Imbalance Between Tissue Inhibitor of Metalloproteinase-4 and Matrix Metalloproteinases During Acute Myoctardial Ischemia-Reperfusion Injury
Background— We have previously reported that matrix metalloproteinase-2 (MMP-2) contributes to myocardial ischemia-reperfusion injury by degradation of troponin I, a regulatory element of the contractile proteins. MMP activities are also tightly regulated by tissue inhibitors of metalloproteinase (TIMPs). The change in TIMPs during acute myocardial ischemia-reperfusion injury is not clear.
Methods and Results— Isolated rat hearts were perfused either aerobically for 75 minutes or subjected to 15, 20, or 25 minutes of global, no-flow ischemia followed by 30 minutes of aerobic reperfusion. During reperfusion after ischemia, there was a rapid, enhanced release of TIMP-4, the most abundant TIMP in the heart, into the coronary effluent, as shown both by reverse zymography and Western blot. There was a negative correlation between the recovery of cardiac mechanical function and the release of TIMP-4 during reperfusion in hearts subjected to different durations of ischemia. Immunogold electron microscopy revealed a close association of TIMP-4 with the sarcomeres in aerobically perfused hearts. Moreover, TIMP-4 was present only in thin myofilaments prepared from aerobically perfused hearts but not in ischemic-reperfused hearts. An enhanced MMP activity was shown in ischemic-reperfused hearts by in situ zymography.
Conclusions— Loss of TIMP-4 from the cardiac myocyte leads to an increase in net myocardial MMP activity that contributes to acute myocardial stunning injury.
Received January 31, 2003; accepted February 4, 2003.
Matrix metalloproteinases (MMPs) are a family of zinc-containing endopeptidases. They are best known for their actions in the long-term remodeling of the extracellular matrix in tissue during various physiological and pathological conditions, such as embryonic development, inflammation, and cancer.1 Recently, a number of acute, non–extracellular matrix–related actions of MMPs have been described that indicate their roles in platelet aggregation,2 the regulation of vascular tone,3,4 and inflammation.5 We recently demonstrated that MMP-2 contributes to acute myocardial ischemia-reperfusion injury.6 In determining the locus of action of MMP-2 in the heart, we found that MMP-2 is localized within the cardiac myocyte in close association with the sarcomeres.7 MMP-2 in fact was found to be colocalized with troponin I, a key regulatory element of the actin-myosin ATPase, and contributes to its proteolytic degradation in acute myocardial ischemia-reperfusion injury.7
MMP activity is tightly controlled by endogenous inhibitors, such as the tissue inhibitors of metalloproteinase (TIMPs).8 Four TIMPs have been identified to date, of which TIMP-4 is the most abundant TIMP in human9 and murine hearts.10 Alteration in the balance between MMP and TIMP, resulting in enhanced MMP activity, has been shown to occur in long-term remodeling processes, such as infarction,11 heart failure,12 and dilated cardiomyopathy.13 The role of TIMPs during acute myocardial ischemia-reperfusion is not clear. We hypothesized that an imbalance between MMP and TIMP leads to increased MMP activity during myocardial ischemia-reperfusion injury.
This investigation conforms to the Guide to the Care and Use of Experimental Animals published by the Canadian Council on Animal Care (revised 1993).
Heart Preparation and Perfusion
Heart perfusions were performed in male Sprague-Dawley rats (250 to 300 g; Bioscience Animal Services, University of Alberta, Edmonton, Canada) as described previously.6 Rat hearts were perfused in the Langendorff mode with Krebs-Henseleit buffer at 37°C at a constant pressure of 60 mm Hg. The composition of the buffer was (in mmol/L): NaCl 118, KCl 4.7, KH2PO4 1.2, MgSO4 1.2, CaCl2 3.0, NaHCO3 25, glucose 11, and EDTA 0.5, and it was gassed continuously with 95% O2/5% CO2 (pH 7.4). Spontaneously beating hearts were used in all experiments. Left ventricular pressures were measured by a latex balloon catheter inserted into the left ventricle. Coronary flow, heart rate, and left ventricular pressure were monitored with MP100 software (BIOPAC). Left ventricular developed pressure was calculated as the difference between systolic and diastolic pressures of the left ventricular pressure trace. The rate-pressure product was calculated as the product of heart rate and left ventricular developed pressure. Using this perfusion protocol, hearts maintained a steady heart rate and left ventricular developed pressure for ≥80 minutes after stabilization.
After 25 minutes of stabilization during aerobic perfusion, hearts were subjected to either 15, 20, or 25 minutes of global, no-flow ischemia (at 37°C) followed by 30 minutes of aerobic reperfusion. Control hearts were perfused aerobically for 75 minutes. Samples of coronary effluent (12 mL) were collected immediately before ischemia and after 1, 5, 10, and 30 minutes of reperfusion. Hearts were freeze-clamped in liquid nitrogen at the end of perfusion and stored at −80°C.
Concentration of the Coronary Effluent
The 12-mL coronary effluent samples were concentrated by use of Centricon-10 concentrating vessels (Amicon Inc). After concentration, the final volume of the samples was measured by gravimetry. The protein level in the concentrate was determined by the bicinchoninic acid assay using BSA as a standard. The protein concentration in the coronary effluent before concentration was then calculated and expressed in μg/mL.
Preparation of Heart Extracts
Frozen hearts were crushed with a mortar and pestle at liquid nitrogen temperature and then homogenized by sonication in 50 mmol/L Tris-HCl (pH 7.4) containing 3.1 mmol/L sucrose, 1 mmol/L dithiothreitol, 10 μg/mL leupeptin, 10 μg/mL soybean trypsin inhibitor, 2 μg/mL aprotinin, and 0.1% Triton X-100. The homogenate was centrifuged at 10 000g at 4°C for 10 minutes, and the supernatant was collected and stored at −80°C.
Three independent preparations of thin myofilaments from both aerobically perfused rat hearts and those subjected to 20 minutes of ischemia followed by 30 minutes of reperfusion (3 hearts per preparation) were isolated as described by Spiess et al.14 SDS-PAGE (15% polyacrylamide) followed by Coomassie blue staining was performed to confirm the protein content of the preparation. TIMP-4 content in these preparations was determined by Western blot analysis.
Western Blot Analysis of TIMP-4
TIMP-4 content in the coronary effluent, myocardium, and thin-filament preparations was detected by immunoblot analysis. Samples were subjected to SDS-PAGE (15% polyacrylamide) and transferred to a polyvinylidene difluoride membrane (Bio-Rad). TIMP-4 was identified by use of a polyclonal rabbit anti-human TIMP-4 antibody (0.25 μg/mL, Chemicon International), and band densities were measured with an HP6100 scanner (Hewlett-Packard) and Sigmagel measurement software (Jandel). To identify the higher-molecular-weight band on the immunoblot of the thin-myofilament fraction, a separate membrane was incubated with rabbit anti–TIMP-4 antibody that was preabsorbed with recombinant rat TIMP-4 in 50 mmol/L Tris-HCl buffer (with 150 mmol/L NaCl, 5 mmol/L CaCl2) at 37°C for 15 minutes. Recombinant rat TIMP-4 was used as a standard (a kind gift from Dr S.C. Tyagi, University of Mississippi Medical Center, Jackson).
Immunogold Electron Microscopy
Control hearts perfused aerobically for 75 minutes (Aerobic 75′) and hearts subjected to 20 minutes of ischemia followed by 30 minutes of reperfusion (20′ I/R hearts, 3 per group) were perfusion-fixed with 4% paraformaldehyde and 0.1% glutaraldehyde. After dehydration in increasing concentrations of ethanol (50% to 100%), specimens were embedded in Unicryl (British BioCell International) and polymerized under ultraviolet radiation at 4°C. Grids with ultrathin sections were floated on 1% glycine and 1% BSA at room temperature. After the grids had been transferred to a Falcon 30340 microtest plate, they were incubated with either rabbit polyclonal anti–human TIMP-4 (4 μg/mL) or isotype control IgG (4 μg/mL, Sigma) for 4 hours at room temperature. Sections were then rinsed with 1% BSA-PBS and placed on drops of anti-rabbit IgG gold conjugate (1:20, Sigma) in BSA-PBS buffer for 2 hours at room temperature. Grids were stained with 2% uranyl acetate and 0.2% of lead citrate. Sections were examined with a Hitachi H-7000 transmission electron microscope at 75 kV.
The level of TIMPs in the coronary effluent samples was determined by reverse zymography. Briefly, concentrated coronary effluent samples were mixed with nonreducing sample loading buffer and applied to 12% polyacrylamide gels copolymerized with 1 mg/mL gelatin (Sigma) and 50 ng/mL human recombinant MMP-2 (Oncogene). Effluent concentrate (30 μL) was loaded in each lane (≈1 to 2 μg protein). After electrophoresis, gels were rinsed in 2.5% Triton X-100 (3×20 minutes) to remove SDS. Then the gels were washed twice in incubation buffer for 20 minutes each at room temperature. The composition of the incubation buffer was (in mmol/L) Tris-HCl 50, CaCl2 5, NaCl 150, and NaN3 0.05%. The gels were then kept in incubation buffer at 37°C for 48 hours. After incubation, gels were stained in staining solution (2% Coomassie brilliant blue G, 25% methanol, and 10% acetic acid) for 2 hours and then destained twice for 30 minutes each in destaining solution (2% methanol, 4% acetic acid). The TIMP inhibitory activity was visualized as dark bands against a clear background.
In Situ Zymography
In situ zymography was used to determine the net tissue gelatinolytic activity from Aerobic 75′ and 20′ I/R hearts as described.15 In brief, left ventricles were trimmed and freeze-embedded in Cryomatrix (Shandon) at the freezing temperature of propanol. Then 4-μm sections were prepared by use of a cryostat. Sections were incubated with 0.5 mg/mL gelatin–Oregon green (Molecular Probes) in 50 mmol/L Tris-HCl (mmol/L: 150 NaCl, 5 CaCl2) at 37°C for 3 hours. Parallel slides of consecutive sections were incubated with the addition of 10 mmol/L EDTA to inhibit MMP activity. The sections were then fixed in 1% paraformaldehyde containing 0.5% d,l-glucose. Sections were examined under a Nikon-Diaphot fluorescent microscope (450 to 550 nm). A green color indicated positive gelatinolytic activity.
Data were expressed as mean±SEM. One-way or 2-way ANOVA with Dunnett’s multiple comparison test was used, as appropriate, for statistical analysis. A value of P<0.05 was considered statistically significant.
Cardiac Mechanical Function Recovery and Release of TIMP-4 Into Coronary Effluent From 20′ I/R Hearts
Figure 1 shows that 20′ I/R hearts showed a depressed recovery of cardiac mechanical function compared with control hearts perfused aerobically for 75 minutes (rate-pressure product at the end of perfusion expressed as a percentage of baseline function at t=0 minutes was 27±8% versus 84±3%, respectively, P<0.05, n=6). Reverse zymography of the coronary effluent during aerobic perfusion demonstrated that there was a single 23-kDa band corresponding to TIMP-4 (Figure 2a). There was no evidence of TIMP-1, -2, or -3 activity. After 20 minutes of ischemia, there was a markedly enhanced release of TIMP-4 into the coronary effluent in the first minute of reperfusion that was maintained over the first 5 minutes and then declined to baseline levels at 10 minutes of reperfusion (Figure 2a). This TIMP-4 activity, as shown by reverse zymography, was verified to be TIMP-4 by Western blot analysis (Figure 2b). There is no discernible change in levels of TIMP-4 in heart tissue extracts prepared from Aerobic 75′ or 20′ I/R groups (data not shown).
Duration of Ischemia, Release of TIMP-4, and Recovery of Cardiac Mechanical Function
We next determined the relationship between the duration of ischemia and the release of TIMP-4 into the coronary effluent in the first minute of reperfusion and how this correlates to the recovery of mechanical function. As reported,6 hearts subjected to 15 minutes of ischemia followed by 30 minutes of reperfusion showed no significant impairment in mechanical function. However, both 20 and 25 minutes of ischemia significantly impaired the recovery of mechanical function (Figure 3a). Whereas 15′ I/R did not increase TIMP-4 release during reperfusion above baseline, either 20 or 25 minutes of ischemia significantly enhanced release of TIMP-4 into the coronary effluent (Figure 3b). There was a significant negative linear correlation between the recovery of cardiac mechanical function and TIMP-4 release in the first minute of reperfusion (r=−0.58, P<0.05, n=13).
Association of TIMP-4 With Sarcomeres in Aerobically Perfused Hearts
We previously found that MMP-2 colocalized with the contractile protein regulatory element troponin I in the sarcomeres of cardiomyocytes.7 We therefore determined the subcellular localization of TIMP-4, an inhibitor of MMP activity. Immunogold electron micrographs show a close association of TIMP-4 with sarcomeres in aerobically perfused rat hearts (Figure 4). There was no positive staining in sections incubated with unrelated IgG. This pattern of positive staining was reduced in 20′ I/R hearts (Figure 4). Two hearts from either the aerobically perfused or 20′ I/R groups were stained, and 10 pictures of different sections from each heart were chosen randomly. The sarcomere-associated TIMP-4–positive staining was quantified by 2 independent observers in a blinded fashion. TIMP-4 staining was significantly less in sections from 20′ I/R hearts than in Aerobic 75′ hearts (29±6 versus 74±9 sarcomere-associated counts per field, respectively, P<0.05).
Loss of TIMP-4 in Thin-Myofilament Preparations From 20′ I/R Hearts
To further investigate the association of TIMP-4 with sarcomeres, highly purified thin-myofilament fractions were prepared from both Aerobic 75′ hearts and 20′ I/R hearts. The purity of these myofilament preparations was demonstrated by Coomassie blue staining, which showed bands for actin, tropomyosin, and troponin I (Figure 5a). TIMP-4 immunoblots of these preparations revealed an ≈25-kDa band in the preparation from aerobically perfused hearts but not in that from 20′ I/R hearts (Figure 5b). Preabsorption of the anti–TIMP-4 antibody with recombinant TIMP-4 abolished its ability to recognize the 25-kDa band seen in the thin myofilaments from Aerobic 75′ hearts (data not shown).
Increased In Situ MMP Activity in 20′ I/R Hearts
With the loss of TIMP-4 from the myocardium as a result of ischemia and reperfusion, we determined whether this was associated with a net increase in tissue gelatinolytic activity. Tissue sections prepared from aerobically perfused control hearts showed very little gelatinolytic activity, as determined by in situ zymography (Figure 6). The intensity of the activity was greatly enhanced in hearts subjected to 20′ I/R. This gelatinolytic activity was significantly reduced in the presence of 10 mmol/L EDTA, a blocker of metalloproteinase activity.
We report here for the first time that myocardial ischemia and reperfusion injury results in the rapid loss of TIMP-4 from the heart into the coronary effluent. This release of TIMP-4 was dependent on the duration of ischemia and was significantly enhanced only when the ischemia was of sufficient duration to reduce the recovery of mechanical function. In the normal heart, TIMP-4 was found to be localized to sarcomeres. Subjecting the heart to ischemia and reperfusion resulted in a loss of TIMP-4 from the thin-myofilament fraction. The loss of this MMP inhibitor protein resulted in enhanced gelatinolytic activity in the myocardium during reperfusion. These results suggest that TIMP-4 plays an important homeostatic role in the normal myocardium and that its release from the heart contributes to myocardial ischemia-reperfusion injury.
Using this same model, we showed previously that there is also a rapid release of MMP-2 from the ischemic-reperfused heart and that pharmacological inhibition of MMP-2 activity protects the heart from ischemia-reperfusion injury.6 We discovered that MMP-2 colocalizes with the thin-myofilament regulatory protein troponin I in the normal heart and contributes to its degradation.7 The degradation of troponin I accounts for diminished contractile function as a result of ischemia-reperfusion injury in rat16 and human17 but not pig18,19 hearts. During ischemia-reperfusion injury, there was an accumulation of MMP-2 within the thin myofilaments.7 We therefore considered whether TIMP-4 may also be associated with the thin myofilaments. Here, we report that there is a loss of TIMP-4 associated with the thin-myofilament fraction during acute myocardial ischemia-reperfusion injury. The enhanced gelatinolytic activity we observed suggests that there is an imbalance between TIMP-4 and MMP-2 and thus increased MMP activity in the myocardium.
Four TIMPs have been characterized, namely, TIMP-1, -2, -3, and -4. They share a high sequence homology but have unique characteristics.20 TIMP-4 is expressed predominantly in the heart compared with other organs.9 We verified by Western blot analysis that there is TIMP-4 in rat hearts. In addition to its inhibitory action on MMPs, the physiological function of TIMP-4 and the reason why it is expressed predominantly in the heart are not well understood. Tummalapalli et al21 suggested that TIMP-4 expression in cardiac fibroblasts regulates their normal cell phenotype. Here, we identified the association of TIMP-4 in thin-myofilament preparations from aerobically perfused rat hearts. The somewhat higher molecular weight of the TIMP-4 band in the thin-myofilament preparations may be a result of an unprocessed 29-amino-acid leader sequence.20 Preabsorption of the anti–TIMP-4 antibody with recombinant TIMP-4 removed the ability of the antibody to recognize this band, precluding the possibility of nonspecific binding. The question thus arises whether such a leader sequence would render TIMP-4 unable to block MMP-2 activity, because the addition of a single amino acid to the N-terminal cysteine of human TIMP-2 abolished its ability to inhibit human MMP-2.22 The rapid processing and removal of the putative signal peptide sequence for export is certainly within the time frame of the 20-minute ischemia and reperfusion period.23 Several independent reports demonstrate that TIMP-4, unlike TIMP-2, does not promote but actually inhibits the activation of MMP-2.24,25 Our data suggest that there is a protective role of TIMP-4 in the normal heart.
An imbalance between TIMPs and MMPs has been suggested to occur in long-term remodeling processes during myocardial infarction,11 heart failure,12 and dilated cardiomyopathy.13 All of these studies showed that TIMPs are reduced in diseased hearts, whereas MMPs are upregulated over days or months. Here, we report that there is a rapid release of TIMP-4 from the hearts in the first minutes during reperfusion after ischemia. The loss of TIMP-4 from the myofilaments was a consequence of this release. Immunogold electron microscopy demonstrated less TIMP-4 staining in ischemic-reperfused hearts than in control hearts. Combined with our previous results of an increase in MMP-2 activity associated with the thin-myofilament fraction prepared in an identical fashion after ischemia-reperfusion,7 it is reasonable to speculate that there may be an increased net MMP activity within cardiomyocytes in ischemic-reperfused hearts. This was confirmed by in situ zymography, which is able to detect the net gelatinolytic activity of the tissue. Compared with aerobically perfused hearts, ischemic-reperfused hearts showed a higher gelatinolytic activity, which is inhibited by EDTA, suggesting that this gelatinolytic activity is indeed a result of MMP activity (Figure 6).
Our results are in good accordance with findings by Mayers et al,26 who investigated biopsies of human myocardium from patients undergoing cardiac bypass surgery, which can result in a type of ischemia-reperfusion injury. They found decreased TIMP-4 in the myocardium immediately after bypass. They speculated that there could be an increased net gelatinolytic activity after ischemia-reperfusion, something that we documented in the present study using in situ zymography.
Acute ischemia-reperfusion of the heart is not a rare situation in clinical practice. The prevention of postischemic cardiac dysfunction is a major challenge for healthcare professionals. Here, we present a new mechanism that leads to depressed cardiac mechanical function in the acute reperfusion phase. In summary, we report that TIMP-4 is expressed in normal myocardium in close association with sarcomeres. During reperfusion after ischemia, there is a rapid release of TIMP-4 into the coronary effluent. Recovery of cardiac mechanical function after ischemia-reperfusion correlates negatively with the amount of TIMP-4 released. Restoring the imbalance between TIMPs and MMPs might be a useful strategy in the prevention or treatment of acute cardiac ischemic syndrome.
This project was funded by a grant to Dr Schulz from the Canadian Institutes of Health Research (MT-14741). Dr Wang is a Graduate Student Trainee, and Dr Schulze is a Fellow of the Heart and Stroke Foundation of Canada. Dr Schulz is a Senior Scholar of the Alberta Heritage Foundation for Medical Research. We thank Dr Ming Chen and Jennifer Kwan for assistance with electron microscopy.
↵*The first 2 authors contributed equally to this work.
Woessner JFJ. The matrix metalloproteinases family. In: Parks WC, Mecham RP, eds. Matrix Metalloproteinases. San Diego, Calif: Academic Press; 1998: 1–14.
Fernandez-Patron C, Radomski MW, Davidge ST. Vascular matrix metalloproteinase-2 cleaves big endothelin-1 yielding a novel vasoconstrictor. Circ Res. 1999; 85: 906–911.
Fernandez-Patron C, Stewart KG, Zhang Y, et al. Vascular matrix metalloproteinase-2–dependent cleavage of calcitonin gene-related peptide promotes vasoconstriction. Circ Res. 2000; 87: 670–676.
McQuibban GA, Gong JH, Tam EM, et al. Inflammation dampened by gelatinase A cleavage of monocyte chemoattractant protein-3. Science. 2000; 289: 1202–1206.
Cheung PY, Sawicki G, Wozniak M, et al. Matrix metalloproteinase-2 contributes to ischemia-reperfusion injury in the heart. Circulation. 2000; 101: 1833–1839.
Wang W, Schulze CJ, Suarez-Pinzon WL, et al. Intracellular action of matrix metalloproteinase-2 accounts for acute myocardial ischemia and reperfusion injury. Circulation. 2002; 106: 1543–1549.
Nagase H, Woessner JF Jr. Matrix metalloproteinases. J Biol Chem. 1999; 274: 21491–21494.
Greene J, Wang M, Liu YE, et al. Molecular cloning and characterization of human tissue inhibitor of metalloproteinase 4. J Biol Chem. 1996; 271: 30375–30380.
Li YY, Feldman AM, Sun Y, et al. Differential expression of tissue inhibitors of metalloproteinases in the failing human heart. Circulation. 1998; 98: 1728–1734.
Senzaki H, Paolocci N, Gluzband YA, et al. Beta-blockade prevents sustained metalloproteinase activation and diastolic stiffening induced by angiotensin II combined with evolving cardiac dysfunction. Circ Res. 2000; 86: 807–815.
McDonough JL, Arrell DK, Van Eyk JE. Troponin I degradation and covalent complex formation accompanies myocardial ischemia/reperfusion injury. Circ Res. 1999; 84: 9–20.
Thomas SA, Fallavollita JA, Lee TC, et al. Absence of troponin I degradation or altered sarcoplasmic reticulum uptake protein expression after reversible ischemia in swine. Circ Res. 1999; 85: 446–456.
Kim SJ, Kudej RK, Yatani A, et al. A novel mechanism for myocardial stunning involving impaired Ca2+ handling. Circ Res. 2001; 89: 831–837.
Wingfield PT, Sax JK, Stahl SJ, et al. Biophysical and functional characterization of full-length, recombinant human tissue inhibitor of metalloproteinases-2 (TIMP-2) produced in Escherichia coli. J Biol Chem. 1999; 274: 21362–21368.
YaDeau JT, Blobel G. Solubilization and characterization of yeast signal peptidase. J Biol Chem. 1989; 264: 2928–2934.
Bigg HF, Morrison CJ, Butler GS, et al. Tissue inhibitor of metalloproteinases-4 inhibits but does not support the activation of gelatinase A via efficient inhibition of membrane type 1-matrix metalloproteinase. Cancer Res. 2001; 61: 3610–3618.