Extracellular Matrix Remodeling in Heart Failure
1997; Lippincott Williams & Wilkins; Volume: 96; Issue: 11 Linguagem: Inglês
10.1161/01.cir.96.11.4065
ISSN1524-4539
Autores Tópico(s)Tissue Engineering and Regenerative Medicine
ResumoHomeCirculationVol. 96, No. 11Extracellular Matrix Remodeling in Heart Failure Free AccessResearch ArticleDownload EPUBAboutView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticleDownload EPUBExtracellular Matrix Remodeling in Heart Failure A Role for De Novo Angiotensin II Generation Karl T. Weber Karl T. WeberKarl T. Weber From the Division of Cardiology, Department of Internal Medicine, University of Missouri Health Sciences Center, Columbia. Originally published2 Dec 1997https://doi.org/10.1161/01.CIR.96.11.4065Circulation. 1997;96:4065–4082Heart failure is a major health problem worldwide. In the United States, it represents the number one hospital discharge diagnosis among elderly persons each year. It appears most commonly in patients with previous MI.1,2 The chronically failing heart of ischemic origin is characterized by iterations in tissue structure, particularly fibrous tissue formation, that appear in infarcted and noninfarcted myocardium of both right and left ventricles.3,4 In other words, fibrosis appears at the site of MI as well as remote from it. Fibrosis remote from the infarct site is considered "the major cause of ventricular remodeling" in ischemic cardiomyopathy.4 Such an adverse accumulation of extracellular matrix initially raises myocardial stiffness; its continued accumulation further increases stiffness and impairs contractile behavior.5–10 Elucidating cellular and molecular mechanisms responsible for accumulation of extracellular matrix is essential to designing cardioprotective and reparative strategies that could prevent or regress fibrosis, respectively, after infarction.11,12ACE inhibition has proved effective in reducing mortal and morbid events, improving symptomatic status, and attenuating the progressive nature of cardiac failure in symptomatic patients with ventricular diastolic and/or systolic dysfunction in whom activation of the circulating RAAS is present.1,13,14 ACE inhibitor–mediated reductions in circulating Ang II and aldosterone no doubt contribute to this salutary response. This would include an attenuation of well-recognized endocrine properties of these hormones, such as altered sodium homeostasis and vascular tonicity, and their adverse influence on matrix structure of atria and ventricles.15–20 Collectively, these adverse responses to RAAS effector hormones contribute to the progressive nature of chronic cardiac failure, which includes recurring bouts of symptomatic failure1,2,21 and reentrant arrhythmias originating in either atria or ventricles.22,23ACEIs have also proved effective in asymptomatic patients with equivalent levels of ventricular systolic dysfunction but in whom chronic RAAS activation is not present.2,24,25 Attention has therefore focused on the importance of Ang II generated de novo within the infarcted heart.26 ACE, a membrane-bound ectoenzyme found on various cells, is central to Ang II generation within an organ.27 Because ACE mRNA expression and activity are each increased in tissue homogenates taken from either the infarcted failing ventricle or the hypertrophied, hypertensive ventricle,28–32 it has been proposed that these responses are a consequence of cardiac myocyte hypertrophy and/or increased systolic wall stress.31,32 Increased ACE activity, however, is most specifically localized to sites of fibrosis, such as ventricular aneurysm in the infarcted heart,29 a finding reinforced by autoradiographic localization of high-density ACE binding at the site of MI.33 Marked ACE binding is also seen at sites of repair remote from the infarct (eg, noninfarcted right ventricle) and sites independent of infarction (eg, fibrosed visceral pericardium),33 where ACE activity (Ang I substrate conversion) is present.30,34 Hence, increased expression of ACE transcript and activity do not appear to be related to hemodynamic factors or to cardiac myocyte hypertrophy. Cells expressing ACE at sites of repair include a phenotypically transformed fibroblast-like cell, the myoFb, because it expresses α-SMA and is contractile.35,36Locally produced Ang II has autocrine and paracrine properties that influence the behavior of constitutive cell populations of the myocardium via Ang II receptor binding. Various paradigms have been suggested, including Ang II–mediated cardiac myocyte hypertrophy37 and apoptosis,38,39 regulation of microvascular blood flow at sites of injury,40 and the process of tissue repair itself.41This report addresses various lines of evidence that implicate Ang II, generated de novo at sites of high collagen turnover, in regulating connective tissue formation. This applies to normal (eg, heart valve leaflets) and pathological sites of tissue repair. On the basis of these collective findings, it is suggested that Ang II, produced de novo at sites of injury in the heart and pericardium, is integral to their repair. Let's set the stage for this Bench to Bedside review with a fictional vignette.Clinical Vignette: Bee Stings and Other ThingsThis warm Tuesday evening in July found internist Mindy Carson at her modest yet comfortable office in central Missouri. Standing by an open window with arms folded, her head resting on the window frame, she pensively gazed at the familiar sight of grazing cattle on neighboring farmland. As was her custom, Mindy reflected on the day's prior events: morning rounds at the hospital, an afternoon of office-based patients. A year had passed since she began her practice; it had proved gratifying and intellectually challenging. Just before closing today, for example, her patient Joseph McKenzie had brought sons Jim and John, ages 10 and 12 years, to the office. John's entire right hand had become swollen, red, and painful after multiple honeybee stings. Jim, more fortunate, had only a single sting with a localized wheal and flare. Antihistamine and topical care for John, comfort and assurance for both. Mindy also examined Joseph's hand. A carpenter by trade, years of hammering and sawing had caused thickened palmar fascia of his right hand to contract, with a consequent retraction of the fourth and fifth digits toward the palm (Dupuytren's contracture). As the McKenzies were leaving, she reminded Joseph that surgery would be needed if his hand was to remain functional and he gainfully employed.Mindy sat at her desk to ponder Joseph's fibrocontractive disorder, her mind struggling to understand its pathophysiological basis. She began drumming the eraser end of her pencil on the desk to an imaginary rhythm. Like a Gene Krupa drum solo from the swing era of jazz, her rendition built in intensity and pace. "Strange," she thought, "a drumhead's stretched membrane of cowhide does not contract. Why would Joseph's palmar fascia?" For that matter, why would scarred rheumatic mitral leaflets retract and become incompetent? There were no contractile proteins in fibrous tissue to mediate such responses. Or were there? She quickly turned to her personal computer for a search of the medical literature in hopes of unraveling the puzzle. As an internist—integrator of basic and clinical sciences—she reminded herself that tissue, defined as a substance of an organ, consists of intercellular material, in this case fibrillar collagen, and cells. "That's it," she exclaimed, "it must be the cells!" Bursting with ideas, her mind racing, she wondered whether such circumstances could also explain Mrs Carver's problem.Mindy had seen Ruth Carver in the ICU earlier in the day. This 67-year-old woman had had an uneventful anterior MI 6 months earlier, without clinical evidence of ventricular dysfunction or arrhythmia. After hospital discharge, she noted the gradual appearance of dyspnea with moderate levels of exertion, particularly evident over the past several months. Last evening, paramedics brought Ruth to the ER because of severe breathlessness. There, Mindy found her tachypneic and orthopneic: blood pressure 180/100 and an irregular heart rate of 140 bpm, with a pulse deficit of 50 bpm. She denied chest pain. Neck veins were not distended, apical impulse not displaced; no S3 gallop or murmur; bibasilar crackles without pleural effusion, hepatomegaly, or pitting ankle edema. The ECG showed atrial fibrillation without acute MI or ischemia. Serum creatine kinase MB fraction likewise did not support MI. Mindy's diagnosis: acute pulmonary edema. She chose electrical cardioversion to restore sinus rhythm and intravenous furosemide to induce a diuresis and clear pulmonary congestion. Thereafter, Mrs Carver's blood pressure returned to its usual normotensive value and she was breathing more comfortably; she was hospitalized for further observation.Mindy was gratified this morning when she noted Ruth's continued recovery. But she was curious. The appearance of atrial fibrillation, with rapid ventricular rate and lost atrial contraction, was inextricably linked to her pulmonary edema. Other patients Mindy followed, without structural heart disease, had tolerated similar episodes of atrial fibrillation without such sequelae. Why had atrial fibrillation precipitated pulmonary edema in Ruth? DD with acute pulmonary venous hypertension seemed most likely. Mindy obtained an echocardiogram from the health center's outreach program later in the day: ejection fraction, 45%; DD as evidenced by an abnormal E/A wave ratio; and absent mitral valve incompetence or ventricular aneurysm. With her clinical impression confirmed, she wondered: What leads to DD after infarction? Mindy pondered whether the hands of the McKenzie family provided clues.Mindy prescribed an ACEI and in subsequent weeks was delighted when Ruth no longer experienced dyspnea on exertion. Her patient's annoying cough would prompt a switch to an AT1 receptor antagonist. A repeat echocardiogram, months later, would demonstrate improved DD and an ejection fraction of 52%.Ventricular Dysfunction and the Clinical Spectrum of Heart FailureThe severity of LVD, which accompanies acute MI, whether expressed in clinical or hemodynamic terms, is an important determinant of short-term survival.42–44 Chronic LVD, seen most frequently in patients with previous MI, is likewise an important prognostic factor.1,2,45–47 A chronic impairment in ventricular systolic or diastolic mechanics of the right and/or left ventricles48–50 will compromise cardiac output reserve to the physiological stress of exercise.51,52 This leads to reduced lactate threshold, increased CO2 production and minute ventilation relative to work performed,53 and associated symptoms of exertional breathlessness and fatigue.54LVD per se does not account for a decompensated clinical state of symptomatic cardiac failure.1,2 This has been clearly underscored by the Studies of Left Ventricular Dysfunction (SOLVD) trial, in which an ejection fraction ≤35% as entrance criterion was satisfied in both symptomatic patients with decompensated failure (treatment arm) and asymptomatic patients with compensated failure (prevention arm).1,2 Decompensated failure is manifested by signs and symptoms of extravascular volume expansion (eg, pleural effusion, pitting edema). It occurs when renal perfusion is significantly impaired and accompanied by marked proximal and distal tubular sodium resorption. This sodium-avid state is mediated via a chronic activation of the circulating RAAS.55–57 Accompanying elevations in circulating Ang II and aldosterone are inappropriate as contrasts to states of sodium deprivation or intravascular volume depletion. Compensated failure, in which patients are asymptomatic, exists when the circulating RAAS has not been activated and as a result renal sodium excretion remains normal.Ventricular diastolic and systolic dysfunction is based on an adverse structural remodeling of the myocardium (see Fig 1). This includes the population of cardiac myocytes, in which, over time, hypertrophy, necrosis, and apoptosis alter myocyte mass, and a progressive accumulation of fibrous tissue. Fibrosis appears in morphologically distinct forms. This includes a reactive form, expressed as a perivascular/interstitial fibrosis, which appears in the absence of myocyte necrosis, and a reparative fibrosis or microscopic scars that replace necrotic myocytes.15 Initially, fibrous tissue will adversely influence tissue stiffness and diastolic mechanics.5,6,9,58–62 This includes recruiting the length-dependent property of cardiac muscle (Frank-Starling mechanism) during volume loading,8,63,64 such as occurs with increased venous return attendant on incremental isotonic exercise. A continued accumulation of matrix further impairs diastolic stiffness and now compromises systolic mechanics, including tissue contractility.6,65 Ultimately, renal perfusion falters. Decompensated failure is associated with increased mRNA expression of matrix proteins (eg, types I and III collagens) and accumulation of corresponding proteins expressed as cardiac fibrosis.66–68Cardiac Remodeling and the Infarcted HeartFibrosis and Ischemic CardiomyopathyIn Robbins' Pathophysiologic Basis of Disease, Cotran et al3 indicate that major histological findings in ischemic cardiomyopathy include "a diffuse myocardial atrophy… of myocytes; a diffuse, mainly perivascular, interstitial fibrosis; patchy (under 1 cm) foci of fibrous tissue; myocytolysis of single cells or clusters of cells; and in some instances large healed scars of previous acute infarctions." Using the explanted failing human heart of ischemic origin, Beltrami et al4 extend these observations, noting the presence of multiple foci of replacement fibrosis in combination with interstitial fibrosis in both right and left ventricular myocardium, and such fibrosis, remote from the infarct, accounts for more than two thirds of fibrous tissue found in the cardiomyopathic heart of ischemic origin, whereas the infarct scar constitutes one third. They conclude that "while the cardiomyopathic process may be initiated by myocyte loss after MI, its evolution appears to be controlled by events occurring in remote noninfarcted myocardium of both ventricles." Factors responsible for the accumulation of collagen in the infarcted and noninfarcted myocardium draw attention to the fibrogenic component of tissue repair.Overview of Tissue RepairHealing is a property common to all vascularized tissues. A reparative process is initiated after cardiac myocyte necrosis. This initially depends on inflammatory cells, such as monocytes and macrophages, that invade the site of injury. Macrophages are activated and thereafter generate peptides integral to repair. Fibroblasts are subsequently attracted to the site of injury, where they convert to myoFbs. Through a complex series of molecular events that includes expression of immediate early response genes (eg, c-fos, c-jun, and egr-1) and activation of multiple second messenger systems, which act synergistically to induce mitosis, these cells proliferate and lay down fibrillar collagen that replaces lost myocytes.69–72 Cells involved in repair are bathed by tissue fluid whose composition regulates their phenotype and behavior. A diverse array of soluble regulatory signals, whose biological properties are expressed via receptor-ligand binding, gain access to tissue fluid of the interstitial space from necrotic myocytes, leukocytes, macrophages, and myoFbs. Monitoring lymphatic73–75 or venous drainage76,77 of the injured heart provides insights to the nature of these signals and their broad-ranging functions that regulate cell migration and differentiation, cell-cell interactions, and gene expression.78Fibrous tissue formation is essential to preserving structural integrity of the infarcted myocardium at the site of myocyte loss. Provided that the degree of parenchymal damage is minor, regulatory signals are confined to the site of necrosis. This resembles the local release of histamine and the subsequent wheal-and-flare response that accompanies a single bee sting. Multiple bee stings evoke a large histamine response widely dispersed within tissue fluid, which then promotes diffuse organ swelling. Such is the case with a large transmural MI, in which signals originating at the site of injury are widely dispersed within the common interstitial space of both ventricles to elicit a fibrogenic response in the noninfarcted portion of the injured ventricle and the noninfarcted ventricle.33,64,79,80 Signals involved in repair after MI, however, are confined to the heart; systemic organs are not involved. When fibrogenic signals gain access to the circulation (see below) and are not neutralized, systemic organs will be involved in an unwanted accumulation of stroma—a wound-healing response gone awry.Collagen Turnover At and Remote From InfarctionCollagen turnover has been studied at the site of infarction and at remote sites after ligation of the rat left coronary artery. At the infarct site, collagen degradation, particularly its neutral salt and acid-soluble fractions,81 exceeds synthesis during the very early phase of repair. MMPs reside in the myocardium in latent form. When activated, MMP-1 (or interstitial collagenase) degrades fibrillar collagen into 1/4-length and 3/4-length fragments; gelatinases (MMP-2 and MMP-9) degrade these smaller fragments. An increase in collagenase activity appears at the infarct site on day 2, peaks by day 7, and declines thereafter, together with increased gelatinase activity.82 An increase in collagenase (MMP-1) mRNA expression appears only at day 7 in the infarcted ventricle, replacing the consumed latent pool. TIMPs neutralize collagenolytic activity. Transcription of TIMP mRNA at the infarct site peaks on day 2 and declines slowly over the course of 14 days. Events related to collagen degradation are not seen remote from the MI. Fibroblast-like cells, not inflammatory or endothelial cells, are responsible for the transcription of MMP-1 and TIMP mRNAs.82A fibrogenic component of healing, including an initial expression of fibronectin mRNA,83 follows early collagen degradation. By Northern blot and in situ hybridization analyses, type III procollagen mRNA at the infarct site is increased by day 2 post-MI, reaching a peak by day 21 and declining thereafter.84 Type I procollagen mRNA increases at day 4 and remains elevated at week 4 (see Fig 2A) and even until day 90 at the site of infarction,84 suggesting that collagen synthesis is an ongoing process in keeping with the persistence of myoFbs at this site.85,86 To a lesser extent than seen at the site of injury, but still evident, is the rise in procollagen mRNA remote from the infarct.84,87 Procollagen I and III mRNAs are increased in the right ventricle and interventricular septum on days 4 and 7, respectively.84 In the septum, closest to the infarct, type I procollagen mRNA remains elevated until day 28 (see Fig 2A) but in the right ventricle only until day 7. Expression of type I collagen mRNA is also increased in the fibrosed visceral pericardium at week 4 post-MI (see Fig 2A). These responses, involving myoFbs at the infarct and remote sites (see below),84,86 are associated with increased expression of TGF-β1 mRNA.87By picrosirius red staining, collagen fibers are morphologically evident at the infarct site by day 7, and an organized assembly of fibers in the form of scar tissue becomes evident by day 14 and continues to accumulate for many weeks (see Fig 2B).33,88 Hydroxyproline concentration at the site of scarring increases progressively from week 1 to week 6, as does collagen crosslinking.89–91 A thinning of infarct scar is evident by week 8. Remote from the infarct, including viable left ventricle, interventricular septum, and right ventricle, fibrillar collagen appears by day 14, continues to accumulate for weeks (see Fig 2B), and is associated with increased pepsin-insoluble collagen.4,62,64,79,80,91MyoFbs and RepairMyoFbs are central to fibrogenesis at sites of repair. Fibroblasts have an extensive clonal heterogeneity, and these phenotypically transformed fibroblast-like cells have considerable diversity.35,92 This includes their synthesis of structural proteins and expression of receptors for Ang II, TGF-β1, and endothelins, which permit their response to these regulatory molecules. In addition, myoFbs express α-SMA and are contractile, and, relevant to this report, they govern fibrogenesis in the heart,86,93,94 pericardium, and systemic organs secondary to diverse forms of injury.35 These α-SMA–positive cells (see Fig 3A) appear at sites of injury within days of cardiac myocyte necrosis33,93–95 and are responsible for increased expression of genes encoding for fibrillar type I/III procollagens and their synthesis.33,96,97 MyoFbs arise from interstitial fibroblasts and/or pericytes, not vascular smooth muscle cells or cardiac myocytes.35,36 MyoFb contraction governs matrix remodeling, including scar thinning.98 The contractile behavior of fibrous tissue is related to myoFbs having α-SMA and their cell-cell (desmosome and gap junction types) and cell-matrix (fibronexus, a transmembrane association between fibronectin fibers and actin microfilaments) connections. Their influence on DD of the infarcted ventricle and abnormal tissue stiffness at remote sites will be addressed later in this report.Signals that determine the appearance of the myoFb phenotype are not entirely certain. TGF-β1 is contributory. Subcutaneous administration of TGF-β1 leads to the appearance of myoFbs within granulation tissue that forms, which is not the case for platelet-derived growth factor or tumor necrosis factor-α.99 Cultured adult skin fibroblasts undergo a phenotype switch and express α-SMA when incubated with TGF-β1.99 The appearance of TGF-β1 at sites of injury is most likely related to necrotic myocytes,100,101 activated macrophages,102 and myoFbs themselves.103A fibrillar fibrin-fibronectin scaffolding forms soon after tissue injury and is the precursor to granulation tissue formation and the attachment of myoFbs via a fibronexus.104 MyoFbs subsequently elaborate type III and then type I collagens, the major fibrillar collagens that constitute fibrous tissue.105–107 At pathological sites of tissue repair in the heart, including the site of MI, in situ hybridization has shown that myoFbs express types I and III collagen transcripts.33,84 MyoFbs elaborate and metabolize various substances that regulate their turnover of collagen and govern fibrous tissue contraction in an autocrine manner (see below).Via apoptosis, or programmed cell death, myoFbs (or a subpopulation of myoFbs) are generally reduced in number at sites of repair involving the heart33,93; with skin injury, they completely disappear.108,109 Apoptosis does not elicit an inflammatory cell response and subsequent fibrosis.110 In the infarcted heart, myoFbs persist at the MI site long after infarct healing has been completed.85 In the kidney injured by experimental glomerulonephritis, persistence of myoFbs is associated with a progressive interstitial fibrosis.103 Irrespective of the location or nature of the inciting stimulus to connective tissue formation in the heart, myoFbs are the dominant cell involved in matrix formation.86Angiotensin II and Tissue Repair in the Heart and Other OrgansThe heart's healing paradigm, while still under investigation, probably does not differ from that of other organs. Regulatory peptides, or cytokines such as TGF-β1, contribute to fibrogenesis. TGF-β1 has numerous actions on the extracellular matrix.111 It also contributes to chemotaxis, phenotype transformation, and proliferation of fibroblasts and scar tissue remodeling.99,112 TGF-β1 mRNA is expressed in infarcted myocardium soon after coronary artery ligation100,101 or catecholamine-induced necrosis.113 In cultured human cardiac fibroblasts, TGF-β1 increases transcription of both type I collagen and TIMP.114 Factors governing expression of TGF-β1 and its receptors in injured heart require investigation. Ang II augments TGF-β1 gene expression via AT1 receptor binding in cultured neonatal or adult rat cardiac fibroblasts and myoFbs,115,116 whereas endogenous elevations in circulating Ang II are associated with upregulation of the TGF-β1 gene in the adult rat heart.117 Macrophages, clustered at sites of tissue injury,118 are a likely source of TGF-β1102 that may be important to the appearance of the myoFb phenotype99 and suppression of further inflammatory cell responses.119 A subsequent elaboration of TGF-β1 by myoFbs is integral to fibrogenesis, whereas persistence of myoFbs elaborating TGF-β1 leads to progressive fibrosis.103 What induces the expression of TGF-β1 by macrophages and myoFbs? Ang II appears to play a central role in the expression of TGF-β1 by activated macrophages and myoFbs in that an AT1Ra, given at the time of tissue injury, prevents these responses.120Localization of ACE and Ang II Receptors and ACE ActivityThe role of Ang II in the fibrogenic phase of tissue repair has recently come under investigation. Requisite factors for the involvement of this peptide in tissue repair include its local production at functionally relevant concentrations and the presence of stereospecific Ang II receptors on cells that regulate collagen turnover.The localization and density of ACE binding in the normal and injured heart have been examined with quantitative in vitro autoradiography and an iodinated derivative of lisinopril (125I-351A).33,86,121–125 In the normal heart, low-density ACE binding is found throughout ventricular myocardium and atria, whereas high-density binding is present at sites of high collagen turnover, including heart valve leaflets and the adventitia of intramyocardial coronary arteries.125–128 High-density ACE binding is likewise found in subcutaneous skin, with its metabolically active fibroblasts, but not skeletal muscle tendon, in which fibrocytes are quiescent.125 Immunolabeling with a monoclonal ACE antibody129 identified cells expressing this ectoenzyme. They include endothelial cells on the surface of each valve leaflet; myoFb-like cells, also called valvular interstitial cells, residing within leaflet matrix, where they are responsible for collagen turnover; and pericytes in the adventitia of intramural vessels. Autoradiography and immunolabeling have further demonstrated ACE binding in cultured intact myoFb-like cells and in their cell membranes.130 RT-PCR with amplification of total RNA has demonstrated the presence of ACE mRNA in these cells. Substrate utilization of membrane-bound ACE in these cells includes Ang I and, in its dual capacity as a kininase II, other chemical mediators of inflammation, such as BK and substance P. Thus, myoFb-like cells of valve leaflets have the potential to regulate local concentrations of Ang II and other mediators of tissue repair, such as TGF-β1, in valve leaflets. The importance of the TGF-β family of peptides in fetal heart development, including endocardial cushion formation and heart valve induction, has recently been reported.131,132 A role for Ang II generated by these cells in contributing to this process should be considered.High-affinity receptors are integral to the biological activity of such ACE-related peptides. The presence of Ang II receptors in heart valves was demonstrated by autoradiography using 125I[Sar,1 Ile8]-Ang II binding.125 Competitive binding using either an AT1 or AT2 receptor antagonist, losartan or PD123177, respectively, provided identification of receptor subtype. Low-density AT1 receptor binding is present throughout the myocardium, whereas high-density binding is present in heart valve leaflets.125 Western immunoblot, as well as binding assay, confirmed these findings in membranes.130 Heart valves and myoFbs membranes likewise contain BK receptors, as seen by 125I[Tyr8]-BK autoradiographic binding125 and binding assay.130Cellular responses to Ang II receptor–ligand binding has been examined in serum-deprived, cultured myoFb-like cells obtained from adult rat heart valve leaflets. By in situ hybridization, these cells express the transcript for type I collagen, and incubation of cultured cells with Ang II in pathophysiological concentrations enhances type I collagen synthesis via AT1 receptor binding.130,133,134 Immunohistochemistry130,135 and electron microscopy136,137 have shown that these cells and pericytes each contain α-SMA microfilaments. These microfilaments confer contractile behavior to these cells, and various substances, including Ang II, promote their contraction.136,137Collectively, these findings implicate a biological role for ACE in the local regulation of Ang II and BK at normal tissue sites where collagen turnover is high. These findings led to additional autoradiographic and morphological studies in the rat in several different models of experimental tissue injury involving the heart, related structures, or systemic organs.High-density autoradiographic ACE binding was found at the site of MI at week 1 and increased progressively over the course of 8 weeks (see Fig 2C) in parallel with morphological evidence of fibrillar collagen accumulation found in serial heart sections.33 ACE activity, as measured by substrate conversion, is increased in aneurysmal tissue of the infarcted LV compared with atrial tissue.29 In addition, a significant correlation was found between ACE activity and the extent of infarction,29 as is also the case for ACE mRNA expression and activity at sites remote from the MI.30 In the rat model of MI, the circulating RAAS is not activated.30,138–141By use of either monoclonal (see Fig 3B) or polyclonal ACE antibodies, ACE-labeled cells at sites of healing were identified.33,84,142 After cardiac myocyte necrosis, they include macrophages, α-SMA–positive myoFbs, and endothelial cells of the neovasculature. High-density ACE binding
Referência(s)