Cardiac interstitium in health and disease: the fibrillar collagen network.

Composed of type I and III collagens, the valve leaflets, chordae tendineae and collagen matrix of the myocardium form a structural continuum. Synthesized by cardiac fibroblasts, these fibrillar collagens support and tether myocytes to maintain their alignment, whereas their respective tensile strength and resilience resist the deformation, maintain the shape and thickness, prevent the rupture and contribute to the passive and active stiffness of the myocardium. An acquired or congenital defect in this collagen network can lead to abnormalities in myocardial architecture, mechanics or valve function. In the hypertrophic process that accompanies a pressure overload, for example, increased collagen synthesis, fibroblast proliferation and a structural and biochemical remodeling of the matrix are seen. This includes distinctive patterns of reparative and reactive myocardial fibrosis, each of which alters diastolic and systolic myocardial stiffness and may lead to pathologic hypertrophy. Alternatively, a loss of collagen tethers or decline in matrix tensile strength can be responsible for regional or global transformations in myocardial architecture and function seen in the reperfused ("stunned") myocardium and in dilated (idiopathic) cardiopathy. Inherited disorders in the transcriptional and posttranslational processing of collagen can also alter the biophysical properties of the network. Future studies into collagen gene regulation, gene switching events and the control of collagen synthesis and degradation are needed to develop a more complete understanding of the relation between the collagen network and acquired and inherited forms of heart disease and to utilize therapeutics that will prevent, retard or regress abnormal collagen matrix remodeling.

Lactate production during maximal and submaximal exercise in patients with chronic heart failure.

In patients with chronic heart failure whose cardiac output response to exercise is impaired, determination of anaerobic threshold may provide a useful and objective approach to grade the severity of heart failure. In such patients performing upright treadmill exercise to exhaustion, this study examined the reproducibility of the response of cardiac output and mixed venous lactate concentration when the exercise test was repeated the same or next day, the nature of this response after rest and exercise cardiac output levels were augmented by the cardiotonic agent amrinone and the response of lactate during symptom-limited submaximal exercise performed at either aerobic or anaerobic levels of work for each patient. Findings were: 1) the response of cardiac output and mixed venous lactate was reproducible (p less than 0.05) when assessed either the same or the next day; 2) when exercise cardiac output was increased (p less than 0.05) by oral amrinone therapy, the increase in lactate was delayed (p less than 0.05) to higher levels of muscular work and this was not true when cardiac output was unchanged; and 3) only submaximal anaerobic exercise was symptom limited and associated with an increase in lactate concentration. Thus, the lactate response and anaerobic threshold determination should prove useful to assess the severity of chronic stable heart failure and its response to pharmacologic intervention. The submaximal anaerobic exercise test may provide additional insights into the effort intolerance these patients experience.

Physiologic correlates of the heart rate response to upright isotonic exercise: relevance to rate-responsive pacemakers.

Rate-responsive cardiac pacing requires a sensitive physiologic variable that is closely correlated with the heart rate-oxygen uptake relation, particularly in patients with heart failure whose cardiac output response to exercise is more dependent on heart rate. Accordingly, the heart rate response to upright exercise was measured in 81 patients with heart failure or hypertension, or both, and in 27 normal subjects. Oxygen uptake (VO2), minute ventilation (VE), cardiac output, right heart pressures and the mixed venous temperature, oxygen saturation (SvO2) and pH were analyzed throughout exercise. Linear regression analysis of these variables with heart rate revealed the following: 1) There was a highly linear heart rate-VO2 relation in each subject (the average slope of this relation was greater [p less than 0.05] in patients with more severe failure). 2) VE was highly correlated with exercise heart rate, and its slope was not different between normal subjects and patients. 3) Mixed venous temperature and pH were poor predictors of exercise heart rate, particularly at low or moderate levels of work; however, SvO2 was highly correlated with heart rate for all levels of work. Thus, in normal subjects and patients with heart failure or hypertension, or both, heart rate increases linearly with isotonic leg exercise. Minute ventilation and mixed venous oxygen saturation are highly correlated with this response and may serve as potential sensors for rate-responsive pacemakers.

Hemodynamic, ventilatory and metabolic effects of light isometric exercise in patients with chronic heart failure.

Light isometric exercise, such as lifting or carrying loads that require 25% of a maximal voluntary contraction, is frequently reported to cause dyspnea in patients with heart failure. The pathophysiologic mechanisms responsible for the appearance of this symptom, however, are unknown. Accordingly, hemodynamic, metabolic and ventilatory responses to 6 min of light isometric forearm exercise were examined and compared in 20 patients with chronic heart failure and abnormal ejection fraction (24 +/- 9%) and 17 normal individuals. In contrast to findings in normal volunteers, exercise cardiac index did not increase whereas exercising forearm and mixed venous lactate concentrations increased (p less than 0.05) above levels at rest in patients with heart failure; at 90 s of recovery, blood lactate concentration remained elevated (p less than 0.05). The venous lactate concentration of the nonexercising arm, unlike that of the exercising forearm, was not altered. Oxygen uptake, carbon dioxide production and minute ventilation increased similarly in patients and normal subjects during exercise, but only in patients did each increase further (p less than 0.05) during recovery. Thus, in patients with heart failure, light isometric forearm exercise represents an anaerobic contraction with lactate production. The subsequent increase in carbon dioxide production leads to a disproportionate increase in minute ventilation and oxygen uptake during recovery that may be perceived as breathlessness.

Effect of hydralazine on nutritive flow to working canine gracilis skeletal muscle.

The direct smooth muscle vasodilator hydralazine has been used to treat exertional fatigue in patients with chronic heart failure. However, prior studies suggest that arteriolar vasodilators such as hydralazine may actually impair nutritive flow to working skeletal muscle by interfering with the distribution of blood flow within muscle. To investigate this possibility, tension development and metabolism were measured in nine vascularly isolated gracilis muscle preparations perfused at 90 mm Hg and stimulated to contract progressively at rates of 1, 3 and 6/s with each stage lasting 3 minutes. Studies were then repeated after 30 minutes of intraarterial hydralazine (0.02 to 0.12 mg/min). At rest, hydralazine decreased mean vascular resistance (+/- SEM) from 15.1 +/- 1.4 to 8.6 +/- 0.9 X 10(2) units (p less than 0.001) and increased blood flow from 6.4 +/- 0.7 to 11.4 +/- 1.2 ml/min (p less than 0.001), but did not change oxygen consumption (VO2) control, 18 +/- 1 versus hydralazine, 17 +/- 2 microliter/min). Hydralazine also decreased vascular resistance and increased flow at a contraction rate of 1/s, but not at 3 and 6/s. Hydralazine had no effect on maximal VO2 (control, 254 +/- 18 versus hydralazine, 236 +/- 19 microliter/min), maximal developed tension (control, 353 +/- 90 versus hydralazine, 334 +/- 74 kg X min) or the response in venous lactate (control, 20.6 +/- 2.3 versus hydralazine, 18.1 +/- 2.0 mg/dl). Hydralazine also did not change muscle metabolism and function at contraction rates of 1 and 3/s. These data suggest that hydralazine does not adversely affect nutritive flow to working skeletal muscle.

Influence of collagen network on left ventricular systolic and diastolic function in aortic valve disease.

OBJECTIVES: The purpose of this study was to evaluate left ventricular structure-function interplay in aortic valve disease. BACKGROUND: An increase in myocardial fibrosis has been demonstrated in aortic valve disease, but changes in the collagen network and their effect on ventricular function have not been defined. METHODS: Left ventricular structure was assessed from left ventricular endomyocardial biopsy specimens obtained in 32 patients with aortic valve disease (aortic stenosis in 25, aortic regurgitation in 7). Total collagen volume fraction, orthogonal collagen fiber meshwork (cross-hatching), endocardial fibrosis, muscle fiber diameter and volume fraction of myofibrils were determined by morphologic-morphometric evaluation. Control biopsy data were obtained from six donor hearts before transplantation. Eleven other patients with normal left ventricular function served as hemodynamic status control subjects. Left ventricular biplane cineangiography and high fidelity pressure measurements were carried out in all patients. Systolic function was assessed from ejection fraction. Diastolic function was evaluated by the time constant of relaxation, early and late peak filling rates and the constant of passive myocardial stiffness. Patients were assigned to three groups according to increasing severity of nonmyocyte tissue alterations. Group 1 comprised 10 patients with elevated total collagen volume fraction. Group 2 comprised 6 patients with normal total collagen volume fraction and the presence of increased cross-hatching or endocardial fibrosis, or both. Group 3 comprised 16 patients with elevated total collagen volume fraction and the presence of cross-hatching or endocardial fibrosis, or both. RESULTS: Muscle fiber diameter was increased in the three groups with aortic valve disease, whereas the volume fraction of myofibrils was comparable in all four study groups. Ejection fraction was depressed in groups 2 and 3 compared with the control group. The time constant of relaxation was prolonged in the three groups with aortic valve disease. No differences in early and late peak filling rate were observed in the four study groups, but the constant of myocardial stiffness increased in groups 2 and 3. CONCLUSIONS: In aortic valve disease, changes in collagen architecture are associated with altered systolic function and passive diastolic properties. The sole increase in total collagen volume fraction without a change in architecture leaves systolic and passive diastolic function unaltered.

Amrinone in the treatment of chronic cardiac failure.

The efficacy and safety of oral amrinone were examined in 17 patients with moderately severe to severe heart failure that was refractory to standard medical therapy and vasodilators. The short-term and 28 week response to open amrinone therapy was assessed first, followed by a placebo-controlled, double-blind withdrawal study of two 13 week stages in nine patients. Rest and exercise ventricular function were determined before and after 32 hours of amrinone; aerobic capacity was serially assessed. After 2 hours, 1.64 mg/kg amrinone produced a 40% (p less than 0.001) increase in cardiac output and a 32% (p less than 0.02) decrease in pulmonary wedge pressure without altering heart rate or blood pressure. The exercise cardiac index-wedge pressure curve obtained 32 hours after the first oral dose was significantly shifted (p less than 0.05) above control values. A sustained improvement in maximal oxygen uptake was noted during long-term open amrinone therapy. Subsequently, seven of the patients randomized to placebo therapy had a significant deterioration of symptoms or exercise tolerance, or both. After 4 weeks of readministration of amrinone, clinical stability was once again established and exercise tolerance was improved by Weeks 8 to 16. Adverse effects of thrombocytopenia (one patient) and hepatic dysfunction (one patient) attributable to amrinone were observed. It is concluded that amrinone is effective in the long-term treatment of chronic cardiac failure.

Myocardial energetics and clinical response to the cardiotonic agent MDL 17043 in advanced heart failure.

Cardiotonic agents may prove useful in the long-term treatment of chronic heart failure provided myocardial efficiency is enhanced and clinical status is improved. Accordingly, the short-term hemodynamic and clinical response to the phosphodiesterase inhibitor, MDL 17043, was evaluated. Intravenous increments of 0.05 mg/kg (maximal total 3 mg/kg) were given to a peak cardiac output response in 13 patients with New York Heart Association functional class IV heart failure secondary to ischemic or myopathic disease. Significant (p less than 0.05) responses at peak effect (1.7 mg/kg) included an increase in cardiac output (3.5 to 4.6 liters/min) and heart rate (86 to 90 beats/min) and a decrease in pulmonary capillary wedge (25 to 17 mm Hg), mean arterial (85 to 78 mm Hg) and right atrial (10 to 7 mm Hg) pressures. Coronary sinus flow (measured in nine patients) increased (122 to 144 ml/min, p less than 0.01) as did myocardial oxygen uptake (14.1 to 15.1 ml/min, p less than 0.01), whereas myocardial extraction of oxygen (78 to 72%, p less than 0.01) and lactate (24 to 9%, p less than 0.01) decreased with three patients producing lactate at the time of their peak cardiac output response. Nine of the 12 patients given long-term oral therapy improved at least one functional class at 2 weeks. This improvement was sustained at 20 weeks in five patients. Thus, MDL 17043 acutely improves the function of the failing heart. However, the decrease in oxygen extraction occurring with increased myocardial oxygen uptake suggests that intracoronary shunting may occur along with an increase in oxygen demand and contribute to myocardial anaerobiosis in some patients.(ABSTRACT TRUNCATED AT 250 WORDS)

Remodeling and reparation of the cardiovascular system.

Growth or altered metabolism of nonmyocyte cells (cardiac fibroblasts, vascular smooth muscle and endothelial cells) alters myocardial and vascular structure (remodeling) and function. However, the precise roles of circulating and locally generated factors such as angiotensin II, aldosterone and endothelin that regulate growth and metabolism of nonmyocyte cells have yet to be fully elucidated. Trials of pharmacologic therapy aimed at preventing structural remodeling and repairing altered myocardial structure to or toward normal in the setting of hypertension, heart failure and diabetes are reviewed. It is proposed that these are therapeutic goals that may reduce cardiovascular morbidity and mortality. Although this hypothesis remains unproved the primary goal of therapy should be to preserve or restore tissue structure and function.

Infarct scar: a dynamic tissue.

Infarct scar, a requisite to the rebuilding of necrotic myocardium following myocardial infarction (MI), has long been considered inert. Earlier morphologic studies suggested healing at the infarct site was complete within 6-8 weeks following MI and resultant scar tissue, albeit necessary, was acellular and simply fibrillar collagen. Utilizing molecular and cellular biologic technologies, recent studies indicate otherwise. Infarct scar is composed of phenotypically transformed fibroblast-like cells, termed myofibroblasts (myoFb) because they express alpha-smooth muscle actin (alpha-SMA) and these microfilaments confer contractile behavior in response to various peptides and amines. These cells are nourished by a neovasculature and are persistent at the MI site, where they are metabolically active expressing components requisite to angiotensin (Ang) peptide generation, including converting enzyme, receptors for AngII and transforming growth factor (TGF)-beta1. They continue to elaborate fibrillar type I collagen. Their generation of these peptides contribute to ongoing scar tissue collagen turnover and to fibrous tissue formation of noninfarcted myocardium. Infarct scar contraction accounts for its thinning and its tonus may contribute to abnormal ventricular chamber stiffness with diastolic dysfunction. Infarct scar is a dynamic tissue: cellular, vascularized, metabolically active and contractile. Pharmacologic interventions with angiotensin converting enzyme inhibitor or AT1 receptor antagonist has proven effective in attenuating scar tissue metabolic activity and minimizing adverse accumulation of fibrous tissue in noninfarcted myocardium.

Angiotensin II receptor binding following myocardial infarction in the rat.

OBJECTIVE: The aim was to determine angiotensin II receptor binding and its relationship to angiotensin converting enzyme (ACE) binding and fibrous tissue formation in the rat heart. METHODS: A model of tissue repair (pericardiotomy and myocardial infarction with left coronary artery ligation) was used together with the following: quantitative in vitro autoradiography to determine angiotensin II receptor (125I[Sar1, Ile8]AngII) and ACE (125I-351A) binding densities. Angiotensin II receptor subtype was determined using an AT1 receptor antagonist (DuP753, losartan) and an AT2 receptor antagonist (PD123177). Five groups were studied: age and sex matched controls receiving this operative procedure without subsequent myocardial infarction (sham operated); rats with coronary artery ligation and myocardial infarction; rats with coronary artery ligation and lisinopril (20 mg.kg-1.d-1 in drinking water); sham operated rats receiving lisinopril; and unoperated rats which served as controls to pericardiotomy. Hearts were collected from each group on postoperative day 3 and weeks 1, 2, 4, and 8. RESULTS: There was (1) low angiotensin receptor binding in normal myocardium; (2) markedly increased angiotensin II receptor binding at the site of left ventricular myocardial infarction and endocardial fibrosis of the interventricular septum at day 3 and weeks 1, 2, 4, and 8; (3) high angiotensin II receptor binding in the pericardial fibrosis that followed pericardiotomy, and in the fibrosis that appeared in response to suture insertion around the left coronary artery, in both infarcted and sham operated rats; (4) total displacement of normal and connective tissue angiotensin II receptor binding by DuP753, but not by PD123177; (5) ACE inhibition by lisinopril, but no change in angiotensin II receptor binding, at all sites of fibrosis; and (6) significant attenuation by lisinopril of collagen formation in the visceral pericardium of sham operated controls. CONCLUSIONS: In this model of tissue repair, increased AT1 receptor binding density is associated with ACE binding and fibrillar collagen formation that appears within sites of fibrous tissue formation, including myocardial infarction, endocardial fibrosis, foreign body (silk suture), and pericardiotomy. AT1 receptors may play a role in mediating the fibrogenic response to tissue injury in the rat heart.

Reactive and reparative myocardial fibrosis in arterial hypertension in the rat.

OBJECTIVE: Myocardial fibrosis is an important determinant of pathological hypertrophy. In two experimental models of arterial hypertension the purpose of the study was (a) to determine total collagen volume fraction and relative contribution of scarring and perivascular/interstitial fibrosis; and (b) to assess the effects of the aldosterone receptor antagonist spironolactone on these patterns of fibrosis. METHODS: 76 eight week old Sprague-Dawley rats weighing 180 to 200 g were used for the studies. Using videodensitometry total collagen volume fraction was separated into the various components for the left and right ventricles in the following experimental models: renovascular hypertension occurring following unilateral renal ischaemia; continuous aldosterone administration via osmotic minipumps with either high (AL) or low (ALLO) sodium diet; renovascular hypertension and AL after pretreatment and continuous treatment with either 20 or 200 mg.kg-1.d-1 subcutaneously of the competitive aldosterone receptor antagonist spironolactone. All groups were compared to age and sex matched controls. RESULTS: After eight weeks, systolic pressure was comparably increased in renovascular hypertension and AL and it remained raised with low dose spironolactone treatment in either model, but was normal with high dose spironolactone or low sodium diet. Left ventricular hypertrophy, expressed as a significant increase in left to right ventricular weight and left ventricle to body weight ratios, was present in renovascular hypertension, AL, and AL + low dose spironolactone compared to control (p < 0.005). In either ventricle: (1) the amount of interstitial/perivascular fibrosis and myocardial scarring was increased (p < 0.005) in renovascular hypertension and AL compared to control; (2) each was reduced (p < 0.005) with either dose of spironolactone; and (3) only scars were seen in ALLO. CONCLUSIONS: Myocardial fibrosis of either ventricle was comparable in renovascular hypertension and AL. Spironolactone was able largely to prevent the perivascular/interstitial fibrosis and scarring in either model irrespective of the development of left ventricular hypertrophy and arterial hypertension. Low sodium diet in hyperaldosteronism prevented hypertension and left ventricular hypertrophy, but not scarring. These findings suggest that a rise in plasma aldosterone, relative to sodium intake, may play a role in mediating collagen accumulation in the heart during the development of experimental arterial hypertension.

Cells expressing angiotensin II receptors in fibrous tissue of rat heart.

OBJECTIVE: Following left coronary artery ligation, markedly increased angiotensin II (AngII) receptor binding appears at the site of myocardial infarction (MI). This is also true for the fibrosed visceral pericardium in rats following pericardiotomy (with or without MI). Cells expressing AngII receptors at these sites remain unknown. In the present study, we sought to identify cells expressing AngII receptors at these sites of fibrosis in the rat heart during both early and late stages of wound healing. METHODS: MI was created by left coronary artery ligation. Sham operation included thoracotomy, pericardiotomy and placement of silk ligature around the left coronary artery without MI. Hearts were collected at postoperative week 1 and 4. In serial sections: autoradiography (125I[Sar1,Ile8]AngII) was used to determine cells expressing AngII receptors; hematoxylin-eosin and alpha smooth muscle actin were used for identification of cell morphology and phenotype, respectively; and picrosirius red for identification of fibrillar collagen. RESULTS: (1) at week 1, necrotic tissue at the site of MI was surrounded by granulation tissue that included macrophages, alpha smooth muscle actin fibroblast-like cells, or myofibroblasts, fibrillar collagen, and new vessels; (2) at week 4, scar tissue had formed and remaining cells were primarily myofibroblasts; (3) pericardial fibrosis was evident at weeks 1 and 4 and contained myofibroblasts, not macrophages or new vessels; (4) at week 1 and 4 myofibroblasts were the predominant cell expressing high-density AngII receptors at the site of MI, while fibroblasts, macrophages and vessels demonstrated low-density AngII receptor binding; and (5) at weeks 1 and 4, myofibroblasts express high-density AngII receptor binding in pericardial fibrosis. CONCLUSION: In a rat model of tissue repair involving either MI or pericardial fibrosis, increased AngII receptor expression is primarily associated with myofibroblasts. This suggests AngII may play a role in mediating the fibrogenic response provided by this wound healing cell at sites of tissue injury in the rat heart.

Endothelin receptors in cultured adult rat cardiac fibroblasts.

OBJECTIVES: Endothelins, released by vascular endothelial cells, are known growth promoters of various mesenchymal cells that contribute to stromal protein accumulation. Whether endothelins could contribute to myocardial fibrosis depends, in part on whether cardiac fibroblasts have endothelin receptors. The identification and binding characteristics of endothelin-1 and endothelin-3 and their ETA and ETB receptor subtypes in cultured adult rat cardiac fibroblasts represented study objectives. METHODS: Cultured rat cardiac fibroblasts (passages 5-10) grown until confluence were used to study radioligand binding assays, receptor subtypes, association and dissociation, effects of agonist and antagonist on binding kinetics, and affinity cross linking. RESULTS: Binding association of 125I-endothelin-1 and 125I-endothelin-3 was rapid, specific, and saturable within 60 minutes. The dissociation of receptor bound 125I-endothelin-1 was slow and partially reversible (30%-40%), suggesting more than one class of endothelin receptors, whereas the dissociation of 125I-endothelin-3 was time dependent and reversible. Competitive displacement with unlabeled endothelin-1, endothelin-3, endothelin-receptor nonselective sarafotoxin (S6b), and ETA receptor selective antagonist PED-3512-PI were used to identify receptor subtypes. Displacement of 125I-endothelin-1 by cold endothelin-1, resulted in a low affinity, high binding site (IC50 5.4 x 10(-9) M; 3.6 x 10(4) binding sites.cell-1) and a high affinity, low binding site (IC50 4.2 x 10(-4) M; 11,830 binding sites.cell-1). With 125I-endothelin-1 the IC50 s for sarafotoxin, endothelin-3, and PED-3512-PI were 1.8 x 10(-10), 1.7 x 10(-9), and 3.7 x 10(-9) M, respectively; for 125I-endothelin-3 these IC50s were 2.28 x 10(-11), 1.9 x 10(-10), and 1.7 x 10(-9) M, respectively. Endothelin receptor subunits of 53, 37, 34, and 24 kDa were identified by affinity cross linking. CONCLUSION: Endothelin-1 and endothelin-3 binding and ETA and ETB receptor subtypes are present in cardiac fibroblasts with ETB predominant. The presence of these receptors support the hypothesis that endothelins may regulate cardiac fibroblast function.

Angiotensin II receptor blockade during gestation attenuates collagen formation in the developing rat heart.

OBJECTIVE: Fetal cardiac development includes rapid formation of a three-dimensional collagen network, composed mainly of type I and III fibrillar collagens. Collagen fibrils have been found in cardiac jelly at very early stages of cardiac development and are thought to have structural and functional properties. In adult rat cardiac tissue, angiotensin II (AngII) via AT1 receptor binding and AngII-regulated expression of transforming growth factor beta-1 (TGF-beta 1) each upregulate collagen transcription. AT1 and AT2 receptor subtypes are developmentally regulated; both have been localized in fetal tissue where the AT2 receptor is considered a determinant of morphogenesis. We sought to determine whether blockade of either receptor would result in attenuation of collagen mRNA expression and fibrillar collagen accumulation and alter TGF-beta 1 mRNA expression in the developing fetal heart examined at birth. METHODS: Pregnant rats were treated either with an AT1 receptor antagonist losartan or an AT2 receptor antagonist PD123319 and compared with untreated age-matched controls. Offspring were studied within 24 h of birth. Type I and type III collagen mRNA expression, as well as TGF-beta 1 mRNA expression, were examined by in situ hybridization. Collagen concentration was determined spectrophotometrically by picrosirius red staining and type I and III collagens were detected by immunoblotting. RESULTS: We found: (1) comparable birth weights in control and PD123319-treated animals, but reduced body weight in newborn losartan-treated animals; (2) compared to untreated animals, type I collagen and TGF-beta 1 mRNA expression in cardiac tissue were each equally reduced in both losartan and PD123319-treated animals; (3) increased type III collagen mRNA expression in both PD123319- and losartan-treated groups; and (4) a significant decrease in total soluble cardiac collagen concentration in both losartan and PD123319-treated groups, confirmed by attenuated immunoreactivity of type I and III collagens in whole heart extracts by Western blotting. CONCLUSIONS: The results of these pharmacologic interventions suggest AngII receptors are expressed in cardiac tissue during gestation, where both AT1 and AT2 receptors are involved in the regulation of type I and III collagen expression and structural protein accumulation. These effects appear to be mediated, in part, by attenuated cardiac TGF-beta 1 levels. The marked decrease in newborn cardiac collagen content has yet undefined functional consequences.

Fibrosis of atria and great vessels in response to angiotensin II or aldosterone infusion.

OBJECTIVE: Myocardial fibrosis, associated with increased expression of angiotensin converting enzyme (ACE) and bradykinin (BK) receptor binding at sites of tissue repair, accompanies chronic elevations in circulating angiotensin II (AngII) and/or aldosterone (ALDO) that simulate chronic cardiac failure. A role for increased ventricular wall stress, associated with arterial hypertension, that can accompany such neurohormonal activation when ventricular function is not compromised, has been held responsible for this structural remodeling. To address this proposition, we monitored morphology of right and left atria and pulmonary artery, where stress is not increased, and compared these structures with hypertensive aorta. METHODS: Experimental groups included: (1) unoperated and untreated controls; (2) intact rats receiving AngII (9 micrograms/h) for 2 weeks and which causes arterial hypertension; (3) uninephrectomized control rats on a high sodium diet for 6 weeks; and (4) uninephrectomized rats receiving ALDO (0.75 micrograms/h) and a high sodium diet for 6 weeks and which results in gradual onset arterial hypertension. Fibrosis was identified by light microscopy in sections stained with collagen specific picrosirius red, while ACE, AngII and BK receptors binding were localized and quantitated by in vitro autoradiography using 125I-351A, 125I[Sar1,Ile8]AngII, and 125I[Tyr8]BK, respectively. AngII receptor subtype was defined by the presence of excess AT1 (losartan) or AT2 (PD123177) receptor antagonists, respectively. RESULTS: With either AngII or ALDO administration, and compared to controls, we found: (1) microscopic scarring that replaced lost myocytes in both left and right atria; (2) an increase in adventitial collagen of both pulmonary artery and aorta (perivascular fibrosis); (3) markedly increased ACE binding at fibrous tissue sites in both atria and great vessels; (4) unchanged atrial and great vessel AT1 receptor binding; and (5) significantly increased BK receptor binding at sites of atrial and perivascular fibrosis. CONCLUSIONS: Thus, the appearance of atrial fibrosis and perivascular fibrosis of aorta and pulmonary artery, together with associated increase in ACE and BK receptor binding, in rats receiving AngII or ALDO suggests these responses are not related to altered ventricular wall stress or arterial hypertension, but rather to these effector hormones of the circulating renin-angiotensin-aldosterone system. Local BK, regulated by ACE found in fibrous tissue and BK receptor binding may play a role in structural remodeling of atria and great vessels in these rat models that stimulate chronic cardiac failure.

Angiotensin II associated cardiac myocyte necrosis: role of adrenal catecholamines.

OBJECTIVE: Cardiac myocyte necrosis is associated with an endogenous increase in angiotensin II or a subpressor dose of angiotensin II. In each case, raised plasma angiotensin II, together with aldosterone, results in cardiac myocyte necrosis followed by scarring. Arterial hypertension is non-contributory. The aim of this study was to determine whether myocyte necrosis is induced directly by angiotensin II, or indirectly by catecholamines or aldosterone released from adrenal glands in response to angiotensin II. METHODS: Right and left ventricular myocardium was examined over a two week period in rats receiving a subpressor yet cardiotoxic dose of angiotensin II (150 ng.min-1.kg-1 subcutaneously) alone or together with total adrenalectomy, bilateral medullectomy, or the aldosterone receptor antagonist spironolactone (150-200 mg.d-1.kg-1). Findings were compared to unoperated age/sex matched controls. Coronal sections obtained from the base, equator and apex of the heart were stained with antifibronectin antibody to detect myocyte injury, picrosirius red for fibrillar collagen, or haematoxylin and eosin. RESULTS: In both right and left ventricles (a) fibronectin labelling was distributed within injured cardiac myocytes and adjacent interstitium of both ventricles in rats receiving angiotensin II or angiotensin II plus spironolactone; (b) fibronectin labelling of myocytes was markedly diminished or absent in the ventricles of rats with total adrenalectomy or bilateral medullectomy; and (c) there was microscopic scarring of both ventricles after two weeks of either angiotensin II or angiotensin II plus spironolactone, but not in medullectomised or adrenalectomised animals. CONCLUSIONS: In the presence of increased circulating levels of angiotensin II, bilateral medullectomy and total adrenalectomy (in contrast to spironolactone) reduced myocyte injury and scarring of the right and left ventricular myocardium, suggesting that the cytotoxic effect of angiotensin II is largely indirect and related to circulating catecholamines released by the adrenal medulla.

Angiotensin II-induced myocardial fibrosis in rats: role of nitric oxide, prostaglandins and bradykinin.

OBJECTIVE: Chronic elevations in plasma angiotensin II (AngII) are associated with an efflux of plasma macromolecules into the perivascular and contiguous interstitial space. This is followed by the appearance of macrophages and type I collagen-producing, fibroblast-like cells that precede the accumulation of fibrous tissue at these sites. Whether this perivascular and interstitial fibrosis is a direct effect of AngII on collagen turnover of these cells or an indirect response mediated by nitric oxide, prostaglandins and/or bradykinin released in response to AngII, is uncertain. METHODS: We measured perivascular and interstitial fibrosis (picrosirius-stained tissue) in response to 14-day infusion of AngII (150 ng/kg/min, s.c.) in male Sprague-Dawley rats. Treated animals were compared to untreated controls and to groups receiving AngII together with either an NO-synthase inhibitor [NG-nitro-L-arginine methyl ester (L-NAME) 10 mg/kg/day in drinking water], a cyclo-oxygenase inhibitor (indomethacin, 2 mg/kg/day in drinking water), or a bradykinin B2 receptor antagonist (Hoe140, 115 ng/kg/min, s.c.). RESULTS: When left and right ventricles of treated rats were compared to untreated controls, AngII led to a respective 68 and 48% increase in perivascular collagen volume fraction (PCVF) and a 54 and 22% increase in interstitial collagen volume fraction (ICVF). Co-administration of AngII + L-NAME did not attenuate either PCVF or ICVF while indomethacin significantly attenuated PCVF by 37 and 33% of left and right ventricle, respectively, but did not alter ICVF in either ventricle when compared to AngII-treated animals. Co-administration of AngII + Hoe140 completely prevented perivascular and interstitial collagen accumulation with the extent of perivascular fibrosis comparable to untreated controls. CONCLUSION: The perivascular and interstitial fibrosis of the rat right and left ventricles seen in association with the exogenous administration of AngII is mediated by the release of bradykinin and prostaglandins, and therefore is an indirect response to elevated circulating AngII.

Extracardiac pressure and ventricular haemodynamics.

To investigate the relative influence of pericardial and intrathoracic pressures on left and right ventricular diastolic and systolic pressures two experimental groups were studied: group 1-10 open chest dogs with a controlled pericardial effusion; and group 2 - five closed chest dogs with a controlled pneumothorax at constant transpulmonary pressure and lung volume. In both groups right and left ventricular diastolic pressures were linear functions of external pressure with respective slopes of 0.71 and 0.39 for group 1 and 0.80 and 0.78 for group 2. The left ventricular slope of 0.39 indicates that left ventricular volume decreased more with increments in pericardial pressure since a slope of approximately 1.00 would be expected if the filling volume was invariant. Accordingly, the fall in left ventricular systolic pressure that occurred in both groups was two to three times greater in group 1. Right ventricular systolic pressure fell in group 1 whereas it increased in group 2. It is concluded that there are no major differences between the influence of either type of external pressure on right ventricular filing. In contrast, the left ventricle, and in particular the filling pressure gradient between the pulmonary circulation and the left ventricle, is more sensitive to pericardial pressure.

Myocardial stiffness and reparative fibrosis following coronary embolisation in the rat.

The structural nature of fibrillar collagen involved in the replacement fibrosis which accompanies discrete areas of cell necrosis remains uncertain, as does its influence on the diastolic and systolic stiffness of the intact myocardium. This study, using 15 micron diameter microsphere embolisation of the rat myocardium, was undertaken to address these issues. Collagen volume fraction (trichrome), fibrillar collagens (picrosirius-polarisation technique), and the stress-strain relations of the intact myocardium (isolated hearts) were determined 30 d after the infusion of microspheres into the left ventricle. Significant differences from controls included: (a) the presence of hypertension secondary to renovascular embolisation; (b) a greater volume fraction of collagen that included not only a meshwork of short, taut appearing, thick and thin collagen fibres, interposed between muscle in areas of cell loss, but also a perivascular fibrosis involving intramyocardial coronary arteries; (c) elevated active stiffness, and (d) a more exponential diastolic stress-strain relation with increased stiffness at strains of 5% or more. These findings suggest that the replacement fibrosis accompanying myocyte necrosis has distinguishing morphological features involving fibrillar collagen and which because of its structure, alignment, and location relative to muscle leads to enhanced myocardial stiffness, including a more exponential rise in the diastolic stress-strain relation. The perivascular accumulation of collagen suggests that additional factors other than microsphere induced necrosis were responsible for this reactive fibrosis.