Am J Physiol Lung Cell Mol Physiol 283: L239–L245, 2002; 10.1152/ajplung.00001.2002.
translational physiology Induction of oxidative stress and disintegrin metalloproteinase in human heart end-stage failure MATTHEW J. HUNT,1 GIORGIO M. ARU,2 MELVIN R. HAYDEN,1 CHARLES K. MOORE,3 BRIAN D. HOIT,4 AND SURESH C. TYAGI1 Departments of 1Physiology and Biophysics, 2Surgery, and 3Medicine, University of Mississippi Medical Center, Jackson, Mississippi 39216; and 4Case Western Reserve University and University Hospitals and Clinics, Cleveland, Ohio 44106 Received 4 January 2002; accepted in final form 26 March 2002
Hunt, Matthew J., Giorgio M. Aru, Melvin R. Hayden, Charles K. Moore, Brian D. Hoit, and Suresh C. Tyagi. Induction of oxidative stress and disintegrin metalloproteinase in human heart end-stage failure. Am J Physiol Lung Cell Mol Physiol 283: L239–L245, 2002; 10.1152/ajplung. 00001.2002.—Collagen degradation is required for the creation of new integrin binding sites necessary for cell survival. However, a complete separation between the matrix and the cell leads to apoptosis, dilatation, and failure. Previous studies have demonstrated increased metalloproteinase activity in the failing myocardium. To test the hypothesis that disintegrin metalloproteinase (DMP) is induced in human heart end-stage failure, left ventricle tissue from ischemic cardiomyopathic (ICM, n ⫽ 10) and dilated cardiomyopathic (DCM, n ⫽ 10) human hearts were obtained at the time of orthotopic cardiac transplant. Normal (n ⫽ 5) tissue specimens were obtained from unused hearts. The levels of reduced oxygen species (ROS) were 12 ⫾ 2, 25 ⫾ 3, and 16 ⫾ 2 nmol (means ⫾ SE, P ⬍ 0.005) in normal, ICM, and DCM, respectively, by spectrofluorometry. The percent levels of endothelial cells were 100 ⫾ 15, 35 ⫾ 19, and 55 ⫾ 11 in normal, ICM, and DCM, respectively, by CD31 labeling. The levels of nitrotyrosine by Western analysis were significantly increased, and endothelial nitric oxide (NO) by the Griess method was decreased in ICM and DCM compared with normal tissue. The synthesis and degradation of 1-integrin and connexin 43 were significantly increased in ICM and DCM compared with normal hearts by Western analysis. Levels of DMP were increased, and levels of cardiac inhibitor of metalloproteinase (CIMP) were decreased. Aggrecanase activity of DMP was significantly increased in ICM and DCM hearts compared with normal. These results suggest that the occurrence of cardiomyopathy is significantly confounded by the increase in ROS, nitrotyrosine, and DMP activity. This increase is associated with decreased NO, endothelial cell density, and CIMP. In vitro, treatment of CIMP abrogated the DMP activity. The treatment with CIMP may
prevent degradation of integrin and connexin and ameliorate heart failure.
Address for reprint requests and other correspondence: S. C. Tyagi, Dept. of Physiology and Biophysics, Univ. of Mississippi Medical Center, 2500 N. State St., Jackson, MS 39216-4505 (E-mail:
[email protected]).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
http://www.ajplung.org
nitric oxide; nitrotyrosine; capillary; endocardial; endothelial
CREATION OF INTEGRIN BINDING sites by collagen degradation in remodeling cardiac tissue is required for survival (16). However, complete loss of integrin signaling due to detachment from extracellular matrix (ECM) leads to myocyte apoptosis (12), cardiac dilatation, and the syndrome of heart failure. The mechanism of integrin-matrix disconnection in heart failure is unclear. We hypothesized that oxidative stress is a critical component of this process. Reduced oxygen species (ROS), both dependent and independent of inflammatory (and/ or) mitochondrial NADH/NAD oxidase (2), mask the activity of superoxide dismutase and catalase (22). This instigates a decrease in endothelial nitric oxide (NO) availability and activates latent resident myocardial metalloproteinase (32) as well as increases the levels of cytokines, growth factors, and neurohormones (8). A vicious cycle of oxidative stress ensues in which neurohormones, such as angiotensin II (30), further increase oxidative stress by lowering the levels of bradykinin and prostaglandins. These latter two molecules mediate antioxidation by increasing NO production (33). In parallel, angiotensin II induces vascular NADH/NAD oxidase (35). Sixteen percent of the myocardium is composed of capillaries, including the lumen and endothelium (11). The capillary cell density is negatively correlated with the muscle/fiber length in hypoxic (oxidative) hypertrophic myocardium (21). During protracted cycles of ischemia-reperfusion (17), volume overload states (19), and chronic heart failure, ROS are generated (5). En-
1040-0605/02 $5.00 Copyright © 2002 the American Physiological Society
L239
L240
CARDIAC INHIBITOR OF METALLOPROTEINASE IN HEART FAILURE
dothelial NO is the primary antioxidant to balance the ROS, and reduced endothelial cell density leads to a decrease in endothelial NO concentration (27). NO, in conjunction with thiols and oxygen radicals, generates nitrotyrosine and nitroarginine via the formation of peroxynitrite (26). By competing for the substrate of endothelial nitric oxide synthase (eNOS) in a feedback mechanism, nitroarginine blocks further NO generation. Therefore, the ability of oxidative stress to play a role in nitration of proteins and ligands is fundamentally important as a potential mechanism that modulates protein and ligand function. The disintegrin metalloproteinase (DMP) is involved in cardiac development (14). It is unclear whether an increase in oxidative stress causes a decrease in endocardial endothelium and an increase in DMP. Therefore, the purpose of this study was to determine the role of oxidative stress, endothelial NO, and DMP in human heart failure. MATERIALS AND METHODS
Source of tissue. Human heart tissue was obtained at the time of orthotopic cardiac transplantation in the operating room at the University of Mississippi Medical Center. Because NO generation is greater in endocardium than in midmyocardium and the role of endocardial endothelium NO in the beating heart has been determined (18), endocardial tissue from failing hearts was examined in 10 patients with documented coronary artery disease and myocardial infarction (ischemic cardiomyopathy, ICM) and in 10 patients with idiopathic dilated cardiomyopathy (DCM). All patients were in either New York Heart Association functional class III or IV. Five normal control samples were obtained from unused hearts. The average age of the infarcts was 4 ⫾ 1 yr. Tissue specimens from ICM, DCM, and normal hearts were collected and processed within 1–2 h (29). An Institutional Review Board waiver was obtained before collection of the tissue from explanted hearts. In situ labeling. The standard deparaffinization protocol was used. Endothelial cells were labeled with mouse antiCD31/platelet/endothelial cell adhesion molecule-1 FITC (1: 200 dilution, Sigma Chemical) and identified by green fluorescence. To avoid tissue variability due to necrosis/ apoptosis/fibrosis, the images were captured from each heart and stored in digital files. The 10–15 grids/heart were analyzed. The number of cells was referenced with normal tissue. Nitrotyrosine was labeled with monoclonal mouse anti-nitrotyrosine antibody (Upstate Biotechnology) and was detected with secondary antibody conjugated with FITC. Levels of ROS and NO2/NO3. Oxidative stress was assessed by measuring left ventricle (LV) tissue ROS generated by incubating the tissue extract in 10 mM Tris-Cl (pH 7.4) with 2⬘,7⬘-dichlorofluorescein (DCFH). DCFH acquires fluorescence properties on reaction with ROS and induces oxidation of DCFH yielding the fluorescent product dichlorofluorescein. Transient oxyradical (O2 䡠), stable ROS, superoxide, O2⫺, and H2O2 (2OH⫺) are thereby produced. The product is detected by an emission at 530 nm when excited at 485 nm as described (36). The concentration of ROS was measured using oxidized DCFH as standard. To determine the levels of NO, total NO2/ NO3 was measured by the Griess method, using a protocol kit (cat. no. #22116) from Oxis Research (Portland, OR). In this protocol, all available NO is converted to NO3 and measured by colorimetry; cadmium-based reduction of nitrate to nitrite is performed, and total nitrite is determined. AJP-Lung Cell Mol Physiol • VOL
Preparation of cardiac tissue homogenate. The cardiac tissue extracts from respective hearts were prepared as described (32). The total protein in the samples was estimated using a Bio-Rad dye-binding assay. Nitrotyrosine Western blot. Normally, NO functions as an antioxidant buffer against increased ROS. However, during oxidative stress, the availability of NO is decreased. This leads to a substantial increase in oxidative byproducts such as nitrotyrosine. Increased nitrotyrosine is associated with impaired cardiovascular function. To measure nitrotyrosine, Laemmli SDS-PAGE was prepared under reducing condition. After electrophoresis, gels were transferred onto nitrocellulose paper. The nonspecific sites were blocked by 5% fat-free milk. Nitrotyrosine was determined using anti-nitrotyrosine antibody (1:200 dilution; Upstate Biotechnology). The secondary alkaline phosphatase conjugate was used as the detection system. To establish the specificity of nitrotyrosine, anti-nitrotyrosine-agarose conjugate (Upstate Biotechnology) was used to immunoprecipitate the total nitrotyrosine content in the sample before loading onto gel/blots. Cardiac inhibitor of metalloproteinase, 1-integrin, and connexin 43. Because tissue inhibitor of metalloproteinase-4 (TIMP-4) is primarily expressed in the heart (9), the levels of cardiac inhibitor of metalloproteinase (CIMP) were measured using anti-TIMP-4 antibody (1:200 dilution; Chemicon). For 1-intregrin, anti-1-integrin (Chemicon) was used. For connexin 43, anti-connexin 43 antibody (Sigma) was used. All experiments were carried out by loading identical amounts of total protein onto each lane. The bands in Western blots were scanned by a Bio-Rad GS-700 densitometer. The relative band intensity is reported. Aggrecanase activity of DMP. Although specific DMP can be identified by immunoblotting, it does not demonstrate precisely whether the DMP is active or latent. Aggrecan is a substrate for disintegrin and metalloproteinase (28) and has been used to measure the activity of DMP. Ten micrograms of aggrecan was incubated with 10 g of total protein from LV tissue extract in PBS containing 0.001% Triton X-100. After 2 h of incubation at 37°C, the incubates were separated on 10% SDS-PAGE. In controls, LV extract was loaded under identical conditions. Because other TIMPs can inhibit DMP (13) to determine whether CIMP can inhibit aggrecanase activity of cardiac DMP, extracts were preincubated with CIMP (10 g, purified in our laboratory from human heart). The incubate was immunoprecipitated with anti-TIMP-4 (Chemicon) followed by conjugation with agarose beads before loading onto the gels. To specifically confirm DMP activity, extracts were incubated with anti-DMP (Chemicon) and secondary conjugated with agarose beads before loading onto the gels. To inhibit metalloproteinase activity of DMP, the extracts were preincubated with 1 mM EDTA before loading onto the gels. To quantitate the activity of DMP, the aggrecan degrada-
Table 1. Clinical data EF on patients with DCM, ICM, and normal controls
Normal ICM DCM
Age, yr
Sex
LV volume, ml
%EF
27 ⫾ 8 50 ⫾ 6 52 ⫾ 7
M/F M/F M/F
81 ⫾ 12 178 ⫾ 15 169 ⫾ 17
N/A 33 ⫾ 4 28 ⫾ 5
Values are means ⫾ SE. Dilated cardiomyopathy (DCM) was documented by echocardiography and by atherosclerosis-free vessels. Ischemic cardiomyopathy (ICM) was documented by echocardiography, atherosclerotic vessels, and by anterolateral, inferior, inferolateral, and inferoposterior infarction. EF, ejection fraction; LV, left ventricle; M, male; F, female; N/A, not applicable. 283 • AUGUST 2002 •
www.ajplung.org
CARDIAC INHIBITOR OF METALLOPROTEINASE IN HEART FAILURE
Fig. 1. Left ventricle (LV) tissue levels of reduced oxygen species (ROS) and nitrate/nitrite in normal, ischemic cardiomyopathy (ICM), and dilated cardiomyopathy (DCM). Freshly isolated human explanted LV tissue was homogenized in 10 mM Tris-Cl (pH 7.4). The levels of ROS were measured by oxidation of dichlorofluorescein by LV homogenates and by spectrofluorometer. The levels of NO2 /NO3 were measured by the Griess method. * P ⬍ 0.005 and ** P ⬍ 0.01, compared with normal.
tion fragments on PAGE were scanned by densitometer, and means ⫾ SE are reported. Statistical analysis. To minimize the variability of patient status within the category, the data are reported as means ⫾ SE. To determine the significance of data, samples from ICM and DCM were compared with normal specimens using unpaired Student’s t-test. P ⬍ 0.05 was considered significant. RESULTS
Clinical data. The ejection fraction of DCM and ICM hearts was ⬍35%. The LV volume was twofold higher in DCM and ICM hearts compared with control hearts. The results suggest there was LV volume overload in DCM and ICM hearts (Table 1). Levels of ROS and NO2/NO3. LV levels of ROS were significantly higher in both ICM and DCM compared with controls, but the increase was greater in ICM
L241
compared with DCM hearts. Similarly, the levels of NO were reduced in both ICM and DCM, but the magnitude of change was smaller in DCM than ICM (Fig. 1). Endocardial endothelial cell density. Positive labeling for CD31 was observed in endocardium of human heart, and labeling was visualized in both vessels and capillaries. The number of CD31-positive cells was significantly decreased in ICM and DCM compared with normal hearts (Fig. 2), suggesting that in ICM and DCM hearts, the capillary endothelial cell density was decreased. Levels of nitrotyrosine. The levels of nitrotyrosine were significantly increased in ICM and DCM hearts. The in situ labeling of nitrotyrosine was associated with the labeling of endothelial cells (Fig. 2 and Fig. 3, A and B). There was a ⬃1.5-fold increase in the nitrotyrosine in ICM and DCM compared with normal hearts (Fig. 3, C and D). These results suggest an increase in protein nitration in failing human heart. Levels of DMP and CIMP. The representative analysis of serial Western blots of DMP and CIMP (Fig. 4A) revealed a negative correlation between the levels of DMP and CIMP. The ratio between DMP and CIMP was robustly increased in ICM heart. The ratio was also increased in DCM heart (Fig. 5B). The results demonstrate an increase in the levels of DMP and a decrease in the levels of CIMP in failing human myocardium. Levels of 1-integrin and connexin 43. To determine whether integrin and cell-cell connections are disrupted in failing human myocardium, the levels of 1-integrin and connexin 43 were measured by Western blot analysis. Although total synthesis of integrin was increased in ICM and DCM hearts compared with normal, the disruption of integrin was also increased, as suggested by the appearance of fragments of 1 ⬍100 kDa (Fig. 5). The synthesis of connexin was also increased in ICM and DCM hearts compared with
Fig. 2. In situ CD31/platelet/endothelial cell adhesion molecule-1 labeling of the LV tissue. A: endothelial cell density was determined by measuring CD31-FITC labeling. The bluegreen fluorescence was measured to determine the presence of endothelial (Endo) cells at ⫻40 magnification. B: number of Endo cells were counted in randomly selected 10–15 grids/slide. Cell counts in DCM and ICM were normalized with control (Normal) tissue. Means ⫾ SE of % of Endo cells are reported. * P ⬍ 0.001 and ** P ⬍ 0.005, compared with normal.
AJP-Lung Cell Mol Physiol • VOL
283 • AUGUST 2002 •
www.ajplung.org
L242
CARDIAC INHIBITOR OF METALLOPROTEINASE IN HEART FAILURE
Fig. 3. Levels of nitrotyrosine. A: normal. B: ICM. Representative in situ labeling of nitrotyrosine. Freshly isolated LV tissue was fixed in paraffin. The deparaffinized tissue was incubated with anti-nitrotyrosine antibody. The secondary FITC-labeled antibody was used as the detection system. The dark yellowish dots represent labeling due to nitrotyrosine at ⫻40 magnification. C: quantitative Western blot analysis of nitrotyrosine. LV tissue homogenates were prepared from normal (N), ICM, and DCM hearts. Identical amounts of total protein, 25 g/lane, were loaded onto a 10% SDS-PAGE. Lane 1, normal; lanes 2–3, ICM; lanes 4–5, DCM. The protein was transferred to nitrocellulose membrane. The membrane was blotted with anti-nitrotyrosine, and blots were detected by alkaline phosphatase-conjugated secondary antibody. D: blots were scanned. The band intensity, means ⫾ SE, is reported. * P ⬍ 0.002 and ** P ⬍ 0.01, compared with normal.
normal tissue. The degradation of connexin was also increased, as suggested by the appearance of fragments of connexin ⬍45 kDa (Fig. 6). These results suggest induction and degradation of integrin and connexin in human heart end-stage failure. Aggrecanase activity of DMP. To determine whether DMP is activated, aggrecan degradation activity of DMP was measured. The fragments of aggrecan were increased in LV extracts incubated with aggrecan from DCM and ICM compared with controls. Aggrecan migrates as a single band at 100 kDa. The incubation with extract produces bands ⬍60 kDa. These bands were not present in the 5⫻ concentrated LV homogenates (Fig. 7A). The incubation with CIMP, anti-DMP antibody, and EDTA abolished the aggrecanase activity in the extracts (Fig. 7B). These results suggest that
DMP is present in human heart and is activated in failing myocardium. DISCUSSION
The results suggest that end-stage heart failure is associated with a robust increase in ROS and activation of DMP. This phenotype is what would be expected if 1) collagen degradation normally supplies integrin ligands during cardiac muscle remodeling, and 2) complete collagen-integrin disconnect ensues and contributes to heart failure. The administration of CIMP may ameliorate ECM-cell detachment and heart failure. In vivo inhibition of NO production by N G-nitro-Larginine methyl ester increases matrix metalloproteinase (MMP) activity (20) and instigates left ventricle hypertrophy (15). Inactivation of MMP by NO may
Fig. 4. Representative serial Western blots for disintegrin metalloproteinase (DMP) and cardiac inhibitor of metalloproteinase (CIMP). A: identical amounts of total protein were loaded onto each lane. Lanes 1–2, ICM; lanes 3–4, DCM; lane 5, normal. The serial blots were developed using anti-DMP and antiCIMP antibodies, respectively. B: accumulative densitometric scanned data of DMP/ CIMP, means ⫾ SE, are reported. * P ⬍ 0.001 and ** P ⬍ 0.002, compared with normal.
AJP-Lung Cell Mol Physiol • VOL
283 • AUGUST 2002 •
www.ajplung.org
CARDIAC INHIBITOR OF METALLOPROTEINASE IN HEART FAILURE
L243
Fig. 5. Quantitative Western blot analysis of 1-integrin. LV tissue homogenates were prepared from normal, ICM, and DCM hearts. A: identical amounts of total protein, 25 g/lane, were loaded onto a 10% SDS-PAGE. Lanes 1–2, DCM; lanes 3–4, ICM; lane 5, normal. The protein was transferred to nitrocellulose membrane. The membrane was blotted by anti-1-integrin, and blots were detected by alkaline phosphatase-conjugated secondary antibody. Both synthesis and degradation of integrin are increased. The specificity of degradative products was confirmed by immunoprecipitating the integrin and its degradative products. B: accumulative scanned data, means ⫾ SE, is reported. * P ⬍ 0.001 and ** P ⬍ 0.001, compared with normal.
result from 1) inhibition of redox-sensitive MMP gene activation (31), 2) direct induction of CIMP (31), or 3) blockade of the metal ion active site in the latent form of MMP. The neutralization of NO by oxyradical may activate MMP (32) and generate nitrotyrosine, suggesting a rate of MMP activation similar to the rate of NO oxidation (Fig. 8). Despite the induction of eNOS and inducible nitric oxide synthase (iNOS) in the failing myocardium (10), the bioavailability of NO is reduced (Fig. 1), nitrotyrosine is generated (Fig. 3), and MMPs are activated (29). In vitro studies have demonstrated that cardiac myocytes also synthesize iNOS (3, 25). Therefore, it may be difficult to differentiate in vivo NO from endothelium and/or myocyte. However, studies have demonstrated the role of impaired endothelium in the reduction of cardiac relaxation (7) as well as suggested that the endothelial NO attenuates the myocyte contractile function (4). Therefore, it is most likely that in vivo endothelial NO modulates myocyte contractile function. Moreover, unlike large epicardial arteries, the capillary endothelium is without a smooth muscle layer, and, therefore, endothelial NO may diffuse faster across the interstitium to inter-
act with ECM components than the NO from large vessels. Also, it is likely that the flow (i.e., volume)mediated oxidative stress may impair endothelial NO before myocyte NO. There is an inverse relationship between muscle/ fiber length and capillary cell density (21). Our results suggest a decrease in capillary endocardial endothelial cell density (Fig. 2) associated with an increase in ROS (Fig. 1). The generation of nitrotyrosine (Fig. 3) exceeds the bioavailability of NO (Fig. 1). Previous studies from our laboratory have demonstrated increased MMP activity in ICM and DCM hearts (29). The increase in ROS and decrease in NO are associated with activation of DMP (Figs. 4 and 7). The synthesis and degradation of 1-integrin are increased (Fig. 5). ECM is composed of integrin and cell-cell junction proteins. Others have suggested that during hypertrophy, more cell junctions are formed (34). This leads to increased communications as compensatory response of cells in pressure and volume overload hypertrophy. We demonstrate that both synthesis and degradation of connexin 43 are induced in heart failure (Fig. 6), suggesting upregulation of gap junction remodeling.
Fig. 6. Quantitative Western blot analysis of connexin 43. LV tissue homogenates were prepared from normal, ICM, and DCM hearts. A: identical amounts of total protein, 25 g/lane, were loaded onto a 10% SDSPAGE. Lanes 1–2, DCM; lanes 3–4, ICM; lane 5, normal. The protein was transferred to nitrocellulose membrane. The membrane was blotted by anti-connexin 43, and blots were detected by alkaline phosphatase-conjugated secondary antibody. Both synthesis and degradation of connexin were increased. B: blots were scanned. The band intensity, means ⫾ SE, is reported. * P ⬍ 0.001 and ** P ⬍ 0.003, compared with normal.
AJP-Lung Cell Mol Physiol • VOL
283 • AUGUST 2002 •
www.ajplung.org
L244
CARDIAC INHIBITOR OF METALLOPROTEINASE IN HEART FAILURE
Fig. 7. A: qualitative analysis of aggrecanase activity of DMP. Identical amounts (10 g) of tissue homogenates were incubated with 10 g of aggrecan. The proteins were analyzed on 10% SDS-PAGE. Lane 1, aggrecan alone; lanes 2–3, aggrecan was incubated with tissue homogenates from ICM extracts; lanes 4–5, aggrecan was incubated with tissue homogenates from DCM extracts; lane 6, aggrecan was incubated with tissue homogenates from normal hearts. Lanes 7–10 are 5⫻ protein SDS-PAGE analysis from respective tissue extracts before incubation with aggrecan. Arrowheads, significant new bands in homogenates incubated with aggrecan compared with 5⫻ protein extracts without aggrecan. Aggrecan and extracts were preincubated at 37°C for 2 h before loading onto gel. B: extracts from normal, ICM (I), DCM (D), ICM extract preincubated with CIMP plus antiCIMP (I⫹C), ICM extract preincubated with anti-DMP (I⫹A), and ICM extract preincubated with EDTA (I⫹E) before loading onto gels. Scanned data of DMP aggrecanase activity (AU, arbitrary unit). The bands marked by arrowhead in A were scanned. The band intensity, means ⫾ SE, is reported. *P ⬍ 0.001 and **P ⬍ 0.002, compared with normal.
Perivascular, microvascular interstitial endocardial fibrosis, and dilatation are primary entities manifested in the chronic failing myocardium. It is a paradox that fibrosis and dilatation track together. This can be explained by the fact that during increases in preload or afterload and especially in the absence of NO, the latent resident myocardial metalloproteinases are activated (Fig. 8) in an attempt to dilate the heart and to reduce the wall stress. Because the metalloproteinase degrades ultrastructural collagen and elastin (1, 24) more rapidly than oxidized collagen, and because the turnover of ultrastructural collagen and elastin is remarkably lower than oxidized stiffer collagen (23), the ultrastructural collagen and elastin are replaced by oxidatively modified stiffer collagen. Consequently, the cardiac wall stress is increased. In addition, the acti-
vation of DMP leads to complete ECM-cell separation and may instigate apoptosis. It is known that the isolation of adult cardiomyocytes by disruption of surrounding ECM by collagenase instigates cell swelling (6). Added, these myocytes do not survive after 2–3 h. It is a paradox that in many remodeling tissues, such as heart, collagen degradation, to provide new integrin-binding sites, is required for survival (16). However, complete disconnection from ECM leads to apoptosis (12). This study contributes to our understanding of the importance of oxidative stress in cardiac remodeling, structure, and function, by defining their link to NO metabolism. Limitation. The hearts from DCM were homogeneous. However, the ICM hearts were heterogeneous. To be consistent, a part of endomyocardium (i.e., inner
Fig. 8. A hypothesis is that increased oxidized-matrix accumulation and endocardial endothelial dysfunction are due to increased levels of metalloproteinase (MP) and collagenolysis. These levels are associated with decreased levels of endocardial endothelial nitric oxide (NO) and CIMP in response to increased levels of ROS during protracted cycles of ischemia-reperfusion and congestive heart failure. A: complex of CIMP, NO, and MP exists in a latent form in tissue. O2⫺ in presence of reactive thiol (HS-R) and Cu2⫹ generates ONOO⫺, produces NO2-tyrosine and NO2-arginine, and activates MP. This leads to oxidized CIMP (Oxy-CIMP). B: NADH/NAD oxidase generates O2⫺ and decreases superoxide dismutase (SOD) and catalase. NO2-arginine inhibits endothelial NO synthase (eNOS) by competing with L-arginine in feedback mechanism. ROS and ONOO⫺ oxidize tetrahydrobiopterin (BH4; cofactor of eNOS). Exogenous CIMP blocks MP activity and ROS generation. ECM, extracellular matrix. AJP-Lung Cell Mol Physiol • VOL
283 • AUGUST 2002 •
www.ajplung.org
CARDIAC INHIBITOR OF METALLOPROTEINASE IN HEART FAILURE
wall) was dissected from DCM hearts. The infarcted, thinned, and dilated regions of endomyocardium were taken from ICM hearts. Although antibodies are made for whole molecules, they also cross-react with degraded fragments. This may suggest that the degradation does not alter the antigen epitope. Moreover, the specificity of anti-nitrotyrosine, integrin, and connexin 43 was established by immunoprecipitating the antigen before loading onto the gel. This eliminated the specific bands related to the respective antigens. The authors greatly appreciate the generous help of Dr. Indria Rao and Yolanda Smith, Department of Pathology, University of Mississippi Medical Center, for in situ labeling and histological staining. This work was supported in part by National Institutes of Health Grants GM-48595 and HL-51971, American Heart Association-Mississippi Affiliate, and Kidney Care Foundation of Mississippi. REFERENCES 1. Aimes RT and Quigley JP. MMP-2 is an interstitial collagenase. J Biol Chem 270: 5872–5876, 1995. 2. Babior BM. NADPH oxidase: an update. Blood 93: 1464–1476, 1999. 3. Balligand JL, Ungureanu-Longrois D, Simmons W, Pimental D, Malinski TA, Kapturczak M, and Taha Z. Cytokineinducible iNOS expression in cardiac myocytes. J Biol Chem 269: 27580–27588, 1994. 4. Brady A, Warren J, Poole-Wilson P, Williams T, and Harding S. Nitric oxide attenuates cardiac myocyte contraction. Am J Physiol Heart Circ Physiol 265: H176–H182, 1993. 5. Diaz-Velez CR, Garcia-Castineiras S, Mendoza-Ramos E, and Hernandez-Lopez E. Increased malondialdehyde in peripheral blood of patients with congestive heart failure. Am Heart J 131: 146–152, 1996. 6. Flegler-Balon C and Behrendt H. Effects of calcium deficiency, collagenase, and mechanical dispersion on the ultrastructure of cardiac myocytes of adult rats. Eur J Cell Biol 27: 262–269, 1992. 7. Gattuso A, Mazza R, Pellegrino D, and Tota B. Endocardial endothelium (EE) mediates luminal acetylcholine-nitric oxide signaling in isolated frog heart. Am J Physiol Heart Circ Physiol 276: H633–H641, 1999. 8. Givertz MM and Colucci WS. New targets for heart-failure therapy: endothelin, inflammatory cytokines, and oxidative stress. Lancet 352: SI34-SI38, 1998. 9. Greene J, Wang M, Liu TE, Raymond LA, Rosen C, and Shi YE. Molecular cloning and characterization of human TIMP-4. J Biol Chem 271: 30375–30380, 1996. 10. Haywood GA and Tsao PS. Expression of inducible NOS in human heart failure. Circulation 93: 1087–1094, 1996. 11. Hoppeler H and Kayar SR. Capillary and oxidative capacity of muscles. News Physiol Sci 3: 113–116, 1988. 12. Juliano RL and Haskill S. Signal transduction from the extracellular matrix. J Cell Biol 120: 577–585, 1993. 13. Kashiwagi M, Tortorella M, Nagase H, and Brew K. TIMP-3 is a potent inhibitor of aggrecanase-1 (ADAM-TS4) and aggrecanase-2 (ADAM-TS5). J Biol Chem 276: 12501–12504, 2001. 14. Loechel F, Gilpin BJ, Engvall E, Albrechtsen R, and Wewer UM. Human ADAM 12 (meltrin ␣) is an active metalloproteinase. J Biol Chem 273: 16993–16997, 1998. 15. Matsuoka H, Nakat M, Keisuke K, Yoshinori K, Nomura G, Toshima H, and Imaizumi T. Chronic L-arginine administration attenuates cardiac hypertrophy in SHR. Hypertension 27: 14–18, 1996. 16. Meredith JE Jr, Fazeli B, and Schwartz MA. The extracellular matrix as a cell survival factor. Mol Biol Cell 4: 953–961, 1993. 17. Nowicki PT. The effects of ischemia-reperfusion on endothelial cell function in postnatal intestine. Pediatr Res 39: 267–274, 1996. AJP-Lung Cell Mol Physiol • VOL
L245
18. Pinsky DJ, Patton S, Mesaros S, Brovkovych V, Kubaszewski E, Grunfeld S, and Malinski T. Mechanical transduction of nitric oxide synthesis in the beating heart. Circ Res 81: 372–379, 1997. 19. Prasad K, Gupta JB, Kalra J, Lee P, Mantha SV, and Bharadwaj B. Oxidative stress as a mechanism of cardiac failure in chronic volume overload in canine model. J Mol Cell Cardiol 28: 375–385, 1996. 20. Radomski A, Sawicki G, Olson DM, and Radomski MW. The role of nitric oxide and metalloproteinases in the pathogenesis of hyperoxia-induced lung injury in newborn rats. Br J Pharmacol 125: 1455–1462, 1998. 21. Roberts JT and Wearn JT. Quantitative changes in the capillary-muscle relationship in human hearts during normal growth and hypertrophy. Am Heart J 16: 617–633, 1941. 22. Roos D, Wening RS, and Wyss SR. Protection of human neutrophils by endogenous catalase: studies with cells from catalase-deficient individuals. J Clin Invest 65: 1515–1522, 1980. 23. Rucklidge GJ, Milne G, McGaw BA, Milne E, and Robins SP. Turnover rates of different collagen types measured by isotope ratio mass spectrometry. Biochim Biophys Acta 11: 1156–1157, 1992. 24. Senior RM, Griffin GL, Eliszar CJ, Shapiro SD, Goldberg GI, and Welgus HG. Human 92- and 72-kilodalton type IV collagenases are elastases. J Biol Chem 266: 7870–7875, 1991. 25. Shindo T, Ikeda U, and Ohkawa F. NOS in rat cardiac myocytes and fibroblasts. Life Sci 55: 1101–1108, 1994. 26. Simon DI, Mullins ME, Jia L, Gaston B, Singel DL, and Stamler JS. Polynitrosylated proteins: characterization, bioactivity and functional consequences. Proc Natl Acad Sci USA 93: 4736–4741, 1996. 27. Treasure CB, Klein JL, Vita JA, Manoukian SV, Renwick GH, Selwyn AP, Ganz P, and Alexander RW. Hypertension and LVH are associated with impaired endothelium-mediated relaxation in human coronary resistance vessels. Circulation 87: 86–93, 1993. 28. Tortorella MD, Burn TC, and Pratta MA. Purification and cloning of aggrecanase-1: a member of the ADAMTS famility of proteins. Science 284: 1664–1666, 1999. 29. Tyagi SC, Haas SJ, Kumar SG, Reddy HK, Voelker DJ, Hayden MR, Demmy TL, Schmaltz RA, and Curtis JJ. Post-transcriptional regulation of extracellular matrix metalloproteinase in human heart end-stage failure secondary to ischemic cardiomyopathy. J Mol Cell Cardiol 28: 1415–1428, 1996. 30. Tyagi SC, Hayden MR, and Hall JE. Role of angiotensin in angiogenesis and cardiac fibrosis in heart failure. Prog Exp Cardiol 2: 537–549, 1998. 31. Tyagi SC, Kumar SG, and Borders S. Reduction-oxidation (redox) state regulation of extracellular matrix metalloproteinases and tissue inhibitors in cardiac normal and transformed fibroblast cells. J Cell Biochem 61: 139–151, 1996. 32. Tyagi SC, Ratajska A, and Weber KT. Myocardial matrix metalloproteinases: localization and activation. Mol Cell Biochem 126: 49–59, 1993. 33. Varin R, Mulder P, Tamion F, Richard V, Henry JP, Lallemand F, Lerebours G, and Thuillez C. Improvement of endothelial function by chronic ACEI in heart failure: role of NO, prostanoids, oxidant stress and bradykinin. Circulation 102: 351–356, 2000. 34. Yamaoto S, James TN, Wawada K, Okabe M, and Kawamura K. Generation of new intercellular junctions between cardiomyocytes: a possible mechanism compensating for mechanical overload in the hypertrophied human adult myocardium. Circ Res 78: 362–370, 1996. 35. Zhang H, Schmeisser A, Garlichs CD, Plotze K, Damme U, Muge A, and Daniel WG. Angiotensin II-induced superoxide anion generation in human vascular endothelial cells: role of membrane-bound NADH/NADPH oxidases. Cardiovasc Res 44: 215–222, 1999. 36. Zhang X, Li H, Jin H, Ebin Z, Brodsky S, and Goligorsky MS. Effects of homocysteine on endothelial nitric oxide production. Am J Physiol Renal Physiol 279: F671–F678, 2000. 283 • AUGUST 2002 •
www.ajplung.org
