1986 Cardiac vasculature and flow during pressure-overload hypertrophy. Am. J. Physiol., 251tH1031-Hl037. Canby, C.A., and R.J. Tomanek 1988 Role of lowering blood pressure (BP) on minimal coronary vascular resistance (MCVR) with and without regression of cardiac hypertrophy. FASEB J., 2tA303. Carey, R.A., C.M. Tipton, and D.R. Lund 1976 Influence of training on myocardial responses of rats subjected to conditions of ischemia and hypoxia. Cardiovasc. Res., 10:359-367. Denker, M.W., R.A. Bergman, and M.M. Nachlas 1969 Ultrastructural changes in myocardium during experimental ischemia. Hopkins Med. J., 124t311-330. Edoute, Y., E. van der Merwe, D. Sanan, J.C.N. Kotze, C. Steinmann, and A. Lochner 1983 Normothermic ischemia cardiac arrest of the isolated working rat heart. Circ. Res., 53r663-678. Feigl, E.O. 1983 Coronary physiology. Physiol. Rev., 63:l-205. Ganote, C.E., R. Seabra-Gomes, W.G. Nayler, and R.B. Jennings 1975 Irreversible myocardial injury in anoxic perfused rat hearts. Am. J . Pathol., 80:419-450. Hearse, D.J., S.M. Humphrey, W.G. Nayler, A. Slade, and D. Border 1975 Ultrastructural damage associated with reoxygenation of the anoxic myocardium. J. Mol. Cell. Cardiol. 7r315-324. Henquell, L., C.L. Odoroff, and C.R. Honig 1977 Intercapillary distance and capillary reserve in hypertrophied rat hearts being in situ. Circ. Res., 41t400-408. Jennings, R.B., J.H. Baum, and P.B. Herdson 1965 Fine structural changes in myocardial ischemic injury. Arch. Pathol., 79t135143. Jennings, R.B., and C.E. Ganote 1974 Structural changes in myocar- 206 C.A. CANBY AND R.J. TOMANEK dium during acute ischemia. Cir. Res., 34 and 35 (Suppl. IIII: 156-168. Lorell, B.H., L.F. Waxler, S. Momomura, E. Wainberg, and C.S. Apstein 1986 The influence of pressure overload left ventricular hypertrophy of diastolic properties during hypoxia in isovolumically contracting rat hearts. Circ. Res., 58:653-663. McCallister, L.P., S. Trapukdi, and J.R. Neely 1979 Morphometric observations on the effects of ischemia in the isolated perfused rat heart. J . Mol. Cell. Cardiol., 11:619-630. OtKeefe, D.D., J.I.E. Hoffman, R. Cheitlin, M.J. ONeill, J.R. Allard, and E. Shapkin 1978 Coronary blood flow in experimental canine left ventricular hypertrophy. Circ. Res., 43:43-51. Rakusan, K., J . Moravec, and P.-Y. Hatt 1980 Regional capillary supply in the normal and hypertrophied rat heart. Microvasc. Res., 20:319-326. Schaper, J., J . Mulch, B. Winkler, and W. Schaper 1979 Ultrastructural, functional, and biochemical criteria for estimation of reversibility of ischemic injury: Study on the effects of global ischemia on the isolated dog heart. J. Mol. Cell. Cardiol., 11:521-541. Tomanek, R.J., J.W. Davis, and S.C. Anderson 1979 The effects of alpha-methyldopa on cardiac hypertrophy in spontaneously hypertensive rats: Ultrastructural, stereological, and morphometric analysis. Cardiovasc. Res., 13:173-182. Tomanek, R.J., J.C. Grimes, and J.N. Diana 1981 Relationship between the magnitude of myocardial ischemia and ultrastructural alterations. Exp. Mol. Pathol., 35:65-83. Tomanek, R.J., R.D. Wangler, and C.A. Bauer 1985 Prevention of coronary vasodilator reserve decrement in spontaneously hypertensive rats. Hypertension, 7533-540. Wangler, R.D., J.M. Hovanec, and R.J. Tomanek 1980 Effect of hypoxemia on the myocardium of normotensive and spontaneously hypertensive rats. Anat. Rec., 196:197-198. Wangler, R.D., K.G. Peters, M.L. Marcus, and R.J. Tomanek 1982 Effects of duration and severity of arterial hypertension and cardiac hypertrophy on coronary vasodilator reserve. Circ. Res., 51: 10-18.