Combination of Tumor Necrosis Factor- Ablation and Matrix Metalloproteinase Inhibition Prevents Heart Failure After Pressure Overload in Tis
http://www.100md.com
循环研究杂志 2005年第8期
the Ontario Cancer Institute (Z.K., O.S., F.F.M., R.K.), University of Toronto, University Health Network, Canada
Heart and Stroke/Richard Lewar Center of Excellence (G.Y.O., F.D., P.P.L., P.H.B.), University of Toronto, Ontario, Canada
the School of Biological Sciences (R.K.N., D.R.E.), University of East Anglia, Norwich, Norfolk, United Kingdom.
Abstract
Cytokine and extracellular matrix (ECM) homeostasis are distinct systems that are each dysregulated in heart failure. Here we show that tissue inhibitor of metalloproteinase (TIMP)-3 is a critical regulator of both systems in a mouse model of left ventricular (LV) dilation and dysfunction. Timp-3eC/eC mice develop precipitous LV dilation and dysfunction reminiscent of dilated cardiomyopathy (DCM), culminating in early onset of heart failure by 6 weeks, compared with wild-type aortic-banding (AB). Timp-3 deficiency resulted in increased TNF converting enzyme (TACE) activity within 6 hours after AB leading to enhanced tumor necrosis factor- (TNF) processing. In addition, TNF production increased in timp-3eC/eC-AB myocardium. A significant elevation in gelatinase and collagenase activities was observed 1 week after AB, with localized ECM degradation in timp-3eC/eC-AB myocardium. Timp-3eC/eC/tnfeC/eC mice were generated and subjected to AB for comparative analyses with timp-3eC/eC-AB mice. This revealed the critical role of TNF in the early phase of LV remodeling, de novo expression of Matrix metalloproteinases (MMP)-8 in the absence of TNF, and highlighted the importance of interstitial collagenases (MMP-2, MMP-13, and MT1-MMP) for cardiac ECM degradation. Ablation of TNF, or limiting MMP activity with a synthetic MMP inhibitor (PD166793), each partially attenuated LV dilation and cardiac dysfunction in timp-3eC/eC-AB mice. Notably, combining TNF ablation with MMP inhibition completely rescued heart disease in timp-3eC/eC-AB mice. This study provides a basis for anti-TNF and MMP inhibitor combination therapy in heart disease.
Key Words: left ventricular dilation and dysfunction extracellular matrix tissue inhibitor of metalloproteinase-3 matrix metalloproteinase tumor necrosis factor-
Introduction
Cardiovascular disease is the major cause of death in the Western world and is predicted to be the leading cause of mortality worldwide by 2020.1 A close relationship between the severity of cardiac dysfunction, development of heart failure, and cardiac expression of tumor necrosis factor- (TNF) has been demonstrated.2,3 TNF is a pleiotropic cytokine and is found elevated in patients with dilated cardiomyopathy (DCM),4,5 ischemic heart disease, and congestive heart failure (CHF).3 Based on the potential importance of TNF in heart disease, anti-TNF therapy has been attempted in patients with heart failure although significant benefits of this therapy remain to be demonstrated.6,7 This suggests that other factors play key roles in the progression of heart failure. Maladaptive extracellular matrix (ECM) remodeling is a common feature of ventricular remodeling in patients with DCM and CHF.8 Matrix metalloproteinases (MMPs) are the primary ECM remodeling enzymes,9 and a disintegrin and metalloproteinase, ADAM-17/TACE (TNF converting enzyme) converts membrane bound TNF to its soluble form.10,11 Furthermore, TNF signaling is known to induce the transcription of metalloproteinases,9,12 evoking a potentially important but overlooked interaction between TNF signaling and ECM remodeling. Whether a direct relation between cytokine and ECM homeostasis operates during the progression of heart failure is currently unknown. Such an interaction could help explain the lack of efficacy of TNF-targeted therapy in heart disease. The discovery of factors that regulate both these systems is critical to designing novel treatments for heart disease.
Among tissue inhibitor of metalloproteinases (TIMPs), TIMP-3 is the only ECM-bound and the most expressed TIMP in the heart.13 The classical role of TIMP-3, as with the other TIMPs, is inhibition of a broad spectrum of MMPs.14 We have recently reported that TIMP-3 is a physiological inhibitor of TACE/ADAM-17, and regulates TNF levels in liver homeostasis and injury.15 Patients with DCM have significantly reduced levels of TIMP-3 levels,16 and elevated levels of TACE and TNF.4,5,17 We have reported that mice deficient in timp-3 develop spontaneous DCM at 21 months of age.18 In the current study, we investigated the consequence of TIMP-3 deletion in progression of heart disease following pressure overload in young mice, and dissected the relationship between cardiac TIMP-3, TACE, TNF, and MMPs. Our study uncovers the concurrent regulation of cytokine bioactivity and matrix remodeling by TIMP-3, as well as the mechanistic integration between these systems that have so far been studied independently in cardiovascular disease. We also demonstrate that deterioration of cardiac structure and function is prevented by simultaneously targeting both TNF and MMP activities. This study forms the basis for anti-TNF and MMPi combination therapy for human heart disease, which individually provide only a partial rescue.
Materials and Methods
Pressure Overload and Cardiac Function
Eight-week-old wild-type (WT), timp-3eC/eC, or TNFeC/eC/timp-3eC/eC mice were subjected to pressure overload by constriction of descending aorta. Sham-operated mice from each group served as controls. In vivo Cardiac function was monitored by echocardiographic imaging, and confirmed by hemodynamic measurements. All animals were cared for in accordance with the Toronto Community Care Assess Centre (CCAC) guidelines. Complete Materials and Methods may be found in the online data supplement at http://circres.ahajournals.org.
Confocal and Electron Microscopy of Matrix Structure and Immunohistochemistry
Picocirius red-stained sections were visualized using 2-photon confocal microscopy. Electron microscopy was performed with a FEI CM100 Biotwin electron microscope. Apoptosis was assessed by TUNEL. Neutrophils were stained using rat anti-neutrophil.
Protein Analysis and Enzymatic Activity
Total gelatinase or collagenase activity was measured in myocardial homogenates using EnzChek assay kit using collagen type I as the substrate for collagenase activity. Pro- and active MMP-2 and MMP-9 were detected by gelatin zymography. TACE activity and Western blotting for TNF was performed as described.15
TaqMan RT-PCR Analysis
RNA levels for the indicated genes were quantified by Real-time TaqMan RT-PCR.13 18S rRNA was used as an endogenous control. Details of sequences are in the online Materials and Methods.
In Vivo MMPi Treatment
An MMP-specific inhibitor, PD166793 (Pfizer Inc) was administered daily by gavage. PD166793 treatment (15 mg/kg/d) began 1 week before aortic-banding and continued until mice were euthanized.
An expanded Materials and Methods section is provided in the online data supplement.
Statistical Analysis
Survival between groups was compared by Kaplan-Meier survival analysis. All other comparisons were performed by ANOVA followed by multiple comparison testing (Student-Neuman Keuls). Values are reported as Mean±SEM. Statistical significance is recognized at P<0.05.
Results
Mortality Attributable to Heart Failure in TIMP-3eCDeficient Mice
Timp-3eC/eC mice were subjected to cardiac pressure overload by aortic-banding (AB). These mice were compared with 3 groups of control littermates throughout the study, aortic-banded WT, sham-operated WT, and sham-operated timp-3eC/eC mice. Aortic-banding generated comparable pressure gradient (59 to 64 mm Hg) independent of the genotype, but caused significantly higher mortality in timp-3eC/eC mice compared with WT (Figure 1A). Six weeks following AB timp-3eC/eC mice exhibited significant morbidity and had to be euthanized. Timp-3eC/eC hearts were grossly enlarged within 1 week after AB and had increased left ventricular (LV) chamber size (Figure 1B).
To characterize the effects of pressure overload on WT and timp-3eC/eC hearts, we analyzed the cardiac function by echocardiography (Table). In WT-AB mice, significant cardiac dysfunction was seen at 6 weeks. In contrast, as early as 1 week after AB we observed dramatic LV dilation, reduced aortic outflow velocity, cardiac contractility, and velocity of circumferential fiber shortening in timp-3eC/eC-AB mice. These parameters deteriorated progressively over 6 weeks (Figure 1C and 1D; Table). Notably, the extent of cardiac dysfunction in timp-3eC/eC mice after 6 weeks was comparable to WT mice after 12 weeks of aortic-banding (Table). Cardiac dysfunction was confirmed by in vivo hemodynamic measurements. Timp-3eC/eC-AB mice had lower LV developed pressure (LVDP), higher end-diastolic pressure (LVEDP), and markedly suppressed LV peak rates of pressure-rise and pressure-fall (±dP/dt) compared with WT-AB (Figure 1E). Reduced eCdP/dt and increased LVEDP suggest diastolic dysfunction in timp-3eC/eC mice. Terminal heart failure in patients is associated with pulmonary congestion. We also observed pulmonary edema and fibrosis in timp-3eC/eC-AB mice at 6 weeks (online Figure I). Altogether, advanced cardiac dysfunction as indicated by the extreme LV dilation and suppressed function in combination with pulmonary congestion demonstrates that unlike the WT mice, timp-3eC/eC mice develop rapid CHF 6 weeks after pressure overload. These features of cardiac dysfunction are reminiscent of DCM in patients.
Excess Hypertrophy with Timp-3 Loss
Next, we examined cardiomyocyte apoptosis that generally underlies LV dilatation. TIMP-3 deletion and overexpression are both linked to apoptosis.20,21 Apoptosis in timp-3eC/eC-AB LV was significantly higher than WT-AB at 6 weeks, suggesting that it is not responsible for the onset of LV dilation in these mice (Figure 2A and 2B). Baseline apoptosis was not different between sham groups. We investigated if myocyte hypertrophy occurred as a compensatory response to biomechanical stress. Heart weight-to-tibial length ratio was far greater in timp-3eC/eC-AB than in WT-AB mice. We also measured myocyte cross-sectional area, length, and expression of prototypic markers of hypertrophy and heart disease (atrial natriuretic factor, ANF; brain natriuretic peptide, BNP). Although WT mice developed significant hypertrophy over the 6-week period, timp-3eC/eC-AB mice showed markedly greater hypertrophy as determined by higher levels of all hypertrophy parameters compared with WT-AB (Figure 2C through 2G). Increased myocyte cross-sectional area and length indicate a combination of eccentric and concentric hypertrophy in timp-3eC/eC-AB hearts, whereas WT-AB hearts exhibit only concentric hypertrophy. This can explain LV dilation in timp-3eC/eC-AB hearts in the absence of reduced LV wall thickness.
Excess TNF and TACE Activity in Timp-3eC/eC-AB Mice
In aged timp-3eC/eC mice, we proposed involvement of TNF in the development of spontaneous DCM.18 In a liver injury model, we have shown TIMP-3 is a physiological regulator of TNF processing through inhibition of ADAM-17/TACE.15 In pursuit of the mechanism underlying the early onset of severe heart disease following pressure overload in young timp-3eC/eC mice, we asked whether cardiac expression of the TIMP3-TACE-TNF axis was affected. In WT hearts, TIMP-3 and TACE mRNA levels underwent parallel temporal changes after aortic-banding peaking at 3 weeks (Figure 3A and 3B). In contrast, in timp-3eC/eC hearts, TACE expression increased immediately after AB (within 6 hours, 1.5-fold, Figure 3B) concomitant with a 4-fold transient rise in TNF mRNA (Figure 3C). TACE activity also increased within 6 hours (1.7-fold) in timp-3eC/eC-AB hearts, whereas it rose much later (3 weeks) in WT-AB hearts (Figure 3D), temporally consistent with the increase in TACE expression. Addition of recombinant TIMP-3 (rTIMP-3), but not rTIMP-1 or a synthetic MMP-specific inhibitor (PD166793), inhibited the enhanced cardiac TACE activity in timp-3eC/eC mice (Figure 3E). TIMP-3-specific inhibition is considered signature of TACE.22 TACE processes membrane-bound 26-kDa TNF to its soluble 17-kDa species. Membrane-bound TNF levels were higher in timp-3eC/eC compared with WT sham hearts. Following pressure overload, elevated membrane-bound and processed TNF proteins were detected in timp-3eC/eC hearts (Figure 3F). Overall, along with a rapid increase in TACE activity, TACE and TNF productions were accelerated and enhanced in timp-3eC/eC-AB compared with WT-AB hearts, leading to amplified activity of this pathway.
Activation of the MMP Axis Perturbs ECM in Timp-3eC/eC Mice
TIMP-3 is known to inhibit a broad range of MMPs.14 We determined whether TIMP-3 loss allowed for altered MMP activity in response to pressure overload. Total gelatinase and collagenase activities markedly increased 1 week post-AB in timp-3eC/eC cardiac extracts compared with WT that displayed only collagenase induction (Figure 4A). The increased gelatinase activity in the timp-3eC/eC-AB group was reduced by rTIMP-3 and PD166793, although collagenase activity was additionally lowered by rTIMP-1 (Figure 4A). In search of specific candidates for the increased MMP activity, we detected active MMP-2 in timp-3eC/eC hearts beginning 1 week after AB, with both latent and active MMP-2 increasing over 6-week period (Figure 4B). Next, we investigated the trimolecular complex, comprised of MMP-2, TIMP-2, and MMP-14/MT1-MMP, typically required for MMP-2 activation.23 mRNA levels of MMP-2 and MT1-MMP were elevated in timp-3eC/eC-AB hearts (Figure 4C). In addition, we found a striking increase of MMP-13 in timp-3eC/eC-AB myocardium (Figure 4C). MMP-7 and MMP-8 were expressed minimally and remained unchanged post-AB (data not shown). MMP-9 mRNA increased over 6 weeks with no significant difference between WT and timp-3eC/eC hearts. Hence, 3 key interstitial collagenases,19,24,25 MMP-2, MMP-13, and MT1-MMP, were elevated in response to pressure overload in timp-3eC/eC mice. Further, soluble myocardial TIMPs (TIMP-1, TIMP-2, or TIMP-4) were not elevated to compensate for the absence of ECM-bound TIMP-3 in sham-operated timp-3eC/eC hearts, although timp-1 expression increased significantly at 6 weeks post-AB (Figure 4C). Altogether, these data indicate that cardiac pressure overload in the absence of timp-3 leads to parallel but temporally differential activation of the 2 metalloproteinase axes, MMPs and ADAMs.
Next, we determined whether the enhanced MMP activity affected the myocardial matrix. Using confocal microscopy on Picocirius red-stained sections and scanning electron microscopy, we examined the integrity of the fibrillar component of myocardial ECM that provides normal structural support. It is primarily comprised of collagen types I and III.26 Intact ECM network and uniform collagen distribution were seen in WT-AB hearts (Figure 4D and 4F), whereas we found areas devoid of or with disrupted fibrillar collagens in timp-3eC/eC-AB myocardium (Figure 4E and 4G). Although WT-AB hearts showed interstitial fibrosis 6 weeks after AB, localized fibrotic areas were noticeably increased in timp-3eC/eC-AB myocardium (data not shown). These findings are consistent with the maladaptive ECM remodeling found in patients with LV dilation.27
Genetic Deletion of TNF Attenuates Cardiac Disease in Timp-3eC/eC-AB Hearts
To determine whether TNF is a critical molecule for LV dilation and dysfunction in timp-3eC/eC mice, we generated timp-3eC/eC/tnfeC/eC double mutants. After AB, disease was delayed and partially rescued (Figure 5), while survival was improved markedly (Figure 6A) in the double knock-outs compared with timp-3eC/eC mice. Specifically, after only 3 weeks LV chamber size increased and cardiac contractility declined in timp-3eC/eC/tnfeC/eC-AB mice, but was significantly attenuated versus timp-3eC/eC-AB mice (Figure 5A through 5C). Cardiac hypertrophy in the double mutants also diverged from WT-AB group at 3 weeks, but remained far less pronounced than timp-3eC/eC-AB hearts until 6 weeks (Figure 5D and 5E). No myocyte apoptosis (Figure 5H) or pulmonary congestion (Figure 5F and 5G) was detected in double mutants. Hence, genetically coupling TNF ablation with TIMP-3 loss prevented CHF and attenuated heart disease in timp-3eC/eC-AB mice.
Dependency of MMP Activity on TNF
Because TNF is a transcriptional inducer of MMPs,9,12,28 we explored the TNF-dependency of MMP expression by comparing MMP activity and expression profiles between timp-3eC/eC/tnfeC/eC-AB and timp-3eC/eC-AB myocardium. Surprisingly, collagenase but not gelatinase activity was significantly higher in timp-3eC/eC/tnfeC/eC-AB than in timp-3eC/eC-AB hearts, with both activities remaining higher than WT-AB (Figure 5I and 5J). We further investigated contribution of specific MMPs in these mice (online Figure II). Overall, MMP-2 activation was abrogated with loss of TNF, but the expression of MMP-13 and MT1-MMP were only partially affected.
De Novo MMP-8 Induction in the Absence of TNF and TIMP-3
Despite the reduced transcriptional induction of specific interstitial collagenases, total myocardial collagenase activity was still higher in timp-3eC/eC/tnfeC/eC-AB than timp-3eC/eC-AB. Screening additional MMPs revealed that expression of MMP-8, another interstitial collagenase,29 was elevated 5-fold in timp-3eC/eC/tnfeC/eC-AB compared with timp-3eC/eC-AB and WT-AB hearts (Figure 5K). Timp-3eC/eC/tnfeC/eC-AB myocardium had significantly higher numbers of neutrophils scattered throughout the LV compared with timp-3eC/eC-AB and WT-AB myocardium (Figure 5L and 5 mol/L). Neutrophils are the primary source of MMP-8.29 The greater increase in MMP-8 RNA levels than in neutrophil numbers may be attributable to the activation state of the neutrophils present in the myocardium. Thus, neutrophil infiltration and de novo MMP-8 induction on loss of TNF likely contributed to elevated collagenase activity in the double mutant mice following pressure overload. We found no increase in macrophage numbers in timp-3eC/eC/tnfeC/eC-AB hearts by immunostaining with Mac3 or F4/80 (data not shown).
MMPi Treatment Attenuates LV Dilation and Dysfunction in Timp-3eC/eC-AB Mice
Because total gelatinase and collagenase activities were elevated in timp-3eC/eC mice following pressure overload, we determined the contribution of this increased MMP activity toward DCM and heart failure, by using a broad-spectrum MMP-specific inhibitor (PD166793, MMPi) that does not inhibit TACE.30 Oral administration of 15 mg/kg/d of PD166793 resulted in plasma drug concentration of 110±4 eol/L, a level reported to be sufficient to inhibit MMP activity.30 This level of PD166793 was sufficient to inhibit the excessive myocardial collagenase and gelatinase activities in timp-3eC/eC-AB cardiac homogenates in vitro (Figure 4A). MMPi treatment of sham-operated WT mice did not affect cardiac function or structure, and no systemic toxicity was detected by histology of liver, kidney or lung (data not shown). During the course of 7-week MMPi treatment, mice in all groups remained active and showed no weight loss. Interestingly, treatment of timp-3eC/eC-AB mice with MMPi abolished the interstitial fibrosis (data not shown), prevented the early heart failure and resulted in partial prevention of LV dilation and dysfunction (Figure 6B).
Combination of MMPi Treatment and TNF Ablation Completely Prevents Heart Disease
We found that TNF ablation or MMP inhibition individually resulted in partial prevention of DCM in timp-3eC/eC-AB mice. Therefore, we reasoned that limiting MMP activity in addition to TNF elimination should completely prevent heart disease in these mice. Treatment of timp-3eC/eC/tnfeC/eC-AB mice with 15 mg/kg/d of MMPi strikingly improved the survival (Figure 6A), completely rescued LV dilation and cardiac dysfunction as indicated by echocardiography and in vivo hemodynamics (Figure 6B and 6C), and hypertrophy (Figure 6D and 6E) up to 6 weeks after AB. These data demonstrate that excessive concomitant activity of both TNF and MMPs are responsible for the cardiac dysfunction and heart failure in timp-3eC/eC-AB mice.
Discussion
Dysregulations in proinflammatory cytokines and structural ECM underlie human heart disease.8,26 TNF has pleiotropic functions in cardiovascular diseases.17,31 Maladaptive matrix remodeling is equally well recognized in promoting abnormal cardiac structure and function. These distinct disciplines have been studied extensively, but mostly independent of each other. We show here that timp-3 provides regulatory crossover for these distinct fields. TIMP-3 serves a dual inhibitory function in the heart against ADAM and MMP metalloproteinases to couple cytokine bioactivity with ECM homeostasis. Timp-3 deficiency results in severe LV dilation and dysfunction in response to cardiac pressure overload, which is partially rescued by TNF deletion or by MMPi treatment, and completely prevented on combining TNF ablation with MMP inhibition. These data demonstrate that TIMP-3 is essential in cardiac recovery from mechanical stress.
Our earlier study suggested involvement of TNF in the development of spontaneous DCM in aged timp-3eC/eC mice as the soluble TNF and its receptor (P75) levels were enhanced in coronary effluent from aged hearts.18 The current study is an in depth investigation of the relationship between TIMP-3, TACE, TNF, and MMPs in a pressure overload model of heart disease. TIMP-3 is the most highly expressed TIMP in the murine heart13 and is the only known physiological inhibitor of ADAM-17/TACE.15,22 TACE and TIMP-3 have parallel baseline expression patterns in murine organs during development.32 We show that on pressure overload, TACE and TIMP-3 undergo parallel temporal inductions in WT hearts, a pattern suggesting that TIMP-3 normally serves to counteract increased TACE levels. In TIMP-3eCdeficient hearts, not only is there a dramatic immediate increase in the transcriptions of both TACE and TNF, but their increased activities go unchecked in the absence of TIMP-3. TNF feeds back positively on its own expression as well as that of TACE.28 TNF also induces TIMP-1 but downregulates TIMP-3 expression in myocyte cultures.33,34 Such downregulation facilitates TNF processing through increased TACE activity. Hence, the TIMP-3-TACE-TNF system provides an inherent regulatory mechanism for controlling TNF bioactivity following cardiac biomechanical stress.
TIMP-3 is classically known to inhibit MMP activity responsible for ECM turnover. Total gelatinase and collagenase activities rise markedly in timp-3eC/eC compared with WT hearts following aortic-banding. Screening of multiple MMPs reveals higher MMP-2, MMP-13 and MT1-MMP mRNA levels and higher MMP-2 activity. MMP-2 degrades several ECM proteins,19,35 whereas MMP-2, MMP-13, and MT1-MMP are established interstitial collagenases.19,25,36 MT1-MMP is also a key molecule for pro-MMP-2 and pro-MMP-13 activation.24,37 MMP-2, MMP-13, and MT1-MMP are elevated in patients with heart disease,38,39 and MMP-2 polymorphisms in human have been linked to CHF.40 Thus, interstitial collagenases, MMP-2, MMP-13, and MT1-MMP appear to be intimately involved in myocardial ECM remodeling following biomechanical stress.
TNF is well recognized for its pathophysiological importance in human heart disease.4,17 It reduces cardiac contractility,41 induces hypertrophy42 and apoptosis.43 In patients with DCM, both TACE and TNF levels are significantly elevated5 as in our aortic-banded timp-3eC/eC mice. Membrane-bound TNF mediates cardiac hypertrophy whereas cleaved TNF can induce DCM,44 both detected in timp-3eC/eC-AB mice. TNF is also a key transcriptional inducer of several MMPs in vitro.9,12 Our in vivo comparison of MMP expression in timp-3eC/eC-AB and timp-3eC/eC/tnfeC/eC-AB hearts provides evidence that TNF induces MMP-2 and MMP-13, but minimally affects MMP-9 and MT1-MMP transcription. The de novo induction of MMP-8 in timp-3eC/eC/tnfeC/eC-AB hearts is consistent with increased neutrophil infiltration in these mice, which intriguingly occurred without an increase in MMP-9, also known to be produced by neutrophils. MMP-8 is found decreased in human heart disease,45 and its role as an interstitial collagenase in the myocardium needs to be further understood. Our data illustrate that TNF bioactivity is intertwined with the amplification of MMP activity. The complete lack of myocardial apoptosis in double mutants suggests a proapoptotic role of TNF in this model of heart disease. Understanding the multiple functions of TNF in timp-3eC/eC myocardium will help us better understand its roles patients with DCM.
We demonstrate that treatment with MMPi in combination with TNF loss completely prevents LV dilation and dysfunction, and heart failure in aortic-banded timp-3eC/eC mice, although only partial rescue is achieved with the individual interventions. This indicates that matrix degradation and cytokines are together responsible for heart disease. The anti-TNF therapy in patients with class III and IV heart failure has so far proven unsuccessful.6,7 Based on our findings, the lack of success with anti-TNF therapy6,7 may be attributable to application at too late a stage or the requirement for concomitant targeting of MMPs. Because TIMP-3 lies upstream of these 2 systems, and is commonly reduced in patients with cardiomyopathy, we propose that a therapeutic strategy that mimics TIMP-3 function in targeting both the MMPs and TNF will be useful in producing an effective cardiac response to mechanical stress. Our findings provide a basis for anti-TNF and MMPi combination therapy in heart disease.
Acknowledgments
Z.K. is supported by Heart and Stroke Foundation of Canada post-doctoral fellowship. This research was supported by Canadian Institutes of Health Research and Heart and Stroke Foundation of Canada to R.K. G.Y.O. is a clinician/scientist supported by Canadian Institutes of Health Research and Tailored Advanced Collaborative Training in Cardiovascular Science fellowships and P.P.L. is a H&S/Polo Chair. The authors thank Geoffrey Woods for critical reading of this manuscript.
References
Olson EN. A decade of discoveries in cardiac biology. Nat Med. 2004; 10: 467eC474.
Feldman AM, McTiernan C. Is there any future for tumor necrosis factor antagonists in chronic heart failure Am J Cardiovasc Drugs. 2004; 4: 11eC19.
Feldman AM, Combes A, Wagner D, Kadakomi T, Kubota T, Li YY, McTiernan C. The role of tumor necrosis factor in the pathophysiology of heart failure. J Am Coll Cardiol. 2000; 35: 537eC544.
Torre-Amione G, Kapadia S, Benedict C, Oral H, Young JB, Mann DL. Proinflammatory cytokine levels in patients with depressed left ventricular ejection fraction: a report from the Studies of Left Ventricular Dysfunction (SOLVD). J Am Coll Cardiol. 1996; 27: 1201eC1206.
Satoh M, Nakamura M, Saitoh H, Satoh H, Maesawa C, Segawa I, Tashiro A, Hiramori K. TACE and TNF in human dilated cardiomyopathy. Circulation. 1999; 99: 3260eC3265.
Chung ES, Packer M, Lo KH, Fasanmade AA, Willerson JT. Randomized, double-blind, placebo-controlled, pilot trial of infliximab, a chimeric monoclonal antibody to TNF, in patients with moderate-to-severe heart failure: results of the anti-TNF Therapy Against Congestive Heart Failure (ATTACH) trial. Circulation. 2003; 107: 3133eC3140.
Mann DL, McMurray JJ, Packer M, Swedberg K, Borer JS, Colucci WS, Djian J, Drexler H, Feldman A, Kober L, Krum H, Liu P, Nieminen M, Tavazzi L, van Veldhuisen DJ, Waldenstrom A, Warren M, Westheim A, Zannad F, Fleming T. Targeted anticytokine therapy in patients with chronic heart failure: results of the Randomized Etanercept Worldwide Evaluation (RENEWAL). Circulation. 2004; 109: 1594eC1602.
Gunja-Smith Z, Morales AR, Romanelli R, Woessner JF Jr. Remodeling of human myocardial collagen in idiopathic dilated cardiomyopathy. Role of metalloproteinases and pyridinoline cross-links. Am J Pathol. 1996; 148: 1639eC1648.
Overall CM, Lopez-Otin C. Strategies for MMP inhibition in cancer: innovations for the post-trial era. Nat Rev Cancer. 2002; 2: 657eC672.
Zheng Y, Saftig P, Hartmann D, Blobel C. Evaluation of the contribution of different ADAMs to TNF shedding and of the function of the TNF ectodomain in ensuring selective stimulated shedding by the TNF convertase (TACE/ADAM17). J Biol Chem. 2004; 279: 42898eC42906.
Black RA, Rauch CT, Kozlosky CJ, Peschon JJ, Slack JL, Wolfson MF, Castner BJ, Stocking KL, Reddy P, Srinivasan S, Nelson N, Boiani N, Schooley KA, Gerhart M, Davis R, Fitzner JN, Johnson RS, Paxton RJ, March CJ, Cerretti DP. A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells. Nature. 1997; 385: 729eC733.
Siwik DA, Colucci WS. Regulation of matrix metalloproteinases by cytokines and reactive oxygen/nitrogen species in the myocardium. Heart Fail Rev. 2004; 9: 43eC51.
Young DA, Phillips BW, Lundy C, Nuttall RK, Hogan A, Schultz GA, Leco KJ, Clark IM, Edwards DR. Identification of an initiator-like element essential for the expression of the tissue inhibitor of metalloproteinases-4 gene. Biochem J. 2002; 364: 89eC99.
Gomez DE, Alonso DF, Yoshiji H, Thorgeirsson UP. Tissue inhibitors of metalloproteinases: structure, regulation and biological functions. Eur J Cell Biol. 1997; 74: 111eC122.
Mohammed FF, Smookler DS, Taylor SE, Fingleton B, Kassiri Z, Sanchez OH, English JL, Matrisian LM, Au B, Yeh WC, Khokha R. Abnormal TNF activity in Timp3eC/eC mice leads to chronic hepatic inflammation and failure of liver regeneration. Nat Genet. 2004; 36: 969eC977.
Li YY, Feldman AM, Sun Y, McTiernan CF. Differential expression of tissue inhibitors of metalloproteinases in the failing human heart. Circulation. 1998; 98: 1728eC1734.
Levine B, Kalman J, Mayer L, Fillit HM, Packer M. Elevated circulating levels of tumor necrosis factor in severe chronic heart failure. N Engl J Med. 1990; 323: 236eC241.
Fedak PW, Smookler DS, Kassiri Z, Ohno N, Leco KJ, Verma S, Mickle DA, Watson KL, Hojilla CV, Cruz W, Weisel RD, Li RK, Khokha R. TIMP-3 deficiency leads to dilated cardiomyopathy. Circulation. 2004; 110: 2401eC2409.
Aimes RT, Quigley JP. Matrix metalloproteinase-2 is an interstitial collagenase. Inhibitor-free enzyme catalyzes the cleavage of collagen fibrils and soluble native type I collagen generating the specific 3/4- and 1/4-length fragments. J Biol Chem. 1995; 270: 5872eC5876.
Fata JE, Leco KJ, Voura EB, Yu HY, Waterhouse P, Murphy G, Moorehead RA, Khokha R. Accelerated apoptosis in the Timp-3-deficient mammary gland. J Clin Invest. 2001; 108: 831eC841.
Baker AH, George SJ, Zaltsman AB, Murphy G, Newby AC. Inhibition of invasion and induction of apoptotic cell death of cancer cell lines by overexpression of TIMP-3. Br J Cancer. 1999; 79: 1347eC1355.
Amour A, Slocombe PM, Webster A, Butler M, Knight CG, Smith BJ, Stephens PE, Shelley C, Hutton M, Knauper V, Docherty AJ, Murphy G. TNF- converting enzyme is inhibited by TIMP-3. FEBS Lett. 1998; 435: 39eC44.
Strongin AY, Collier I, Bannikov G, Marmer BL, Grant GA, Goldberg GI. Mechanism of cell surface activation of 72-kDa type IV collagenase. Isolation of the activated form of the membrane metalloprotease. J Biol Chem. 1995; 270: 5331eC5338.
Knauper V, Will H, Lopez-Otin C, Smith B, Atkinson SJ, Stanton H, Hembry RM, Murphy G. Cellular mechanisms for human procollagenase-3 (MMP-13) activation. Evidence that MT1-MMP (MMP-14) and gelatinase a (MMP-2) are able to generate active enzyme. J Biol Chem. 1996; 271: 17124eC17131.
Knauper V, Lopez-Otin C, Smith B, Knight G, Murphy G. Biochemical characterization of human collagenase-3. J Biol Chem. 1996; 271: 1544eC1550.
Kassiri Z, Khokha R. Myocardial ECM and its regulation by metalloproteinases and their inhibitors. Thromb Haemost. 2005; 93: 212eC219.
Lindsey ML, Mann DL, Entman ML, Spinale FG. Extracellular matrix remodeling following myocardial injury. Ann Med. 2003; 35: 316eC326.
Bzowska M, Jura N, Lassak A, Black RA, Bereta J. TNF stimulates expression of TNF- converting enzyme in endothelial cells. Eur J Biochem. 2004; 271: 2808eC2820.
Owen CA, Hu Z, Lopez-Otin C, Shapiro SD. Membrane-bound MMP-8 on activated polymorphonuclear cells is a potent, TIMP-resistant collagenase and serpinase. J Immunol. 2004; 172: 7791eC7803.
Peterson JT, Hallak H, Johnson L, Li H, O’Brien PM, Sliskovic DR, Bocan TM, Coker ML, Etoh T, Spinale FG. Matrix metalloproteinase inhibition attenuates left ventricular remodeling and dysfunction in a rat model of progressive heart failure. Circulation. 2001; 103: 2303eC2309.
Torre-Amione G, Kapadia S, Lee J, Durand JB, Bies RD, Young JB, Mann DL. Tumor necrosis factor-alpha and tumor necrosis factor receptors in the failing human heart. Circulation. 1996; 93: 704eC711.
Nuttall RK, Sampieri CL, Pennington CJ, Gill SE, Schultz GA, Edwards DR. Expression analysis of the entire MMP and TIMP gene families during mouse tissue development. FEBS Lett. 2004; 563: 129eC134.
Alexander JP, Acott TS. Involvement of protein kinase C in TNF regulation of trabecular matrix metalloproteinases and TIMPs. Invest Ophthalmol Vis Sci. 2001; 42: 2831eC2838.
Li YY, McTiernan CF, Feldman AM. Proinflammatory cytokines regulate tissue inhibitors of metalloproteinases and disintegrin metalloproteinase in cardiac cells. Cardiovasc Res. 1999; 42: 162eC172.
Matrisian LM. The matrix-degrading metalloproteinases. Bioessays. 1992; 14: 455eC463.
Holmbeck K, Bianco P, Birkedal-Hansen H. MT1-mmp: a collagenase essential for tumor cell invasive growth. Cancer Cell. 2003; 4: 83eC84.
Pei D, Weiss SJ. Transmembrane-deletion mutants of the MT1-MMP process progelatinase A and express intrinsic matrix-degrading activity. J Biol Chem. 1996; 271: 9135eC9140.
Spinale FG, Coker ML, Heung LJ, Bond BR, Gunasinghe HR, Etoh T, Goldberg AT, Zellner JL, Crumbley AJ. A matrix metalloproteinase induction/activation system exists in the human left ventricular myocardium and is upregulated in heart failure. Circulation. 2000; 102: 1944eC1949.
Spinale FG. Matrix metalloproteinases: regulation and dysregulation in the failing heart. Circ Res. 2002; 90: 520eC530.
Vasku A, Goldbergova M, Holla LI, Spinarova L, Spinar J, Vitovec J, Vacha J. Two MMP-2 promoter polymorphisms (-790T/G and -735C/T) in chronic heart failure. Clin Chem Lab Med. 2003; 41: 1299eC1303.
Oral H, Dorn GW 2nd, Mann DL. Sphingosine mediates the immediate negative inotropic effects of tumor necrosis factor-alpha in the adult mammalian cardiac myocyte. J Biol Chem. 1997; 272: 4836eC4842.
Dibbs ZI, Diwan A, Nemoto S, DeFreitas G, Abdellatif M, Carabello BA, Spinale FG, Feuerstein G, Sivasubramanian N, Mann DL. Targeted overexpression of transmembrane tumor necrosis factor provokes a concentric cardiac hypertrophic phenotype. Circulation. 2003; 108: 1002eC1008.
Engel D, Peshock R, Armstong RC, Sivasubramanian N, Mann DL. Cardiac myocyte apoptosis provokes adverse cardiac remodeling in transgenic mice with targeted TNF overexpression. Am J Physiol Heart Circ Physiol. 2004; 287: H1303eC11.
Diwan A, Dibbs Z, Nemoto S, DeFreitas G, Carabello BA, Sivasubramanian N, Wilson EM, Spinale FG, Mann DL. Targeted overexpression of noncleavable and secreted forms of TNF provokes disparate cardiac phenotypes. Circulation. 2004; 109: 262eC268.
Wilson EM, Gunasinghe HR, Coker ML, Sprunger P, Lee-Jackson D, Bozkurt B, Deswal A, Mann DL, Spinale FG. Plasma MMPs and inhibitor profiles in patients with heart failure. J Card Fail. 2002; 8: 390eC398.(Zamaneh Kassiri, Gavin Y.)
Heart and Stroke/Richard Lewar Center of Excellence (G.Y.O., F.D., P.P.L., P.H.B.), University of Toronto, Ontario, Canada
the School of Biological Sciences (R.K.N., D.R.E.), University of East Anglia, Norwich, Norfolk, United Kingdom.
Abstract
Cytokine and extracellular matrix (ECM) homeostasis are distinct systems that are each dysregulated in heart failure. Here we show that tissue inhibitor of metalloproteinase (TIMP)-3 is a critical regulator of both systems in a mouse model of left ventricular (LV) dilation and dysfunction. Timp-3eC/eC mice develop precipitous LV dilation and dysfunction reminiscent of dilated cardiomyopathy (DCM), culminating in early onset of heart failure by 6 weeks, compared with wild-type aortic-banding (AB). Timp-3 deficiency resulted in increased TNF converting enzyme (TACE) activity within 6 hours after AB leading to enhanced tumor necrosis factor- (TNF) processing. In addition, TNF production increased in timp-3eC/eC-AB myocardium. A significant elevation in gelatinase and collagenase activities was observed 1 week after AB, with localized ECM degradation in timp-3eC/eC-AB myocardium. Timp-3eC/eC/tnfeC/eC mice were generated and subjected to AB for comparative analyses with timp-3eC/eC-AB mice. This revealed the critical role of TNF in the early phase of LV remodeling, de novo expression of Matrix metalloproteinases (MMP)-8 in the absence of TNF, and highlighted the importance of interstitial collagenases (MMP-2, MMP-13, and MT1-MMP) for cardiac ECM degradation. Ablation of TNF, or limiting MMP activity with a synthetic MMP inhibitor (PD166793), each partially attenuated LV dilation and cardiac dysfunction in timp-3eC/eC-AB mice. Notably, combining TNF ablation with MMP inhibition completely rescued heart disease in timp-3eC/eC-AB mice. This study provides a basis for anti-TNF and MMP inhibitor combination therapy in heart disease.
Key Words: left ventricular dilation and dysfunction extracellular matrix tissue inhibitor of metalloproteinase-3 matrix metalloproteinase tumor necrosis factor-
Introduction
Cardiovascular disease is the major cause of death in the Western world and is predicted to be the leading cause of mortality worldwide by 2020.1 A close relationship between the severity of cardiac dysfunction, development of heart failure, and cardiac expression of tumor necrosis factor- (TNF) has been demonstrated.2,3 TNF is a pleiotropic cytokine and is found elevated in patients with dilated cardiomyopathy (DCM),4,5 ischemic heart disease, and congestive heart failure (CHF).3 Based on the potential importance of TNF in heart disease, anti-TNF therapy has been attempted in patients with heart failure although significant benefits of this therapy remain to be demonstrated.6,7 This suggests that other factors play key roles in the progression of heart failure. Maladaptive extracellular matrix (ECM) remodeling is a common feature of ventricular remodeling in patients with DCM and CHF.8 Matrix metalloproteinases (MMPs) are the primary ECM remodeling enzymes,9 and a disintegrin and metalloproteinase, ADAM-17/TACE (TNF converting enzyme) converts membrane bound TNF to its soluble form.10,11 Furthermore, TNF signaling is known to induce the transcription of metalloproteinases,9,12 evoking a potentially important but overlooked interaction between TNF signaling and ECM remodeling. Whether a direct relation between cytokine and ECM homeostasis operates during the progression of heart failure is currently unknown. Such an interaction could help explain the lack of efficacy of TNF-targeted therapy in heart disease. The discovery of factors that regulate both these systems is critical to designing novel treatments for heart disease.
Among tissue inhibitor of metalloproteinases (TIMPs), TIMP-3 is the only ECM-bound and the most expressed TIMP in the heart.13 The classical role of TIMP-3, as with the other TIMPs, is inhibition of a broad spectrum of MMPs.14 We have recently reported that TIMP-3 is a physiological inhibitor of TACE/ADAM-17, and regulates TNF levels in liver homeostasis and injury.15 Patients with DCM have significantly reduced levels of TIMP-3 levels,16 and elevated levels of TACE and TNF.4,5,17 We have reported that mice deficient in timp-3 develop spontaneous DCM at 21 months of age.18 In the current study, we investigated the consequence of TIMP-3 deletion in progression of heart disease following pressure overload in young mice, and dissected the relationship between cardiac TIMP-3, TACE, TNF, and MMPs. Our study uncovers the concurrent regulation of cytokine bioactivity and matrix remodeling by TIMP-3, as well as the mechanistic integration between these systems that have so far been studied independently in cardiovascular disease. We also demonstrate that deterioration of cardiac structure and function is prevented by simultaneously targeting both TNF and MMP activities. This study forms the basis for anti-TNF and MMPi combination therapy for human heart disease, which individually provide only a partial rescue.
Materials and Methods
Pressure Overload and Cardiac Function
Eight-week-old wild-type (WT), timp-3eC/eC, or TNFeC/eC/timp-3eC/eC mice were subjected to pressure overload by constriction of descending aorta. Sham-operated mice from each group served as controls. In vivo Cardiac function was monitored by echocardiographic imaging, and confirmed by hemodynamic measurements. All animals were cared for in accordance with the Toronto Community Care Assess Centre (CCAC) guidelines. Complete Materials and Methods may be found in the online data supplement at http://circres.ahajournals.org.
Confocal and Electron Microscopy of Matrix Structure and Immunohistochemistry
Picocirius red-stained sections were visualized using 2-photon confocal microscopy. Electron microscopy was performed with a FEI CM100 Biotwin electron microscope. Apoptosis was assessed by TUNEL. Neutrophils were stained using rat anti-neutrophil.
Protein Analysis and Enzymatic Activity
Total gelatinase or collagenase activity was measured in myocardial homogenates using EnzChek assay kit using collagen type I as the substrate for collagenase activity. Pro- and active MMP-2 and MMP-9 were detected by gelatin zymography. TACE activity and Western blotting for TNF was performed as described.15
TaqMan RT-PCR Analysis
RNA levels for the indicated genes were quantified by Real-time TaqMan RT-PCR.13 18S rRNA was used as an endogenous control. Details of sequences are in the online Materials and Methods.
In Vivo MMPi Treatment
An MMP-specific inhibitor, PD166793 (Pfizer Inc) was administered daily by gavage. PD166793 treatment (15 mg/kg/d) began 1 week before aortic-banding and continued until mice were euthanized.
An expanded Materials and Methods section is provided in the online data supplement.
Statistical Analysis
Survival between groups was compared by Kaplan-Meier survival analysis. All other comparisons were performed by ANOVA followed by multiple comparison testing (Student-Neuman Keuls). Values are reported as Mean±SEM. Statistical significance is recognized at P<0.05.
Results
Mortality Attributable to Heart Failure in TIMP-3eCDeficient Mice
Timp-3eC/eC mice were subjected to cardiac pressure overload by aortic-banding (AB). These mice were compared with 3 groups of control littermates throughout the study, aortic-banded WT, sham-operated WT, and sham-operated timp-3eC/eC mice. Aortic-banding generated comparable pressure gradient (59 to 64 mm Hg) independent of the genotype, but caused significantly higher mortality in timp-3eC/eC mice compared with WT (Figure 1A). Six weeks following AB timp-3eC/eC mice exhibited significant morbidity and had to be euthanized. Timp-3eC/eC hearts were grossly enlarged within 1 week after AB and had increased left ventricular (LV) chamber size (Figure 1B).
To characterize the effects of pressure overload on WT and timp-3eC/eC hearts, we analyzed the cardiac function by echocardiography (Table). In WT-AB mice, significant cardiac dysfunction was seen at 6 weeks. In contrast, as early as 1 week after AB we observed dramatic LV dilation, reduced aortic outflow velocity, cardiac contractility, and velocity of circumferential fiber shortening in timp-3eC/eC-AB mice. These parameters deteriorated progressively over 6 weeks (Figure 1C and 1D; Table). Notably, the extent of cardiac dysfunction in timp-3eC/eC mice after 6 weeks was comparable to WT mice after 12 weeks of aortic-banding (Table). Cardiac dysfunction was confirmed by in vivo hemodynamic measurements. Timp-3eC/eC-AB mice had lower LV developed pressure (LVDP), higher end-diastolic pressure (LVEDP), and markedly suppressed LV peak rates of pressure-rise and pressure-fall (±dP/dt) compared with WT-AB (Figure 1E). Reduced eCdP/dt and increased LVEDP suggest diastolic dysfunction in timp-3eC/eC mice. Terminal heart failure in patients is associated with pulmonary congestion. We also observed pulmonary edema and fibrosis in timp-3eC/eC-AB mice at 6 weeks (online Figure I). Altogether, advanced cardiac dysfunction as indicated by the extreme LV dilation and suppressed function in combination with pulmonary congestion demonstrates that unlike the WT mice, timp-3eC/eC mice develop rapid CHF 6 weeks after pressure overload. These features of cardiac dysfunction are reminiscent of DCM in patients.
Excess Hypertrophy with Timp-3 Loss
Next, we examined cardiomyocyte apoptosis that generally underlies LV dilatation. TIMP-3 deletion and overexpression are both linked to apoptosis.20,21 Apoptosis in timp-3eC/eC-AB LV was significantly higher than WT-AB at 6 weeks, suggesting that it is not responsible for the onset of LV dilation in these mice (Figure 2A and 2B). Baseline apoptosis was not different between sham groups. We investigated if myocyte hypertrophy occurred as a compensatory response to biomechanical stress. Heart weight-to-tibial length ratio was far greater in timp-3eC/eC-AB than in WT-AB mice. We also measured myocyte cross-sectional area, length, and expression of prototypic markers of hypertrophy and heart disease (atrial natriuretic factor, ANF; brain natriuretic peptide, BNP). Although WT mice developed significant hypertrophy over the 6-week period, timp-3eC/eC-AB mice showed markedly greater hypertrophy as determined by higher levels of all hypertrophy parameters compared with WT-AB (Figure 2C through 2G). Increased myocyte cross-sectional area and length indicate a combination of eccentric and concentric hypertrophy in timp-3eC/eC-AB hearts, whereas WT-AB hearts exhibit only concentric hypertrophy. This can explain LV dilation in timp-3eC/eC-AB hearts in the absence of reduced LV wall thickness.
Excess TNF and TACE Activity in Timp-3eC/eC-AB Mice
In aged timp-3eC/eC mice, we proposed involvement of TNF in the development of spontaneous DCM.18 In a liver injury model, we have shown TIMP-3 is a physiological regulator of TNF processing through inhibition of ADAM-17/TACE.15 In pursuit of the mechanism underlying the early onset of severe heart disease following pressure overload in young timp-3eC/eC mice, we asked whether cardiac expression of the TIMP3-TACE-TNF axis was affected. In WT hearts, TIMP-3 and TACE mRNA levels underwent parallel temporal changes after aortic-banding peaking at 3 weeks (Figure 3A and 3B). In contrast, in timp-3eC/eC hearts, TACE expression increased immediately after AB (within 6 hours, 1.5-fold, Figure 3B) concomitant with a 4-fold transient rise in TNF mRNA (Figure 3C). TACE activity also increased within 6 hours (1.7-fold) in timp-3eC/eC-AB hearts, whereas it rose much later (3 weeks) in WT-AB hearts (Figure 3D), temporally consistent with the increase in TACE expression. Addition of recombinant TIMP-3 (rTIMP-3), but not rTIMP-1 or a synthetic MMP-specific inhibitor (PD166793), inhibited the enhanced cardiac TACE activity in timp-3eC/eC mice (Figure 3E). TIMP-3-specific inhibition is considered signature of TACE.22 TACE processes membrane-bound 26-kDa TNF to its soluble 17-kDa species. Membrane-bound TNF levels were higher in timp-3eC/eC compared with WT sham hearts. Following pressure overload, elevated membrane-bound and processed TNF proteins were detected in timp-3eC/eC hearts (Figure 3F). Overall, along with a rapid increase in TACE activity, TACE and TNF productions were accelerated and enhanced in timp-3eC/eC-AB compared with WT-AB hearts, leading to amplified activity of this pathway.
Activation of the MMP Axis Perturbs ECM in Timp-3eC/eC Mice
TIMP-3 is known to inhibit a broad range of MMPs.14 We determined whether TIMP-3 loss allowed for altered MMP activity in response to pressure overload. Total gelatinase and collagenase activities markedly increased 1 week post-AB in timp-3eC/eC cardiac extracts compared with WT that displayed only collagenase induction (Figure 4A). The increased gelatinase activity in the timp-3eC/eC-AB group was reduced by rTIMP-3 and PD166793, although collagenase activity was additionally lowered by rTIMP-1 (Figure 4A). In search of specific candidates for the increased MMP activity, we detected active MMP-2 in timp-3eC/eC hearts beginning 1 week after AB, with both latent and active MMP-2 increasing over 6-week period (Figure 4B). Next, we investigated the trimolecular complex, comprised of MMP-2, TIMP-2, and MMP-14/MT1-MMP, typically required for MMP-2 activation.23 mRNA levels of MMP-2 and MT1-MMP were elevated in timp-3eC/eC-AB hearts (Figure 4C). In addition, we found a striking increase of MMP-13 in timp-3eC/eC-AB myocardium (Figure 4C). MMP-7 and MMP-8 were expressed minimally and remained unchanged post-AB (data not shown). MMP-9 mRNA increased over 6 weeks with no significant difference between WT and timp-3eC/eC hearts. Hence, 3 key interstitial collagenases,19,24,25 MMP-2, MMP-13, and MT1-MMP, were elevated in response to pressure overload in timp-3eC/eC mice. Further, soluble myocardial TIMPs (TIMP-1, TIMP-2, or TIMP-4) were not elevated to compensate for the absence of ECM-bound TIMP-3 in sham-operated timp-3eC/eC hearts, although timp-1 expression increased significantly at 6 weeks post-AB (Figure 4C). Altogether, these data indicate that cardiac pressure overload in the absence of timp-3 leads to parallel but temporally differential activation of the 2 metalloproteinase axes, MMPs and ADAMs.
Next, we determined whether the enhanced MMP activity affected the myocardial matrix. Using confocal microscopy on Picocirius red-stained sections and scanning electron microscopy, we examined the integrity of the fibrillar component of myocardial ECM that provides normal structural support. It is primarily comprised of collagen types I and III.26 Intact ECM network and uniform collagen distribution were seen in WT-AB hearts (Figure 4D and 4F), whereas we found areas devoid of or with disrupted fibrillar collagens in timp-3eC/eC-AB myocardium (Figure 4E and 4G). Although WT-AB hearts showed interstitial fibrosis 6 weeks after AB, localized fibrotic areas were noticeably increased in timp-3eC/eC-AB myocardium (data not shown). These findings are consistent with the maladaptive ECM remodeling found in patients with LV dilation.27
Genetic Deletion of TNF Attenuates Cardiac Disease in Timp-3eC/eC-AB Hearts
To determine whether TNF is a critical molecule for LV dilation and dysfunction in timp-3eC/eC mice, we generated timp-3eC/eC/tnfeC/eC double mutants. After AB, disease was delayed and partially rescued (Figure 5), while survival was improved markedly (Figure 6A) in the double knock-outs compared with timp-3eC/eC mice. Specifically, after only 3 weeks LV chamber size increased and cardiac contractility declined in timp-3eC/eC/tnfeC/eC-AB mice, but was significantly attenuated versus timp-3eC/eC-AB mice (Figure 5A through 5C). Cardiac hypertrophy in the double mutants also diverged from WT-AB group at 3 weeks, but remained far less pronounced than timp-3eC/eC-AB hearts until 6 weeks (Figure 5D and 5E). No myocyte apoptosis (Figure 5H) or pulmonary congestion (Figure 5F and 5G) was detected in double mutants. Hence, genetically coupling TNF ablation with TIMP-3 loss prevented CHF and attenuated heart disease in timp-3eC/eC-AB mice.
Dependency of MMP Activity on TNF
Because TNF is a transcriptional inducer of MMPs,9,12,28 we explored the TNF-dependency of MMP expression by comparing MMP activity and expression profiles between timp-3eC/eC/tnfeC/eC-AB and timp-3eC/eC-AB myocardium. Surprisingly, collagenase but not gelatinase activity was significantly higher in timp-3eC/eC/tnfeC/eC-AB than in timp-3eC/eC-AB hearts, with both activities remaining higher than WT-AB (Figure 5I and 5J). We further investigated contribution of specific MMPs in these mice (online Figure II). Overall, MMP-2 activation was abrogated with loss of TNF, but the expression of MMP-13 and MT1-MMP were only partially affected.
De Novo MMP-8 Induction in the Absence of TNF and TIMP-3
Despite the reduced transcriptional induction of specific interstitial collagenases, total myocardial collagenase activity was still higher in timp-3eC/eC/tnfeC/eC-AB than timp-3eC/eC-AB. Screening additional MMPs revealed that expression of MMP-8, another interstitial collagenase,29 was elevated 5-fold in timp-3eC/eC/tnfeC/eC-AB compared with timp-3eC/eC-AB and WT-AB hearts (Figure 5K). Timp-3eC/eC/tnfeC/eC-AB myocardium had significantly higher numbers of neutrophils scattered throughout the LV compared with timp-3eC/eC-AB and WT-AB myocardium (Figure 5L and 5 mol/L). Neutrophils are the primary source of MMP-8.29 The greater increase in MMP-8 RNA levels than in neutrophil numbers may be attributable to the activation state of the neutrophils present in the myocardium. Thus, neutrophil infiltration and de novo MMP-8 induction on loss of TNF likely contributed to elevated collagenase activity in the double mutant mice following pressure overload. We found no increase in macrophage numbers in timp-3eC/eC/tnfeC/eC-AB hearts by immunostaining with Mac3 or F4/80 (data not shown).
MMPi Treatment Attenuates LV Dilation and Dysfunction in Timp-3eC/eC-AB Mice
Because total gelatinase and collagenase activities were elevated in timp-3eC/eC mice following pressure overload, we determined the contribution of this increased MMP activity toward DCM and heart failure, by using a broad-spectrum MMP-specific inhibitor (PD166793, MMPi) that does not inhibit TACE.30 Oral administration of 15 mg/kg/d of PD166793 resulted in plasma drug concentration of 110±4 eol/L, a level reported to be sufficient to inhibit MMP activity.30 This level of PD166793 was sufficient to inhibit the excessive myocardial collagenase and gelatinase activities in timp-3eC/eC-AB cardiac homogenates in vitro (Figure 4A). MMPi treatment of sham-operated WT mice did not affect cardiac function or structure, and no systemic toxicity was detected by histology of liver, kidney or lung (data not shown). During the course of 7-week MMPi treatment, mice in all groups remained active and showed no weight loss. Interestingly, treatment of timp-3eC/eC-AB mice with MMPi abolished the interstitial fibrosis (data not shown), prevented the early heart failure and resulted in partial prevention of LV dilation and dysfunction (Figure 6B).
Combination of MMPi Treatment and TNF Ablation Completely Prevents Heart Disease
We found that TNF ablation or MMP inhibition individually resulted in partial prevention of DCM in timp-3eC/eC-AB mice. Therefore, we reasoned that limiting MMP activity in addition to TNF elimination should completely prevent heart disease in these mice. Treatment of timp-3eC/eC/tnfeC/eC-AB mice with 15 mg/kg/d of MMPi strikingly improved the survival (Figure 6A), completely rescued LV dilation and cardiac dysfunction as indicated by echocardiography and in vivo hemodynamics (Figure 6B and 6C), and hypertrophy (Figure 6D and 6E) up to 6 weeks after AB. These data demonstrate that excessive concomitant activity of both TNF and MMPs are responsible for the cardiac dysfunction and heart failure in timp-3eC/eC-AB mice.
Discussion
Dysregulations in proinflammatory cytokines and structural ECM underlie human heart disease.8,26 TNF has pleiotropic functions in cardiovascular diseases.17,31 Maladaptive matrix remodeling is equally well recognized in promoting abnormal cardiac structure and function. These distinct disciplines have been studied extensively, but mostly independent of each other. We show here that timp-3 provides regulatory crossover for these distinct fields. TIMP-3 serves a dual inhibitory function in the heart against ADAM and MMP metalloproteinases to couple cytokine bioactivity with ECM homeostasis. Timp-3 deficiency results in severe LV dilation and dysfunction in response to cardiac pressure overload, which is partially rescued by TNF deletion or by MMPi treatment, and completely prevented on combining TNF ablation with MMP inhibition. These data demonstrate that TIMP-3 is essential in cardiac recovery from mechanical stress.
Our earlier study suggested involvement of TNF in the development of spontaneous DCM in aged timp-3eC/eC mice as the soluble TNF and its receptor (P75) levels were enhanced in coronary effluent from aged hearts.18 The current study is an in depth investigation of the relationship between TIMP-3, TACE, TNF, and MMPs in a pressure overload model of heart disease. TIMP-3 is the most highly expressed TIMP in the murine heart13 and is the only known physiological inhibitor of ADAM-17/TACE.15,22 TACE and TIMP-3 have parallel baseline expression patterns in murine organs during development.32 We show that on pressure overload, TACE and TIMP-3 undergo parallel temporal inductions in WT hearts, a pattern suggesting that TIMP-3 normally serves to counteract increased TACE levels. In TIMP-3eCdeficient hearts, not only is there a dramatic immediate increase in the transcriptions of both TACE and TNF, but their increased activities go unchecked in the absence of TIMP-3. TNF feeds back positively on its own expression as well as that of TACE.28 TNF also induces TIMP-1 but downregulates TIMP-3 expression in myocyte cultures.33,34 Such downregulation facilitates TNF processing through increased TACE activity. Hence, the TIMP-3-TACE-TNF system provides an inherent regulatory mechanism for controlling TNF bioactivity following cardiac biomechanical stress.
TIMP-3 is classically known to inhibit MMP activity responsible for ECM turnover. Total gelatinase and collagenase activities rise markedly in timp-3eC/eC compared with WT hearts following aortic-banding. Screening of multiple MMPs reveals higher MMP-2, MMP-13 and MT1-MMP mRNA levels and higher MMP-2 activity. MMP-2 degrades several ECM proteins,19,35 whereas MMP-2, MMP-13, and MT1-MMP are established interstitial collagenases.19,25,36 MT1-MMP is also a key molecule for pro-MMP-2 and pro-MMP-13 activation.24,37 MMP-2, MMP-13, and MT1-MMP are elevated in patients with heart disease,38,39 and MMP-2 polymorphisms in human have been linked to CHF.40 Thus, interstitial collagenases, MMP-2, MMP-13, and MT1-MMP appear to be intimately involved in myocardial ECM remodeling following biomechanical stress.
TNF is well recognized for its pathophysiological importance in human heart disease.4,17 It reduces cardiac contractility,41 induces hypertrophy42 and apoptosis.43 In patients with DCM, both TACE and TNF levels are significantly elevated5 as in our aortic-banded timp-3eC/eC mice. Membrane-bound TNF mediates cardiac hypertrophy whereas cleaved TNF can induce DCM,44 both detected in timp-3eC/eC-AB mice. TNF is also a key transcriptional inducer of several MMPs in vitro.9,12 Our in vivo comparison of MMP expression in timp-3eC/eC-AB and timp-3eC/eC/tnfeC/eC-AB hearts provides evidence that TNF induces MMP-2 and MMP-13, but minimally affects MMP-9 and MT1-MMP transcription. The de novo induction of MMP-8 in timp-3eC/eC/tnfeC/eC-AB hearts is consistent with increased neutrophil infiltration in these mice, which intriguingly occurred without an increase in MMP-9, also known to be produced by neutrophils. MMP-8 is found decreased in human heart disease,45 and its role as an interstitial collagenase in the myocardium needs to be further understood. Our data illustrate that TNF bioactivity is intertwined with the amplification of MMP activity. The complete lack of myocardial apoptosis in double mutants suggests a proapoptotic role of TNF in this model of heart disease. Understanding the multiple functions of TNF in timp-3eC/eC myocardium will help us better understand its roles patients with DCM.
We demonstrate that treatment with MMPi in combination with TNF loss completely prevents LV dilation and dysfunction, and heart failure in aortic-banded timp-3eC/eC mice, although only partial rescue is achieved with the individual interventions. This indicates that matrix degradation and cytokines are together responsible for heart disease. The anti-TNF therapy in patients with class III and IV heart failure has so far proven unsuccessful.6,7 Based on our findings, the lack of success with anti-TNF therapy6,7 may be attributable to application at too late a stage or the requirement for concomitant targeting of MMPs. Because TIMP-3 lies upstream of these 2 systems, and is commonly reduced in patients with cardiomyopathy, we propose that a therapeutic strategy that mimics TIMP-3 function in targeting both the MMPs and TNF will be useful in producing an effective cardiac response to mechanical stress. Our findings provide a basis for anti-TNF and MMPi combination therapy in heart disease.
Acknowledgments
Z.K. is supported by Heart and Stroke Foundation of Canada post-doctoral fellowship. This research was supported by Canadian Institutes of Health Research and Heart and Stroke Foundation of Canada to R.K. G.Y.O. is a clinician/scientist supported by Canadian Institutes of Health Research and Tailored Advanced Collaborative Training in Cardiovascular Science fellowships and P.P.L. is a H&S/Polo Chair. The authors thank Geoffrey Woods for critical reading of this manuscript.
References
Olson EN. A decade of discoveries in cardiac biology. Nat Med. 2004; 10: 467eC474.
Feldman AM, McTiernan C. Is there any future for tumor necrosis factor antagonists in chronic heart failure Am J Cardiovasc Drugs. 2004; 4: 11eC19.
Feldman AM, Combes A, Wagner D, Kadakomi T, Kubota T, Li YY, McTiernan C. The role of tumor necrosis factor in the pathophysiology of heart failure. J Am Coll Cardiol. 2000; 35: 537eC544.
Torre-Amione G, Kapadia S, Benedict C, Oral H, Young JB, Mann DL. Proinflammatory cytokine levels in patients with depressed left ventricular ejection fraction: a report from the Studies of Left Ventricular Dysfunction (SOLVD). J Am Coll Cardiol. 1996; 27: 1201eC1206.
Satoh M, Nakamura M, Saitoh H, Satoh H, Maesawa C, Segawa I, Tashiro A, Hiramori K. TACE and TNF in human dilated cardiomyopathy. Circulation. 1999; 99: 3260eC3265.
Chung ES, Packer M, Lo KH, Fasanmade AA, Willerson JT. Randomized, double-blind, placebo-controlled, pilot trial of infliximab, a chimeric monoclonal antibody to TNF, in patients with moderate-to-severe heart failure: results of the anti-TNF Therapy Against Congestive Heart Failure (ATTACH) trial. Circulation. 2003; 107: 3133eC3140.
Mann DL, McMurray JJ, Packer M, Swedberg K, Borer JS, Colucci WS, Djian J, Drexler H, Feldman A, Kober L, Krum H, Liu P, Nieminen M, Tavazzi L, van Veldhuisen DJ, Waldenstrom A, Warren M, Westheim A, Zannad F, Fleming T. Targeted anticytokine therapy in patients with chronic heart failure: results of the Randomized Etanercept Worldwide Evaluation (RENEWAL). Circulation. 2004; 109: 1594eC1602.
Gunja-Smith Z, Morales AR, Romanelli R, Woessner JF Jr. Remodeling of human myocardial collagen in idiopathic dilated cardiomyopathy. Role of metalloproteinases and pyridinoline cross-links. Am J Pathol. 1996; 148: 1639eC1648.
Overall CM, Lopez-Otin C. Strategies for MMP inhibition in cancer: innovations for the post-trial era. Nat Rev Cancer. 2002; 2: 657eC672.
Zheng Y, Saftig P, Hartmann D, Blobel C. Evaluation of the contribution of different ADAMs to TNF shedding and of the function of the TNF ectodomain in ensuring selective stimulated shedding by the TNF convertase (TACE/ADAM17). J Biol Chem. 2004; 279: 42898eC42906.
Black RA, Rauch CT, Kozlosky CJ, Peschon JJ, Slack JL, Wolfson MF, Castner BJ, Stocking KL, Reddy P, Srinivasan S, Nelson N, Boiani N, Schooley KA, Gerhart M, Davis R, Fitzner JN, Johnson RS, Paxton RJ, March CJ, Cerretti DP. A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells. Nature. 1997; 385: 729eC733.
Siwik DA, Colucci WS. Regulation of matrix metalloproteinases by cytokines and reactive oxygen/nitrogen species in the myocardium. Heart Fail Rev. 2004; 9: 43eC51.
Young DA, Phillips BW, Lundy C, Nuttall RK, Hogan A, Schultz GA, Leco KJ, Clark IM, Edwards DR. Identification of an initiator-like element essential for the expression of the tissue inhibitor of metalloproteinases-4 gene. Biochem J. 2002; 364: 89eC99.
Gomez DE, Alonso DF, Yoshiji H, Thorgeirsson UP. Tissue inhibitors of metalloproteinases: structure, regulation and biological functions. Eur J Cell Biol. 1997; 74: 111eC122.
Mohammed FF, Smookler DS, Taylor SE, Fingleton B, Kassiri Z, Sanchez OH, English JL, Matrisian LM, Au B, Yeh WC, Khokha R. Abnormal TNF activity in Timp3eC/eC mice leads to chronic hepatic inflammation and failure of liver regeneration. Nat Genet. 2004; 36: 969eC977.
Li YY, Feldman AM, Sun Y, McTiernan CF. Differential expression of tissue inhibitors of metalloproteinases in the failing human heart. Circulation. 1998; 98: 1728eC1734.
Levine B, Kalman J, Mayer L, Fillit HM, Packer M. Elevated circulating levels of tumor necrosis factor in severe chronic heart failure. N Engl J Med. 1990; 323: 236eC241.
Fedak PW, Smookler DS, Kassiri Z, Ohno N, Leco KJ, Verma S, Mickle DA, Watson KL, Hojilla CV, Cruz W, Weisel RD, Li RK, Khokha R. TIMP-3 deficiency leads to dilated cardiomyopathy. Circulation. 2004; 110: 2401eC2409.
Aimes RT, Quigley JP. Matrix metalloproteinase-2 is an interstitial collagenase. Inhibitor-free enzyme catalyzes the cleavage of collagen fibrils and soluble native type I collagen generating the specific 3/4- and 1/4-length fragments. J Biol Chem. 1995; 270: 5872eC5876.
Fata JE, Leco KJ, Voura EB, Yu HY, Waterhouse P, Murphy G, Moorehead RA, Khokha R. Accelerated apoptosis in the Timp-3-deficient mammary gland. J Clin Invest. 2001; 108: 831eC841.
Baker AH, George SJ, Zaltsman AB, Murphy G, Newby AC. Inhibition of invasion and induction of apoptotic cell death of cancer cell lines by overexpression of TIMP-3. Br J Cancer. 1999; 79: 1347eC1355.
Amour A, Slocombe PM, Webster A, Butler M, Knight CG, Smith BJ, Stephens PE, Shelley C, Hutton M, Knauper V, Docherty AJ, Murphy G. TNF- converting enzyme is inhibited by TIMP-3. FEBS Lett. 1998; 435: 39eC44.
Strongin AY, Collier I, Bannikov G, Marmer BL, Grant GA, Goldberg GI. Mechanism of cell surface activation of 72-kDa type IV collagenase. Isolation of the activated form of the membrane metalloprotease. J Biol Chem. 1995; 270: 5331eC5338.
Knauper V, Will H, Lopez-Otin C, Smith B, Atkinson SJ, Stanton H, Hembry RM, Murphy G. Cellular mechanisms for human procollagenase-3 (MMP-13) activation. Evidence that MT1-MMP (MMP-14) and gelatinase a (MMP-2) are able to generate active enzyme. J Biol Chem. 1996; 271: 17124eC17131.
Knauper V, Lopez-Otin C, Smith B, Knight G, Murphy G. Biochemical characterization of human collagenase-3. J Biol Chem. 1996; 271: 1544eC1550.
Kassiri Z, Khokha R. Myocardial ECM and its regulation by metalloproteinases and their inhibitors. Thromb Haemost. 2005; 93: 212eC219.
Lindsey ML, Mann DL, Entman ML, Spinale FG. Extracellular matrix remodeling following myocardial injury. Ann Med. 2003; 35: 316eC326.
Bzowska M, Jura N, Lassak A, Black RA, Bereta J. TNF stimulates expression of TNF- converting enzyme in endothelial cells. Eur J Biochem. 2004; 271: 2808eC2820.
Owen CA, Hu Z, Lopez-Otin C, Shapiro SD. Membrane-bound MMP-8 on activated polymorphonuclear cells is a potent, TIMP-resistant collagenase and serpinase. J Immunol. 2004; 172: 7791eC7803.
Peterson JT, Hallak H, Johnson L, Li H, O’Brien PM, Sliskovic DR, Bocan TM, Coker ML, Etoh T, Spinale FG. Matrix metalloproteinase inhibition attenuates left ventricular remodeling and dysfunction in a rat model of progressive heart failure. Circulation. 2001; 103: 2303eC2309.
Torre-Amione G, Kapadia S, Lee J, Durand JB, Bies RD, Young JB, Mann DL. Tumor necrosis factor-alpha and tumor necrosis factor receptors in the failing human heart. Circulation. 1996; 93: 704eC711.
Nuttall RK, Sampieri CL, Pennington CJ, Gill SE, Schultz GA, Edwards DR. Expression analysis of the entire MMP and TIMP gene families during mouse tissue development. FEBS Lett. 2004; 563: 129eC134.
Alexander JP, Acott TS. Involvement of protein kinase C in TNF regulation of trabecular matrix metalloproteinases and TIMPs. Invest Ophthalmol Vis Sci. 2001; 42: 2831eC2838.
Li YY, McTiernan CF, Feldman AM. Proinflammatory cytokines regulate tissue inhibitors of metalloproteinases and disintegrin metalloproteinase in cardiac cells. Cardiovasc Res. 1999; 42: 162eC172.
Matrisian LM. The matrix-degrading metalloproteinases. Bioessays. 1992; 14: 455eC463.
Holmbeck K, Bianco P, Birkedal-Hansen H. MT1-mmp: a collagenase essential for tumor cell invasive growth. Cancer Cell. 2003; 4: 83eC84.
Pei D, Weiss SJ. Transmembrane-deletion mutants of the MT1-MMP process progelatinase A and express intrinsic matrix-degrading activity. J Biol Chem. 1996; 271: 9135eC9140.
Spinale FG, Coker ML, Heung LJ, Bond BR, Gunasinghe HR, Etoh T, Goldberg AT, Zellner JL, Crumbley AJ. A matrix metalloproteinase induction/activation system exists in the human left ventricular myocardium and is upregulated in heart failure. Circulation. 2000; 102: 1944eC1949.
Spinale FG. Matrix metalloproteinases: regulation and dysregulation in the failing heart. Circ Res. 2002; 90: 520eC530.
Vasku A, Goldbergova M, Holla LI, Spinarova L, Spinar J, Vitovec J, Vacha J. Two MMP-2 promoter polymorphisms (-790T/G and -735C/T) in chronic heart failure. Clin Chem Lab Med. 2003; 41: 1299eC1303.
Oral H, Dorn GW 2nd, Mann DL. Sphingosine mediates the immediate negative inotropic effects of tumor necrosis factor-alpha in the adult mammalian cardiac myocyte. J Biol Chem. 1997; 272: 4836eC4842.
Dibbs ZI, Diwan A, Nemoto S, DeFreitas G, Abdellatif M, Carabello BA, Spinale FG, Feuerstein G, Sivasubramanian N, Mann DL. Targeted overexpression of transmembrane tumor necrosis factor provokes a concentric cardiac hypertrophic phenotype. Circulation. 2003; 108: 1002eC1008.
Engel D, Peshock R, Armstong RC, Sivasubramanian N, Mann DL. Cardiac myocyte apoptosis provokes adverse cardiac remodeling in transgenic mice with targeted TNF overexpression. Am J Physiol Heart Circ Physiol. 2004; 287: H1303eC11.
Diwan A, Dibbs Z, Nemoto S, DeFreitas G, Carabello BA, Sivasubramanian N, Wilson EM, Spinale FG, Mann DL. Targeted overexpression of noncleavable and secreted forms of TNF provokes disparate cardiac phenotypes. Circulation. 2004; 109: 262eC268.
Wilson EM, Gunasinghe HR, Coker ML, Sprunger P, Lee-Jackson D, Bozkurt B, Deswal A, Mann DL, Spinale FG. Plasma MMPs and inhibitor profiles in patients with heart failure. J Card Fail. 2002; 8: 390eC398.(Zamaneh Kassiri, Gavin Y.)