Matrix Metalloproteinase-7 Affects Connexin-43 Levels, Electrical Conduction, and Survival After Myocardial Infarction
Background— Matrix metalloproteinases (MMPs) contribute to left ventricular remodeling after myocardial infarction (MI). Specific causative roles of particular MMPs, however, remain unclear. MMP-7 is abundant in cardiomyocytes and macrophages, but MMP-7 function after MI has not been defined.
Methods and Results— Wild-type (WT; n=55) and MMP-7–null (MMP-7−/−; n=32) mice underwent permanent coronary artery ligation for 7 days. MI sizes were similar, but survival was greatly improved in MMP-7−/− mice. The survival difference could not be attributed to differences in left ventricular dilation because end-diastolic volumes increased similarly. ECG analysis revealed a prolonged PR interval in WT but not in MMP-7−/− post-MI mice. Post-MI conduction velocity, determined by optically mapping electrical wavefront propagation, decreased to 78±6% of control for WT and was normalized in MMP-7−/− mice. In WT mice, slower conduction velocity correlated with a 53% reduction in the gap junction protein connexin-43. Direct binding of MMP-7 to connexin-43, determined by surface plasmon resonance technology, occurred in a dose-dependent manner. Connexin-43 processing by MMP-7 was confirmed by in silico and in vitro substrate analyses and MMP-7 infusion induced arrhythmias in vivo.
Conclusions— MMP-7 deletion results in improved survival and myocardial conduction patterns after MI. This is the first report to implicate MMP-7 in post-MI remodeling and to demonstrate that connexin-43 is a novel MMP-7 substrate.
Received January 9, 2006; revision received April 17, 2006; accepted April 24, 2006.
Left ventricular (LV) remodeling after myocardial infarction (MI) is a necessary component of the healing process that ensures adequate scar formation.1 MI-stimulated LV remodeling, however, is now the primary cause of congestive heart failure.2 Matrix metalloproteinases (MMPs) are a family of enzymes that increase in human and animal models after MI.3 A cause and effect relationship between MMP activity and changes in LV geometry and function after MI has been established.3 Although >25 MMPs have been identified, the roles of individual MMP types during post-MI LV remodeling remain poorly defined.
Clinical Perspective p 2928
MMP-7 is expressed in macrophages4,5 and cardiomyocytes6 after MI. Pro-MMP-7 is 28 kDa; active-MMP-7 is 19 kDa. In vitro, MMP-7 can cleave an extensive portfolio of extracellular matrix substrates, including collagen IV, laminin, and fibronectin,7,8 as well as bioactive molecules such as tumor necrosis factor-α5 and other MMPs, including MMP-1, MMP-2, and MMP-9.9 The list of in vitro substrates suggests that MMP-7 may influence LV remodeling, particularly during the acute phase after MI.7 The goal of the present study was to test the hypothesis that MMP-7 gene deletion alters post-MI LV remodeling. Interestingly, the absence of MMP-7 resulted in improved post-MI survival through mechanisms involving the processing of connexin-43, which in turn influenced electrical remodeling. This is the first report to demonstrate that MMP-7 may mediate electrical remodeling after MI.
All animal procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals (National Research Council, National Academy Press, Washington, DC, 1996) and were approved by the Institutional Animal Care and Use Committee at the Medical University of South Carolina.
MMP-7−/− mice were generated with the use of homologous recombination in embryonic stem cells and backcrossed >10 generations into the C57BL/6 strain, as previously described.10 Homozygous wild-type C57BL/6 mice (WT mice; n=55; 11.5±0.4 weeks of age) and MMP-7−/− C57BL/6 mice (MMP-7−/− mice; n=32; 11.7±0.3 weeks of age) of both genders were used for the MI protocols. Unoperated WT (n=27; 11.7±0.4 weeks of age) and MMP-7−/− (n=15; 11.7±0.4 weeks of age) mice (The Jackson Laboratory, Bar Harbor, Me) were used as reference controls.
Subsets of mice were randomized to the terminal studies and downstream analyses, with sample sizes as shown in Table 1. No mice were excluded from any subset analysis. For all analyses, the evaluator was blinded to genotype.
In Vivo Procedures
Coronary artery ligation, echocardiography, and ECG procedures were performed as detailed in the online-only Data Supplement. For optical mapping of electrical activation, the aorta was cannulated, and the heart was perfused with Tyrode’s solution at 37°C, at a constant pressure of 100 mm Hg. The heart was placed in a thermostatically controlled tissue chamber (37±1°C) on the stage of a dissecting microscope (MZ FL III, Leica, Deerfield, Ill). The potentiometric dye di-4-ANEPPS (12 μmol/L; Molecular Probes/Invitrogen, Eugene, Ore) was perfused for 5 minutes. Cytochalasin D (20 μmol/L) was added to the perfusate to uncouple mechanical contraction.11 Hearts were paced with the use of bipolar electrodes from the apex at a rate of 300 bpm. The stimulation pulses were 2 ms in duration, and the voltage was adjusted to ×2 threshold. Changes in membrane voltage during activation of the myocyte action potentials were recorded as chromatographic changes in di-4-ANNEPPS excitation with the use of a high-speed 80×80-pixel CCD camera (Red Shirt Imaging, Fairfield, Conn). Activation sequences were recorded at a rate of 1 kHz. Action potential upstrokes were marked with the use of Cardioplex software (Red Shirt Imaging).12 The peak of the first derivative of the action potential upstroke was used to mark activation time. Conduction velocities were derived from the gradients of activation times and are presented as discrete local velocity estimations.13
Surface Plasmon Resonance Protein-Protein Binding Studies
Surface plasmon resonance is an optical technique that characterizes macromolecular interactions.14 The binding of a soluble analyte to an immobilized ligand changes the refractive index, which can be monitored in real time. Interaction kinetics, including association rates, are measured precisely.14 A BIACore 3000 Instrument was used with CM5 research grade sensor chips (BIACore, Piscataway, NJ).15 Experiments were performed at 25°C with the use of BIACore buffer (10 μmol/L HEPES, pH 7.4, 150 mmol/L NaCl, 0.005% surfactant P20) supplemented with 2 mmol/L CaCl2 (HBS-P-Ca). A flow rate of 20 μL/min was maintained throughout the experiments. Connexin-43 (30 mmol/L) was immobilized on the sensor chip, and MMP-7 (0.1 to 1.6 μmol/L) was injected as analyte. Buffer-only (no MMP-7) runs were used to subtract out background. Collagen IV (32 mmol/L), a known MMP-7 substrate, was the positive control ligand. Collagen I (37 mmol/L), known not to be a MMP-7 substrate, was the negative control ligand. MMP-2 (1.6 μmol/L) was the negative control analyte. Increasing amounts of MMP-7 from 0.1 to 1.6 μmol/L were used to generate a dose-response curve. At least 4 injections per ligand and analyte were performed. Protein-protein interactions between analyte (MMP-7) and immobilized ligand (connexin-43) are plotted as relative units versus time (sensorgrams). Increased relative units indicate increased concentration of analyte at the surface of the sensor chip due to specific interaction between analyte and ligand.
Histological and Biochemical Analyses
Tissue collection, histology, immunoblotting, real-time reverse transcriptase–polymerase chain reaction, and cleavage analyses are described in the online-only Data Supplement.
In Vivo Infusion of MMP-7
Recombinant pro-MMP-7 (30 ng/μL diluted in saline; Chemicon International, Temecula, Calif) was infused into mice (n=3) via a jugular vein catheter (PE-10 tubing; Becton Dickinson, Sparks, Md) attached to a 2.5-mL Bee Stinger gas-tight syringe (Bioanalytical Systems, West Lafayette, Ind). This concentration was the half-maximum dose determined by in vitro cleavage analyses. The Baby Bee Syringe Drive was controlled by the MD-1020 Bee Hive controller (Bioanalytical Systems) to maintain a constant flow rate of 0.8 μL/min. Saline infusions served as negative controls (n=4). Within 60 minutes of infusion, all mice infused with MMP-7 developed heart block. The mice were euthanized when heart block developed, as described above, and tissue was collected for histology and immunoblotting.
Data are presented as mean±SEM. Statistical analyses were performed with the use of Intercooled Stata 8.0 for Windows (Stata Corporation, College Station, Tex). Survival rates were analyzed by Fisher exact test. Echocardiographic and ECG data were analyzed by repeated-measures ANOVA, with comparisons between groups made by Bonferroni-corrected t test. Immunoblotting and electrical conduction parameters were compared with the use of a 1-sample t test. A 2-tailed value of P<0.05 was considered statistically significant.
The authors had full access to the data and take full responsibility for its integrity. All authors have read and agree to the manuscript as written.
MMP-7 Influences Post-MI Survival
Six WT and 5 MMP-7−/− mice died perioperatively (P=NS) and were not included in the postsurgical survival statistics (Table 2). The 2 predominant causes of acute death after MI are LV rupture and sudden death due to arrhythmias.16 The 7-day postoperative survival rate was significantly higher in MMP-7−/− mice than in WT mice. The similar post-MI rupture incidences between WT and MMP-7−/− groups suggest that the significant improvement in survival may be attributed to a change in sudden death due to arrhythmias.
LV Geometry and Myocardial Morphology
Compared with pre-MI values, LV end-diastolic volumes increased and ejection fractions decreased in both post-MI groups (Table 2). Infarct sizes for WT and MMP-7−/− mice were 45±2% (n=16) and 48±2% (n=15; P=0.28), respectively. Therefore, the extent of necrosis and the MI-induced dilation response were equivalent in the 2 groups.
MMP-7 protein levels, detected by immunohistochemistry and immunoblotting, were elevated in both remote and infarct regions of WT mice after MI (Figure 1). Cardiomyocytes were the predominant cell type stained in the remote region, whereas macrophages were the predominant cell type stained in the infarct region. No MMP-7 protein was detected in the MMP-7−/− mice. By immunoblotting (Figure 1E), MMP-7 levels were &3-fold higher in both remote and MI regions compared with control values.
To determine whether MMP-2 or MMP-9 levels were upregulated as a result of MMP-7 deletion, MMP-2 and MMP-9 immunoblottings were performed on WT and MMP-7−/− mice pre-MI (control) samples and compared with remote and MI regions of WT and MMP-7−/− mice (n=4 in each group). MMP-2 levels were not different between any of the 6 groups (P>0.05). MMP-9 levels were significantly increased in both remote and MI regions compared with control samples in both WT and MMP-7−/− mice (P<0.01 for remote and MI versus control for both WT and MMP-7−/− mice). MMP-2 and MMP-9 levels, therefore, were not different between WT and MMP-7−/− mice.
Myocyte Cross-sectional Areas and Macrophage Infiltration
Additional parameters to assess the extent of LV remodeling after MI include the hypertrophic response of still viable myocytes in the remote region and macrophage infiltration into the infarcted region. Myocyte cross-sectional areas for unoperated reference control mice were 147±7 μm2 for WT (n=9) and 134±5 μm2 for MMP-7−/− (n=6; P=NS) mice. At 7 days after MI, myocyte cross-sectional areas increased to 269±16 μm2 for WT (n=8) and 269±11 μm2 for MMP-7−/− mice (n=11; P=NS). Both WT and MMP-7−/− myocyte cross-sectional areas were significantly higher than corresponding unoperated reference control values (P<0.001 for both). The area of maximum macrophage infiltration was 20±3% for WT mice (n=8) and was increased to 25±1% for MMP-7−/− mice (n=13; P=0.045).
MMP-7 Influences Electrical Remodeling Parameters
To determine whether changes in electrical remodeling provided a mechanism for improved survival in the MMP-7−/− mice, ECG and action potential parameters were evaluated. The PR interval for WT unoperated mice (n=17) was 36.2±1.9 ms. The PR interval was prolonged in WT post-MI mice (47.3±3.2 ms; n=11; P<0.01) but not in MMP-7−/− post-MI mice (41.0±2.9 ms; n=4; P=0.97). The QRS duration was not different between groups (15.1±0.6, 17.8±1.7, and 15.8±2.3 ms for WT unoperated mice (n=17), WT post-MI mice (n=11), and MMP-7−/− post-MI mice (n=4), respectively; P=0.23). High-resolution optical maps of myocardial depolarization were constructed to determine electrical conduction patterns and to generate isochrones of wavefront propagation (Figure 2). In WT control hearts, the depolarization wavefront progression was rapid and homogeneous. With MI, depolarization propagation was slower, and the conduction pattern was heterogeneous. In the MMP-7−/− post-MI hearts, in contrast, LV epicardial conduction from apex to base was faster than WT post-MI hearts. The conduction velocity of WT mice after MI, but not MMP-7−/− mice, was decreased compared with control (Figure 2D). The conduction velocity for MMP-7−/− unoperated mice was 0.97±0.09 m/s (n=3; P=0.76 versus WT unoperated mice). Therefore, these findings suggest that MMP-7 may mediate electrical processes in the post-MI myocardium.
Connexin-43, the most abundant connexin in the mammalian ventricular myocardium, is a major determinant of myocardial electrical properties (Figure 3).17,18 To determine whether qualitative and/or quantitative changes in connexin-43 levels explained differences in synctitial conduction, connexin-43 levels were evaluated by immunohistochemistry and immunoblotting with the antibody that recognizes the C-terminus of connexin-43. Qualitatively, a discrete pattern of connexin-43 staining could be observed in control unoperated LV sections, predominantly localized to the myocyte-myocyte border and consistent with previous reports.19,20 In post-MI WT LV sections, connexin-43 staining was decreased in the remote region, whereas the post-MI MMP-7−/− LV sections showed preservation of connexin-43 staining in the remote region. By immunoblotting, connexin-43 levels in the remote region of WT mice at 7 days after MI were 47±10% of control values (P<0.05, remote versus control WT). There were no differences among the remote region of MMP-7−/− mice, MI regions of WT or MMP-7−/− mice, and control samples. Together, these data demonstrate a loss of staining for the C-terminal domain of connexin-43 in WT mice after MI that is attenuated in MMP-7−/− mice.
To determine whether the loss of connexin-43 in the remote region of WT mice was due to a downregulation at the mRNA levels (rather than a proteolytic event), connexin-43 mRNA levels were quantified by reverse transcriptase–polymerase chain reaction. Connexin-43 levels, normalized to 18S RNA, were 0.22±0.09 units for WT control (n=3), 0.45±0.13 units for MMP-7−/− control (n=3), 0.49±0.14 units for WT remote region (n=6), 0.34±0.13 units for WT MI region (n=6), 0.34±0.04 units for MMP-7−/− remote region (n=5), and 0.39±0.19 units for MMP-7−/− MI region (n=5). None of the groups were statistically different, indicating that connexin-43 translation was not depressed in the absence of MMP-7.
Specific and Dose-Dependent Binding of MMP-7 to Connexin-43
To determine whether MMP-7 was capable of binding to connexin-43, we performed surface plasmon resonance binding studies. Figure 4 demonstrates the direct and specific protein-protein interaction between connexin-43 and MMP-7. Compared with collagen IV (a positive control ligand) and collagen I (a negative control ligand), connexin-43 was a very specific ligand for MMP-7. There were no significant signals generated between MMP-7 and the buffer only (no ligand) or collagen I ligand negative controls. Additionally, MMP-2 served as a negative control analyte and did not show binding to connexin-43. The classic protein association phase (black line in Figure 4B) occurred during the MMP-7 pulse and was followed by dissociation during the buffer only injection. The on rate (ka), determined by dose-response curves, was 1.5×104 (1/ms), indicating high specificity and affinity of MMP-7 binding to connexin-43.
In Silico Analysis of MMP-7 Cleavage Sites in Connexin-43
Two sites with sequence homology to known MMP-7 cleavage sequences were detected within the connexin-43 sequence (Figure 5A, 5B, and 5C). Residues 354 to 361 of connexin-43 matched the MMP-7 cleavage sequence GPQAIAGQ in 5 of 8 positions. Residues 375 to 382 of connexin-43 matched an additional MMP-7 cleavage sequence, PPEELKFQ, in 5 of 8 positions. Potential cleavage at either or both sites would generate small peptide sequences of 3 and 22 amino acids, with apparent molecular weights of 0.37 and 2.44 kDa, respectively. In addition, cleavage at either or both sites would remove the C-terminal antibody binding site and explain the loss in connexin-43 staining observed in vivo and in vitro.
MMP-7 Cleaves Connexin-43 at the C-Terminus
To determine whether connexin-43 was an MMP-7 substrate, in vitro substrate analysis was performed. As shown in Figure 5D, concentration-dependent decreased levels of recombinant connexin-43 protein fragment were detected with the use of the C-terminal connexin-43 antibody when increasing amounts of recombinant active MMP-7 were added.
To evaluate the ability of MMP-7 to process connexin-43 within the myocardium, in vitro substrate analysis was also performed with myocardial extracts. As shown in Figure 5E and 5F, concentration-dependent decreased levels of the C-terminal domain (amino acids 363 to 382) were also detected in myocardial extracts when increasing amounts of recombinant active MMP-7 were added. There was no difference in connexin-43 levels when an additional antibody, which recognizes upstream amino acids 252 to 270, was used (Figure 5G; P=0.84; n=6). Thus, MMP-7 was able to process connexin-43 at a C-terminus cleavage site.
MMP-7 Infusion In Vivo Induces Heart Block
Within 60 minutes of MMP-7 infusion, all mice (n=3) developed heart block (Figure 6). No ECG abnormalities were noted in the saline-infusion control group. Histological staining demonstrated reduced staining for connexin-43 in myocardial sections from the MMP-7–infused mice with the use of the C-terminus antibody. By immunoblotting, connexin-43 levels in the MMP-7–treated group were 56±22% lower than in the saline-treated group (P=0.23).
MMP-7 is a low-molecular-weight MMP expressed by macrophages4,5 and cardiomyocytes,6 and proteolytic substrates of MMP-7 are diverse. We examined the functional consequences of MMP-7 gene deletion on post-MI remodeling with respect to survival, echocardiographic and morphometric parameters, macrophage infiltration, and electrical conduction properties. The significant and unique findings were that MI induction in MMP-7−/− mice resulted in (1) increased early survival compared with WT mice and (2) favorable alterations in electrical conduction patterns and connexin-43 levels. Because myocardial conduction patterns were altered, along with changes in gap junction proteins, the contributing factor for differences in survival was likely a difference in electrical remodeling due to MMP-7 deletion. Surface plasmon resonance experiments revealed that MMP-7 could bind to connexin-43, and in silico, in vitro, and in vivo analyses each determined that MMP-7 is capable of mediating connexin-43 cleavage. This is the first study to demonstrate an association between changes in a particular MMP type and myocardial electrical activation patterns after MI.
MMP-7 has a large proteolytic portfolio, which includes matrix proteins, cytokines, and chemokines.21 Interestingly, we observed an increased, rather than decreased, staining for macrophages in MMP-7–null mice after MI. Increased macrophage density in MMP-7–null mice suggests that MMP-7 deletion interrupts a feedback loop that would normally curtail the continued accrual of macrophages within the infarct. Alternatively, increased macrophages in MMP-7–null mice may indicate that normal exit pathways are altered and that the increase occurs as a result of impaired regression. Future studies demonstrating the exact role of MMP-7 in mediating macrophage accumulation after MI are warranted.
In the present study MMP-7 gene deletion was associated with improved post-MI survival compared with age-matched WT mice. The WT survival rate was similar to that previously reported for C57/BL6 mice.22 Arrhythmias are an important electrogenic mechanism for sudden death after MI. Intercellular uncoupling due to a loss of cardiac gap junctions contributes to arrhythmogenesis, and slow conduction and heterogeneous repolarization are known to produce an arrhythmogenic substrate.23 Intercellular uncoupling occurs 3 to 7 days after MI in mice,24 consistent with the timing of sudden death in this study and others.25–27 A link between decreased connexin-43 levels and decreased conduction velocity has been demonstrated previously, as mice with cardiac-restricted deletion of connexin-43 display increased incidence of arrhythmias and sudden death.28 Myocardial connexin-43 expression also decreases in mice after MI19,20,23 and in patients with chronic ischemic heart disease29 or end-stage congestive heart failure.30 The fact that the MMP-7 infusion study showed a different arrhythmogenic pattern than what is typically seen after MI suggests that the infusion experiment did not fully recapitulate the post-MI setting. Because arrhythmias were not documented in the post-MI mice, a direct connection between MMP-7 and electrical remodeling remains presumed. Another limitation of the present study is that effective refractory period was not measured to determine whether MMP-7 deletion prolonged the effective refractory period after MI. Because velocity is a function of distance over time, conduction velocity measurements are potentially influenced by lines of conduction block. An increase in the time of conduction between 2 fixed points, therefore, would alter the calculated conduction velocity. Monitoring conduction time, rather than velocity, may provide a more robust measurement of altered conduction.
The present study builds on past findings by demonstrating a potential 2-fold role for MMP-7 in adverse electrical remodeling after MI. First, connexin-43 levels were higher and more localized to myocyte ends after MI in MMP-7–null mice compared with WT mice. Differential post-MI changes in connexin-43 were associated with normalization of electrical wavefront propagation in the absence of MMP-7. Second, in vitro incubation of recombinant connexin-43 fragment or myocardial extracts with MMP-7 demonstrated concentration-dependent proteolysis of connexin-43. Additionally, surface plasmon resonance was used to document a direct MMP-7 interaction with connexin-43. Although past studies document connexin-43 degradation by proteasomal- and lysosomal-dependent pathways,31 this report is the first to demonstrate connexin-43 cleavage by MMP-7. The data presented herein indicate that connexin-43 cleavage results in generation of a larger membrane-associated polypeptide and small free peptides derived from the extreme C-terminus of connexin-43. Although C-terminal truncations of connexin-43 are still competent to form connexon channels,32,33 such channels no longer interact with proteins binding the C-terminus of connexin-43. One such protein, zona occludens-1, mediates the interaction between gap junctions and the actin cytoskeleton.34,35 Therefore, signaling cascades initiating at the gap junction and dependent on zona occludens-1 interaction may be affected.
Consideration should also be given to the potential bioactivity of peptides liberated from the extreme C-terminus of connexin-43 by MMP-7 processing. Hunter et al36 demonstrated recently that a synthetic peptide based on the last 9 amino acids of connexin-43 was able to bind to zona occludens-1, inhibiting interactions with endogenous connexin-43. In turn, disruption of the connexin-43/zona occludens-1 interaction by the inhibitory peptide had striking effects on gap junction size and distribution, possibly via shifting the balance between connexons distributed in nonjunctional hemichannels and those aggregated within gap junction plaques. The present data indicate that a peptide of structure similar to the synthetic analogue characterized as bioactive by Hunter et al may be generated by MMP-7 processing of connexin-43 in vivo. MMP-7 could also affect connexin-43 levels indirectly through activation of other proteases that directly cleave connexin-43 or through disruption of matrix structure and function. This in turn would facilitate a loss of normal myocyte-myocyte interface and subsequent dissolution of connexin-43. One limitation of this study is that relatively small sample sizes were used in a few of the subgroups. Despite the fact that we noted statistical differences, more robust sample sizes with quantitative measures of connexin-43 levels and conduction parameters will be necessary. Nevertheless, this is the first study to identify a potential relationship between a specific MMP type, connexin remodeling, and electrical activation patterns.
Global and selective MMP inhibition strategies attenuate LV remodeling parameters after MI by altering inflammatory, angiogenic, and hypertrophic indices.3 That increases in LV dimensions and compensatory hypertrophy were similar between WT and MMP-7–null mice after MI suggests that additional parameters of LV remodeling, beyond what are traditionally assessed, may elucidate unique roles for individual MMPs. The model that emerges from this study involves complex interactions between the macrophage and myocyte, electrical remodeling, and connexin-43. Future studies incorporating additional MMPs and tissue inhibitors of MMPs will be necessary to fully understand MMP role(s) in the myocardial response to MI.
The authors thank David M. McClister, Jr, Joseph T. Mingoia, Jeffrey A. Sample, Julie McLean, Abigail Painter, and Sarah E. Camens for excellent technical assistance and Leslie L. Clark for statistical consultation.
Sources of Funding
This study was supported by the following grants: HL-10337 and HL-75360 (M.L.L.); HL-66029 (R.M.); HL-56728, HL-36059, and HD-39946 (R.G.G.); HL-45024, HL-97012, P01-48788, and a VA Career Development Award (F.G.S.).
Weber KT. Extracellular matrix remodeling in heart failure. Circulation. 1997; 96: 4065–4082.
Agnihotri R, Crawford HC, Haro H, Matrisian LM, Havrda MC, Liaw L. Osteopontin, a novel substrate for matrix metalloproteinase-3 (stromelysin-1) and matrix metalloproteinase-7 (matrilysin). J Biol Chem. 2001; 276: 28261–28267.
Wilson CL, Heppner KJ, Labosky PA, Hogan BL, Matrisian LM. Intestinal tumorigenesis is suppressed in mice lacking the metalloproteinase matrilysin. Proc Natl Acad Sci U S A. 1997; 94: 1402–1407.
Girouard SD, Pastore JM, Laurita KR, Gregory KW, Rosenbaum DS. Optical mapping in a new guinea pig model of ventricular tachycardia reveals mechanisms for multiple wavelengths in a single reentrant circuit. Circulation. 1996; 93: 603–613.
Thomas SP, Kucera JP, Bircher-Lehmann L, Rudy Y, Saffitz JE, Kleber AG. Impulse propagation in synthetic strands of neonatal cardiac myocytes with genetically reduced levels of connexin43. Circ Res. 2003; 92: 1209–1216.
Peters NS, Coromilas J, Severs NJ, Wit AL. disturbed connexin43 gap junction distribution correlates with the location of reentrant circuits in the epicardial border zone of healing canine infarcts that cause ventricular tachycardia. Circulation. 1997; 95: 988–996.
Matsushita T, Oyamada M, Fujimoto K, Yasuda Y, Masuda S, Wada Y, Oka T, Takamatsu T. Remodeling of cell-cell and cell-extracellular matrix interactions at the border zone of rat myocardial infarcts. Circ Res. 1999; 85: 1046–1055.
Hayashidani S, Tsutsui H, Ikeuchi M, Shiomi T, Matsusaka H, Kubota T, Imanaka-Yoshida K, Itoh T, Takeshita A. Targeted deletion of MMP-2 attenuates early LV rupture and late remodeling after experimental myocardial infarction. Am J Physiol. 2003; 285: H1229–H1235.
Berul CI. Electrophysiological phenotyping in genetically engineered mice. Physiol Genom. 2003; 13: 207–216.
Lerner DL, Beardslee MA, Saffitz JE. The role of altered intercellular coupling in arrhythmias induced by acute myocardial ischemia. Cardiovasc Res. 2001; 50: 263–269.
Feng Q, Lu X, Jones DL, Shen J, Arnold JM. Increased inducible nitric oxide synthase expression contributes to myocardial dysfunction and higher mortality after myocardial infarction in mice. Circulation. 2001; 104: 700–704.
Barandon L, Couffinhal T, Ezan J, Dufourcq P, Costet P, Alzieu P, Leroux L, Moreau C, Dare D, Duplaa C. Reduction of infarct size and prevention of cardiac rupture in transgenic mice overexpressing FrzA. Circulation. 2003; 108: 2282–2289.
Hwang MW, Matsumori A, Furukawa Y, Ono K, Okada M, Iwasaki A, Hara M, Miyamoto T, Touma M, Sasayama S. Neutralization of interleukin-1beta in the acute phase of myocardial infarction promotes the progression of left ventricular remodeling. J Am Coll Cardiol. 2001; 38: 1546–1553.
Gutstein DE, Morley GE, Tamaddon H, Vaidya D, Schneider MD, Chen J, Chien KR, Stuhlmann H, Fishman GI. Conduction slowing and sudden arrhythmic death in mice with cardiac-restricted inactivation of connexin43. Circ Res. 2001; 88: 333–339.
Saffitz JE, Kleber AG. Effects of mechanical forces and mediators of hypertrophy on remodeling of gap junctions in the heart. Circ Res. 2004; 94: 585–591.
Fernandes R, Girao H, Pereira P. High glucose down-regulates intercellular communication in retinal endothelial cells by enhancing degradation of connexin 43 by a proteasome-dependent mechanism. J Biol Chem. 2004; 279: 27219–27224.
Fishman GI, Hertzberg EL, Spray DC, Leinwand LA. Expression of connexin43 in the developing rat heart. Circ Res. 1991; 68: 782–787.
Barker RJ, Price RL, Gourdie RG. Increased association of ZO-1 with connexin43 during remodeling of cardiac gap junctions. Circ Res. 2002; 90: 317–324.
Barker RJ, Gourdie RG. Connexin interacting proteins. In: De Mello WC, Janse MJ, eds. Heart Cell Coupling and Impulse Propagation in Health and Disease. Boston, Mass: Kluwer; 2003: 25–50.
Hunter AW, Barker RJ, Zhu C, Gourdie RG. Zonula occludens-1 alters connexin43 gap junction size and organization by influencing channel accretion. Mol Biol Cell. 2005; 16: 5686–5698.
Adverse remodeling of the left ventricle after myocardial infarction (MI) remains a leading cause of congestive heart failure. The matrix metalloproteinase (MMP) family of proteolytic enzymes plays a key role in regulating the post-MI remodeling process. Although past animal studies in which global and selective MMP inhibition strategies were used attenuated left ventricular remodeling parameters after MI, key issues remain to be investigated. For example, particular MMPs influence the inflammatory response, whereas other MMPs modulate the angiogenic or hypertrophic reactions. More than 25 MMPs have been identified to date, and not every MMP has been characterized in terms of post-MI roles in the left ventricle. In the present report, we demonstrate a novel role for MMP-7 in modulating connexin-43 levels. Connexin-43 is a gap junction protein with important roles in electrical conduction. Future studies incorporating additional MMPs and tissue inhibitors of MMPs will be necessary to fully understand MMP role(s) in the myocardial response to injury. Clinical trials in which MMP inhibitors are used will likely not be successful until full consideration is given to the specificity, selectivity, and timing of inhibition of each individual MMP. Because the net effect of selectively inhibiting a particular MMP species will depend on which substrates for that MMP are also present, some of these issues may not be fully reconciled until a more complete catalog of MMP substrates is available.
The online-only Data Supplement, which includes an expanded Methods section, can be found at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.106.612960/DC1.