Skip Navigation

Cardiovascular Research 2004 63(1):87-97; doi:10.1016/j.cardiores.2004.03.010
© 2004 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Stawowy, P.
Right arrow Articles by Graf, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stawowy, P.
Right arrow Articles by Graf, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2004, European Society of Cardiology

Regulation of matrix metalloproteinase MT1-MMP/MMP-2 in cardiac fibroblasts by TGF-β1 involves furin-convertase

Philipp Stawowy*,a, Christian Margetaa, Heike Kallischa, Nabil G Seidahb, Michel Chrétienc, Eckart Flecka and Kristof Grafa

aDepartment of Medicine/Cardiology, Deutsches Herzzentrum Berlin, Augustenburger Platz 1, D-13353 Berlin, Germany
bLaboratory of Biochemical Neuroendocrinology, Clinical Research Institute, Montréal, QC, Canada
cDiseases of Aging and Regional Protein Chemistry Centers, Ottawa Health Research Institute, University of Ottawa, ON, Canada

*Corresponding author. Tel.: +49-30-4593-2413; fax: +49-30-4593-2415. Email address: stawowy{at}dhzb.de

Received 18 December 2003; revised 1 March 2004; accepted 9 March 2004


   Abstract
Top
 Abstract
1. Materials and methods
 2. Results
 3. Discussion
 Acknowledgments
 References
 
Objective: Heart failure is characterized by an imbalance of matrix synthesis/turnover, finally resulting in fibrosis. Cardiac myocytes and fibroblasts play a pivotal role in the remodeling process. Cardiac remodeling involves the expression of TGF-β1 and matrix metalloproteinases (MMPs) in cardiac fibroblasts (CFBs). Furin, a subtilisin/kexin-like proprotein convertase (PC), activates TGF-β1 and membrane-bound MT1-MMP, which facilitates pro-gelatinase A (MMP-2) activation. Even though several reports identified TGF-β1 as a pro-fibrotic cytokine in the heart, it increases MMP-activity and cell migration/invasion in several cell types. The present study was done to investigate the contribution of TGF-β1 and furin to CFBs MMP-activity and motility. Methods and results: Stimulation of CFBs from adult Sprague–Dawley rats with TGF-β1 (20 ng/ml) induced furin, but had no effect on the closely related PC5. Inhibition of furin inhibited angiotensin II-induced TGF-β1 activation, indicating that TGF-β1 amplifies its activating convertase in CFBs. Pretreatment of CFBs with TGF-β1 (20 ng/ml, 24 h) increased their migration by about two-fold (p<0.05), which was accompanied by an enhanced expression and activity of MT1-MMP and MMP-2. Brefeldin A (BFA), a Golgi-disturbing agent, inhibited MT1-MMP activation, indicating that it occurs in the trans-Golgi network (TGN), where furin is concentrated and colocalized with MT1-MMP. Inhibition of furin significantly inhibited TGF-β1-induced MT1-MMP/MMP-2 activation. Furthermore, inhibition of furin attenuated TGF-β1-enhanced migration on gelatin-coated membranes (p<0.05). This was comparable to the effects of the MMP-inhibitor GM6001, pointing out that MMPs are major mediators of TGF-β1-enhanced CFB motility. Conclusion: We demonstrate that TGF-β1 induces MMP-activity in CFBs, thereby facilitating CFBs motility. Furthermore, TGF-β1 amplifies its activating convertase furin, which is also required for MT1-MMP/MMP-2 activation in CFBs. Thus, furin is central for TGF-β1 and MT1-MMP activation and might be a novel target in cardiac remodeling.

KEYWORDS Growth factors; Matrix metalloproteinases; Extracellular matrix; Remodelling

Chronic heart failure, irrespectively of its cause, is characterized by cardiac remodeling, an imbalance of extracellular matrix (ECM) synthesis and turnover, eventually leading to fibrosis [1]. This involves growth factors, such as transforming growth factor (TGF)-β1 which facilitate collagen production, as well as the expression of matrix metalloproteinases (MMPs), which are responsible for tissue breakdown [1,2]. Both cardiac myocytes and fibroblasts play a pivotal role in the remodeling process [1]. Cardiac fibroblasts (CFBs) and myofibroblasts, which are phenotypically transformed fibroblasts characterized by expression of {alpha}-smooth muscle actin ({alpha}-SMA) [3], are not only the major contributor to collagen biosynthesis, but also participate in cardiac remodeling via expression and release of MMPs [2]. In vitro, TGF-β1 leads to myofibroblast differentiation of CFBs and increases collagen synthesis [4]. Blocking of TGF-β1 function can prevent cardiac fibrosis in pressure-overloaded rats [5]. TGF-β1 also promotes cell adhesion and migration in many cell types, including fibroblasts [6] and myofibroblasts [7]. MMPs, which are potential target in heart failure [8], are diversely affected by TGF-β1. In fibroblasts, TGF-β1 inhibits MMP-1 [9], but induces MMP-2 [10], indicating that this cytokine may differentially regulate the MMP repertoire of cells. A diverse regulation of MMPs has been found in cardiac remodeling. Reduction in MMP-1 [11] and tissue inhibitor of metalloproteinase (TIMP)-3 [12], but increased levels of MMP-2 and membrane-bound MT1-MMP (MMP-14) [11], are reported in the terminally failing human myocardium, a setting when ECM turnover [13] and TGF-β1 [14] are increased. MMP-2 is released from the cells as inactive precursor, which is activated by MT1-MMP (MMP-14), involving formation of a trimer with TIMP-2 in a multistep pathway at the cell surface [15]. In contrast, MT-MMPs are expressed at the cell surface as locally active enzymes and are activated intracellular by limited endoproteolytic cleavage of their propeptide involving furin-convertase [16–18]. Furin is a member of the family of subtilisin/kexin-like proprotein convertases (PCs), which are responsible for the activation of precursor pro-proteins to become biologically active [19,20]. Furin activates protein precursors with narrow specificity following basic R-Xaa-K/R-R-like motifs [21]. Using overexpression experiments of the prototype membrane-anchored MT1-MMP and furin in furin-deficient LoVo cells, Yana and Weiss [16] demonstrated that furin controls the functional activity of MT1-MMP, which is the major activator of MMP-2 [15]. Whereas inhibition of furin prevented MT1-MMP maturation and activity, inhibitors of plasmin, trypsin or urokinase had no effect [16]. Among the PCs, furin, PC5 and PC7 display a widespread cell and tissue distribution, and are typically localized in the trans-Golgi network (TGN) [20]. Furin, which is detected in normal tissue at low levels, has been shown to promote human cancer cell migration/invasion involving pro-MT1-MMP and/or pro-TGF-β1 activation [22,23]. Furin is the principal TGF-β1 convertase and is positively regulated by TGF-β1 [24]. In furin knock-out mice, failure of the heart tube to fuse and undergo looping has been ascribed to the importance of furin for TGF and TGF-signaling [25]. Thus, a tight relationship between TGF-β1–furin–MT1-MMP exists. The aim of this study was to investigate the contribution of TGF-β1 and furin to CFB MMP-activity and motility.


   1. Materials and methods
Top
 Abstract
 1. Materials and methods
2. Results
 3. Discussion
 Acknowledgments
 References
 
1.1. Materials
Rat TGF-β1 and TNF-{alpha} were purchased from Prepro Tech, angiotensin II and the furin-like proprotein convertase inhibitor decanoyl-RVKR-chloromethylketone (dec-CMK) were from Bachem. The broad spectrum hydroxamate class MMP-inhibitor GM6001 (Ilomastat) was from Chemicon. All other chemicals were from Sigma. Cell culture media and materials were from GIBCO. The following primary antibodies were used: anti furin (MON139; Alexis Biochem.), anti PC5 (Clinical Research Institute) and anti trans-Golgi network marker protein (TGN38; Affinity Bioreagents). The antibodies to {alpha}-SMA, actin and desmin were from Sigma. Antibodies to vimentin and factor VIII were from Dako. For MT1-MMP detection, a rabbit polyclonal antibody against the hinge region of human MT1-MMP (BIOMOL) and a monoclonal antibody directed against the hemopexin-like domain (clone 113-5B7; Chemicon) were used. The antibodies directed against the hemopexin domain of MMP-2 and against TIMP-2 were from Sigma, anti TGF-β1 antibody directed against the C-terminus was from Santa Cruz. Secondary antibodies for immunofluorescence were from Vector.

1.2. Isolation and culture of rat cardiac fibroblasts
Animals used in this study were maintained in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institute of Health (NIH publication no. 85-23, revised 1996). Cardiac ventricular fibroblasts (CFBs) were prepared from adult male Sprague–Dawley rats (body weight about 300 g) as described by Iwami et al. [26]. Briefly, hearts were dissected free of atria, minced, and subject to trypsin and DNase II digestion. The isolated cells were preplated for 30 min in DMEM/low-glucose (1% penicillin/streptomycin and 1% L-glutamine) with 10% FCS. During this period, the nonmyocytes attached to the plate while the myocytes remained floating, thus separating both populations. The attached nonmyocytes were than grown in DMEM/low-glucose (1% penicillin/streptomycin and 1% L-glutamine) with 10% FCS until they reached confluency and than detached by accutase treatment (PAA Laboratories, Austria) and split 1:4. Cell cultures were routinely immunophenotyped by their absence of desmin, factor VIII and {alpha}-SMA, but presence of vimentin as CFBs. Cell cultures contained more than 95% CFBs. All experiments were performed in the second to fourth passage after starvation in serum-free DMEM/low-glucose (1% penicillin/streptomycin and 1% L-glutamine) for 24 h. In inhibitor experiments, CFBs were pretreated with the pharmacological inhibitor for 12 h, followed by growth factor stimulation in the presence of the inhibitor (concentrations as indicated). Cell viability was assessed by trypan-blue staining. Experiments were done at least in triplicates with different preparations of cells.

1.3. Western blot analysis
Immunoblotting was done as described [27]. Briefly, proteins were extracted in RIPA-buffer (1x PBS, 1% Nonidet P-40, 0.5% sodium deoxycholate and 0.1% SDS) containing freshly dissolved protease inhibitors. Up to 50 µg of proteins were separated by reducing SDS-PAGE (8% and 12%). Immunodetection was by chemiluminescence (Amersham). Primary antibodies were used at the following dilutions: furin 1:250, PC5 1:500, {alpha}-SMA 1:500, actin 1:250, MT1-MMP 1:500 (both antibodies), MMP-2 1:250 and TIMP-2 1:500. Semiquantitative densitometry was done using the NIH program 1.62 and is expressed in arbitrary units (AU).

1.4. Gelatin zymography
Conditioned media were collected at the indicated time points and centrifuged for 5 min at 500xg to remove cells and debris. Supernatants were than concentrated using Vivaspin 2 (Vivascience) according to the manufacture. Protein content was normalized between samples (final 5 µg protein loaded each sample) and samples mixed 1:1 with 2x Laemmli buffer without reducing agents (0.125 M Tris–HCl, pH 6.8, 4% SDS, 20% glycerol, 0.04% bromophenol blue). Samples were electrophoresed in 10% SDS-PAGE containing 0.1% gelatin. Following migration, gels were renaturated by exchanging SDS to Triton-X100 (2.5%) and than incubated for 24 h at 37 °C in activation buffer (50 mM Tris-base, pH 7.6, 5 mM CaCl2, 0.2 M NaCl and 0.02% Brij). Gels were than stained with Coomassie staining solution (0.5% Coomassie R250, 30% MeOH, 10% acetic acid) for 2 h, followed by destaining (50% MeOH and 10% acetic acid). Supernatants from the human fibrosarcoma cell line (HT1080) where used as controls for zymography. In inhibitor experiments, the ratio of active (68 kDa intermediate MMP-2 and 62 kDa full-active MMP-2) to inactive (72 kDa pro-MMP-2) was used as an index of MMP-2 activation. Semiquantitative densitometry was done using the NIH program 1.62 and is expressed in arbitrary units.

1.5. Immunofluorescence
CFBs were platted on 0.2% gelatin coated plastic camber slides (Nunc), let to adhere and than synchronized in serum-deficient DMEM/low-glucose (1% penicillin/streptomycin and 1% L-glutamine) overnight, followed by stimulation with TGF-β1 (20 ng/ml, 24 h). Cells were fixed with 4% buffered formaline (10 min) at room temperature. Non-specific binding was blocked with 10% normal-goat serum. Cells were washed with PBS/0.01% Triton-X100 several times and incubated with the primary antibodies overnight at 4 °C. After washing with PBS/0.01% Triton-X100 several times, FITC- and Texas Red (TXR)-conjugated secondary antibodies were used. In double-labeling experiments, species-specific antibodies were used to distinguish primary antibodies. All primary antibodies were used at a dilution of 1:50. Nuclei were stained with DAPI. Microscopy was carried out on a Olympus BX61, using analySIS imagine software.

1.6. Migration assay
Chemotaxis experiments with CFBs were done as described recently [28] using transwell cell culture chambers with a gelatin-coated (0.2%) polycarbonate membrane with 8 µm pores. The number of CFBs per high power field (HPF, magnification x320) that migrated after 4 h to the lower surface of the filters was determined microscopically. Four randomly chosen HPFs were counted per filter. Experiments were performed in triplicates and were repeated at least three times.

1.7. Statistical analysis
ANOVA and paired or unpaired t-test were performed for statistical analysis as appropriate. Statistical significance was designated at a probability value of less than 0.05. Values are expressed as mean±S.D.


   2. Results
Top
 Abstract
 1. Materials and methods
 2. Results
3. Discussion
 Acknowledgments
 References
 
2.1. TGF-β1 specifically amplifies its activating convertase furin in cardiac fibroblasts
To investigate the effects of TGF-β1 on furin levels, CFBs were stimulated with TGF-β1 (20 ng/ml) for 24 h initially. Immunoblotting revealed a significant increase of furin (87 kDa) upon TGF-β1 stimulation (Figs. 1A and B, *p<0.05 vs. controls). In contrast, levels of PC5 (117 kDa), a closely related PC [20], were not significantly altered by TGF-β1. Induction of furin by TGF-β1 was time- and concentration dependent, with a significant increase following 2 h stimulation (Figs. 2A and B, TGF-β1 20 ng/ml, *p<0.05 vs. controls) or a significant induction at 0.5 ng/ml TGF-β1 (Fig. 2C and D, 24 h stimulation, *p<0.05 vs. controls).


Figure 1
View larger version (32K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 (A) Stimulation of CFBs with TGF-β1 (20 ng/ml, 24 h) resulted in a significant increase of furin (87 kDa), whereas levels of PC5 (117 kDa) were not affected. Reblotting with actin was done to demonstrate equal protein loading. (B) Densitometry of furin and PC5 levels upon TGF-β1 stimulation is depicted (*p<0.05 vs. controls). n=3.

 

Figure 2
View larger version (30K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 TGF-β1 increased furin in a time- and concentration-dependent manner. A significant induction was found following 2 h TGF-β1 (20 ng/ml) stimulation (A and B, *p<0.05 vs. controls) or at a concentration of 0.5 ng/ml (B and D, *p<0.05 vs. controls). Actin reblotting is done to demonstrate equal protein loading. n=3.

 
TGF-β1 is synthesized as a 55-kDa precursor, which is converted into propepide (44 kDa) and mature TGF-β1 (12.5 kDa). Endoproteolytic activation of pro-TGF-β1 occurs within the TGN [29] and involves furin [24]. To investigate the contribution of furin to TGF-β1 activation in CFBs, cells were treated with angiotensin II (Ang II; 1 µmol/l, 24 h), which stimulates TGF-β1 synthesis in CFBs and myofibroblasts [30]. Following Ang II stimulation, levels of mature TGF-β1 (12.5 kDa) increased on reducing SDS-PAGE. Inhibition of furin with the furin-like proprotein convertase inhibitor dec-CMK (100 µmol/l) inhibited Ang II-induced TGF-β1 activation, evident by the increases of the 55-kDa precursor (Fig. 3).


Figure 3
View larger version (46K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 To investigate the role of furin for TGF-β1 activation, furin was inhibited with dec-CMK (100 µmol/l) and CFBs stimulated with Ang II (1 µmol/l, 24 h). Extracts were subject to reducing SDS-PAGE. The furin-inhibitor attenuated Ang II-induced TGF-β1 maturation, apparently by the increase in the TGF-β1 precursor (55 kDa). Reblotting with actin is done to demonstrate protein loading. n=3.

 
2.2. TGF-β1 enhances cardiac fibroblast migration and induces MT1-MMP/MMP-2
Studies have demonstrated that TGF-β1 promotes cell motility in several cell types [2,6,7,23]. First, we investigated whether TGF-β1 itself is a chemoatractant to CFBs. Using a checker-box experiment, we found that TGF-β1 itself is not significantly chemotactic to serum-starved CFBs (Table 1). In contrast, pre-treatment with TGF-β1 (20 ng/ml, 24 h) increased FCS (10%)-directed CFB migration on gelatin-coated membranes by about two-fold (Fig. 4A, *p<0.05 vs. controls). TGF-β1-enhanced migration was concentration-dependent and paralleled by increased expression of {alpha}-SMA, indicating myofibroblast differentiation [3] of CFBs (Figs. 4B,E,H). Immunofluorescence with a monoclonal antibody to {alpha}-SMA revealed that non-TGF-β1-treated CFB cultures consisted of about 90% fibroblasts and about 10% myofibroblasts, whereas following TGF-β1 stimulation (20 ng/ml, 24 h) almost 100% CFBs were myofibroblasts (Figs. 4B and E). Both controls and TGF-β1-treated CFBs expressed vimentin (Figs. 4C and F), but not desmin (Figs. 4D and G).


View this table:
[in this window]
[in a new window]

  		Table 1 TGF-β1 is not a significant chemoatractant to CFBs

 

Figure 4
View larger version (58K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 (A) Pretreatment of CFBs with TGF-β1 (20 ng/ml, 24 h) increased CFB migration on gelatin (0.2%)-coated membranes towards 10% FCS in a concentration-dependent manner (*p<0.05 vs. controls). n=4. Immunofluorescence revealed that non-TGF-β1-treated CFB cultures (controls) consisted of about 90% fibroblasts and about 10% myofibroblasts, whereas following TGF-β1 stimulation (20 ng/ml, 24 h) almost 100% CFBs were myofibroblasts, evident by the expression of {alpha}-SMA (B and E). Both controls and TGF-β1-treated CFBs expressed vimentin (C and F), but were negative for desmin ((D and G, bar in G=200 µm). Nuclei are stained with DAPI. (H) TGF-β1 induced {alpha}-SMA expression in CFBs in concentration-dependent manner. Membrane is reblotted with an antibody to total actin. n=3. (I) TGF-β1-enhanced CFBs motility was accompanied by increased levels of mature MT1-MMP (63 kDa, pro-MT1-MMP=68 kDa) and MMP-2 (68 kDa, pro-MMP-2=72 kDa), whereas levels of TIMP-2 (21 kDa) did not change at 24 h TGF-β1 stimulation. n=3. (J) Both latent (72 kDa) and intermediate/active (68/62 kDa) MMP-2 forms were significantly upregulated upon TGF-β1 (20 ng/ml) stimulation, comparable to the effect of TNF-{alpha} (100 ng/ml, co.=controls). n=3.

 
TGF-β1 promoted CFBs motility was also accompanied by increased levels of mature MT1-MMP (63 kDa) and MMP-2 (68 kDa) on immunoblotting (Fig. 4I). In contrast, no significant changes of the endogenous MMP-2 inhibitor [15], TIMP-2 were detected at 24 h stimulation (Fig. 4I). Induction of MMP-2 activity by TGF-β1 was confirmed by gelatin zymography. Comparable to TNF-{alpha} (100 ng/ml, 24 h) [31], TGF-β1 (20 ng/ml, 24 h) increased both, latent (72 kDa) and intermediate/active (68/62 kDa) MMP-2 species on gelatin zymography (Fig. 4J). Detailed analysis revealed that the induction of latent and mature forms of MMP-2 by TGF-β1 occurred time- and concentration-dependent. A significant increase of MMP-2 activity was found at 24 h TGF-β1 stimulation (20 ng/ml, Figs. 5A and B, *p<0.05 vs. controls) or at a concentration of 2.5 ng/ml TGF-β1 (24 h, Figs. 5C and D, *p<0.05 vs. controls). In contrast, levels of TIMP-2 increased on SDS-PAGE at much later time points (72 h) than MMP-2 upon TGF-β1 stimulation (Figs. 5E and F, *p<0.05 vs. controls).


Figure 5
View larger version (39K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 MMP-2 gelatinolytic activity of latent (72 kDa) and intermediate/active (68/62 kDa) MMP-2 species increased upon TGF-β1 stimulation in a time- and concentration-dependent manner. A significant increase was found at 24 h stimulation (A; densitometry depicted in B; *p<0.05 vs. controls) or at a concentration of 2.5 ng/ml TGF-β1 (C; densitometry depicted in D; *p<0.05 vs. controls). HT denotes supernatants of HT1080 cells. n=3. (E) In contrast to MMP-2, levels of TIMP-2 increased delayed by TGF-β1 (20 ng/ml) at 72 h stimulation. Reblotting of the membrane with actin was done to confirm equal loading. Densitometry of TIMP-2 levels is depicted (F, *p<0.05 vs. controls). n=3.

 
2.3 MT1-MMP is activated in the trans-Golgi network in CFBs
Precursor proprotein activation by PCs occurs typically in the TGN [20]. To define the intracellular compartment of pro-MT1-MMP activation in CFBs, cells were treated with the Golgi-disturbing agent brefeldin A (BFA) [32] and extracts subject to SDS-PAGE. Following BFA-pretreatment (15 µg/ml, 6 h) and TGF-β1 stimulation (20 ng/ml, 24 h) in the presence of BFA, maturation of pro-MT1-MMP (68 kDa) to active MT1-MMP (63 kDa) was inhibited (Fig. 6A), indicating that MT1-MMP activation occurs in the TGN were furin is highly concentrated [19,20]. To further identify the subcellular localization of furin and its putative substrate MT1-MMP, double-labeling immunofluorescence was used. TGF-β1 stimulated CFBs were treated with BFA to inhibit Golgi-trafficking (15 µg/ml, 12 h before fixation). Following BFA treatment, furin concentrated in the condensed perinuclear compartment (Fig. 6B), characteristic for the TGN. Double-labeling with an antibody to the TGN-marker TGN38 (Fig. 6D) confirmed that furin staining is predominantly localized in the TGN (Fig. 6C=merge). Identical results were obtained for MT1-MMP. Following BFA-treatment, MT1-MMP was trapped in perinuclear regions as well (Fig. 6E). Again, double-labeling with TGN38 (Fig. 6G) confirmed localization of MT1-MMP in the TGN of BFA-treated CFBs (Fig. 6F), whereas without BFA treatment, MT1-MMP localized to the cell surface (Figs. 6H and K). TGF-β1-treated CFBs expressed {alpha}-SMA (Fig. 6I) and vimentin (Fig. 6J).


Figure 6
View larger version (52K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 (A) Following treatment with the Golgi-disturbing agent BFA (15 µg/ml), TGF-β1-induced (20 ng/ml, 24 h) maturation of pro-MT1-MMP (68 kDa) to active MT1-MMP (63 kDa) was inhibited, indicating that MT1-MMP activation occurs in the TGN. Membrane is reblotted with actin. co.=control. (n=3). Double-labeling immunofluorescence revealed that following BFA treatment, furin colocalized with the TGN-marker TGN38 (B=furin, C=merge, D=TGN38). Similar results were obtained for MT1-MMP (E=MT1-MMP, F=merge, G=TGN38). Without BFA, MT1-MMP was localized at the cell membrane (arrows in H and higher magnification in K). Cells are treated with TGF-β1 (20 ng/ml). TGF-β1-treated CFBs expressed {alpha}-SMA (I) and vimentin (J).

 
2.4. Inhibition of furin attenuates TGF-β-enhanced CFB migration via inhibition of pro-MT1-MMP and subsequent pro-MMP-2 activation
To investigate the contribution of furin to pro-MT1-MMP activation, CFBs were pretreated with the furin-like proprotein convertase inhibitor dec-CMK (100 µmol/l) and than stimulation with TGF-β1 (20 ng/ml, 24 h). Upon furin inhibition, MT1-MMP self-proteolysis to the 43-kDa non-active species was inhibited. This was comparable to the reported effect of GM6001 (50 µmol/l) [33], a broad-spectrum MMP-inhibitor (Fig. 7A, immunodetection with a monoclonal MT1-MMP antibody, clone 113-5B7). The dependency of pro-MT1-MMP activation on furin activity was further confirmed with a polyclonal antibody directed against the hinge region of MT1-MMP, which recognizes latent and active MT1-MMP forms. Following furin inhibition with dec-CMK (100 µmol/l), TGF-β1-induced MT1-MMP maturation from pro- (68 kDa) to active MT1-MMP (63 kDa) species was significantly inhibited (Fig. 7B). Furthermore, inhibition of pro-MT1-MMP activation by dec-CMK was accompanied by an inhibition of MMP-2 activation on gelatin zymography, which is an indicator of MT1-MMP activity. The ratio of active (68 kDa intermediate MMP-2 and 62 kDa full-active MMP-2) to inactive (72 kDa pro-MMP-2) was used as an index of MMP-2 activation. Whereas TGF-β1 alone significantly increased MMP-2 gelatinolytic activity (Figs. 7C and D, *p<0.05 vs. controls), furin inhibition with dec-CMK resulted in a significantly decreased ratio of active/inactive MMP-2 forms (Figs. 7C and D, #p<0.05 vs. TGF-β1). Taken together, our data demonstrates that furin is a major activator of MT1-MMP activation in CFBs. Inhibition of furin with dec-CMK (100 µmol/l) also significantly inhibited TGF-β1-enhanced CFB migration towards FCS on gelatin-coated membranes. Inhibition of migration by dec-CMK was comparable to the inhibition of migration achieved with GM6001 (50 µmol/l), further supporting that MMPs are major mediators of TGF-β1-enhanced CFB migration (Fig. 7E, *p<0.05 vs. controls, #p<0.05 vs. TGF-β1).


Figure 7
View larger version (38K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 (A) Using a monoclonal antibody to MT1-MMP (clone 113-5B7), we found inhibition of MT1-MMP self-proteolysis to the non-active 43 kDa species by dec-CMK (100 µmol/l), comparable to the effects of GM6001 (50 µmol/l). Cell were treated with TGF-β1 (20 ng/ml, 24 h); co.=controls. Actin reblotting of membrane is shown. n=3. (B) Results are confirmed with a polyclonal antibody to the hinge-region, which recognizes pro- and active MT1-MMP forms. Upon furin inhibition with dec-CMK (100 µmol/l), TGF-β1 (20 ng/ml, 24 h) induced activation of pro-MT1-MMP (68 kDa) to mature MT1-MMP (63 kDa) was inhibited. co.=controls. Membrane is reblotted with actin. n=3. (C) Dec-CMK inhibited MMP-2 activation in supernatants of TGF-β1-treated CFBs in a concentration-dependent (50 and 100 µM) manner. HT1080=supernatants from HT1080 cells. Pro-MMP-2=72 kDa, intermediate/active MMP-2=68/62 kDa. (D) The ratio of active (68 kDa intermediate MMP-2 and 62 kDa full-active MMP-2) to inactive (72 kDa pro-MMP-2) was used as an index of MMP-2 activation. Furin inhibition with dec-CMK resulted in a significantly decreased ratio of active/inactive MMP-2 forms (*p<0.05 vs. controls, #p<0.05 vs. TGF-β1). co.=controls. n=3. (E) Dec-CMK (100 µmol/l) inhibited TGF-β1 (20 ng/ml, 24 h)-enhanced CFB migration, comparable to GM6001 (50 µmol/l, *p<0.05 vs. controls, #p<0.05 vs. TGF-β1). co.=controls. n=5.

 

   3. Discussion
Top
 Abstract
 1. Materials and methods
 2. Results
 3. Discussion
Acknowledgments
 References
 
The present study demonstrates that TGF-β1 promotes CFB migration via increased expression and activity of MT1-MMP and MMP-2. Furthermore, we found that TGF-β1 specifically induces furin, which is central for both, TGF-β1 and MT1-MMP activation in CFBs.

CFBs are major mediators in cardiac remodeling, synthesizing both, ECM and MMPs [2]. They are also a major source of TGF-β1 [34], which contributes to collagen synthesis in vitro and in vivo [1,2,4,5]. Furthermore, the pro-fibrotic effects of Ang II on CFBs are at least partially mediated via TGF-β1 [35], stressing the importance of this cytokine in the remodeling process. TGF-β1 is synthesized as a precursor proprotein, which is activated by endoproteolytic cleavage within the TGN [29], followed by its release in a larger multiprotein complex associated with its NH2-terminal propeptide (LAP, latency-associated peptide) [36]. Whereas furin seems to be always required for its intracellular activation, release of mature TGF-β1 from its latency complex involves cell–cell interactions [37] or extracellular proteolysis by other enzymes, including MMP-2 [38]. Using the furin-like proprotein convertase inhibitor dec-CMK [39], the present study demonstrates that furin activity is required for Ang II-induced TGF-β1 activation in CFBs. Moreover, we found that TGF-β1, which self amplifies its synthesis in CFBs [40], amplifies its activating convertase furin in a positive feed-back loop. Interestingly, Dubois et al. [24] recently identified furin as the authentic TGF-β1-convertase among several PCs tested. Comparable to their study [24], levels of the closely related PC5 [20] were not significantly altered by TGF-β1 stimulation in our study, indicating a specific effect of TGF-β1 on furin in CFBs.

Our study further demonstrates that increases in furin by TGF-β1 are accompanied the acquisition of a myofibroblastic phenotype [3], as well as an enhancement of CFB motility and MT1-MMP/MMP-2 activity. We found that TGF-β1 pretreatment of CFBs resulted in an about two-fold increase in migration, paralleled by the induction of MT1-MMP and MMP-2. Whereas levels of MMP-2 gelatinolytic activity were significantly upregulated at 24 h stimulation, levels of TIMP-2 increased at 72 h stimulation, indicating a differential effects of TGF-β1 on MMP-2 and its endogenous inhibitor in CFBs.

MT1-MMP, which is the major MMP-2 activator [15], is activated intracellular and than expressed on the cell surface as locally active enzyme [16]. Although endocytosis has been shown as a short-term function of MT1-MMP regulation [41], synthesis remains a major regulatory step. Furin has been demonstrated to activate pro-MT1-MMP [16,22], but MT1-MMP may also function as a self-convertase [33]. The present study reveals that furin is required for MT1-MMP activation in CFBs. Using the Golgi-disturbing agent BFA [32], we found that MT1-MMP activation occurs in the TGN, where furin is concentrated [19,20]. With immunofluorescence, we demonstrate colocalization of furin and MT1-MMP in the TGN of CFBs. Finally, inhibition of furin with dec-CMK attenuated TGF-β1-induced MT1-MMP and subsequent MMP-2 activation in CFBs. Thus, furin is a physiological convertase for MT1-MMP activation in CFBs.

In humans and animal models, MT1-MMP and MMP-2 are still increased at late stages of cardiac remodeling and accompany increases in TGF-β, collagen and ECM turnover [11–14,42]. Animal studies revealed that targeted deletion of MMP-2 in mice attenuated not only early complications, but also late cardiac remodeling [43]. Furthermore, type I collagen, the principal component of the myocardial interstitium [1], may add to the complex nature of the remodeling process via induction of MT1-MMP/MMP-2 activity itself [44]. Moreover, beside its function as major MMP-2 activator and its importance for cell migration, MT1-MMP may also participate in the remodeling process via direct cleavage of critical matrix components as well as activation of growth factor and adhesion molecules [15,45]. Thus MMPs potentially contribute to cardiac remodeling in a number of ways and stages and are promising targets in heart failure [8]. The present study demonstrates that inhibition of furin is a novel target to inhibit MT1-MMP-dependent CFB functions.

Furin has been shown to modulate cell migration/invasion. In cancer cell lines, inhibition of furin results in the inhibition of their invasiveness, which is due to a decreased MT1-MMP/MMP-2 [22] and/or TGF-β1 [22,23] activation by furin. Here, we show that inhibition of furin attenuates TGF-β1-enhanced CFB migration. Even though we cannot completely exclude the contribution of PCs to the activation of other signaling pathways [46], we demonstrate that inhibition of furin by dec-CMK resulted in a comparable inhibition of CFB migration as achieved by the broad-spectrum MMP-inhibitor GM6001. This points out to MMPs as major mediator of TGF-β1-enhanced CFB migration.

While our manuscript was in preparation, Guo and Piacentini [47] demonstrated a three-dimensional type I collagen-induced MMP-2 activation in CFBs. Incubation of CFBs embedded in the lattice with dec-CMK attenuated collagen type I-induced MT1-MMP/MMP-2 activation and CFB invasion [47], supporting that furin is critical for MMP activation in CFB.

In conclusion, the present study describes a tight relationship between TGF-β1–furin–MT1-MMP in CFBs. Furin could play a central role in cardiac remodeling, since it is required for the activation of both TGF-β1 and MT1-MMP. Thus, furin-mediated activation of MT1-MMP is required for the recruitment of cells to their site of action, but furin-mediated TGF-β1 activation may be crucial for ECM synthesis as well. We propose that furin, which has been found to increase following myocardial infarction in rodents [48], may be a novel target in cardiac remodeling.


   Acknowledgments
Top
 Abstract
 1. Materials and methods
 2. Results
 3. Discussion
 Acknowledgments
References
 
This work was supported by a grant from the Bundesministerium für Bildung und Forschung (BMBF) (CAN02/005) to PS, EF and KG. CM was supported by the Deutscher Akademischer Austauschdienst (DAAD). MC and NGS were supported by Canadian Institutes of Health Research (CIHR) grants MOP-44362 and MGP-44363, respectively.


   Notes
 
Time for primary review 20 days


   References
Top
 Abstract
 1. Materials and methods
 2. Results
 3. Discussion
 Acknowledgments
 References
 

  1. Swynghedauw B. Molecular mechanisms of myocardial remodeling. Physiol. Rev. (1999) 79:215–262.[Abstract/Free Full Text]
  2. Manabe I, Shindo T, Nagai R. Gene expression in fibroblasts and fibrosis. Involvement in cardiac hypertrophy. Circ. Res. (2002) 91:1103–1113.[Abstract/Free Full Text]
  3. Gabbiani G. Evolution and clinical implications of the myofibroblast concept. Cardiovasc. Res. (1998) 38:545–548.[Free Full Text]
  4. Petrov V, Fagard R, Lijnen P. Stimulation of collagen production by transforming growth factor-β1 during differentiation of cardiac fibroblast to myofibroblasts. Hypertension (2002) 39:258–263.[Abstract/Free Full Text]
  5. Kuwahara F, Kai H, Tokuda K, et al. Transforming growth factor-β function blocking prevents myocardial fibrosis and diastolic dysfunction in pressure-overloaded rats. Circulation (2002) 106:130–135.[Abstract/Free Full Text]
  6. Mogford J.E, Tawil N, Chen A, Gies D, Xia Y, Mustoe T.A. Effects of age and hypoxia in TGFbeta1 receptor expression and signal transduction in human dermal fibroblasts: impact on cell migration. J. Cell. Physiol. (2002) 190:259–265.[CrossRef][ISI][Medline]
  7. Leeb S.N, Vogl D, Falk W, Scholmerich J, Rogler G, Gelbmann C.M. Regulation of migration of human colonic myofibroblasts. Growth Factors (2002) 20:81–91.[CrossRef][ISI][Medline]
  8. Spinale F, Coker M, Bond B, Zellner J. Myocardial matrix degradation and metalloproteinase activation in the failing heart: a potential target. Cardiovasc. Res. (2000) 46:225–238.[Abstract/Free Full Text]
  9. Yuan W, Varga J. Transforming growth factor-β repression of matrix metalloproteinase-1 in dermal fibroblasts involves Smad3. J. Biol. Chem. (2001) 276:38502–38510.[Abstract/Free Full Text]
  10. Overall C, Wrana J, Sodek J. Transcriptional and post-transcriptional regulation of 72-kDa gelatinase/type IV collagenase by transforming growth factor-β1 in human fibroblasts. J. Biol. Chem. (1991) 266:14064–14071.[Abstract/Free Full Text]
  11. Spinale F.G, Coker M.L, Heung L.J, et al. A matrix metalloproteinase induction/activation system exists in the human left ventricular myocardium and is upregulated in heart failure. Circulation (2000) 102:1944–1949.[Abstract/Free Full Text]
  12. Fedak P, Altamentova S, Weisel R, et al. Matrix remodeling in experimental and human heart failure: a possible regulatory role for TIMP-3. Am. J. Physiol, Heart Circ. Physiol. (2003) 284:H626–H634.[Abstract/Free Full Text]
  13. Klappacher G, Franzen P, Haab D, et al. Measuring extracellular matrix turnover in the serum of patients with idiopathic or ischemic dilated cardiomyopathy and impact on diagnosis and prognosis. Am. J. Cardiol. (1995) 75:913–918.[CrossRef][ISI][Medline]
  14. Hein S, Arnon E, Kostin S, et al. Progression from compensated hypertrophy to failure in the pressure-overloaded human heart. Circulation (2003) 107:984–991.[Abstract/Free Full Text]
  15. Visse R, Nagase H. Matrix metalloproteinases and tissue inhibitor of metalloproteinases. Structure, function and biochemistry. Circ. Res. (2003) 92:827–839.[Abstract/Free Full Text]
  16. Yana I, Weiss S.J. Regulation of membrane type-1 matrix metalloproteinase activation by proprotein convertases. Mol. Biol. Cell (2000) 11:2387–2401.[Abstract/Free Full Text]
  17. Wang X, Pei D. Shedding of membrane type matrix metalloproteinase 5 by a furin-type convertase. J. Biol. Chem. (2001) 276:35953–35960.[Abstract/Free Full Text]
  18. Kang T, Nagase H, Pei D. Activation of membrane-type matrix metalloproteinase 3 zymogen by the proprotein convertase furin in the trans-Golgi network. Cancer Res. (2002) 62:675–681.[Abstract/Free Full Text]
  19. Thomas G. Furin at the cutting edge: from protein traffic to embryogenesis and disease. Nat. Rev. Mol. Cell Biol. (2002) 3:753–766.[CrossRef][ISI][Medline]
  20. Seidah N.G, Mbikay M, Marcinkiewicz M, Chrétien M. Proteolytic and cellular mechanisms in prohormone and proprotein processing. Hook V.Y.H, ed. (1998) Georgetown, TX, USA: R.G. Landes. 49–76.
  21. Henrich S, Cameron A, Bourenkov G.P, et al. The crystal structure of the proprotein convertase furin explains its stringent specificity. Nat. Struct. Biol. (2003) 10:520–526.[CrossRef][ISI][Medline]
  22. Bassi D, Lopez De Cicco R, Mahloogi H, Zucker S, Thomas G, Klein-Szanto A. Furin inhibition results in absent or decreased invasiveness and tumorgenicity of human cancer cells. Proc. Natl. Acad. Sci. U. S. A. (2001) 98:10326–10331.[Abstract/Free Full Text]
  23. McMahon S, Laprise M.-H, Dubois C.M. Alternative pathway for the role of furin in tumor cell invasion process. Enhanced MMP-2 levels through bioactive TGFβ. Exp. Cell Res. (2003) 291:326–339.[CrossRef][ISI][Medline]
  24. Dubois C.M, Blanchette F, Laprise M.H, Leduc R, Grondin F, Seidah N.G. Evidence that furin is an authentic transforming growth factor-beta1-converting enzyme. Am. J. Pathol. (2001) 158:305–316.[Abstract/Free Full Text]
  25. Roebroek A, Umans L, Pauli I, et al. Failure of ventral closure and axial rotation in embryos lacking the proprotein convertase furin. Development (1998) 125:4863–4876.[Abstract]
  26. Iwami K, Ashizawa N, Do Y.S, et al. Comparison of ang II with other growth factors on egr-1 and matrix gene expression in cardiac fibroblasts. Am. J. Physiol, Heart Circ. Physiol. (1996) 270:H2100–H2107.[Abstract/Free Full Text]
  27. Stawowy P, Blaschke F, Kilimnik A, et al. Proprotein convertase PC5 regulation by PDGF-BB involves PI3-kinase/p70s6-kinase activation in vascular smooth muscle cells. Hypertension (2002) 39:399–404.[Abstract/Free Full Text]
  28. Graf K, Neuss M, Stawowy P, et al. Angiotensin II and {alpha}vβ3 integrin expression in rat neonatal cardiac fibroblasts. Hypertension (2000) 35:978–984.[Abstract/Free Full Text]
  29. Gentry L.E, Lioubin M.N, Purchio A.F, Marquardt H. Molecular events in the processing of recombinant type I pre-pro-transforming growth factor beta to the mature polypeptide. Mol. Cell. Biol. (1988) 8:4162–4168.[Abstract/Free Full Text]
  30. Campbell S.E, Katawa L.C. Angiotensin II stimulated expression of transforming growth factor-beta1 in cardiac fibroblasts and myofibroblasts. J. Mol. Cell. Cardiol. (1997) 7:1947–1958.
  31. Siwik D, Chang D, Colucci W. Interleukin-1β and tumor necrosis factor-{alpha} decrease collagen synthesis and increase matrix metalloproteinase activity in cardiac fibroblasts in vitro. Circ. Res. (2000) 86:1259–1265.[Abstract/Free Full Text]
  32. Dinter A, Berger E.G. Golgi-disturbing agents. Histochem. Cell Biol. (1998) 109:571–590.[CrossRef][ISI][Medline]
  33. Rozanov D.V, Strongin A.Y. Membrane type-1 matrix metalloproteinase functions as a proprotein self-convertase. Expression of the latent zymogen in pichia pastoris, autolytic activation, and the peptide sequence of the cleavage forms. J. Biol. Chem. (2003) 278:8257–8260.[Abstract/Free Full Text]
  34. Eghbali M. Cellular origin and distribution of transforming growth factor-beta in the normal rat myocardium. Cell Tissue Res. (1989) 256:553–556.[ISI][Medline]
  35. Sun Y, Zhang J.Q, Zhang J, Ramires F.J. Angiotensin II transforming growth factor-β1 and repair in the infarcted heart. J. Mol. Cell. Cardiol. (1998) 30:1559–1569.[CrossRef][ISI][Medline]
  36. Öklü R, Hesketh R. The latent transforming growth factor β binding protein (LTBP) family. Biochem. J. (2000) 352:601–610.[CrossRef][ISI][Medline]
  37. Antonelli-Orlidge A, Saunders K.B, Smith S.R, D'Amore P.A. An activated form of transforming growth factor beta is produced by coculture of endothelial cells and pericytes. Proc. Natl. Acad. Sci. U. S. A. (1989) 86:4544–4548.[Abstract/Free Full Text]
  38. Cheng S, Lovett D.H. Gelatinase A (MMP-2) is necessary and sufficient for renal tubular cell epithelial-mesenchymal transformation. Am. J. Pathol. (2003) 162:1937–1949.[Abstract/Free Full Text]
  39. Hallenberger S, Bosch V, Angeliker H, et al. Inhibition of furin-mediated cleavage activation of HIV-1 glycoprotein gp160. Nature (1992) 360:358–361.[CrossRef][Medline]
  40. Flanders K.C, Holder M.G, Winokur T.S. Autoinduction of mRNA and protein expression for transforming growth factor-beta 1 in cultured cardiac cells. J. Mol. Cell. Cardiol. (1995) 27:805–812.[CrossRef][ISI][Medline]
  41. Jiang A, Lehti K, Wang X, Weiss S, Keski-Oja J, Pei D. Regulation of membrane-type matrix metalloproteinase 1 activity by dynamin-mediated endocytosis. Proc. Natl. Acad. Sci. U. S. A. (2001) 98:13693–13698.[Abstract/Free Full Text]
  42. Deten A, Holzl A, Leicht M, Barth W, Zimmer H.G. Changes in extracellular matrix and in transforming growth factor beta isoforms after coronary artery ligation in rats. J. Mol. Cell. Cardiol. (2001) 33:1191–1207.[CrossRef][ISI][Medline]
  43. Hayashidani S, Tsutsui H, Ikeuchi M, et al. Targeted deletion of MMP-2 attenuates early LV rupture and late remodeling after experimental myocardial infarction. Am. J. Physiol, Heart Circ. Physiol. (2003) 285:H1229–H1235.[Abstract/Free Full Text]
  44. Zigrino P, Drescher C, Mauch C. Collagen-induced pro-MMP-2 activation by MT1-MMP in human dermal fibroblasts and the possible role of {alpha}2β1 integrins. Eur. J. Cell Biol. (2001) 80:68–77.[CrossRef][ISI][Medline]
  45. Ratnikov B.I, Rozanov D.V, Postnova T.I, et al. An alternative processing of integrin {alpha}v subunit in tumor cells by membrane type-1 matrix metalloproteinase. J. Biol. Chem. (2002) 277:7377–7385.[Abstract/Free Full Text]
  46. Stawowy P, Kallisch H, Veinot J.P, et al. Endoproteolytic activation of {alpha}v integrin by proprotein convertases PC5 is required for VSMC adhesion and integrin-dependent signaling. Circulation (2004) 109:770–776.[Abstract/Free Full Text]
  47. Guo C, Piacentini L. Type I collagen-induced MMP-2 activation coincidents with up-regulation of membrane type 1-matrix metalloproteinase and TIMP-2 in cardiac fibroblasts. J. Biol. Chem. (2003) 278:46699–46708.[Abstract/Free Full Text]
  48. Sawada Y, Inoue M, Kanda T, et al. Co-elevation of brain natriuretic peptide and proprotein-processing endoprotease furin after myocardial infarction in rats. FEBS Lett. (1997) 400:177–182.[CrossRef][ISI][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Stawowy, P.
Right arrow Articles by Graf, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stawowy, P.
Right arrow Articles by Graf, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?