Top of page

Introduction

Cutaneous wound repair is a complex physiological process in which various growth factors and cytokines regulate the overlapping phases of healing, that is, inflammation, re-epithelialization, granulation tissue formation, matrix remodeling, and wound contraction. Despite effective remodeling of newly formed extracellular matrix (ECM), the end result of wound healing in skin is usually a relatively acellular and collagenous scar, which obtains about 70% of the original tensile strength of intact skin (Singer and Clark, 1999). The cells responsible for production of new collagenous dermal ECM in wound are fibroblasts, which also remodel new collagenous stroma by producing matrix degrading proteinases (Ravanti and Kähäri, 2000). Matrix metalloproteinases (MMPs) are a family of zinc-dependent endopeptidases, which as a group are capable of cleaving all ECM molecules and numerous non-matrix substrates (Ravanti and Kähäri, 2000; Sternlicht and Werb, 2001). MMPs are involved in several physiological and pathological events, such as embryonal tissue development, angiogenesis, and cancer, as well as in different phases of wound repair by stimulating cell migration and activating growth factors, and other proteinases (Ravanti and Kähäri, 2000; Sternlicht and Werb, 2001).

Fibrillar collagens of types I, II, and III are cleaved by members of the collagenase subgroup of MMPs: collagenase-1 (MMP-1) (Goldberg et al., 1986), collagenase-2 (MMP-8) (Hasty et al., 1990), and collagenase-3 (MMP-13) (Freije et al., 1994). Also gelatinase-A (MMP-2) and membrane-type-1 MMP (MMP-14) are capable of cleaving fibrillar collagens (Aimes and Quigley, 1995; Ohuchi et al., 1997). After dermal wounding, the expression of MMP-1 is temporarily induced in migrating keratinocytes in response to contact with type I collagen of dermis (Saarialho-Kere et al., 1992; Pilcher et al., 1997). In addition, MMP-1 is produced by dermal fibroblasts in normal and chronic wounds (Vaalamo et al., 1997). Other MMPs, including stromelysin-1 (MMP-3), stromelysin-2 (MMP-10), metalloelastase (MMP-12), and MMP-14, as well as MMP-2 and gelatinase-B (MMP-9), are also produced in mammalian wounds by cells participating in re-epithelialization, vascularization, and inflammation (Vaalamo et al., 1999). Accordingly, inhibition of MMP activity suppresses dermal wound healing in vivo (Ågren et al., 2001; Beare et al., 2003).

Collagenase-3 (MMP-13) can cleave a wide selection of substrates, including fibrillar collagens and other ECM components, as well as non-matrix components, such as transforming growth factor- (Fosang et al., 1996; Knäuper et al., 1996; Ashworth et al., 1999; D'Angelo et al., 2001). MMP-13 is characterized by limited physiological expression pattern, and it is not expressed by any cells during normal wound healing in adult human skin (Vaalamo et al., 1997). However, MMP-13 is abundantly expressed by fibroblasts in adult gingival wounds (Ravanti et al., 1999b) and fetal skin wounds (Ravanti et al., 2001), which heal with minimal scarring, suggesting a role for MMP-13 in effective remodeling of collagenous granulation tissue resulting in scarless wound repair. Moreover, the expression of MMP-13 by fibroblasts in chronic dermal ulcers suggests a role for MMP-13 in the pathogenesis of chronic ulcers (Vaalamo et al., 1997).

Here, we have examined the role of MMP-13 in fibroblast-mediated remodeling of three-dimensional (3D) collagenous matrix in a floating collagen lattice, a widely used model for dermal wound contraction and matrix reorganization by fibroblasts (Bell et al., 1979). Our results show that adenoviral expression of MMP-13 enhances collagen gel contraction by HSFs and results in altered morphology of their actin-containing extensions. Furthermore, expression of MMP-13 results in activation of Akt and ERK1/2 signaling pathways, and inhibits apoptosis and stimulates proliferation of fibroblasts in 3D collagen. These results suggest a role for MMP-13 in collagen remodeling and contraction by affecting cell interaction with surrounding 3D matrix and increasing cell number in collagen by promoting survival and proliferation.

Top of page

Results

MMP-13 enhances collagen gel contraction by skin fibroblasts

MMP-13 is not expressed in normally healing adult human skin wounds (Vaalamo et al., 1997). However, it is produced by fibroblasts in adult gingival wounds (Ravanti et al., 1999b) and in fetal skin wounds (Ravanti et al., 2001), both characterized by rapid healing with minimal scar formation (Häkkinen et al., 2000; Bullard et al., 2003), suggesting a role for MMP-13 in effective remodeling of collagenous tissue involved in scarless healing. To study in detail the role of MMP-13 in fibroblast-mediated collagenous matrix remodeling, we used adenoviral gene delivery to obtain high-level expression of MMP-13 in primary adult human skin fibroblasts (HSFs), which normally express MMP-13 at low level only when cultured within 3D collagen (Ravanti et al., 1999a). HSFs were transduced with adenovirus encoding human MMP-13 (RAdMMP-13) or with empty control adenovirus (RAdpCA3) and subsequently cultured in floating type I collagen gels for 4 days. Transduction with RAdMMP-13 resulted in marked dose-dependent production of proMMP-13 protein by HSFs in collagen (Figure 1a). In comparison, production of MMP-1, the expression of which is highly stimulated in fibroblasts by ligation of 21-integrin with type I collagen (Langholz et al., 1995), was not markedly altered by RAdMMP-13-transduced HSFs (Figure 1a). Adenoviral expression of MMP-13 resulted in marked and dose-dependent increase in collagen gel contraction (by up to 60%), as compared to cultures transduced with control virus RAdpCA3 (Figure 1b). Collagen gel contraction by RAdpCA3 and RAdMMP-13-transduced HSFs was first noted after culturing for 24 hours in floating collagen gel and it continued up to 4 days (data not shown).

In further experiments, the effect of exogenously added recombinant human MMP-13 (rMMP-13) on collagen gel contraction by HSFs was also examined. Addition of rMMP-13 to culture medium enhanced collagen contraction by HSFs in a dose-dependent manner, the maximal effect noted with a concentration of 600 ng/ml (Figure 1c and d). Production of proMMP-1 by dermal fibroblasts in collagen gel was not altered by rMMP-13 (Figure 1d).

Adenovirally produced MMP-13 is activated by fibroblasts

MMPs are generally secreted as latent zymogens and activated in the pericellular space (Ravanti and Kähäri, 2000; Sternlicht and Werb, 2001). To investigate whether proMMP-13 produced by dermal fibroblasts in 3D collagen after adenoviral infection is activated and exerts collagenolytic activity, we used an in vitro assay where MMP-13 was harvested from the sample media with specific antibody and incubated with a Europium- (Eu) and quencher-labeled substrate peptide for MMPs. The substrate peptide was cleaved with high efficiency in a time-dependent manner, when incubated with 4-aminophenylmercuric acetate activated MMP-13 from conditioned medium of RAdMMP-13-infected HSFs. Also, an increase in Eu signal was detected in this medium without external activation of MMP-13, indicating the presence of endogenously activated MMP-13 (Figure 2a). The activated form of MMP-13 in the corresponding sample medium was also detected by Western immunoblot analysis (Figure 2b).

To further study the cleavage of fibrillar collagen in 3D matrix by adenovirally produced MMP-13, HSFs were infected with RAdMMP-13 or with a control adenovirus RAdLacZ and cultured in fluorescein-labeled type I collagen gel for 1–2 days. The gels were fixed, the fibroblasts were stained for filamentous actin (f-actin) to visualize the cell shape, and examined with confocal microscopy. After 2 days of incubation, fluorescent collagen had disappeared around the MMP-13-expressing fibroblasts, indicating pericellular collagenolysis by active MMP-13 in RAdMMP-13-infected fibroblast cultures. In the control adenovirus-infected cultures, the fluorescent collagen was not cleaved in significant amounts (Figure 2c). This was further verified by measuring the released fluorescence in cell culture media, which showed about 20% increase (n=6, P<0.005, Mann–Whitney U-test) of fluorescence release in RAdMMP-13-infected fibroblast cultures. The expression levels of MMP-1 were not altered by MMP-13 as detected by Western immunoblot (data not shown).

MMP-13 inhibitor and expression of TIMP-1 inhibit MMP-13-enhanced collagen gel contraction by HSFs

To examine whether MMP-13-enhanced contraction of collagen is related to its actual MMP activity, we infected HSFs with RAdMMP-13 or with control adenovirus RAdLacZ and cultured the cells in floating 3D collagen supplemented with specific MMP-13 inhibitor. MMP-13 inhibitor (2 M) reduced MMP-13-enhanced collagen gel contraction potently by 60% (P<0.05), whereas the basal contraction by RAdLacZ-infected cells was not markedly affected (Figure 3a). The MMP-13 inhibitor did not interfere with cell viability as detected by nuclear Hoechst staining (data not shown). Inhibition of MMP-13-enhanced collagen contraction by inhibitors of MMP-13 activity was further verified by coinfecting HSFs with adenovirus RAdTIMP-1 for tissue inhibitor of metalloproteinases-1 (TIMP-1) together with RAdMMP-13 or with empty control adenovirus RAdpCA3. MMP-13-enhanced collagen gel contraction was inhibited by adenoviral coexpression of TIMP-1 (Figure 3b). These results indicate that MMP-13 possesses enzymatic activity that augments the contraction process. Transduction of HSFs with RAdMMP-13 and RAdTIMP-1 resulted in potent expression of MMP-13 and TIMP-1, respectively, as detected by Western blot analysis (Figure 3c). The level of endogenous TIMP-1 expression was not altered by MMP-13 inhibitor or by MMP-13. However, the level of endogenous MMP-1 expression was not altered by MMP-13 inhibitor but was slightly downregulated by MMP-13 and TIMP-1 (Figure 3c).

MMP-13 alters actin cytoskeleton and promotes clustering of f-actin in fibroblasts in 3D collagen

Collagen gel contraction by fibroblasts is dependent on the functional structure of cellular f-actin, attachment to surrounding matrix, and to other cells (Cooke et al., 2000; Ehrlich et al., 2000; Grinnell and Ho, 2002). It is regulated by various factors, such as transforming growth factor-, lysophosphatidic acid and platelet-derived growth factor, cell number, collagen concentration, and mechanical stress generated by the surrounding matrix (Grinnell and Ho, 2002; Tamariz and Grinnell, 2002). Fibroblasts cultured in floating 3D collagen gel acquire a dendritic phenotype in contrast to myofibroblastic phenotype typical for fibroblasts in mechanically loaded collagen (Grinnell et al., 2003). To examine the morphology of MMP-13-expressing HSFs in more detail, cells were cultured for 2 days in floating collagen gel, fixed, and stained for f-actin with fluorescent-labeled phalloidin. In accordance with the previous data, control fibroblasts showed dendritic morphology with fine and long f-actin containing projections with few stress fiber-like structures (Figure 4a). Similarly, few stress fibers were observed in the presence of MMP-13 (Figure 4b). However, f-actin of MMP-13-expressing HSFs was organized in numerous brush- or hair-like structures reaching from cell body to adjacent cells and surrounding collagen. Compared to actin structures of control cells, these filaments were clearly shorter and often very complex in structure (Figure 4b). With higher magnification, numerous actin-containing patch-like structures were also detected along the cell projections of RAdMMP-13-infected fibroblasts, suggesting altered interaction between the cells and the surrounding collagen, whereas in control cells f-actin was evenly distributed at the cortical areas of the cells (Figure 4c and d). Quantitation of the relative number of fibroblasts with normal and altered f-actin morphology in three corresponding experiments showed that the majority of cells in 3D collagen-expressing MMP-13 displayed altered f-actin morphology, as compared to RAdLacZ-infected control cultures (Table 1). Previous studies have shown that collagen gel contraction by fibroblasts involves collagen receptor integrins 11 and 21, as well as Arg-Gly-Asp-binding integrin V3 (Gullberg et al., 1990; Schiro et al., 1991; Langholz et al., 1995; Cooke et al., 2000). Incubation with function blocking 2 integrin antibody reduced basal contraction by control adenovirus-infected fibroblasts by 20%, whereas MMP-13-enhanced collagen gel contraction was inhibited by up to 40%, indicating that MMP-13-enhanced contraction is dependent on functional cell attachment to collagen via 21 integrin (data not shown). The antibody against 1-integrin had no effect on collagen contraction at the concentration used. The antibody against V-integrin had no effect on MMP-13-enhanced contraction, whereas it partly inhibited basal collagen contraction. The overall levels of 1 and 2 integrins were relatively low under both culture conditions and no marked differences were detected in the expression levels of 1 or 2 integrins at 24–72 hours in cells expressing MMP-13 (data not shown).

MMP-13 promotes survival and proliferation of fibroblasts in 3D collagen

Fibroblasts have been reported to acquire a quiescent phenotype and undergo apoptosis in floating or mechanically unloaded collagen gel (Fluck et al., 1998; Rosenfeldt and Grinnell, 2000; Niland et al., 2001). In this context, we examined the effect of MMP-13 expression on survival and proliferation of HSFs in floating 3D collagen gel. In accordance with the previous observations, numerous apoptotic fibroblasts with fragmented DNA and condensed nuclei were detected with TUNEL and Hoechst staining, respectively, among uninfected and RAdLacZ-infected control cells after culturing in collagen for 48 hours (Figure 5a). In contrast, the RAdMMP-13-infected HSFs showed normal nuclear morphology and minimal number of apoptotic TUNEL-positive cells (Figure 5a).

To determine, whether increased survival of MMP-13-expressing fibroblasts in collagen gel resulted in increased cell proliferation, we determined DNA synthesis by BrdU labeling. The results showed that MMP-13 expression also resulted in increased DNA synthesis of HSFs by over 20% (P<0.01) (Figure 5b), indicating enhanced cell proliferation and increased cell number in the gels. There was no alteration in DNA synthesis when fibroblasts correspondingly infected with control adenovirus or with RAdMMP-13 were cultured in monolayer on plastic cell culture wells (data not shown).

To further study the mechanism of enhanced cell survival and proliferation by MMP-13, we analyzed phosphorylation of Akt, extracellular signal-regulated kinase1/2 (ERK1/2), and p38, signal molecules involved in the regulation of cell survival and proliferation (Franke et al., 1997; Chang and Karin, 2001). Immunoblot analysis of cell lysates of MMP-13-expressing HSFs showed marked increase in the levels of phosphorylated (serine 473) Akt, indicating activation of PI3K pathway enhancing cell survival in 3D collagen (Figure 5c). In addition, expression of MMP-13 resulted in phosphorylation and activation of ERK1/2, associated with increased cell proliferation. In contrast, phosphorylation of p38 MAPK was unaltered (Figure 5c).

Top of page

Discussion

Fibroblast-mediated contraction of collagenous ECM is an important phase of normal wound healing in adult skin, and it accelerates wound closure by bringing wound edges closer to each other (Tomasek et al., 2002). Furthermore, contraction contributes to remodeling of newly formed collagenous dermal matrix, in which collagen molecules secreted by fibroblasts are assembled into fibrils and oriented in a direction that provides sufficient tensile strength to skin (Prockop and Kivirikko, 1995; Tomasek et al., 2002). Collagen remodeling by fibroblasts is regulated by various factors such as fibroblast–myofibroblast transformation (Tomasek et al., 2002), transforming growth factor-, platelet-derived growth factor, lysophosphatidic acid, mechanical stimulus by the surrounding matrix as well as by cell number in collagen (Grinnell and Ho, 2002; Tamariz and Grinnell, 2002). There is plenty of evidence indicating that collagen remodeling is also dependent on the function of MMPs. This is demonstrated in vitro and in vivo using wide-spectrum chemical MMP inhibitors, which reduced fibroblast-mediated collagen contraction and wound contraction (Scott et al., 1998; Phillips et al., 2003; Mirastschijski et al., 2004). In addition, mice lacking MMP-3 show reduced wound contraction (Bullard et al., 1999) and mice with targeted mutation in collagenase cleavage site in mouse type I collagen 1 chain display impaired contraction and re-epithelialization of the skin wounds (Beare et al., 2003). Furthermore, reports showing elevated levels of several MMPs relative to TIMP expression in scarless healing of wounds in fetal mouse skin (Dang et al., 2003) and increased activation of MMP-2 and collagen gel contraction by oral fibroblasts (Stephens et al., 2001) emphasize the role of MMPs in efficient collagen remodeling.

We have previously noted that human fetal skin fibroblasts express MMP-13 during wound healing (Ravanti et al., 2001), and that MMP-13 is also expressed by fibroblasts during wound repair in adult human gingiva (Ravanti et al., 1999b). Both types of wounds are characterized by rapid closure, efficient remodeling of newly formed collagenous dermis, and scarless healing. In addition, the expression of MMP-13 by human gingival and fetal skin fibroblasts is potently upregulated by transforming growth factor-, whereas adult HSFs normally express MMP-13 at low level only when cultured within 3D collagen (Ravanti et al., 1999a, 1999b, 2001). In the present study, we noted that primary adult HSFs expressing high levels of MMP-13, as a result of adenoviral MMP-13 gene delivery, remodel collagen matrix with high efficiency demonstrated by potent collagen gel contraction. Increased collagen gel remodeling was also noted after addition of rMMP-13 to the culture medium of fibroblasts. However, collagen gel remodeling induced by rMMP-13 was not as potent as that achieved by adenovirally produced MMP-13. It is hypothesized that MMP-13 enhances collagen contraction primarily and ultimately by partial cleavage of collagen fibers, although it is possible that other mechanisms, for example, activation of other factors, may be involved. Our results show that restricted collagenolysis by adenovirally expressed MMP-13 takes place in the proximity of the cell. It is therefore possible that exogenously added latent rMMP-13 is not distributed in high enough concentration to the pericellular areas to be activated and cleave collagen as potently as MMP-13 produced by the fibroblasts.

It has been well documented that fibroblasts cultured in mechanically non-stressed, freely floating collagen gel do not acquire myofibroblastic phenotype characterized by expression of -smooth muscle actin and stress fibers, but they exhibit dendritic morphology with long and thin cell protrusions (Grinnell et al., 2003). Several studies have shown that these fibroblasts are still able to promote collagen remodeling, resulting in collagen gel contraction (Grinnell, 1994). In the present study, staining of HSFs in fluorescent-labeled collagen attached on cover glasses for f-actin revealed numerous stress fibers generated by mechanically loaded matrix. In accordance with the previous observations (Grinnell, 1994), no obvious stress fibers were detected in fibroblasts cultured in floating collagen gel. However, f-actin staining of HSFs expressing MMP-13 revealed that the extensions of these cells were generally shorter and structurally more complex than in control fibroblasts, possibly owing to exposure of cryptic binding sites from collagen by MMP-13 for cell adhesion and active retraction of the extended protrusions.

Interestingly, f-actin of the control fibroblasts cultured in 3D gel was evenly distributed in the dendritic cell protrusions, whereas f-actin of MMP-13-expressing HSFs was condensed to patch-like structures located along the cell extensions, suggesting altered cell contact to surrounding collagen cleaved by MMP-13. Resembling these structures, fibroblasts cultured on a rigid substrate generate focal adhesion sites, in which integrins are clustered on cell membrane together with cellular signaling proteins and cytoskeletal components to integrate intracellular compartment with ECM (Petit and Thiery, 2000). However, when fibroblasts are cultured in mechanically unloaded matrix (e.g. floating collagen gel), the large focal adhesions are absent, and focal adhesion complex proteins, focal adhesion kinase (FAK), talin, paxillin, and p130^cas are downregulated (Wang et al., 2003). It remains to be elucidated whether the actin-containing clusters detected at the cell extensions of MMP-13-expressing HSFs can be distinguished as focal adhesions. However, these structures were detected already after culturing HSFs in collagen gel for 2 days, when the degree of collagen contraction was not yet very extensive, making it unlikely that focal adhesions were formed as a result of increased rigidity of the strongly remodeled collagen. Furthermore, the cell surface levels of collagen receptor integrins 11 and 21, which mediate collagen contraction and are recruited to focal adhesion sites (Schiro et al., 1991; Ivaska et al., 1999), were not altered by MMP-13 expression of HSFs cultured in floating 3D collagen.

A number of studies have demonstrated that fibroblasts in floating or in attached and released collagen gels acquire quiescence involving disruption in ERK signaling (Rosenfeldt and Grinnell, 2000; Fringer and Grinnell, 2003) and subsequent apoptosis (Fluck et al., 1998; Niland et al., 2001; Tian et al., 2002; Xia et al., 2004). Cell surface integrins function as mechanoreceptors for cells in 3D collagen, and focal adhesion kinase and Akt are phosphorylated as a result of collagen ligation with 1 integrin (Carlson et al., 2004; Xia et al., 2004). However, during contraction of collagen gel, 21 integrin appears to serve as a receptor for alterations in mechanical stimulus generated by surrounding collagen and it initiates a signaling cascade leading to dephosphorylation of Akt and subsequent apoptosis (Ivaska et al., 2002; Tian et al., 2002; Carlson et al., 2004; Xia et al., 2004). This process may also involve activation of the function of p53 (Carlson et al., 2004). In contrast to 21 integrin, V-integrin, which binds to Arg-Gly-Asp motif in denatured collagen, has been shown to promote survival of melanoma cells in 3D collagen via inactivation of p53 and activation of MEK1 and ERK1/2 pathway (Bao and Strömblad, 2004).

Here, we show that adenoviral expression of human MMP-13 by HSFs promotes their escape from cell quiescence and apoptosis induced by contractile floating collagen gel, and despite simultaneous enhancement of collagen gel contraction increases their survival and proliferation. This is expected to result in increased number of HSFs in 3D collagen, and efficiently augment collective collagen gel contraction. We detected a potent phosphorylation of Akt in HSFs expressing MMP-13, suggesting disrupted signaling via mechanoreceptor 21-integrin. This could be owing to a shift of collagen ligation via 21-integrin to V-integrin, or altered out-side-in signaling initiated by reduced affinity of 21 to cleaved and denatured type I collagen. In accordance with increased proliferative activity, we also observed activation of ERK1/2 in HSFs expressing human MMP-13. ERK1/2 signaling pathway can be influenced by cell adhesion, cell shape, and cytoskeletal organization (Huang and Ingber, 1999).

To conclude, our results demonstrate a multifunctional role for human MMP-13 in remodeling of collagenous ECM, an important process in normal acute wound healing. Our results show that MMP-13 expression potently enhances collagen gel contraction by adult skin fibroblasts in a cell culture model, which mimics dermal compartment of skin. We also detected various phenotypic chances in these fibroblasts providing mechanisms how human MMP-13 promotes matrix remodeling. We provide evidence showing that MMP-13-mediated collagenolysis promotes escape of fibroblasts from cell quiescence and apoptosis normally induced by floating and contractile collagen gel. This may be an important survival mechanism for dermal fibroblasts embedded in fibrotic collagen in chronic dermal ulcers (Vaalamo et al., 1997). Finally, our results suggest that human MMP-13 may also enhance wound contraction in vivo and promote scarless healing of adult gingival and fetal skin wounds by enhancing matrix remodeling.

Top of page

Materials and Methods

Cell cultures and reagents

Primary adult HSFs were established from a skin punch biopsy obtained from the arm of a healthy male volunteer (age 27 years). All studies were approved by the ethical committee of the University of Turku. Participants gave their written informed consent, and the study was conducted according to the Declaration of Helsinki Principles. HSFs were cultured in DMEM (Flow Laboratories, Irvine, UK) supplemented with 10% fetal calf serum (FCS), 2 mM L-glutamine, 100 IU/ml penicillin G, and 100 g/ml streptomycin. HSFs were used between passages 6 and 15. MMP-13 inhibitor pyrimidine-4,6-dicarboxylic acid, bis-(4-fluoro-3-methyl-benzylamide) was purchased from Calbiochem (Merck Biosciences, Darmstadt, Germany). Function blocking antibodies against 1- and 2-integrins (SR-84 and AK-7, respectively) were purchased from BD Biosciences Pharmingen (San Diego, CA). Hybridoma cell line producing function blocking mAb L230 against V-integrin was obtained from ATCC (Manassus, VA). The FITC-conjugated type I collagen from bovine skin (D-12060) was obtained from Molecular Probes Inc. (Eugene, OR).

Production of rMMP-13

The recombinant baculovirus for human proMMP-13 was generated using the Bac-To-Bac baculovirus expression system (GIBCO BRL, Gaithersburg, MD). To produce the recombinant pFastBAC1 donor plasmid, the region corresponding to nucleotides 10–1,429 of MMP-13 cDNA, originally derived from TNF--treated human SCC cells (Ala-aho et al., 2002), in pCI-neo vector was amplified using PCR by simultaneously introducing NheI and SpeI sites at the 5' and 3' ends, correspondingly. Tev protease consensus sequence (ENLYFQS) and the myc tag sequence (EQKLISEEDL) were introduced to the 3' SpeI site by primer annealing and ligation proceeded by T4 kinase treatment of the primers. The sequence was verified by sequencing. The recombinant baculovirus was generated according to the manufacturer's instructions. For production of proMMP-13, recombinant baculovirus was added in Sf9 cell cultures (ATCC CRL 1711) in SF-900 II SFM serum-free culture medium (GIBCO BRL), supplemented with 0.5 PSN antibiotic mixture (GIBCO BRL) for 3 days at 27°C in a shaker at 125 r.p.m. The cells were pelleted by centrifugation (1,000 g, 5 minutes) and the supernatant was analyzed by Western blotting for proMMP-13 production using myc and MMP-13-specific antibodies. Catalytic activity of MMP-13 was analyzed by radioactive collagen cleavage assay following activation by ammonium polymethacrylate (Ala-aho et al., 2002). The concentration of MMP-13 in the supernatant was determined using Biotrak™ ELISA system according to the manufacturer's instructions (Amersham Pharmacia Biotech, UK).

Adenoviral infection of primary skin fibroblasts

Recombinant replication-deficient adenovirus RAdLacZ (Wilkinson and Akrigg, 1992), which contains the E. coli -galactosidase (lacZ) gene under the regulation of the cytomegalovirus immediate early (CMV IE) promoter has been described previously (Wilkinson and Akrigg, 1992). The construction of control adenovirus RAdpCA3 and recombinant adenoviruses for human MMP-13 (RAdMMP-13) and TIMP-1 (RAdTIMP-1) has been described previously (Wilkinson and Akrigg, 1992; Baker et al., 1996; Ala-aho et al., 2002; Leivonen et al., 2002). HSFs were infected in suspension with adenoviruses at multiplicity of infection (MOI) 50–500 for 16 hours in the presence of 0.5% FCS. Thereafter, culture medium was replaced with fresh DMEM containing 0.5% FCS and HSFs were incubated for an additional 24 hours, detached with trypsin, and suspended into collagen gel.

Construction of 3D collagen gels and analysis of collagen gel contraction

3D collagen gels were prepared from native bovine dermal collagen containing 95% type I collagen and 5% type III collagen (Cellon, Strassen, France) as described previously (Ravanti et al., 1999a). Briefly, acid-soluble Cellon was neutralized by mixing 8 volumes of Cellon with 1 volume of 10 concentrated DMEM and 1 volume of NaOH in 0.2 M HEPES buffer (pH 7.4) to obtain a final collagen concentration of 2.4 mg/ml. To examine the effect of MMP-13 and TIMP-1 expression on collagen gel remodeling by HSFs, uninfected or adenovirus-infected fibroblasts (5 10⁵) were trypsinized and re-suspended into 600 l of neutralized Cellon gel, suspension was cast on pre-solidified Cellon in 24-well plates, and incubated at 37°C for collagen polymerization. Thereafter, the gels were detached from the wells and 600 l of 0.5% FCS DMEM was added. In certain experiments, recombinant human proMMP-13 was added to the medium twice: immediately after collagen gel preparation and 48 hours later. A chemical inhibitor of MMP-13 or vehicle was added to 3D cell cultures simultaneously with medium in 2 M concentration. In experiments with function-blocking antibodies against 2- and 1-integrins, 1 g of either antibody was administered to cell suspension before addition of collagen and 2 g of antibody was added simultaneously with culture medium in final concentration 5 g/ml. For V-integrin, 5 g of antibody was added to cell suspension before collagen addition and 10 g simultaneously with culture medium to get the final concentration of 25 g/ml. Collagen gel contraction by HSFs was observed for 1–4 days, the gels were photographed and their areas were measured using microcomputer imaging device Image Analysis software (Imaging Research Inc., St Catharines, Ontario, Canada).

Analysis of collagenolytic activity

Proteolytic activity of MMP-13 in the conditioned media of HSFs cultured in 3D collagen was analyzed in vitro. Briefly, equal aliquots of conditioned media were incubated in the 96-well plate coated with anti-MMP-13 antibody (clone 181-15A12; Calbiochem, San Diego, CA). After extensive washing, the specifically bound MMP-13 was incubated at room temperature with Eu- and quencher-labeled peptide containing a cleavage site for MMPs. In a subset of samples, MMP-13 was activated with 1 mM 4-aminophenylmercuric acetate for 30 minutes in 37°C. The time-resolved Eu signal produced by peptide cleavage was detected with Victor2™ 1423 Multilabel Counter (Perkin-Elmer Life and Analytical Sciences, Wallac Finland OY, Turku, Finland).

To detect the collagenolytic activity in situ in cell cultures in 3D collagen, RAdLacZ-, and RAdMMP-13-infected fibroblasts were seeded in type I collagen gel containing 3% of fluorescein-conjugated collagen, cast on sterile cover glasses, and cultured for 2 days anchored on glass in the presence of DMEM containing 0.5% FCS. The gels were then fixed with 4% paraformaldehyde at 37°C overnight, and degradation of fluorescein-conjugated collagen from the pericellular areas was detected with Zeiss LSM 510 confocal microscope. Released fluorescence (485/535 nm) in the cell culture media was quantified using Victor2™ 1423 Multilabel Counter (Perkin-Elmer Life and Analytical Sciences, Wallac Finland OY, Turku, Finland).

Staining of f-actin and detection of apoptotic cells

The fibroblast-containing collagen gels were washed with phosphate-buffered saline at 37°C and fixed with 4% paraformaldehyde at 37°C overnight. The gels were permeabilized with 1% Triton X-100 in phosphate-buffered saline. For actin staining, the gels were washed and incubated with tetramethyl rhodamine isothiocyanate-conjugated phalloidin (P-1951, Sigma Chemical Co., St Louis, MO) at concentration 1 g/ml overnight at 4°C. After washing with phosphate-buffered saline, the gels were mounted on object glasses with Mowiol solution and examined and photographed with the Zeiss LSM 510 confocal microscope. To detect the apoptotic cells, the permeabilized cells were stained with Hoechst 33258 and with TUNEL kit (Roche, Mannheim, Germany) according to the manufacturer's instructions, mounted on object glasses, and visualized with fluorescence microscope.

Immunoblot analysis

HSFs were released from collagen by short treatment with bacterial collagenase (type II; Sigma Chemical Co., St Louis, MO) 0.2 mg/ml in phosphate-buffered saline (pH 7.4) containing 1 mM CaCl₂, pelleted, and lysed with SDS sample buffer. Equal aliquots of cell culture media or cell lysates were fractionated electrophoretically in SDS-PAGE and transferred to nitrocellulose membranes. To detect the human MMP-13 and MMP-1 in the media by immunoblotting, mouse mAb against human MMP-13 (181-15A12; Calbiochem, San Diego, CA) and rabbit polyclonal antibody against MMP-1 (AB806; Chemicon International Inc., Temecula, CA) at 0.2 g/ml concentrations were used. To determine the levels of TIMP-1 in the medium, aliquots of conditioned media were reduced with 5% -mercaptoethanol before electrophoretic fractionation and analyzed by Western blotting with rabbit polyclonal antibody (AB800, 0.2 g/ml) (Chemicon International Inc., Temecula, CA). For analyzing cell lysates, polyclonal antibody against -actin (A1978) was purchased from Sigma Chemical Co (St Louis, MO), and for phospho-ERK1/2, ERK1/2, phospho-p38, p38, and phospho-Akt (Ser473) from Cell Signaling Technology (Beverly, MA). Anti-Akt antibody was purchased from Santa Cruz Biotechnology Inc. (sc-1618, Santa Cruz, CA). The specific binding of primary antibodies was detected with peroxidase-conjugated secondary antibodies and visualized by enhanced chemiluminescence (Amersham Biosciences).

Analysis of DNA synthesis

DNA synthesis was measured with colorimetric immunoassay quantifying the incorporation of BrdU into DNA according to the manufacturer's instructions (Roche, Mannheim, Germany). BrdU was added in culture media 24 hours after casting fibroblasts in collagen gel or when cultured in monolayer on plastic cell culture wells, 24 hours after infection, and incubated for 24 hours. For analysis, the fibroblasts were released from gels with bacterial collagenase treatment as described above.

Top of page

Conflict of Interest

The authors state no conflict of interest.

Top of page

References

1. Ågren MS, Mirastschijski U, Karlsmark T, Saarialho-Kere UK (2001) Topical synthetic inhibitor of matrix metalloproteinases delays epidermal regeneration of human wounds. Exp Dermatol 10:337–348 | Article | PubMed | ISI | ChemPort |
2. Aimes RT, Quigley JP (1995) 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 270:5872–5876 | Article | PubMed | ISI | ChemPort |
3. Ala-aho R, Johansson N, Baker AH, Kähäri VM (2002) Expression of collagenase-3 (MMP-13) enhances invasion of human fibrosarcoma HT-1080 cells. Int J Cancer 97:283–289 | Article | PubMed | ISI | ChemPort |
4. Ashworth JL, Murphy G, Rock MJ, Sherratt MJ, Shapiro SD, Shuttleworth CA et al. (1999) Fibrillin degradation by matrix metalloproteinases: implications for connective tissue remodelling. Biochem J 340:171–181 | Article | PubMed | ISI | ChemPort |
5. Baker AH, Wilkinson GW, Hembry RM, Murphy G, Newby AC (1996) Development of recombinant adenoviruses that drive high level expression of the human metalloproteinase-9 and tissue inhibitor of metalloproteinase-1 and -2 genes: characterization of their infection into rabbit smooth muscle cells and human MCF-7 adenocarcinoma cells. Matrix Biol 15:383–395 | Article | PubMed | ISI | ChemPort |
6. Bao W, Strömblad S (2004) Integrin v-mediated inactivation of p53 controls a MEK1-dependent melanoma cell survival pathway in three-dimensional collagen. J Cell Biol 167:745–756 | Article | PubMed | ChemPort |
7. Beare AH, O'Kane S, Krane SM, Ferguson MW (2003) Severely impaired wound healing in the collagenase-resistant mouse. J Invest Dermatol 120:153–163 | Article | PubMed | ISI | ChemPort |
8. Bell E, Ivarsson B, Merrill C (1979) Production of a tissue-like structure by contraction of collagen lattices by human fibroblasts of different proliferative potential in vitro. Proc Natl Acad Sci USA 76:1274–1278 | Article | PubMed | ChemPort |
9. Bullard KM, Longaker MT, Lorenz HP (2003) Fetal wound healing: current biology. World J Surg 27:54–61 | Article | PubMed |
10. Bullard KM, Lund L, Mudgett JS, Mellin TN, Hunt TK, Murphy B et al. (1999) Impaired wound contraction in stromelysin-1-deficient mice. Ann Surg 230:260–265 | Article | PubMed | ISI | ChemPort |
11. Carlson MA, Longaker MT, Thompson JS (2004) Modulation of FAK, Akt, and p53 by stress release of the fibroblast-populated collagen matrix. J Surg Res 120:171–177 | Article | PubMed | ChemPort |
12. Chang L, Karin M (2001) Mammalian MAP kinase signalling cascades. Nature 410:37–40 | Article | PubMed | ISI | ChemPort |
13. Cooke ME, Sakai T, Mosher DF (2000) Contraction of collagen matrices mediated by 21 and v3 integrins. J Cell Sci 113:2375–2383 | PubMed | ChemPort |
14. Dang CM, Beanes SR, Lee H, Zhang X, Soo C, Ting K (2003) Scarless fetal wounds are associated with an increased matrix metalloproteinase-to-tissue-derived inhibitor of metalloproteinase ratio. Plast Reconstr Surg 111:2273–2285 | PubMed |
15. D'Angelo M, Billings PC, Pacifici M, Leboy PS, Kirsch T (2001) Authentic matrix vesicles contain active metalloproteases (MMP). A role for matrix vesicle-associated MMP-13 in activation of transforming growth factor-. J Biol Chem 276:11347–11353 | Article | PubMed | ChemPort |
16. Ehrlich HP, Gabbiani G, Meda P (2000) Cell coupling modulates the contraction of fibroblast-populated collagen lattices. J Cell Physiol 184:86–92 | Article | PubMed | ChemPort |
17. Fluck J, Querfeld C, Cremer A, Niland S, Krieg T, Sollberg S (1998) Normal human primary fibroblasts undergo apoptosis in three-dimensional contractile collagen gels. J Invest Dermatol 110:153–715 | Article | PubMed | ISI | ChemPort |
18. Fosang AJ, Last K, Knäuper V, Murphy G, Neame PJ (1996) Degradation of cartilage aggrecan by collagenase-3 (MMP-13). FEBS Lett 380:17–20 | Article | PubMed | ChemPort |
19. Franke TF, Kaplan DR, Cantley LC (1997) PI3K: downstream AKTion blocks apoptosis. Cell 88:435–437 | Article | PubMed | ISI | ChemPort |
20. Freije JM, Diéz-Itza I, Balbín M, Sánchez LM, Blasco R, Tolivia J et al. (1994) Molecular cloning and expression of collagenase-3, a novel human matrix metalloproteinase produced by breast carcinomas. J Biol Chem 269:16766–16773 | PubMed | ISI | ChemPort |
21. Fringer J, Grinnell F (2003) Fibroblast quiescence in floating collagen matrices: decrease in serum activation of MEK and Raf but not Ras. J Biol Chem 278:20612–20617 | Article | PubMed | ChemPort |
22. Goldberg GI, Wilhelm SM, Kronberger A, Bauer EA, Grant GA, Eisen AZ (1986) Human fibroblast collagenase. Complete primary structure and homology to an oncogene transformation-induced rat protein. J Biol Chem 261:6600–6605 | PubMed | ISI | ChemPort |
23. Grinnell F (1994) Fibroblasts, myofibroblasts, and wound contraction. J Cell Biol 124:401–404 | Article | PubMed | ISI | ChemPort |
24. Grinnell F, Ho CH (2002) Transforming growth factor stimulates fibroblast-collagen matrix contraction by different mechanisms in mechanically loaded and unloaded matrices. Exp Cell Res 273:248–255 | Article | PubMed | ISI | ChemPort |
25. Grinnell F, Ho CH, Tamariz E, Lee DJ, Skuta G (2003) Dendritic fibroblasts in three-dimensional collagen matrices. Mol Biol Cell 14:384–395 | Article | PubMed | ISI | ChemPort |
26. Gullberg D, Tingström A, Thuresson AC, Olsson L, Terracio L, Borg TK et al. (1990) 1 integrin-mediated collagen gel contraction is stimulated by PDGF. Exp Cell Res 186:264–272 | Article | PubMed | ISI | ChemPort |
27. Häkkinen L, Uitto VJ, Larjava H (2000) Cell biology of gingival wound healing. Periodontol 24:127–152 | Article |
28. Hasty KA, Pourmotabbed TF, Goldberg GI, Thompson JP, Spinella DG, Stevens RM et al. (1990) Human neutrophil collagenase. A distinct gene product with homology to other matrix metalloproteinases. J Biol Chem 265:11421–11424 | PubMed | ISI | ChemPort |
29. Huang S, Ingber DE (1999) The structural and mechanical complexity of cell-growth control. Nat Cell Biol 1:E131–E138 | Article | PubMed | ISI | ChemPort |
30. Ivaska J, Nissinen L, Immonen N, Eriksson JE, Kähäri VM, Heino J (2002) Integrin 21 promotes activation of protein phosphatase 2A and dephosphorylation of Akt and glycogen synthase kinase 3. Mol Cell Biol 22:1352–1359 | Article | PubMed | ISI | ChemPort |
31. Ivaska J, Reunanen H, Westermarck J, Koivisto L, Kähäri VM, Heino J (1999) Integrin 21 mediates isoform-specific activation of p38 and upregulation of collagen gene transcription by a mechanism involving the 2 cytoplasmic tail. J Cell Biol 147:401–416 | Article | PubMed | ISI | ChemPort |
32. Knäuper V, López-Otín C, Smith B, Knight G, Murphy G (1996) Biochemical characterization of human collagenase-3. J Biol Chem 271:1544–1550 | Article | PubMed | ISI | ChemPort |
33. Langholz O, Rockel D, Mauch C, Kozlowska E, Bank I, Krieg T et al. (1995) Collagen and collagenase gene expression in three-dimensional collagen lattices are differentially regulated by 11 and 21 integrins. J Cell Biol 131:1903–1915 | Article | PubMed | ISI | ChemPort |
34. Leivonen SK, Chantry A, Häkkinen L, Han J, Kähäri VM (2002) Smad3 mediates transforming growth factor--induced collagenase-3 (matrix metalloproteinase-13) expression in human gingival fibroblasts. Evidence for cross-talk between Smad3 and p38 signaling pathways. J Biol Chem 277:46338–46346 | Article | PubMed | ISI | ChemPort |
35. Mirastschijski U, Haaksma CJ, Tomasek JJ, Ågren MS (2004) Matrix metalloproteinase inhibitor GM 6001 attenuates keratinocyte migration, contraction and myofibroblast formation in skin wounds. Exp Cell Res 299:465–475 | Article | PubMed | ChemPort |
36. Niland S, Cremer A, Fluck J, Eble JA, Krieg T, Sollberg S (2001) Contraction-dependent apoptosis of normal dermal fibroblasts. J Invest Dermatol 116:686–692 | Article | PubMed | ChemPort |
37. Ohuchi E, Imai K, Fujii Y, Sato H, Seiki M, Okada Y (1997) Membrane type 1 matrix metalloproteinase digests interstitial collagens and other extracellular matrix macromolecules. J Biol Chem 272:2446–2451 | Article | PubMed | ISI | ChemPort |
38. Petit V, Thiery JP (2000) Focal adhesions: structure and dynamics. Biol Cell 92:477–494 | Article | PubMed | ISI | ChemPort |
39. Phillips JA, Vacanti CA, Bonassar LJ (2003) Fibroblasts regulate contractile force independent of MMP activity in 3D-collagen. Biochem Biophys Res Commun 312:725–732 | Article | PubMed | ChemPort |
40. Pilcher BK, Dumin JA, Sudbeck BD, Krane SM, Welgus HG, Parks WC (1997) The activity of collagenase-1 is required for keratinocyte migration on a type I collagen matrix. J Cell Biol 137:1445–1457 | Article | PubMed | ISI | ChemPort |
41. Prockop DJ, Kivirikko KI (1995) Collagens: molecular biology, diseases, and potentials for therapy. Annu Rev Biochem 64:403–434 | Article | PubMed | ISI | ChemPort |
42. Ravanti L, Häkkinen L, Larjava H, Saarialho-Kere U, Foschi M, Han J et al. (1999b) Transforming growth factor- induces collagenase-3 expression by human gingival fibroblasts via p38 mitogen-activated protein kinase. J Biol Chem 274:37292–37300 | Article | PubMed | ISI | ChemPort |
43. Ravanti L, Heino J, López-Otín C, Kähäri VM (1999a) Induction of collagenase-3 (MMP-13) expression in human skin fibroblasts by three-dimensional collagen is mediated by p38 mitogen-activated protein kinase. J Biol Chem 274:2446–2455 | Article | PubMed | ISI | ChemPort |
44. Ravanti L, Kähäri V-M (2000) Matrix metalloproteinases in wound repair (review). Int J Mol Med 6:391–407 | PubMed | ISI | ChemPort |
45. Ravanti L, Toriseva M, Penttinen R, Crombleholme T, Foschi M, Han J et al. (2001) Expression of human collagenase-3 (MMP-13) by fetal skin fibroblasts is induced by transforming growth factor via p38 mitogen-activated protein kinase. FASEB J 15:1098–1100 | PubMed | ChemPort |
46. Rosenfeldt H, Grinnell F (2000) Fibroblast quiescence and the disruption of ERK signaling in mechanically unloaded collagen matrices. J Biol Chem 275:3088–3092 | Article | PubMed | ISI | ChemPort |
47. Saarialho-Kere UK, Chang ES, Welgus HG, Parks WC (1992) Distinct localization of collagenase and tissue inhibitor of metalloproteinases expression in wound healing associated with ulcerative pyogenic granuloma. J Clin Invest 90:1952–1957 | PubMed | ChemPort |
48. Schiro JA, Chan BM, Roswit WT, Kassner PD, Pentland AP, Hemler ME et al. (1991) Integrin 2 1 (VLA-2) mediates reorganization and contraction of collagen matrices by human cells. Cell 67:403–410 | Article | PubMed | ISI | ChemPort |
49. Scott KA, Wood EJ, Karran EH (1998) A matrix metalloproteinase inhibitor which prevents fibroblast-mediated collagen lattice contraction. FEBS Lett 441:137–140 | Article | PubMed | ISI | ChemPort |
50. Singer AJ, Clark RA (1999) Cutaneous wound healing. N Engl J Med 341:738–746 | Article | PubMed | ISI | ChemPort |
51. Stephens P, Davies KJ, Occleston N, Pleass RD, Kon C, Daniels J et al. (2001) Skin and oral fibroblasts exhibit phenotypic differences in extracellular matrix reorganization and matrix metalloproteinase activity. Br J Dermatol 144:229–237 | Article | PubMed | ISI | ChemPort |
52. Sternlicht MD, Werb Z (2001) How matrix metalloproteinases regulate cell behavior. Annu Rev Cell Dev Biol 17:463–516 | Article | PubMed | ISI | ChemPort |
53. Tamariz E, Grinnell F (2002) Modulation of fibroblast morphology and adhesion during collagen matrix remodeling. Mol Biol Cell 13:3915–3929 | Article | PubMed | ISI | ChemPort |
54. Tian B, Lessan K, Kahm J, Kleidon J, Henke C (2002) 1 integrin regulates fibroblast viability during collagen matrix contraction through a phosphatidylinositol 3-kinase/Akt/protein kinase B signaling pathway. J Biol Chem 277:24667–24675 | Article | PubMed | ISI | ChemPort |
55. Tomasek JJ, Gabbiani G, Hinz B, Chaponnier C, Brown RA (2002) Myofibroblasts and mechano-regulation of connective tissue remodelling. Nat Rev Mol Cell Biol 3:349–363 | Article | PubMed | ISI | ChemPort |
56. Vaalamo M, Leivo T, Saarialho-Kere U (1999) Differential expression of tissue inhibitors of metalloproteinases (TIMP-1, -2, -3, and -4) in normal and aberrant wound healing. Hum Pathol 30:795–802 | Article | PubMed | ISI | ChemPort |
57. Vaalamo M, Mattila L, Johansson N, Kariniemi A-L, Karjalainen-Lindsberg M-L, Kähäri V-M et al. (1997) Distinct populations of stromal cells express collagenase-3 (MMP-13) and collagenase-1(MMP-1) in chronic ulcers but not in normally healing wounds. J Invest Dermatol 109:96–101 | Article | PubMed | ISI | ChemPort |
58. Wang YK, Wang YH, Wang CZ, Sung JM, Chin WT, Lin SH et al. (2003) Rigidity of collagen fibrils controls collagen gel-induced down-regulation of focal adhesion complex proteins mediated by 21 integrin. J Biol Chem 278:21886–21892 | Article | PubMed | ChemPort |
59. Wilkinson GW, Akrigg A (1992) Constitutive and enhanced expression from the CMV major IE promoter in a defective adenovirus vector. Nucleic Acids Res 20:2233–2239 | PubMed | ISI | ChemPort |
60. Xia H, Nho RS, Kahm J, Kleidon J, Henke CA (2004) Focal adhesion kinase is upstream of phosphatidylinositol 3-kinase/Akt in regulating fibroblast survival in response to contraction of type I collagen matrices via a 1 integrin viability signaling pathway. J Biol Chem 279:33024–33034 | Article | PubMed | ISI | ChemPort |

Top of page

Acknowledgments

The expert technical assistance of Sari Pitkänen, Johanna Markola, and Marjo Hakkarainen is gratefully acknowledged. This work has been supported by grants from the Academy of Finland (project 45996), Sigrid Jusélius Foundation, The Cancer Research Foundation of Finland, Turku University Central Hospital (project 13336), European Union Framework Programme 6 (CANCERDEGRDOME, LSHC-CT-2003-503297), Turku Graduate School of Biomedical Sciences, Turku University Foundation, and by Farmos Research and Science Foundation.