| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Center for Transgene Technology and Gene Therapy,* Flanders Interuniversity Institute, Leuven, Belgium; Molecular and Experimental Cardiology, CARIM, Maastricht University, Maastricht, The Netherlands; Oklahoma Medical Research Foundation, Oklahoma City, Oklahoma; Sanofi Recherche, Haemobiology Research Department, Toulouse Cedex, France; and the BHF Blood Pressure Group,¶ Department of Medicine and Therapeutics, Western Infirmary, Glasgow, Scotland, United Kingdom
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
LV hypertrophy is a natural adaptation of the heart to increased pressure loading. This response may, however, progress to maladaptation and congestive heart failure when complicated by cardiomyocyte degeneration, LV fibrosis, and dilatation.2-7 The structural changes in myocardial extracellular matrix (ECM) during LV remodeling constitute a milestone in the progression to heart failure. At the myocyte level, laminin, collagen type-IV, and fibronectin in the basement membrane not only attach the cardiomyocyte to the interstitial ECM but also induce transmembrane signaling through integrin receptors.8 At the tissue level, interstitial collagen type-I and -III fibers interconnect, align, and organize myocytes and muscle fibers, allowing optimal force generation and transmission in the myocardium. When collagen fibrils and struts are lost, myocyte slippage, LV dilation, and progressive contractile dysfunction may occur.9
Remodeling of the myocardial ECM requires proteolysis. Support for the importance of the activity of matrix metalloproteinases (MMP) in the development of chronic heart failure has been demonstrated both in animal models of heart disease and in humans. Numerous papers indeed described the altered expression of interstitial collagenase (MMP-1), stromelysin (MMP-3), 72-kd gelatinase (MMP-2), 92-kd gelatinase (MMP-9), TIMP-1, and TIMP-2, while also neutrophil-elastase (MMP-8), membrane-type-1 MMPs, and TIMP-4 seem substantially up-regulated during cardiac hypertrophy, remodeling, and failure.7,10-32 The functional importance of the MMP system in LV remodeling has been evidenced by genetic and pharmacological studies. Direct genetic evidence for a role in cardiac remodeling has been provided for MMP-1, MMP-2, MMP-9, TIMP-1, and TIMP-3. Chronic myocardial overexpression of MMP-1 in mice produced loss of interstitial collagen associated with compensatory cardiac hypertrophy and cardiac dysfunction at older age,33 while MMP-9 as well as MMP-2 deficiency in mice prevented cardiac rupture, and decreased cardiac dilatation and dysfunction after acute myocardial infarction.34-36 In addition, we recently demonstrated that increased MMP-2 activity in absence of TSP-2 predisposed to cardiac rupture and dilated cardiomyopathy during hypertension.37 Instead, gene deletion of TIMP-1 caused spontaneous LV dilatation38 and amplified the hypertrophic response and adverse LV remodeling after myocardial injury, thereby emphasizing the importance of local endogenous control of cardiac MMP activity by TIMP-1.39,40 Also, loss of TIMP-3 function triggered spontaneous LV dilatation, cardiomyocyte hypertrophy, and contractile dysfunction.41 Broad-spectrum pharmacological inhibition of MMPs significantly attenuated myocardial remodeling associated with chronic volume overload, hypertension, myocardial infarction, or other conditions in animal models,26,42-50 while selective MMP inhibition, eg, sparing MMP-1, attenuated LV remodeling after MI or pacing.51-54
Also, other non-MMP proteinases, including serine elastase, the cysteine protease, caspase-1, the major serine proteases released by mast cells, chymase and tryptase, and the lysosomal cysteine peptidase, cathepsin, have been implicated in cardiac ECM remodeling in response to myocardial infarction or during viral myocarditis.55-62 Particularly relevant is the plasminogen system, with its plasminogen activators (PA), eg, tissue-type PA (t-PA) and urokinase-type PA (u-PA) and its PA-inhibitor (PAI-1), as plasmin activates pro-MMPs in the cardiovascular system.34,63 The implication of this system in myocardial tissue degradation, leading to congestive heart failure, has received little attention thus far.34,64,65 Although u-PA activity in the LV is increased after pressure overload,66,67 a functional implication in LV remodeling has not been demonstrated. A single study provided circumstantial evidence that reduced inactivation of PAI-1 in myeloperoxidase-deficient mice diminished LV remodeling after myocardial infarction,68 while another recent paper showed that increased expression of PAI-1 contributes to myocardial fibrosis after myocardial infarction.69 Thus, the role of t-PA or u-PA in the progression of the initially adaptive to maladaptative cardiac response to acute pressure overload, and the potential of PAI-1 to attenuate LV remodeling, remain to be elucidated. Similarly, while LV remodeling after myocardial infarction has been studied in MMP-9-deficient mice, the role of MMP-9 or TIMP-1 in LV remodeling after acute pressure overload has not been studied at the genetic level. We therefore addressed these outstanding issues in the present study.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Ten-to fourteen-week-old male C57Bl6(75%)/129/Sv(25%) mice lacking u-PA or t-PA, CD1(50%)/129Sv(50%) mice lacking MMP-9 (gift from Dr. S.D. Shapiro, Washington University School of Medicine, St. Louis, MO) and their genetically identical wild-type (WT) mice, weighing 27 to 30 g, were used. Animals were maintained in an open animal facility and experiments were performed according to the guidelines for the care and use of laboratory animals approved by the institutional ethical animal care committee.
Transverse Aortic Banding
Transverse aortic banding (TAB) was performed as previously described.70
Briefly, after anesthetizing by isoflurane inhalation (induction with 3% isoflurane in 100% O2 and maintenance with 1.5% isoflurane), mice were intubated, ventilated, and a sternotomy was performed. The transverse aortic arch was ligated between the innominate and left common carotid arteries with an overlying 27-gauge needle, which after removal of the needle left a reproducible discrete region of stenosis. Sham-operation included all procedures, except constriction of the aorta. Perioperative (24 hours) and 1-week mortalities were each 
20% in all genotypes.
PAI-1 and TIMP-1 Gene Transfer
Two days before constriction of the transverse aorta, 100 µl of 1.3 x 109 plaque-forming units (pfu) of the replication-deficient adenovirus expressing human PAI-1 (AdPAI-1),71 AdTIMP-1,72 or control AdRR573 virus was injected in the tail vein of C57BL6(75%)/129/Sv(25%) WT mice. Five days after virus injection, PAI-1 or TIMP-1 plasma levels were measured in 100 µl blood, sampled from the retro-orbital plexus, using a murine monoclonal antibody-based enzyme-linked immunosorbent assay for PAI-1,71 or using a commercially available ELISA for TIMP-1 (Amersham Biosciences, Amersham, UK).
Morphology and Histology
After the study period, constricted or sham-operated mice were anesthetized and perfused via the abdominal aorta with a 0.9% saline solution until the blood was removed. Lungs, left, and right ventricle were dissected, blotted dry, and weighed. To measure heart weight/body weight ratio, actual left ventricular mass was normalized to body mass just before TAB. Left and right ventricles were then post-fixed in 1% phosphate-buffered paraformaldehyde overnight, dehydrated, and embedded in paraffin. Four-µm thick transverse sections were made for histological analysis or for immunostaining as described.74 Immunostaining was performed using antibodies against thrombomodulin (gift from Dr. R. Jackman, Department of Health Sciences, Boston University, Boston, MA), CD45 (Pharmingen, Becton Dickinson, Aalst, Belgium), ANF (Santa Cruz Biotechnology, Telri Bio, Bouchout, Belgium), and laminin (Sigma, Bornem, Belgium) as previously described.34 Morphometric analysis was performed using a Quantimet Q 600 image analysis system (Leica, Brussels, Belgium). The average cardiomyocyte cross-sectional area was evaluated in the subendocardial layer, the central and the subepicardial layer of the mid-septum, and the mid-free wall of the left ventricle, on laminin-stained sections. Suitable cross sections of myocytes were defined as containing nuclei and having nearly circular capillary profiles. Approximately 300 cells were counted per sample and the average was used for analysis. Thrombomodulin-stained capillaries and CD45-stained leukocytes were counted in different layers through the left ventricle in at least 20 optical fields at x40 magnification. Collagen type-I and -III was stained using Sirius red, and the amount of collagen was quantified as percentage Sirius red stained area per total cardiac area, excluding fibrosis around large cardiac vessels.34 Ultrastructural analysis was done as described.34,74
Biochemical and Molecular Analysis
Levels of angiotensin-converting enzyme (ACE) activity were measured in serum and in left ventricular tissue extracts, using a spectrophotometric assay, with 3-(2-furylacryloyl)-L-phenylalanyl-glycyl-glycine (FAPGG) as substrate (Sigma). For ACE-activity measurement in cardiac tissue, the left ventricle was extracted in 400 µl extraction buffer containing 50 mmol/L HEPES, 150 mmol/L NaCl, pH 7.4, 0.5% Triton-X 100, 25 µmol/L ZnCl2 and 1 mmol/L phenylmethyl sulfonyl fluoride (PMSF), and corrected for total protein level (Biorad). LV levels of latent and active TGF-ß1 were measured with a commercially available ELISA (R&D Systems).34
Quantitative RT-PCR for atrial natriuretic factor (ANF) and 1
pro-collagen type I (Col 1a1) was performed on RNA isolated from left ventricular tissue, as described.34,75
The following forward (F) and reverse (R) oligonucleotides, and probes (P) were used for real-time RT-PCR of ANF: (F: 5' GTG TAC AGT GCG GTG TCC AA 3'. R: 5' TCC CAG TCT GTG TCC CAC TTC 3'; P: 5' TGG CCC TAC CTT GAA ATC CAT CAG ATC TG 3'), and of pro-Coll-1: (F: 5' GCT TTG TGG ATA CGC GGA CT 3'. R: 5' CGT ACT GAT CCC GAT TGC AA 3'; P: 5' CAG TAA CTT CGT GCC TAG CAA CAT GCC AAT 3'). The relative expression levels of these genes were calculated by dividing their signals by the signals obtained for the HPRT gene.
Hemodynamic Measurements
Pressure measurements and M-mode echocardiography in mice were performed as described.34,76 Briefly, a 1.4 French high-fidelity catheter-tip micromanometer (SPR-671; Millar Instruments, Houston, TX) was inserted through the right carotid artery into the left ventricular cavity to measure carotid and LV pressure. For M-mode echocardiography, animals were sedated with intraperitoneal ketamine (30 mg/kg) and diazepam (2.0 mg/kg) and transthoracic echocardiography was performed with a 12-MHz transducer (Hewlett Packard, Brussels, Belgium) on a Sonos 5500 (Hewlett Packard) echocardiograph. Views in the parasternal short-axis were obtained for guided M-mode measurements.
Statistical Analysis
All data are expressed as mean ± SEM. The Mann-Whitney U-test assessed statistical significant difference between groups, using two-sided P values.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Transverse aortic banding (TAB) was performed in wild-type (WT) mice and in mice lacking tissue-type plasminogen activator (t-PA–/–), urokinase-type plasminogen activator (u-PA–/–), or MMP-9 (MMP-9–/–). Compared to sham-operated mice, left ventricular (LV) systolic pressure and right carotid arterial pressure increased by 40% in all genotypes after TAB (Table 1)
. Increased loading resulted in significant LV hypertrophy in WT and t-PA–/– mice, as evidenced by the 35% increase of the LV/body weight ratio and the 40% increase in the LV cardiomyocyte size (Figure 1, a and d
; Table 2
). Hypertrophy occurred primarily during the first 2 weeks and persisted for up to 7 weeks after TAB. Loss of u-PA did not abolish LV hypertrophy, but provided a significant protection against LV hypertrophy for at least 7 weeks, ie, the LV/body weight ratio increased by only 20% and LV cardiomyocytes enlarged by only 22% in u-PA–/– mice (Figure 1, b and e
; Table 2
). LV hypertrophy was intermediate in hypertensive MMP-9–/– mice (about 30% increase in cardiomyocyte size) (Figure 1, c and f
; Table 2
). Measurements of the thickness of the LV posterior and septal wall by M-mode echocardiography confirmed the genotypic differences in cardiac hypertrophy at 2 and 7 weeks after TAB (Table 3)
. Posterior and septal wall thickness increased in all genotypes after aortic banding but, compared to the other genotypes, the increase in wall thickness was the smallest in u-PA–/– mice and intermediate in MMP-9–/– mice (Table 3)
.
|
|
|
|
. At 4 days after TAB, the majority of LV cardiomyocytes in hypertensive WT mice appeared hypertrophied and contained prominent sarcomeres. There were signs of focal cardiomyocyte degeneration, atrophy, and myofibril disorganization, especially in the subendocardium and at the junction between the septum and anterior wall, ie, at sites of maximal mechanical stress (Figure 2, c, e, and f)
. ECM degradation and disorganization with interstitial edema were pronounced in WT mice after TAB (Figure 2, e and f)
. In contrast, in banded u-PA–/– mice, moderate cardiomyocyte hypertrophy but only minimal degenerative changes were detectable (Figure 2d)
.
|
Loss of u-PA Attenuates LV Inflammation and Fibrosis after Pressure Overload
Cardiac remodeling after TAB is associated with infiltration of inflammatory cells.80,81
Very few CD45+ leukocytes (<5 per mm2) were present in the sham-operated LV of WT and u-PA–/– mice (not shown). After banding, numerous leukocytes infiltrated the LV in WT mice, while fewer leukocytes infiltrated the myocardium in u-PA–/– mice (CD45+ leukocytes/mm2 at 4 and 7 days after TAB: 120 ± 20 and 90 ± 8 in WT mice versus 46 ± 15 and 35 ± 7 in u-PA–/– mice; n = 6; P < 0.05; Figure 3, a and b
). Leukocyte infiltration was especially prominent at sites of focal myocyte degeneration and matrix fragmentation in the subendocardium and septal/anterior wall border (Figure 3c)
.
|
. Quantification of Sirius red-stained area revealed that the myocardial collagen content significantly increased (by 50%) in WT, t-PA–/–, and MMP-9–/– mice within 2 weeks after TAB (Table 2)
. Hypertrophic LV from WT mice contained focal accumulation of numerous spindle-shaped cells, with ultrastructural characteristics of collagen-producing myofibroblasts, entrapped in abundant amounts of interstitial collagen and other matrix components (Figure 3, f and g)
. In contrast, the increase in collagen-rich ECM and spindle-shaped myofibroblasts was much smaller in the LV of hypertensive u-PA–/– mice (Figure 2d
; Table 2
). Reduced interstitial fibrosis in u-PA–/– mice was also shown by RT-PCR data for 1 pro-collagen type I (Col 1a1) at 4 days after TAB. Indeed, 1 pro-collagen type I (Col 1a1) transcript levels, while comparable in sham-operated WT and uPA–/– mice (1 pro-collagen type I (Col 1a1)/1000 HPRT mRNA copies: 125 ± 9 in WT mice versus 104 ± 14 in uPA–/– mice; n = 5, P = NS), were up-regulated in u-PA–/– hearts to only 40% of the levels in WT hearts at 4 days after TAB (pro-Coll-1/1000 HPRT mRNA copies: 2340 ± 406 in WT mice versus 940 ± 129 in uPA–/– mice; n = 5, P < 0.05). Attenuated Increase of Cardiac Levels of TGF-ß1 and ACE in u-PA–/– Mice
Additional evidence that the program of cardiac fibrosis was only minimally activated in hypertensive u-PA–/– mice was provided by measuring the myocardial levels of the angiotensin-converting enzyme (ACE) and transforming growth factor-ß1 (TGF-ß1), both implicated in cardiac fibrosis.2,82 At 14 days after TAB, myocardial ACE activity levels were increased 3.6-fold in WT mice (U/ml/mg protein: 3.3 ± 0.5 after sham versus 12 ± 2.2 after TAB; n = 7 to 10; P < 0.05), but only 1.9-fold in u-PA–/– mice (U/ml/mg protein: 2.8 ± 0.6 after sham versus 5.4 ± 0.7 after TAB; n = 10; P < 0.05). Cardiac levels of active TGF-ß1 (pg TGF-ß1/µg protein) were 16 ± 2 in WT mice versus 4 ± 1 in u-PA–/– mice (n = 8; P < 0.05) at 7 days after TAB. The lower myocardial levels of ACE and TGFß-1 in hypertensive u-PA–/– mice may be attributable to the reduced accumulation or activation of cardiac fibroblasts and leukocytes, which produce these factors.2,82,83
Preserved Myocardial Vascularization in u-PA–/– Mice after Pressure Overload
To evaluate the consequences of the increased pressure loading on the coronary vasculature, vessels in WT and u-PA–/– mice were immunostained for the endothelial cell marker thrombomodulin (TM) at 2 weeks after TAB. A comparable capillary density was present in the LV of both genotypes after sham-surgery (capillaries per mm2: 6100 ± 260 in WT versus 6000 ± 310 in u-PA–/– mice; n = 6; P = NS). Capillary density was more extensively reduced in WT than in u-PA–/– mice after TAB (capillaries per mm2: 3600 ± 380 in WT versus 4800 ± 190 in u-PA–/– mice; n = 6; P < 0.05). The decrease in capillary density was strongest in the subendocardial area (capillaries per mm2: 3200 ± 230 in WT versus 4100 ± 210 in u-PA–/– mice; n = 6; P < 0.05). In addition, intercapillary distances were larger in WT than in u-PA–/– mice after banding (9.4 ± 0.5 µm in WT versus 9.6 ± 0.6 µm in u-PA–/– mice after sham-surgery; n = 6; P = NS; 13.4 ± 0.8 µm in WT versus 10.4 ± 0.8 µm in u-PA–/– mice after TAB; n = 6; P < 0.05). The capillary-to-cardiomyocyte ratio, however, remained constant (1.2 ± 0.2 in WT versus 1.1 ± 0.2 in u-PA–/– mice after sham-surgery; n = 6; P = NS; 1.0 ± 0.3 in WT versus 1.2 ± 0.1 in u-PA–/– mice at 14 days after TAB; n = 6; P = NS), indicating that the reduced capillary density was due to the cardiomyocyte hypertrophy and not to capillary loss. Fibrosis around arterioles and larger coronary arteries was comparable in WT and u-PA–/– mice (not shown). These findings suggest that WT cardiomyocytes, because of their increased hypertrophy, are at greater risk than u-PA–/– cardiomyocytes to suffer insufficient oxygen delivery (ischemia).
Loss of u-PA Attenuates LV Dysfunction and Remodeling after Pressure Overload
When stressed by pressure overload, the heart initially improves its function compensatorily, but subsequently, the heart cannot sustain the overload and cardiac function progressively deteriorates, ultimately resulting in pump failure.5,6
In healthy WT mice, the initial improvement of cardiac function occurs rapidly, ie, within the first days, and is evidenced by an increased LV contractility (measured as dP/dtmax) which is capable of maintaining normal levels of fractional shortening.84,85
In the subsequent weeks, the initial gain of LV contractility is lost and LV contractility deteriorates below normal levels, resulting in progressively lower levels of fractional shortening.84,85
Consistent with these previous findings, we observed that LV contractility (dP/dtmax) was preserved at 2 weeks but subsequently impaired at 7 weeks after TAB, and was accompanied with a progressive loss of fractional shortening, as measured by M-mode echocardiography (Tables 1 and 3)
. In contrast, in u-PA–/– mice, LV contractility was improved and fractional shortening levels were maintained at 2 weeks and even at 7 weeks after TAB (Tables 1 and 3)
, thereby resembling the compensatorily improved cardiac response, as it initially occurs in normal WT mice.84,85
Cardiac function was also better preserved in MMP-9–/– mice, though the response was intermediate (Tables 1 and 3)
.
After 7 weeks of high blood pressure, WT mice exhibited obvious signs of LV systolic dysfunction and LV dilatation (Tables 1 and 3)
, with 3 of 10 WT mice even exhibiting signs of severe respiratory distress and pulmonary congestion (lung/body weight: 8.8 ± 1.1 mg/g at 7 weeks after TAB versus 5.3 ± 0.3 mg/g in sham-operated mice; n = 11; P < 0.05). In contrast, at 7 weeks after TAB, LV systolic dysfunction and dilatation were only marginally detectable in u-PA–/– mice (Tables 1 and 3)
without signs of pulmonary edema (lung/body weight: 5.7 ± 0.2 mg/g in u-PA–/– mice after TAB versus 5.4 ± 0.3 mg/g in u-PA–/– mice after sham-operation; n = 10; P = NS). The cardiac response of MMP-9–/– mice to pressure overload was intermediate to that of WT and u-PA–/– mice: LV dilatation was minimal and fractional shortening was preserved at 7 weeks after TAB (Table 3)
, with no signs of pulmonary edema (lung/body weight ratio in mg/g: 5.4 ± 0.3 in MMP-9–/– mice after TAB versus 5.5 ± 0.4 in MMP-9–/– mice after sham-operation; n = 10; P = NS).
Thus, despite the greater cardiomyocyte hypertrophy in WT than u-PA–/– mice, cardiac function was, overall, better preserved in u-PA–/– than in WT mice. In WT mice, the increase in cardiomyocyte size was accompanied by an increase in intercapillary distance (and thus likely an impaired delivery of oxygen to the center of the myocyte) and by hemodynamically undesirable structural changes (eg, myolysis, fibrosis). In contrast, in u-PA–/– mice, the modest cardiomyocyte hypertrophy, together with the preserved oxygen delivery and the minimal myocyte degeneration and interstitial fibrosis seemed to suffice to allow cardiac function to respond to the increased loading conditions. Loss of MMP-9 caused similar but less striking changes as deficiency of u-PA.
PAI-1 or TIMP-1 Gene Transfer Preserve Cardiac Function after Pressure Overload
An adenovirus expressing human PAI-1 (AdPAI-1), human TIMP-1 (AdTIMP-1), or a control AdRR5 adenovirus was injected intravenously 48 hours before TAB; this procedure is known to result in markedly elevated plasma levels of PAI-1 and TIMP-1 within 10 to 14 days.34 AdPAI-1 elevated plasma PAI-1 levels to 48 ± 5 µg/ml and AdTIMP-1 increased plasma TIMP-1 levels to 32 ± 9 µg/ml within 5 days after gene transfer, while human PAI-1 or TIMP-1 were undetectable after control virus transfer.
In general, overexpression of PAI-1 or TIMP-1 yielded a comparable degree of protection against TAB-induced LV dysfunction. Moreover, PAI-1 gene transfer was as efficient as u-PA deficiency, while TIMP-1 gene transfer was more effective than MMP-9 deficiency, likely because TIMP-1 also inhibits other MMPs than MMP-9 alone. Because PAI-1 and TIMP-1 levels were only elevated for 10 to 14 days after adenoviral gene transfer, we limited our analysis to the 2-week time point only. In mice expressing elevated plasma levels of PAI-1 or TIMP-1, TAB induced only a modest degree of LV hypertrophy (see LV/body weight ratio, cardiomyocyte size, and LV wall thickness; Tables 2 and 3
) and myocardial fibrosis (Table 2)
. In contrast to AdRR5-injected mice, where leukocytes massively infiltrated the heart after TAB, significantly fewer leukocytes accumulated in hearts of AdPAI-1- or AdTIMP-1-treated mice (CD45-stained leukocytes per mm2: 29 ± 6 in AdPAI-1; 27 ± 4 in AdTIMP-1 versus 95 ± 10 in AdRR5-treated mice at 7 days after TAB; n = 6; P < 0.05). Moreover, because of the more modest cardiomyocyte hypertrophy, myocardial capillary densities were higher after AdTIMP-1 and AdPAI-1 than after AdRR5 treatment (capillaries per mm2: 4700 ± 430 in AdPAI-1, 4900 ± 220 in AdTIMP-1 versus 3700 ± 350 in AdRR5-treated mice at 14 days after TAB; n = 6; P < 0.05). PAI-1 or TIMP-1 overexpression did not, however, alter the capillary-to-cardiomyocyte ratio (not shown). Importantly, elevated PAI-1 or TIMP-1 plasma levels not only prevented the development of cardiac dilatation, they also allowed the stressed heart to improve its contractility (dP/dtmax) and largely preserved the fractional shortening index (Tables 1 and 3)
. Taken together, gene transfer of PA or MMP proteinase inhibitors permitted the stressed heart to respond in a more beneficial way to the increased pressure loading, thereby mimicking the gene-inactivation studies.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
How does u-PA affect the cardiac adaptation to pressure overload? One likely mechanism is that u-PA itself or via generation of plasmin is involved in the remodeling of the myocardial ECM.64 MMPs are secreted as inactive pro-enzymes and first need to be activated; plasmin is a possible activator of pro-MMPs.34,63 u-PA may indeed affect the remodeling of the interstitial collagen matrix, which plays an important role in maintaining the interaction of cardiomyocytes with each other and their three-dimensional organization in the entire heart.2,20 When this interstitial collagen network is broken down, cardiomyocyte slippage may occur, resulting in LV dilatation. Our ultrastructural analysis revealed that the interstitial matrix in WT mice was broken down after pressure overload, preceding subsequent LV dilatation. In contrast, the interstitial matrix was better preserved in u-PA–/– mice, which were protected against LV remodeling.
Another mechanism whereby u-PA might affect the cardiac response to pressure overload is by facilitating the infiltration of inflammatory cells and collagen-producing cardiac fibroblasts, as both cell types require u-PA to migrate into the damaged myocardium.34 Fewer leukocytes and fibroblasts were present in u-PA–/– mice, underscoring the critical role of this proteinase. We and others previously documented that u-PA may have a dual function; one related to proteolytic matrix breakdown, the other related to matrix production via proteolytic activation of growth factors, such as the fibrogenic TGF-ß1. This dual role of u-PA may explain why, at one side, ECM degradation and, at the other side, collagen deposition both were attenuated in the absence of u-PA.
Deficiency of MMP-9 and gene transfer of TIMP-1 also attenuated LV remodeling and dysfunction after pressure overload, with TIMP-1 gene transfer being more efficient. This may not be surprising, since TIMP-1 inhibits other MMPs in addition to MMP-9. Genetic and pharmacological studies indeed support the involvement of other MMPs in the development of chronic heart failure (see Introduction). All these findings suggest that various MMPs, each possibly having distinct functions, cooperate with each other in the cardiac response to pressure overload. Our results are also consistent with previous studies demonstrating a prominent role of MMP-9 in myocardial ECM remodeling in cardiac rupture34 and LV remodeling.35 The similar phenotypes after TIMP-1 and PAI-1 gene transfer are also not surprising, when considering that u-PA-mediated plasmin activates the inactive pro-MMPs.86
The role of u-PA in the cardiac response to pressure overload has not been previously reported. However, our findings about the role of u-PA are consistent with and extend previous observations on the role of MMPs, though some differences exist. In general, inhibition or genetic deficiency of MMPs attenuates LV maladaptation to conditions such as pressure or volume overload, myocardial infraction, and pacing,26,35,36,42-49,51,52 while the opposite phenotype develops when MMP activities are enhanced.33,38-40 Our findings also indicate that proteinase inhibition promotes a favorable adaptation. But, there are also differences, such as in the hypertrophy of cardiomyocytes after proteinase inhibition, which was attenuated in our study and in others,26 unaffected in response to volume overload47 or during spontaneous hypertension,46 and even increased in a pacing-induced model of cardiac failure in pigs.52 These differences are likely attributable, at least in part, to differences in animal models. For instance, the cardiomyocyte cross-sectional area is increased after TAB in the mouse model, but reduced in the pacing-induced heart failure pig model.52 Proteinase inhibition resulted, however, in both models in a favorable adaptive response. Some of the quantitative differences in phenotypes may also be due to additional differences in the type of LV loading (eg, acute pressure overload, volume overload,47 chronic spontaneous hypertension,46 rapid pacing52 ), the species (eg, mouse, rat, pig), and the type and dose of proteinase inhibitor and its method of delivery (adenoviral gene transfer yields very high proteinase inhibitor levels; TIMP-1 is a natural efficient inhibitor of several MMPs; pharmacological MMP inhibitors differ greatly in their ability to block various MMPs).46,47,52
In conclusion, this study demonstrates that inactivation or inhibition of u-PA attenuates LV hypertrophy, remodeling, and dysfunction after pressure overload. These genetic insights should encourage future analysis as to whether long-term pharmacological inhibition of u-PA may offer any therapeutic opportunities for the treatment of maladaptative LV remodeling, often leading to fatal heart failure.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported in part by the European Union (Biomed BMH4 CT98 3380), by FWO, Belgium grant G0125.00, by an unrestricted Bristol-Myers-Squibb grant, by a Concerted Research Activities, Belgium grant GOA2001/09, by a Belgian Science Policy grant IAP-P5/02 and by a grant of the Netherlands Heart Foundation (NHS 2003To36).
Accepted for publication September 14, 2004.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  H. Li, A. Mittal, P. K. Paul, M. Kumar, D. S. Srivastava, S. C. Tyagi, and A. Kumar Tumor Necrosis Factor-related Weak Inducer of Apoptosis Augments Matrix Metalloproteinase 9 (MMP-9) Production in Skeletal Muscle through the Activation of Nuclear Factor-{kappa}B-inducing Kinase and p38 Mitogen-activated Protein Kinase: A POTENTIAL ROLE OF MMP-9 IN MYOPATHY J. Biol. Chem., February 13, 2009; 284(7): 4439 - 4450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Heymans, E. Hirsch, S. D. Anker, P. Aukrust, J.-L. Balligand, J. W. Cohen-Tervaert, H. Drexler, G. Filippatos, S. B. Felix, L. Gullestad, et al. Inflammation as a therapeutic target in heart failure? A scientific statement from the Translational Research Committee of the Heart Failure Association of the European Society of Cardiology Eur J Heart Fail, February 1, 2009; 11(2): 119 - 129. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Marchant, C. Cheung, E. K.Y. Walker, H. Luo, Z. Luo, J. Zhang, B. Yanagawa, M. Rahmani, B. M. McManus, J. Cox, et al. Response to Letter Regarding Article, "Ablation of Matrix Metalloproteinase-9 Increases Severity of Viral Myocarditis in Mice" Circulation, November 11, 2008; 118(20): e698 - e698. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Biolo, S. Ramamurthy, L. H. Connors, C. J. O'Hara, H. K. Meier-Ewert, P. T. Soo Hoo, D. B. Sawyer, D. S. Seldin, and F. Sam Matrix Metalloproteinases and Their Tissue Inhibitors in Cardiac Amyloidosis: Relationship to Structural, Functional Myocardial Changes and to Light Chain Amyloid Deposition Circ Heart Fail, November 1, 2008; 1(4): 249 - 257. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Tanabe, Y. Sato, T. Suzuki, K. Suzuki, T. Nagao, and T. Yamaguchi Gene Expression Profiling of Human Mesenchymal Stem Cells for Identification of Novel Markers in Early- and Late-Stage Cell Culture J. Biochem., September 1, 2008; 144(3): 399 - 408. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-H. Chang, W.-W. Kuo, R.-J. Chen, M.-C. Lu, F.-J. Tsai, W.-H. Kuo, L.-Y. Chen, W.-J. Wu, C.-Y. Huang, and C.-H. Chu IGF-II/mannose 6-phosphate receptor activation induces metalloproteinase-9 matrix activity and increases plasminogen activator expression in H9c2 cardiomyoblast cells J. Mol. Endocrinol., August 1, 2008; 41(2): 65 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Vinet, P. Rouet-Benzineb, X. Marniquet, N. Pellegrin, L. Mangin, L. Louedec, J.-L. Samuel, and J.-J. Mercadier Chronic doxycycline exposure accelerates left ventricular hypertrophy and progression to heart failure in mice after thoracic aorta constriction Am J Physiol Heart Circ Physiol, July 1, 2008; 295(1): H352 - H360. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Cheung, D. Marchant, E. K.-Y. Walker, Z. Luo, J. Zhang, B. Yanagawa, M. Rahmani, J. Cox, C. Overall, R. M. Senior, et al. Ablation of Matrix Metalloproteinase-9 Increases Severity of Viral Myocarditis in Mice Circulation, March 25, 2008; 117(12): 1574 - 1582. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Van den Bergh, A. Vanderper, P. Vangheluwe, F. Desjardins, I. Nevelsteen, W. Verreth, F. Wuytack, P. Holvoet, W. Flameng, J.-L. Balligand, et al. Dyslipidaemia in type II diabetic mice does not aggravate contractile impairment but increases ventricular stiffness Cardiovasc Res, January 15, 2008; 77(2): 371 - 379. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. G. Spinale Myocardial Matrix Remodeling and the Matrix Metalloproteinases: Influence on Cardiac Form and Function Physiol Rev, October 1, 2007; 87(4): 1285 - 1342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Shimano, Y. Tsuji, Y. Yoshida, Y. Inden, N. Tsuboi, T. Itoh, H. Suzuki, T. Muramatsu, T. Okada, S. Harata, et al. Acute and chronic effects of cardiac resynchronization in patients developing heart failure with long-term pacemaker therapy for acquired complete atrioventricular block Europace, October 1, 2007; 9(10): 869 - 874. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Yamaguchi, J. M. Lopez-Guisa, X. Cai, S. J. Collins, D. M. Okamura, and A. A. Eddy Endogenous urokinase lacks antifibrotic activity during progressive renal injury Am J Physiol Renal Physiol, July 1, 2007; 293(1): F12 - F19. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. K. Graham and A. W. Trafford Spatial disruption and enhanced degradation of collagen with the transition from compensated ventricular hypertrophy to symptomatic congestive heart failure Am J Physiol Heart Circ Physiol, March 1, 2007; 292(3): H1364 - H1372. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Vanhoutte, M. W.M. Schellings, M. Gotte, M. Swinnen, V. Herias, M. K. Wild, D. Vestweber, E. Chorianopoulos, V. Cortes, A. Rigotti, et al. Increased Expression of Syndecan-1 Protects Against Cardiac Dilatation and Dysfunction After Myocardial Infarction Circulation, January 30, 2007; 115(4): 475 - 482. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. W. Cheng, K. Obata, M. Kuzuya, H. Izawa, K. Nakamura, E. Asai, T. Nagasaka, M. Saka, T. Kimata, A. Noda, et al. Elastolytic Cathepsin Induction/Activation System Exists in Myocardium and Is Upregulated in Hypertensive Heart Failure Hypertension, November 1, 2006; 48(5): 979 - 987. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Heymans, M. Pauschinger, A. De Palma, A. Kallwellis-Opara, S. Rutschow, M. Swinnen, D. Vanhoutte, F. Gao, R. Torpai, A. H. Baker, et al. Inhibition of Urokinase-Type Plasminogen Activator or Matrix Metalloproteinases Prevents Cardiac Injury and Dysfunction During Viral Myocarditis Circulation, August 8, 2006; 114(6): 565 - 573. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Stempien-Otero, A. Plawman, J. Meznarich, T. Dyamenahalli, G. Otsuka, and D. A. Dichek Mechanisms of Cardiac Fibrosis Induced by Urokinase Plasminogen Activator J. Biol. Chem., June 2, 2006; 281(22): 15345 - 15351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. N. Panek and M. Bader Matrix Reloaded: The Matrix Metalloproteinase Paradox Hypertension, April 1, 2006; 47(4): 640 - 641. [Full Text] [PDF]
|
|
|
|
|
|
H. Matsusaka, T. Ide, S. Matsushima, M. Ikeuchi, T. Kubota, K. Sunagawa, S. Kinugawa, and H. Tsutsui Targeted Deletion of Matrix Metalloproteinase 2 Ameliorates Myocardial Remodeling in Mice With Chronic Pressure Overload Hypertension, April 1, 2006; 47(4): 711 - 717. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Manso, L. Elsherif, S.-M. Kang, and R. S. Ross Integrins, membrane-type matrix metalloproteinases and ADAMs: Potential implications for cardiac remodeling Cardiovasc Res, February 15, 2006; 69(3): 574 - 584. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Janssens and H. R. Lijnen What has been learned about the cardiovascular effects of matrix metalloproteinases from mouse models? Cardiovasc Res, February 15, 2006; 69(3): 585 - 594. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Rutschow, J. Li, H.-P. Schultheiss, and M. Pauschinger Myocardial proteases and matrix remodeling in inflammatory heart disease Cardiovasc Res, February 15, 2006; 69(3): 646 - 656. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Janicki, G. L. Brower, J. D. Gardner, M. F. Forman, J. A. Stewart Jr., D. B. Murray, and A. L. Chancey Cardiac mast cell regulation of matrix metalloproteinase-related ventricular remodeling in chronic pressure or volume overload Cardiovasc Res, February 15, 2006; 69(3): 657 - 665. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ichihara, A. Noda, K. Nagata, K. Obata, J. Xu, G. Ichihara, S. Oikawa, S. Kawanishi, Y. Yamada, and M. Yokota Pravastatin increases survival and suppresses an increase in myocardial matrix metalloproteinase activity in a rat model of heart failure Cardiovasc Res, February 15, 2006; 69(3): 726 - 735. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Menon, M. Singh, R. S. Ross, J. N. Johnson, and K. Singh {beta}-Adrenergic receptor-stimulated apoptosis in adult cardiac myocytes involves MMP-2-mediated disruption of {beta}1 integrin signaling and mitochondrial pathway Am J Physiol Cell Physiol, January 1, 2006; 290(1): C254 - C261. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. V. Ovechkin, N. Tyagi, W. E. Rodriguez, M. R. Hayden, K. S. Moshal, and S. C. Tyagi Role of matrix metalloproteinase-9 in endothelial apoptosis in chronic heart failure in mice J Appl Physiol, December 1, 2005; 99(6): 2398 - 2405. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ponten, E. Bergsten Folestad, K. Pietras, and U. Eriksson Platelet-Derived Growth Factor D Induces Cardiac Fibrosis and Proliferation of Vascular Smooth Muscle Cells in Heart-Specific Transgenic Mice Circ. Res., November 11, 2005; 97(10): 1036 - 1045. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Heymans, B. Schroen, P. Vermeersch, H. Milting, F. Gao, A. Kassner, H. Gillijns, P. Herijgers, W. Flameng, P. Carmeliet, et al. Increased Cardiac Expression of Tissue Inhibitor of Metalloproteinase-1 and Tissue Inhibitor of Metalloproteinase-2 Is Related to Cardiac Fibrosis and Dysfunction in the Chronic Pressure-Overloaded Human Heart Circulation, August 23, 2005; 112(8): 1136 - 1144. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
