Acute cardiac allograft rejection in nitric oxide synthase-2(-/-) and nitric oxide synthase-2(+/+) mice: effects of cellular chimeras on myocardial inflammation and cardiomyocyte damage and apoptosis

BACKGROUND

The contribution of nitric oxide synthase (NOS)-2 to myocardial inflammation and cardiomyocyte necrosis and apoptosis during allograft rejection was investigated through heterotopic cardiac transplantation in mice.

METHODS AND RESULTS

In the first experiments, hearts from C3H donor mice were transplanted into NOS-2(-/-) and NOS-2(+/+) C57BL/6J.129J recipients. A second series of experiments included NOS-2(-/-) donor hearts transplanted into NOS-2(-/-) recipients and wild-type NOS-2(+/+) donor hearts transplanted into wild-type NOS-2(+/+) recipients. (All donors were C57BL/6J and recipients were C57BL/6J.129J.) In the first series of experiments, no significant differences were observed in allograft survival, rejection score, total number of apoptotic nuclei (TUNEL), total number of apoptotic cardiomyocytes, or graft NOS-2 mRNA and protein. Positive NOS-2 immunostaining occurred in endothelial cells and cardiomyocytes in the allografts; the inflammatory infiltrate was NOS-2 positive only when recipients were NOS-2(+/+). In the second series of experiments, cardiac allograft survival was significantly increased in the NOS-2(-/-) mice (26+/-13 versus 17+/-8 days, P<0.05), along with significant reductions in inflammatory infiltrate, rejection score, and total number of apoptotic nuclei (23.5+/-9.5 versus 56.4+/-15.3, P<0.01) and of apoptotic cardiomyocytes (2.9+/-1.6 versus 6.9+/-2.7, P<0.05). No NOS-2 or nitrotyrosine, a marker of peroxynitrite exposure, was detected in NOS-2(-/-) allografts transplanted into NOS-2(-/-) recipients.

CONCLUSIONS

The data suggest that NO derived from NOS-2 contributes to the inflammatory response and to cardiomyocyte damage and apoptosis during acute cardiac allograft rejection.

Inducible nitric oxide synthase expression in smooth muscle cells and macrophages of human transplant coronary artery disease

BACKGROUND

The inducible isoform of the nitric oxide synthase (iNOS) produces large amounts of nitric oxide in response to cytokine stimulation. Previous investigations have demonstrated iNOS expression in the setting of acute and chronic rejection in experimental cardiac transplant models. The goal of this study was to investigate whether iNOS is upregulated in human transplant coronary artery disease (TCAD), a major cause of late mortality after cardiac transplantation.

METHODS AND RESULTS

We studied 15 patients with TCAD and 10 with normal coronary arteries. In situ hybridization and immunohistochemistry were used in tissue sections to localize iNOS mRNA and protein, respectively. The presence of peroxynitrite was indirectly assessed by immunostaining with an anti-nitrotyrosine antibody. Normal coronary arteries had no evidence of iNOS expression. In contrast, 30 of 36 coronary artery segments with TCAD (83%) were immunostained by the iNOS antibody. The presence of iNOS mRNA was demonstrated in these vessels by in situ hybridization. Specific cell markers identified iNOS-positive cells as neointimal macrophages and smooth muscle cells. Nitrotyrosine immunoreactivity colocalized with iNOS expression in arteries with TCAD, distributed in macrophages and smooth muscle cells.

CONCLUSIONS

iNOS mRNA and protein are expressed in human arteries with TCAD, where they are associated with extensive nitration of protein tyrosines. These findings indicate that the high-output nitric oxide pathway and possibly the oxidant peroxynitrite might be involved in the process leading to the development of TCAD.

Increased immunoreactive endothelin-1 in human transplant coronary artery disease

BACKGROUND

The pathogenesis of transplant coronary artery disease (TCAD) is unknown, but it is thought to derive from an interaction between immune and nonimmune factors, leading to smooth muscle cell proliferation and accumulation in the expanded neointima. Endothelin-1 (ET-1), a potent vasoconstrictor with mitogenic properties for vascular smooth muscle cells, has recently been demonstrated in native vessel atherosclerosis. The present study used immunohistochemistry to investigate the role of ET-1 in TCAD.

METHODS AND RESULTS

ET-1 immunoreactivity and cellular localization were assessed in human coronary arteries with TCAD (n = 13) and in normal coronary arteries (n = 10) with single- and double-label immunohistochemistry. The intensity of immunostaining was determined by a semiquantitative method. Diffuse and intense ET-1 immunoreactivity was found in 11 of 13 patients with TCAD (85%), mainly in myointimal cells and, in lesser amounts, in macrophages and endothelial cells. In contrast, normal coronary arteries had only faint immunostaining localized to the endothelial layer. Mean semiquantitative grade was significantly higher in TCAD than in normal arteries (1.8 versus 0.7; P < .05). ET-1 was more frequently present in lipid-rich, atheromatous lesions than in lipid-poor, proliferative ones. Intimal neovessels consistently immunostained for ET-1.

CONCLUSIONS

Immunoreactivity for ET-1 is significantly increased in TCAD, possibly as a result of stimulatory cytokines and growth factors that are upregulated in the posttransplant state. The results suggest a role for this mitogenic peptide in the pathogenesis of graft arteriosclerosis.

Nonadrenergic noncholinergic innervation. Anatomic distribution of calcitonin gene-related peptide-immunoreactive tissue in the dog heart

Calcitonin gene-related peptide (CGRP) has inotropic and chronotropic effects in rat and guinea pig hearts. It also may mediate nonadrenergic noncholinergic regulation of canine cardiac electrophysiology. In this study, immunohistochemistry was used to determine the anatomic distribution of CGRP in mature dog heart and autonomic ganglia controlling cardiac function. The stellate ganglia had scattered CGRP-immunoreactive cells and nerve processes; intracardiac ganglia contained stained nerve processes but no CGRP-immunoreactive cells. Although the extramural coronary arteries were modestly innervated by varicose individual nerve processes, the great majority of CGRP-immunoreactive neural tissue in the heart existed adjacent to the sinoatrial node where varicose nerve processes coursed in numerous large nerve bundles. Each bundle contained only a few stained processes, however, indicating that CGRP-immunoreactive nerve processes were accompanying another type of autonomic tissue. Double staining and immunoultrastructure confirmed that the nerve bundles were heterogeneous. Similar nerve bundles were fewer in the left atrium, the region of the atrioventricular node, atrioventricular bundle, and the ventricles. In contrast to the distribution of sympathetic neural tissue, CGRP-immunoreactive nerve processes virtually were nonexistent among muscle fibers. We conclude that 1) CGRP-immunoreactive neural tissue likely affects sympathetic and parasympathetic ganglia that control cardiac function, 2) the preponderance of this nonadrenergic noncholinergic tissue near regions of specialized muscle (especially the sinoatrial node) suggests an efferent function in the canine heart, and 3) the presence of varicosities along CGRP-immunoreactive nerve processes within heterogeneous nerve bundles may indicate that direct axo-axonal contact is the mechanism by which these nonadrenergic noncholinergic nerve processes modulated other autonomic neural tissue.

Electrophysiology and ultrastructure of canine subendocardial Purkinje cells isolated from control and 24-hour infarcted hearts

Ventricular arrhythmias that accompany myocardial infarction in dogs may be secondary to the altered electrophysiological properties of the subendocardial Purkinje fibers that survive 24 hours after the coronary occlusion. To better understand the ionic mechanisms that underlie the altered electrical activity of these fibers, we have dispersed, using an enzymatic technique, Purkinje cells from the subendocardium of the infarcted ventricle (IZPCs) and compared their electrical and structural properties to Purkinje cells dispersed from fiber strands (SPCs) and from the subendocardium of the noninfarcted ventricle (NZPCs). Ultrastructural analysis of these cells shows that IZPCs contain an increased number of lipid droplets when compared with the SPCs and NZPCs. In addition, transmembrane action potentials of IZPCs have reduced resting potentials, action potential amplitudes, and upstroke velocity and are increased in duration when compared with either SPCs or NZPCs. Input resistance of IZPCs is increased over that measured in control cells (SPCs and NZPCs). Furthermore, the time course of the process of electrical restitution of action potential duration is altered in IZPCs with long action potentials. Finally, using K+-sensitive microelectrode techniques, we have determined that intracellular free K+ activity (aKi) in IZPCs (93.7 +/- 15 mM) is not significantly different from control aKi measurements (SPC, 106 +/- 13 mM; NZPC, 103 +/- 12 mM). Thus a reduction in aKi does not provide a basis for the reduced resting potentials observed in IZPCs. By studying the relation between the resting potential and log [K+]o we determined that in IZPCs with reduced resting potentials, there is a significant increase in the PNa/PK ratio when compared with control. In summary, to better understand the cellular basis of ventricular arrhythmias postinfarction, we have developed a single cell model that will allow for more rigorous electrophysiological studies of the specific ionic currents that underlie the abnormal electrophysiology.

Structural and electrophysiological changes in the epicardial border zone of canine myocardial infarcts during infarct healing

Structural and electrophysiological properties of the epicardial muscle which survives on the surface of transmural infarcts of the canine heart (epicardial border zone) were studied at different times after occlusion of the left anterior coronary artery (LAD). Isolated preparations were superfused in vitro, transmembrane potentials recorded, and impulse propagation mapped. In preparations from subacute infarcts (1 and 5 days), resting potential, action potential amplitude, upstroke velocity, and duration were all significantly reduced. Well-defined directional differences in propagation occurred. Propagation was more rapid in the direction perpendicular to the left anterior coronary artery than in the direction perpendicular to the base of the heart, because of the uniform anisotropic structure of the surviving muscle fibers which were arranged in tightly packed bundles oriented perpendicular to the left anterior coronary artery. The only ultrastructural abnormalities found in these muscle fibers was an accumulation of large amounts of lipid droplets. As the infarcts healed, resting potential, action potential amplitude, and upstroke velocity returned to normal by 2 weeks, although action potential duration decreased further. Lipid droplets had disappeared, and connective tissue had invaded the epicardial border zone, separating the muscle bundles. By 2 months, action potentials were normal, but the muscle fibers were widely separated and disoriented by the connective tissue (parallel bundles no longer were found). In these regions with a nonuniform anisotropic structure, the well-defined directional differences in impulse propagation were lost. However, activation was very slow, perhaps because of diminished connections between cells. The persistence of slow conduction in healed infarcts may contribute to the occurrence of chronic arrhythmias.

Mechanisms for atrial arrhythmias associated with cardiomyopathy: a study of feline hearts with primary myocardial disease

The cellular electrophysiologic and structural characteristics of arrhythmic and non-arrhythmic atria isolated from feline hearts with spontaneously occurring cardiomyopathy were studied. The animals were divided into three groups according to the degree of left atrial enlargement: mild (group I), moderate (group II), and severe (group III). The right atria were of relatively normal size. Microelectrode recordings showed that inexcitable cells were present in both left and right atria of all groups but were most numerous in the left atria of group III animals. Most inexcitable cells had low resting membrane potentials. There was also a significant reduction in resting membrane potentials, maximum rate of phase 0 depolarization, and action potential amplitude of excitable cells in left atria of animals in groups II and III, whereas action potentials of excitable cells in the right atria were normal. Acetylcholine or norepinephrine often restored excitability to cells that originally did not generate action potentials. Norepinephrine also caused slow-response action potentials as well as abnormal automaticity and triggered activity due to delayed afterpotentials. The diseased atria showed marked structural abnormalities, which were most pronounced in group III cats, including large amounts of interstitial fibrosis, cellular hypertrophy and degeneration, and thickened basement membranes. Therefore electrophysiologic abnormalities and concurrent changes in cell structure may be involved in the genesis of atrial tachyarrhythmias caused by cardiomyopathy.

Malignant neurogenic tumor of the heart

This is the first report of the fine structure of a malignant neurogenic tumor in the heart. The tumor, arising in the infundibulum of the right ventricle of the heart, was resected and examined by both light and electron microscopy. The histologic appearance of the originally resected tumor was consistent with that of a neurofibroma, while that of the recurrent tumor suggested a malignant schwannoma. Ultrastructurally, the presence of basement membrane investing some cells confirmed the neurogenic nature of the tumor. The discontinuous nature of the basement membrane in these cells may provide a clue as to the malignant nature of this tumor.

Diagnosis of acute rheumatic carditis by endomyocardial biopsy

This report is the first to document the definitive diagnosis of acute rheumatic carditis by endomyocardial biopsy. An adolescent girl with mitral regurgitation and biventricular congestive heart failure thought to be secondary to rheumatic heart disease was admitted to the hospital for mitral valve replacement. An endomyocardial biopsy performed at cardiac catheterization demonstrated and Aschoff's body in the endocardium and thus confirmed a clinical suspicion of active carditis. Following a course to steroids, the mitral valve was replaced. The surgically excised valve showed gross and microscopic features of acute and chronic valvulitis consistent with rheumatic endocarditis.

Cardiac fibromatosis: an ultrastructural study

The fine structure of a surgically resected cardiac fibroma is described. The tumor consisted primarily of fibroblasts set in a matrix composed of acid mucopolysaccharide and collagen. Many of the fibroblasts contained abundant myofilaments and dense bodies, suggesting differentiation or transformation of these cells to myofibroblasts. No striated muscle was identified. This study confirms the fact that the cardiac fibroma is derived from connective tissue elements and should not be considered either a hamartoma or a mesoblastic tumor. Structurally, as well as biologically, the cardiac fibroma closely resembles soft tissue fibromatoses and should be considered an example of cardiac fibromatosis.

Time course of infarct growth toward the endocardium after coronary occlusion

Transmembrane potentials and ultrastructure of subendocardial Purkinje and ventricular muscle fibers, isolated 1, 3, 5, 6, 14, and 24 h after coronary occlusion were investigated. Action potentials were recorded from progressively fewer layers of muscle cells as the age of the infarct increased. At 14 h little viable muscle remained. The decrease in the number of electrophysiologically viable muscle fibers correlated with structural evidence that the infarct moved with time toward the endocardial surface until only viable Purkinje fibers remained. Purkinje and surviving ventricular muscle fibers demonstrated a progressive decrease in resting potential, action potential amplitude, and Vmax and a progressive increase in action potential duration. Spontaneous diastolic depolarizations were found in Purkinje fibers only in 24-h infarcts and occasionally in cells deep to the endocardial surface, which may have been muscle cells. We hypothesize that during the first 24 h after coronary occlusion arrhythmias originate near the interface of infarcted and ischemic myocardium. As this interface moves toward the endocardium, this site of origin of arrhythmias moves with it until the Purkinje network is reached.

Structural basis of ventricular arrhythmias in human myocardial infarction: a hypothesis

The present study was undertaken using light and electron microscopic techniques to determine whether Purkinje fibers survive in the subendocardial region of anteroseptal infarcts in humans. Tissue was obtained for this purpose from 11 patients with 12 documented infarctions at the time of autopsy; six patients died within 72 hours of the infarction and five had healed infarcts. Seven of the 11 patients had ventricular arrhythmias. Light microscopic study indicated that intact cells with a normal appearance remained on the subendocardial surface, although the underlying ventricular muscle either was necrotic or was replaced by fibrous tissue. Electron microscopy demonstrated that these intact surviving cells over the surface of the infarct had few randomly oriented myofibrils, abundant glycogen, and other characteristics of Purkinje fibers. These cells could be readily distinguished from normal or infarcted ventricular muscle cells. Purkinje fibers, the most peripheral part of the conduction system, survive in extensive anteroseptal infarcts and may be the site of origin of ventricular arrhythmias.

Hemoglobin uptake by rat hepatocytes and its breakdown within lysosomes

The peroxidatic activity of hemoglobin permitted visualization of its uptake by rat hepatocytes by means of the Graham-Karnovsky 3,3'-diaminobenzidine (DAB) procedure. Lysosomes were visualized by their acid phosphatase, beta-glucuronidase, and glucosaminidase activities. When large doses of rat, cow, or human hemoglobin are intravenously injected, or when hemoglobinemia is induced by injection of distilled water, DAB-positive hemoglobin is engulfed by pinocytosis. Pinocytotic vacuoles become digestive vacuoles ("phagolysosomes") by fusion with lysosomes of the dense body type that have moved from their pericanalicular position. By 16-24 hr after even massive amounts of hemoglobin (400 mg/100 g), the protein is barely demonstrable in hepatocytes. At the lowest doses of injected hemoglobin (15 mg/100 g body weight), DAB-positive vacuoles are demonstrable only in the Kupffer cells.

Ultrastructural and cytochemical observations on B-16 and Harding-Passey mouse melanomas. The origin of premelanosomes and compound melanosomes

The B-16 and Harding-Passey mouse melanomas were studied by light microscopy (tyrosinase, acid phosphatase, aryl sulfatase, thiamine pyrophosphatase and inosine diphosphatase activities) and electron microscopy (morphology and tyrosinase and acid phosphatase activities). Lysosomal enzyme activity is present in individual premelanosomes and melanosomes as well as in compound melanosomes. Acid phosphatase and tyrosinase activities are present in a Golgi-associated system of smooth endoplasmic reticulum (GERL) and small vesicles related to it. The acid phosphatase and tyrosinase activities of premelanosomes, and morphologic appearances, support the hypothesis that the granules arise from GERL. On the basis of the evidence presented, it is suggested that compound melanosomes arise within melanoma cells by autophagy.