Dipyridamole-induced capillary growth in normal and hypertrophic hearts RONALD PATRICE Department
J. TORRY, DIANE M. O’BRIEN, M. CONNELL, AND ROBERT J. TOMANEK of Anatomy
and The Cardiovascular
Torry, Ronald J., Diane M. O’Brien, Patrice M. Connell, and Robert J. Tomanek. Dipyridamole-induced capillary growth in normal and hypertrophic hearts. Am. J. Physiol. 262 (Heart Circ. Physiol. 31): H980-H986, 1992.-Chronic increases in myocardial blood flow have been shown to stimulate capillary proliferation in normal growing hearts. It is unknown, however, if elevated myocardial blood flow stimulates precapillary and/or capillary growth in hearts undergoing hypertrophy. Accordingly, renal hypertension was produced in rabbits (Page, l-kidney, l-wrap model) in which one group of Page (n = 9) and one group of normotensive sham (n = 10) rabbits were given dipyridamole (4.0 mg/kg SC)twice daily for 2 mo. Another group of Page (n = 7) and sham (n = 12) rabbits received vehicle injections. In separate acute studies performed on conscious rabbits, this dose of dipyridamole increased myocardial blood flow 3560% over time without altering transmural distribution of flow or systemic blood pressure. Two months later, minimal coronary vascular resistance (MCVR/ 100 g) was calculated from perfusion during maximal coronary vasodilation in conscious animals. Histomorphometric methods were then utilized to evaluate various indexes of capillarity in perfuse-fixed hearts. Systolic pressure and left ventricle weight to-body weight ratios were significantly higher in Page vs. sham rabbits; dipyridamole treatment did not alter these parameters within either group. Similarly, dipyridamole treatment did not significantly alter MCVR/lOO g values in either normotensive or hypertensive rabbits. In contrast, dipyridamole treatment increased endomyocardial capillary length density by 33% in the hypertensive group (P < 0.05) and 11% in the sham group (P not significant) compared with the respective vehicle-treated rabbits. In addition, intercapillary distance was significantly reduced in the endomyocardial region of both groups receiving dipyridamole injections. Thus our data support the conclusion that chronic dipyridamole treatment induces left ventricular capillary growth but does not promote resistance vessel growth in either normal or hypertrophic hearts. coronary angiogenesis; hypertension; hypertrophy; microcirculation MECHANICAL
FACTORS associated with chronic
increases intravascular
in myocardial blood flow (i.e., increased pressure, stretch, shear stress, and so on) have been shown to induce capillary angiogenesis in both skeletal and cardiac muscle in vivo (reviewed in Ref. 13). It is unknown, however, whether chronically elevated myocardial blood flow (MBF) can produce sufficient proliferation of capillary and precapillary vessels in the pressure-overloaded, hypertrophic heart so as to return minimal coronary vascular resistance (MCVR) and capillarity to normal levels. Chronic administration of dipyridamole (a potent coronary vasodilator) has been shown to stimulate myocardial capillary endothelial cell proliferation (29) and increase capillary length and surface density (19), as well as capillary-to-fiber ratio (28) in the hearts of normotensive rats. Long-term vasodilation with adenosine or H980
0363-6135/92
Center, University
of Iowa, Iowa City, Iowa 52242
HWA-285 (a xanthine derivative) results in substantial increases in both cardiac and skeletal muscle capillary densities of rabbits (33). Ethanol administration, which increases MBF (l), also increases myocardial capillary density (18). However, these studies did not assess the response of the precapillary vessels to elevated MBF. Considering the evidence that arterioles may develop from capillaries exposed to high flow rates (6) and that chronic dipyridamole treatment increases myocardial collateral capacity of dogs (24)) there is a good foundation for the hypothesis that precapillary growth may be stimulated by increased MBF. Although growth of coronary vessels in some models of cardiac hypertrophy has been considered negligible, there is now good evidence that significant angiogenesis occurs in some models and under certain conditions (for review see Ref. 26). In the present study, we assessed the coronary vascular response after 2 mo of dipyridamole treatment in both normotensive and renal hypertensive rabbits. MBF and MCVR (an index of resistance vessel cross-sectional area) were measured in conscious rabbits. Perfuse-fixed hearts were used to assess quantitative indexes
of capillarity
(i.e., capillary
length
densities,
diameters, intercapillary distances, and volume densities). Using these two approaches we were able to determine the growth response of capillary and precapillary vessels in normal and hypertrophic hearts after dipyridamole treatment. METHODS Protocol. The study was performed on 38 adult, New Zealand White male rabbits, each of which was randomly assigned to one of the following groups: Page hypertensive + dipyridamole (n = 9), Page hypertensive + vehicle (n = 7), sham normotensive + dipyridamole (n = lo), and sham normotensive + vehicle (n = 12).
Renal hypertension was produced in 16 rabbits according to the procedure described by Page (23). Briefly, the technique consisted of the removal of one kidney while the contralateral kidney was securely wrapped in cellophane. Sham normotensive controls consisted of animals that underwent unilateral nephrectomy; the contralateral kidney was exposed but not wrapped. All surgery was performed under sterile conditions after the rabbit had been anesthetized with a ketamine (40 mg/ kg)-Rompun (4 mg/kg)-acepromazine (1 mg/kg) cocktail (im). Maintenance doses of the anesthetic were given as needed throughout the procedure. Standard postoperative care including antibiotics (tetracycline) were provided until the wounds healed. Seven to ten days after surgery, 9 Page and 10 sham rabbits began receiving dipyridamole (4.0 mg/kg SC)twice daily, 7 days/wk, for 2 mo. The efficacy of this dosage in elevating MBF was established by measuring myocardial perfusion in an additional seven conscious rabbits at various time points after drug administration. Seven Page and 12 sham rabbits received
$2.00 Copyright 0 1992 the American Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.
DIPYRIDAMOLE-INDUCED
vehicle injections in equal volume (per body weight basis) as the dipyridamole rabbits. The amount of dipyridamole and vehicle given was adjusted weekly based on body weight gain for each animal. At the end of the 2 mo, MBF was measured with radioactive microspheres and the heart was fixed by vascular perfusion. After quantification of radioactivity, tissue specimens were excised and processed for histomorphometric analysis. Measurement of regional MBF in conscious rabbits. Dipyridamole injections were stopped at least 24 h before the blood flow study. One day before the study, the rabbit was anesthetized with the ketamine-Rompun-acepromazine cocktail and cannulas were placed in the left ventricle (LV; via the right common carotid artery), both femoral arteries, and the jugular vein. All catheters were secured and, under xylocaine infiltration, tunneled subcutaneously, and exposed between the scapulae of the animal. All wounds were carefully sutured, and the animal was allowed to recover overnight. The next day, MBF was determined in the conscious rabbit with 15 pm microspheres (New England Nuclear) labeled with either g5Nb, s5Sr, ““Gd, or 46Scand suspended in 10% dextran with 0.1% Tween 80 added to prevent clumping. The microspheres were dispersed with a mechanical mixer for a minimum of 5 min before injection of l-2 x lo6 spheres into the left ventricular catheter. The catheter was then flushed with 2 ml of saline. The arterial reference sample was withdrawn from a femoral catheter at a constant rate of 0.62 ml/min beginning 20 s before and continuing for 2 min after flushing the left ventricular catheter. MBF was assessedat rest and during maximal coronary vasodilation with intravenous dipyridamole infusion. Previous unpublished studies from our laboratory have shown that infusion of 0.4 mg kg*‘. min-’ dipyridamole maximally vasodilates the rabbit coronary vasculature. Two doses of the drug were used in this study (0.4 and 0.7 mg* kg-‘amin-‘), and the highest attained flow was utilized as “maximal coronary flow.” After the measurements of hemodynamics and myocardial perfusion, the animal was anesthetized with pentobarbital sodium (30 mg/kg iv), and the chest was opened rapidly during artificial ventilation. After injection of heparin (-10,000 U), the heart was arrested in diastole with 2% procaine and was immediately removed from the animal. Each heart was then perfused in vitro with Locke’s solution (37°C) followed by glutaraldehyde fixative solution at room temperature. After the atria were removed, the right ventricle (RV) and LV were separated, blotted dry, and weighed. Subsequently, the LV was separated into anterior, posterior, and septal regions and further subdivided into endomyocardial and epimyocardial portions. These portions, along with the RV, were weighed, and the radioactivity in each was assessedwith a Beckman gamma counter (model 8000). Standard calculations were used to estimate MBF (ml. min-’ . 100 g-‘) to both the RV and LV (12). Left and right ventricular MCVR, used as an index of arteriolar cross-sectional area, were calculated by dividing mean arterial pressure by MBF/lOO g during maximal coronary vasodilation. Total MCVR for each ventricle was derived by dividing MCVR/ 100 g by ventricular weight (g). Capillary morphometrics. Several specimens from the epimyocardium, midmyocardium, and endomyocardium of the LV, and the endomyocardial region of the RV were dissected free, washed in buffer, post-fixed in Os04, dehydrated in ethanol, and embedded in Spurr’s plastic. Cross-sections (1 pm in thickness) were then cut from these specimens, subsequently stained with Richardson’s stain, and used for morphometric analysis of capillaries. Capillary morphology was assessedin only those sections that showed uniform perfusion-fixation, as judged by a dilated appearance of capillaries with an absence of plasma proteins. Tissue images from these sections were projected onto drawing paper at a final magnification of x1.440. Three to five l
H981
ANGIOGENESIS
fields from each specimen were randomly selected, and the capillaries in each field were counted. To minimize the inclusion of venules, only vesselswith short-axis diameters
J. TORRY, DIANE M. O’BRIEN, M. CONNELL, AND ROBERT J. TOMANEK of Anatomy
and The Cardiovascular
Torry, Ronald J., Diane M. O’Brien, Patrice M. Connell, and Robert J. Tomanek. Dipyridamole-induced capillary growth in normal and hypertrophic hearts. Am. J. Physiol. 262 (Heart Circ. Physiol. 31): H980-H986, 1992.-Chronic increases in myocardial blood flow have been shown to stimulate capillary proliferation in normal growing hearts. It is unknown, however, if elevated myocardial blood flow stimulates precapillary and/or capillary growth in hearts undergoing hypertrophy. Accordingly, renal hypertension was produced in rabbits (Page, l-kidney, l-wrap model) in which one group of Page (n = 9) and one group of normotensive sham (n = 10) rabbits were given dipyridamole (4.0 mg/kg SC)twice daily for 2 mo. Another group of Page (n = 7) and sham (n = 12) rabbits received vehicle injections. In separate acute studies performed on conscious rabbits, this dose of dipyridamole increased myocardial blood flow 3560% over time without altering transmural distribution of flow or systemic blood pressure. Two months later, minimal coronary vascular resistance (MCVR/ 100 g) was calculated from perfusion during maximal coronary vasodilation in conscious animals. Histomorphometric methods were then utilized to evaluate various indexes of capillarity in perfuse-fixed hearts. Systolic pressure and left ventricle weight to-body weight ratios were significantly higher in Page vs. sham rabbits; dipyridamole treatment did not alter these parameters within either group. Similarly, dipyridamole treatment did not significantly alter MCVR/lOO g values in either normotensive or hypertensive rabbits. In contrast, dipyridamole treatment increased endomyocardial capillary length density by 33% in the hypertensive group (P < 0.05) and 11% in the sham group (P not significant) compared with the respective vehicle-treated rabbits. In addition, intercapillary distance was significantly reduced in the endomyocardial region of both groups receiving dipyridamole injections. Thus our data support the conclusion that chronic dipyridamole treatment induces left ventricular capillary growth but does not promote resistance vessel growth in either normal or hypertrophic hearts. coronary angiogenesis; hypertension; hypertrophy; microcirculation MECHANICAL
FACTORS associated with chronic
increases intravascular
in myocardial blood flow (i.e., increased pressure, stretch, shear stress, and so on) have been shown to induce capillary angiogenesis in both skeletal and cardiac muscle in vivo (reviewed in Ref. 13). It is unknown, however, whether chronically elevated myocardial blood flow (MBF) can produce sufficient proliferation of capillary and precapillary vessels in the pressure-overloaded, hypertrophic heart so as to return minimal coronary vascular resistance (MCVR) and capillarity to normal levels. Chronic administration of dipyridamole (a potent coronary vasodilator) has been shown to stimulate myocardial capillary endothelial cell proliferation (29) and increase capillary length and surface density (19), as well as capillary-to-fiber ratio (28) in the hearts of normotensive rats. Long-term vasodilation with adenosine or H980
0363-6135/92
Center, University
of Iowa, Iowa City, Iowa 52242
HWA-285 (a xanthine derivative) results in substantial increases in both cardiac and skeletal muscle capillary densities of rabbits (33). Ethanol administration, which increases MBF (l), also increases myocardial capillary density (18). However, these studies did not assess the response of the precapillary vessels to elevated MBF. Considering the evidence that arterioles may develop from capillaries exposed to high flow rates (6) and that chronic dipyridamole treatment increases myocardial collateral capacity of dogs (24)) there is a good foundation for the hypothesis that precapillary growth may be stimulated by increased MBF. Although growth of coronary vessels in some models of cardiac hypertrophy has been considered negligible, there is now good evidence that significant angiogenesis occurs in some models and under certain conditions (for review see Ref. 26). In the present study, we assessed the coronary vascular response after 2 mo of dipyridamole treatment in both normotensive and renal hypertensive rabbits. MBF and MCVR (an index of resistance vessel cross-sectional area) were measured in conscious rabbits. Perfuse-fixed hearts were used to assess quantitative indexes
of capillarity
(i.e., capillary
length
densities,
diameters, intercapillary distances, and volume densities). Using these two approaches we were able to determine the growth response of capillary and precapillary vessels in normal and hypertrophic hearts after dipyridamole treatment. METHODS Protocol. The study was performed on 38 adult, New Zealand White male rabbits, each of which was randomly assigned to one of the following groups: Page hypertensive + dipyridamole (n = 9), Page hypertensive + vehicle (n = 7), sham normotensive + dipyridamole (n = lo), and sham normotensive + vehicle (n = 12).
Renal hypertension was produced in 16 rabbits according to the procedure described by Page (23). Briefly, the technique consisted of the removal of one kidney while the contralateral kidney was securely wrapped in cellophane. Sham normotensive controls consisted of animals that underwent unilateral nephrectomy; the contralateral kidney was exposed but not wrapped. All surgery was performed under sterile conditions after the rabbit had been anesthetized with a ketamine (40 mg/ kg)-Rompun (4 mg/kg)-acepromazine (1 mg/kg) cocktail (im). Maintenance doses of the anesthetic were given as needed throughout the procedure. Standard postoperative care including antibiotics (tetracycline) were provided until the wounds healed. Seven to ten days after surgery, 9 Page and 10 sham rabbits began receiving dipyridamole (4.0 mg/kg SC)twice daily, 7 days/wk, for 2 mo. The efficacy of this dosage in elevating MBF was established by measuring myocardial perfusion in an additional seven conscious rabbits at various time points after drug administration. Seven Page and 12 sham rabbits received
$2.00 Copyright 0 1992 the American Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.
DIPYRIDAMOLE-INDUCED
vehicle injections in equal volume (per body weight basis) as the dipyridamole rabbits. The amount of dipyridamole and vehicle given was adjusted weekly based on body weight gain for each animal. At the end of the 2 mo, MBF was measured with radioactive microspheres and the heart was fixed by vascular perfusion. After quantification of radioactivity, tissue specimens were excised and processed for histomorphometric analysis. Measurement of regional MBF in conscious rabbits. Dipyridamole injections were stopped at least 24 h before the blood flow study. One day before the study, the rabbit was anesthetized with the ketamine-Rompun-acepromazine cocktail and cannulas were placed in the left ventricle (LV; via the right common carotid artery), both femoral arteries, and the jugular vein. All catheters were secured and, under xylocaine infiltration, tunneled subcutaneously, and exposed between the scapulae of the animal. All wounds were carefully sutured, and the animal was allowed to recover overnight. The next day, MBF was determined in the conscious rabbit with 15 pm microspheres (New England Nuclear) labeled with either g5Nb, s5Sr, ““Gd, or 46Scand suspended in 10% dextran with 0.1% Tween 80 added to prevent clumping. The microspheres were dispersed with a mechanical mixer for a minimum of 5 min before injection of l-2 x lo6 spheres into the left ventricular catheter. The catheter was then flushed with 2 ml of saline. The arterial reference sample was withdrawn from a femoral catheter at a constant rate of 0.62 ml/min beginning 20 s before and continuing for 2 min after flushing the left ventricular catheter. MBF was assessedat rest and during maximal coronary vasodilation with intravenous dipyridamole infusion. Previous unpublished studies from our laboratory have shown that infusion of 0.4 mg kg*‘. min-’ dipyridamole maximally vasodilates the rabbit coronary vasculature. Two doses of the drug were used in this study (0.4 and 0.7 mg* kg-‘amin-‘), and the highest attained flow was utilized as “maximal coronary flow.” After the measurements of hemodynamics and myocardial perfusion, the animal was anesthetized with pentobarbital sodium (30 mg/kg iv), and the chest was opened rapidly during artificial ventilation. After injection of heparin (-10,000 U), the heart was arrested in diastole with 2% procaine and was immediately removed from the animal. Each heart was then perfused in vitro with Locke’s solution (37°C) followed by glutaraldehyde fixative solution at room temperature. After the atria were removed, the right ventricle (RV) and LV were separated, blotted dry, and weighed. Subsequently, the LV was separated into anterior, posterior, and septal regions and further subdivided into endomyocardial and epimyocardial portions. These portions, along with the RV, were weighed, and the radioactivity in each was assessedwith a Beckman gamma counter (model 8000). Standard calculations were used to estimate MBF (ml. min-’ . 100 g-‘) to both the RV and LV (12). Left and right ventricular MCVR, used as an index of arteriolar cross-sectional area, were calculated by dividing mean arterial pressure by MBF/lOO g during maximal coronary vasodilation. Total MCVR for each ventricle was derived by dividing MCVR/ 100 g by ventricular weight (g). Capillary morphometrics. Several specimens from the epimyocardium, midmyocardium, and endomyocardium of the LV, and the endomyocardial region of the RV were dissected free, washed in buffer, post-fixed in Os04, dehydrated in ethanol, and embedded in Spurr’s plastic. Cross-sections (1 pm in thickness) were then cut from these specimens, subsequently stained with Richardson’s stain, and used for morphometric analysis of capillaries. Capillary morphology was assessedin only those sections that showed uniform perfusion-fixation, as judged by a dilated appearance of capillaries with an absence of plasma proteins. Tissue images from these sections were projected onto drawing paper at a final magnification of x1.440. Three to five l
H981
ANGIOGENESIS
fields from each specimen were randomly selected, and the capillaries in each field were counted. To minimize the inclusion of venules, only vesselswith short-axis diameters
