Distinct biological events generated by ECM proteolysis by two homologous collagenases

Significance

Extracellular matrix (ECM) proteolysis is an abundant biochemical process. Here we describe the multilayered biological complexity generated by structurally homologous collagenases (matrix metalloproteinase-1 and matrix metalloproteinase-13) in collagen-rich, native ECM, that may prove central to tissue homeostasis and pathology. The events induced by these two collagenases generate microenvironments characterized by distinct chemical, biomechanical, and morphological ECM properties that lead to different cellular behaviors. Our findings improve the fundamental understanding of selective ECM degradation by homologous collagenases and its impact on cell behavior.

Abstract

It is well established that the expression profiles of multiple and possibly redundant matrix-remodeling proteases (e.g., collagenases) differ strongly in health, disease, and development. Although enzymatic redundancy might be inferred from their close similarity in structure, their in vivo activity can lead to extremely diverse tissue-remodeling outcomes. We observed that proteolysis of collagen-rich natural extracellular matrix (ECM), performed uniquely by individual homologous proteases, leads to distinct events that eventually affect overall ECM morphology, viscoelastic properties, and molecular composition. We revealed striking differences in the motility and signaling patterns, morphology, and gene-expression profiles of cells interacting with natural collagen-rich ECM degraded by different collagenases. Thus, in contrast to previous notions, matrix-remodeling systems are not redundant and give rise to precise ECM–cell crosstalk. Because ECM proteolysis is an abundant biochemical process that is critical for tissue homoeostasis, these results improve our fundamental understanding its complexity and its impact on cell behavior.

The function and integrity of the extracellular matrix (ECM) is vital for cell behavior as well as for whole-tissue homeostasis (1–4). The ECM undergoes constant remodeling during health and disease states. Its components are regularly being deposited, degraded, or otherwise modified. The highly stable fibrillar collagen type I (Col I) is abundant in many organ-derived ECMs and connective tissues (5, 6); it serves as a tissue scaffold, determining ECM mechanical properties and anchoring other ECM proteins necessary for cell function (7). These processes are orchestrated by multiple remodeling enzymes, among which the matrix metalloproteinase (MMP) family plays an important role. Only a few members of this proteinase family, the collagenases, are able to degrade the resistant fibrillar collagens, i.e., Col I, as well as other ECM molecules (8, 9). The collagenases have conserved amino acids in their zinc-containing catalytic domain (8, 10) and display high structural similarities, as reflected by their functional domain organization. Nevertheless, the complex effects exerted by different MMPs on ECM and cells in vivo remain poorly understood.

The enzymatic activity of MMPs, and specifically of collagenases, in vivo is tightly regulated (11), with enzymatic dysregulation causing irreversible damage associated with a variety of diseases. Abnormally elevated levels of MMP-1 or of both MMP-1 and MMP-13 have been associated with different types of cancers and with inflammatory diseases (12–21).

Here, we explored the degradation patterns of natural collagen-rich ECM by two homologous MMPs and tested the response of cells to intact vs. selectively degraded ECM. We collectively profiled the unique remodeling events caused by two secreted collagenases, MMP-1 and MMP-13, by using biochemical, biophysical, and proteomics tools. We show that these proteases drive morphological, biochemical, and viscoelastic changes in the ECM leading to unique ECM–cell crosstalk. We reveal that MMP-1 and MMP-13 cause distinct ECM degradation, bringing about significantly distinct cellular phenotypes. Our findings show the complexity and selectivity of collagenase-associated degradation mechanisms during tissue remodeling; these mechanisms could be used as a tool for future therapeutic interventions.

Results

Selective Degradation of ECM by Collagenases Determines Fibroblast Behavior.

We set out to characterize the specific influences of the highly abundant collagenases on fibroblast–ECM crosstalk. In this study we used natural collagen fascicles (termed “ECM”) from tendons of 6-mo-old rats (22, 23). We examined the effect of this ECM on rat-1 fibroblasts, which are known to be sensitive to ECM composition and rigidity. Moreover, these cells, much like epithelial cells, can exhibit different modes of migration toward different substrates and engage in cell–cell interactions (24, 25). Because fibroblasts inherently express ECM proteins and remodeling enzymes, we conducted our experiments at early stages of interaction (up to 4 h); in this time frame, no collagen deposition or MMP-1 and MMP-13 secretion was detected (Fig. S1 A and B).

Fig. S1.

Detection of collagen deposition and MMP-1 and MMP-13 secretion in rat-1 fibroblasts. (A) Rat-1 fibroblasts were seeded and grown for 24 or 72 h and were imaged using a two-photon microscope in second harmonic generation mode. No collagen deposition was detected in comparison with the positive control (collagen deposition after 21 d). (B) Western blot analysis of medium collected from rat-1 cells demonstrates minimal secretion of MMP-1 and MMP-13. Cells were seeded and grown for 4 or 24 h; then medium was collected (2 mL) and concentrated using an Amicon ultra centrifugal tube (200 µL). As a positive control, 50 ng of recombinant protein was loaded (MMP-1 or MMP-13). Quantification of protein secretion performed using ImageJ detected ∼0.5 ng of MMP-13 at 24 h after cell seeding. (C) Quantification of the motility of cells moving toward natural ECM or degraded by MMP-1 or MMP-13 (n = 50 in each group, *P < 0.05).

Cell motility was characterized using real-time optical imaging. Although fibroblasts demonstrated a clustered, directional motility toward native ECM (control) with a measured velocity of 40 ± 14 µm/h, they showed individual modes of motility with a velocity of 37 ± 5 µm/h when moving toward MMP-1–degraded ECM. In the presence of MMP-13–degraded ECM, the cells exhibited reduced or arrested motility with a measured velocity of 2 ± 2 µm/h (Fig. 1 A–C, Fig. S1C, and Movies S1–S3).

Fig. 1.

Cell–ECM interactions. (A–C) Real-time in vivo imaging demonstrates the morphological features of the cells interacting with natural ECM (A) or ECM degraded by MMP-1 (B) or MMP-13 (C) at different time points. Within the same treatment, colors identify representative cells at different time points. (Scale bars: 15 µm.) (D–F) SEM images of fibroblasts adhered to natural ECM (D) or to ECM degraded by MMP-1 (E) or MMP-13 (F) 4 h after cell seeding. (Scale bars: 20 µm.) (G) Kinetics of cell aspect ratio calculated from real-time in vivo imaging; n = 50 cells per treatment. (H) Cell aspect ratios quantified from SEM images; n = 200 cells per treatment; *P < 0.05; **P < 0.01, t test.

To analyze the overall shape of the cells, we quantified the cell aspect ratio (the length divided by the width) as a function of time using our live-cell imaging data (Fig. 1G). This analysis allowed us to compare precisely the shape of cells migrating on differentially remodeled collagen-rich ECM. Cells imaged on control ECM showed aspect ratios ranging from 1.36 to 1.42, exhibiting flat and stable morphologies. Cells imaged on MMP-1–degraded ECM showed increasing cell aspect ratios over time ranging from 1.49 to 4.5, exhibiting an elongated cell shape. Remarkably, cells imaged on MMP-13–degraded ECM showed cell aspect ratios of 1.34–1.51, exhibiting round and stable morphologies over time. In agreement with the live-cell images reported in Fig. 1 A–C, these results demonstrate that cells migrating on MMP-1–degraded ECM adopted an elongated shape with time, whereas cells migrating on MMP-13–degraded ECM adopted round, symmetrical morphologies. To validate these morphological changes, we further quantified the aspect ratios of cells that adhered to differentially degraded ECM using our SEM images at 4 h after seeding (Fig. 1 D–F). This analysis yielded a trend in cell aspect ratio similar to that obtained with live-cell imaging at the measured time point (Fig. 1H). Overall, these analyses provide cell-shape statistics indicating that rat-1 cells adopt different morphologies upon interaction with differentially degraded Col I-rich ECM.

The differential phenotypes acquired by fibroblasts interacting with degraded ECM suggested a typical and unique cellular response driven by signal modulations transmitted by the ECM (26, 27). Because cells regulate migration, proliferation, and adhesion mainly through the activation of the ERK1/2 cascade (28), we examined the effect of ECM remodeling on this cascade. Fibroblasts adhering to native ECM demonstrate a sustained mode of ERK1/2 activation; in contrast, a transient activation, peaking at 30 or 60 min, was detected in cells adhering to ECM degraded by either MMP-1 or MMP-13. Moreover, cells interacting with degraded ECM exhibited higher levels of total protein (represented by ERK1/2), indicating improved cell adhesion (Fig. S2 A–C). These results are in good agreement with published work (29, 30) demonstrating that adhesion to the ECM permits efficient growth factor-mediated activation of ERK1/2 in fibroblasts. We further quantified the Western blot experiments and calculated the overall adherence rate by applying a linear regression algorithm (Fig. S2D). The results show an increase in the number of cells adhering to degraded ECM in comparison with natural ECM. Remarkably, an increase of ∼34% in cell adherence was observed in cells that adhered to MMP-1– versus MMP-13–degraded ECM. In correlation with the observed changes in cell shape in Fig. 1, these results suggest that degradation of the ECM by MMP-1 results in improved adherence manifested through cell elongation.

Fig. S2.

Cell response to different ECM environments. (A) Western blot analysis demonstrates time-dependent ERK1/2 activation (pERK, Upper) and total protein (ERK1/2, Lower) in rat-1 fibroblasts adhered to natural or MMP-degraded ECM. (B and C) Time-dependent quantification of protein levels of ERK1/2 (B) or ERK1/2 activity (C) from lysates of rat-1 fibroblasts adhered to natural or degraded ECM. Data are presented as mean ± SD of three independent experiments. (D) The overall adherence rate was calculated by quantifying ERK1/2 bands in Fig. 2A as a function of time and applying a linear regression algorithm. (E) Differential gene-expression profile of rat-1 fibroblasts interacting with natural, MMP-1–, or MMP-13–degraded ECM 4 h after cell seeding. We applied a log₂ transformation, floor to 3, and subtracted each entry by the average of control sample genes. The top changing genes were clustered by k-means (n = 15). The color coding refers to global analysis of the entire dataset and shows the minimum and maximum values. Expression levels were considered significantly different from the control when a change of >50% was recorded. (F–H) mRNA expression levels show the induction of genes included in adhesion, proliferation, and morphogenesis in rat-1 fibroblasts interacting with MMP-1– or MMP-13–degraded ECM 4 h after cell seeding. Data are presented as mean ± SD of three repeats.

By profiling fibroblast transcriptional responses, we found 3,163 genes that were differentially expressed in cells interacting with ECM remodeled by MMP-1 or MMP-13 compared with control. The transcriptional responses of fibroblasts interacting with degraded ECM clustered into several groups of genes involved in adhesion (P < 10⁻⁹), proliferation (P < 10⁻⁵), and morphogenesis (P < 10⁻⁵) (Fig. S2E and Datasets S1 and S2). Induction was observed in protocadherins (Pcdhga3, Pcdhga5, and Pcdhga8), which are involved in cell adhesion and have been shown to interact with a wide range of binding partners, resulting in cytoskeleton changes (31, 32) as well as transcription factor 3 (Tcf3), which is involved in cell migration and wound healing. The elevated expression of the Pcdhga3 and Pcdhga5 genes can be related to direct and indirect cellular adhesion properties corroborated by a large number of other interacting proteins, including phosphatases, kinases, and adhesion molecules.

In addition, we found induction of genes involved in proliferation, such as cyclin-dependent kinases (Cdk2, Cdk9, and Cdk14). Induction also was found in genes that are involved in morphogenesis such as discoidin domain receptor tyrosine kinase 1 (Ddr1), which is activated by various types of collagen and is involved in cell growth and communication (33), and tropomyosin 1 (Tpm1) and cadherins (Cdh2, Cdh3). Our gene-expression data were validated using quantitative PCR (qPCR) analysis of representative genes (Fig. S2 F–H and Dataset S3). Overall, these analyses indicate that the responses of rat-1 cells differ upon their interactions with differentially degraded Col I-rich matrices.

MMP-1 and MMP-13 Produce Distinct Microscale Morphologies and Viscoelastic Alterations of ECM.

Because collagenases are highly potent proteases that are capable of irreversibly cleaving and reshaping the ECM landscape, we next focused on identifying the morphological changes exerted upon the ECM as a result of specific collagenase activity. SEM images demonstrate that natural ECM consists mainly of collagen fibrils aligned along the fiber axis. Upon degradation by MMP-1 or MMP-13, the spatial organization of the collagen fibrils is changed: the fibril alignment is disrupted, producing specific and robust digestion patterns. MMP-1 produces widely distributed broken and bent fibrils exhibiting multiple orientations, whereas MMP-13 causes the splitting of the native collagen fibrils into thinner ones, as opposed to the straight and aligned intact fibrils (Fig. 2). The unique ECM microscale morphologies produced by MMP-1 and MMP-13 may lead to changes in ECM biomechanical properties on the macroscale level.

Fig. 2.

MMP-1 and MMP-13 produce distinct microscale ECM morphologies. Shown are SEM images of natural ECM (A and B), ECM degraded by MMP-1 (C and D), and ECM degraded by MMP-13 (E and F). The ECM degradation was done using 500 nM MMP-1 or MMP-13 at 30 °C for 24 h. The detailed analysis of more than 400 SEM images shows that normal collagen fibrils lying underneath the degraded fibers are observed in all treatments. At chosen conditions the ratio of degraded to undegraded fibrils was estimated to be 0.13:1. Proteolysis could be observed up to ∼80 microns deep within the collagen fascicles. (Scale bars: 1 µm.)

By applying rheology measurements, we determined the frequency dependence of the elastic (G′) and viscous (G′′) moduli, measuring the stress response of the ECM with frequencies varying from 1 to 100 Hz (Fig. S3A). All samples exhibited gel-like behavior: G′ was higher than G′′, and both parameters increased slightly with frequency. A comparative analysis of G′ values points to intact ECM as being stiffer (∼37 kPa) than degraded ECMs (G′ of ∼1.5 kPa for MMP-1 and ∼14 kPa for MMP-13). In addition, the G′′ values revealed that intact ECM has the highest viscosity (∼1.75 kPa), whereas ECM altered by MMP-1 and MMP-13 is less viscous (G′′ ∼0.6 kPa). Taken together, these findings demonstrate that selective degradation results in distinct ECM microscale morphologies and viscoelastic properties that may lead to differential regulation of cell behavior (34, 35).

Fig. S3.

Macrorheological properties of natural and MMP-degraded ECMs. (A) The frequency dependence of the averaged elastic G′ and viscous G″ moduli (filled and empty shapes, respectively) of natural ECM (squares) or ECM degraded by MMP-1 or MMP-13 (triangles or circles, respectively). Degraded ECM samples were prepared by incubation of the fascicles in 500 nM MMP-1 or MMP-13 in TNC buffer at 30 °C for 24 h. The frequency varied from 1 to 100 Hz, and measurements were made in triplicates. (B) Representative TEM images of negatively stained Col I of natural ECM (Top), ECM degraded by MMP-1 (Middle), and ECM degraded by MMP-13 (Bottom). (C) TEM images of Col I bands by negative staining (Upper) and cryoTEM (Lower). Alignment of the two allows the assignment of cryoTEM-imaged bands by the notation of Hodge and Schmitt (36) and led to the identification of the N- and C-telopeptide regions as well as other MMP cleavage sites on cryo-TEM images.

Collagenolysis Is Driven by Distinct Structural Mechanisms.

We used transmission electron microscopy (TEM), cryoTEM, and negative staining to visualize the degradation products present in supernatants after MMP degradation. TEM images of native ECM supernatants revealed extremely low quantities of individual fibrils. The images display empty background areas around highly ordered fibrils, confirming the near absence of degradation, as expected because Col I is very stable and abundantly cross-linked (Fig. 3 A and D and Fig. S3B) (5). In contrast, samples treated with MMP-1 or MMP-13 displayed highly abundant, ruffled fibrils surrounded by unique degradation products, strongly suggesting that these fibrils are formed during MMP degradation (Fig. 3 B, C, E, and F and Fig. S3B). The distinct banding of 67 nm observed in Col I fibrils has been used to correlate protein sequence location to the bands (36), and these assignments were correlated to the bands observed by cryoTEM (Fig. S3C). This correlation led to the identification of the N- and C-telopeptide regions and the site of MMP cleavage (Fig. 3). From these assignments, we observed structural anisotropicity of Col I cleavage in the cases of both proteases. The images reveal the “peeling” of degraded fragments, fringing off from the C-to-N terminus direction of the Col I fibrils (Fig. 4 B, C, E, and F). This directionality in MMP degradation may be dictated by the natural polarity of Col I fibrils, in which C and N termini of collagen molecules directed toward different poles of the fibrils. The anisotropicity of collagen degradation is also confirmed by comparing fibril termini, which display distinct morphologies (Fig. S4). The N-terminal ends of the degraded fibrils are more compact than their C-terminal counterparts, suggesting that fibril degradation generally progresses from the C to the N terminus of the fibril. Most importantly, the cryoTEM images of degraded fibrils show that cross-linked C-telopeptides are not degraded during MMP-1 processing, as indicated by their presence in the background of proteolyzed morphologies extending out from the fibrils (Fig. 3B, green arrows). In comparison, cross-linked C-telopeptides are not present in fibrils degraded by MMP-13 (Fig. 3C), indicating the existence of a highly selective degradation mechanism on the Col I fibril. TEM images of negatively stained samples reveal that both proteases produce heterogeneous populations of degraded products, with triangular microfibril morphologies prevalent in MMP-1–treated ECM and rod-like fragments prevalent in MMP-13–treated ECM (Fig. 3 E and F and Figs. S5 and S6). The normalized distribution of fragment lengths for MMP-1 and MMP-13 showed that the most abundant lengths were 223 ± 15 nm and 82 ± 13 nm for MMP-1 and 207 ± 15 nm and 83 ± 15 nm for MMP-13 (Fig. S7). These distributions reflect the signature cleavage positions at 3/4 and 1/4 of the monomeric length of collagen, as well as other nonclassical cleavage sites marked by the broad Gaussian peak. The individual rod-like fragments resulting from MMP-13 degradation observed in the TEM images (Fig. 3F and Figs. S5 and S6) had a diameter of ∼4 nm, corresponding to the proposed diameters of individual microfibrils (five-molecule bundles) from TEM and diffraction studies (37, 38). We interpret the triangular morphologies present in MMP-1–degraded samples as being formed by bundles of microfibrils that are connected at the C-telopeptide terminus. Our observations strongly suggest that one microfibril is processed as a single cleavage incident. This conclusion is supported by degradation kinetics studies showing a processive burst of 15 ± 4 cleavage events occurring within one cut (39), corresponding to five triple-helical molecules in a single microfibril (5 × 3 = 15 cleavage events).

Fig. 3.

Structural features of fibrillar collagen degradation by MMP-1 and MMP-13. We accumulated more than 2,500 TEM images from more than 300 samples per treatment and analyzed the highly reproducible images. Shown are TEM images of cryo-preserved and negatively stained supernatants of control (A and D), MMP-1–treated (B and E), and MMP-13–treated (C and F) samples. All fibrils show characteristic of the Col I banding pattern of 67 nm. Arrows indicate Col I fibril polarity from the C to the N termini. (Scale bars: 100 nm in A–C; 200 nm in D–F.)

Fig. 4.

Mass spectrometry-based proteomics data of ECM degraded by MMP-1 or MMP-13 determined from silver-stained SDS/PAGE and analyzed by nano-LC-ESI-MS/MS. (A) Relative abundances of matrisome proteins released during ECM degradation by MMP-1 or MMP-13. (B) Zoom-in of relative abundances of core- and matrisome-associated proteins. The color-coded proteins are (1) Col I; (2) collagen type VI; (3) collagen type XV; (4) decorin; (5) fibromodulin; (6) isoform 2 of aggrecan core protein; (7) proteoglycan 4; (8) fibulin 5; (9) tenascin-C; (10) transforming growth factor-β–induced protein Ig-h3 precursor; (11) long isoform of hyaluronan and proteoglycan link protein; (12) lactadherin; (13) myocilin; (14) procollagen C-endopeptidase enhancer; (15) annexin A1; (16) short isoform of annexin A2; (17) annexin A5; (18) serine peptidase inhibitor clade F, member 1; (19) inter–α-trypsin inhibitor heavy chain H3. (C) Col I cleavage sites identified under proteolytic degradation of natural ECM by MMP-1 or MMP-13. Green indicates cleavage sites common to both MMPs. Red indicates cleavage sites detected under degradation of Col I by MMP-1 or MMP-13. Single and double asterisks indicate specific native triple-helical Col I cleavage sites. Spectral counts represent the absolute number identified for each Col I semitryptic peptide and demonstrate the differential efficiency of each MMP at any detected cleavage site. (Data reproducible in five experiments.)

Fig. S4.

TEM images of degraded Col I fibrils. TEM images of negatively stained (A and D) and cryo-preserved (B, C, E, and F) Col I fibrils formed during MMP-1 (A–C) or MMP-13 (D–F) processing. The anisotropicity of Col I degradation by both MMPs may be detected by comparing fibril termini, which display distinct morphologies. The N-terminal ends of the degraded fibrils are more compact than the C-terminal ends, suggesting that fibril degradation progresses mostly from the C to the N terminus of the fibril.

Fig. S5.

Proteolytic Col I fragments detected upon degradation by collagenases. TEM images of negatively stained degraded Col I fragments observed in supernatants of specimens treated by MMP-1 (A–C) or MMP-13 (D–F). Triangular morphologies predominate in the MMP-1–treated samples, and rod-like fragments are prevalent in the MMP-13–treated samples.

Fig. S6.

Small (1/4) degraded Col I fragments visualized in supernatants using TEM. Degradation by MMP-1 (A and B) and MMP-13 (C and D). The small fragments (1/4 of the monomeric length) degraded by MMP-13 are marked by asterisks. The images further confirm that the C-termini telopeptides remain intact under MMP-1 proteolysis and are degraded in the presence of MMP-13.

Fig. S7.

Statistical analysis of the length of degradation products. Histogram of normalized distribution of the length of Col I degraded fragments shows that both collagenases mainly digest Col I at the specific cleavage sites corresponding to ∼3/4 and ∼1/4 of the monomeric length. The broadness of the Gaussian peaks indicates the existence of multiple cleavage sites in addition to the classical one (Gly791–Ile792 in α1 and Gly784–Ile785 in α2). n = 300 degradation products per treatment.

Differential Proteomic Profiles Are Generated During ECM Degradation by Collagenases.

Mass spectrometry analysis [nano-LC-electrospray ionization (ESI)-MS/MS] was used to examine the proteomic profiles of supernatants of ECM degraded either by MMP-1 or MMP-13. This analysis revealed distinctly different degradation patterns for MMP-1 and MMP-13, whereas the control samples, as expected, contained a minimal amount of degradation products (Fig. S8). Fig. 4 A and B and Dataset S4 list the total relative abundances of matrisome proteins released from treated ECM, in which Col I is the most abundantly degraded protein. Because trypsin digestion is highly specific (40), we correlated semitryptic peptides detected by MS with the proteolytic activity of MMPs and determined Col I (and several noncollagenous proteins) unique and common cleavage sites for MMP-1 and MMP-13 (Fig. 4C and Fig. S9). Principal component analysis (PCA) of Col I tryptic peptides resulted in three distinctly isolated, closely clustered populations (Fig. S10), indicating that each collagenase degrades Col I fibrils using a distinct mechanism.

Fig. S8.

Representative silver-stained gel of supernatants of the samples treated by MMPs or natural ECM. Typical silver-stained SDS/PAGE patterns obtained from supernatants of control and treated (500 nM MMP-1 or MMP-13) collagen fascicles after 24 h of incubation at 30 °C. The gel reveals that, in contrast to control, the supernatants of MMP-1– and MMP-13–treated samples show multiple bands with molecular masses lower than 130 kDa, corresponding to degraded collagen fragments and/or other proteins. For MS analysis 1.5-mm gels were used, and 30 µL of samples were loaded. Lines 0 and 21 show the border of the gel, which was used further for MS analysis. These lanes were divided into 21 horizontal slices which were individually analyzed by nano-LC-ESI-MS/MS. The full list of proteins recovered from each of 21 lines for MMP-1 and MMP-13 is presented in Dataset S4.

Fig. S9.

Cleavage sites identified in Col I-rich ECM under proteolytic degradation by MMP-1 or MMP-13. (A) Sporadic cleavage sites of Col I produced by MMP-1 and MMP-13. Green indicates common sites; red indicates individual cleavage sites. (B) Cleavage sites identified in ECM proteins under proteolytic degradation of Col I ECM by MMP-1 or MMP-13. The absolute number of spectral counts identified for each semitryptic peptide from silver-stained SDS/PAGE is shown. The absolute number of spectral counts identified for each Col I semitryptic peptide demonstrating differential efficiency of each MMP to any detected cleavage site.

Fig. S10.

A 3D distribution of the principal component scores of mass spectra of Col I tryptic peptides detected from in-solution digestion. PCA shows the significant differences among mass spectra detected in supernatants of triplicates of the three kinds of samples. The results show the close clustering of samples within each group, indicating low experimental variability within specific groups.

The most striking differences between the two supernatant profiles are the content and relative abundance of noncollagenous components such as proteoglycans (PG), glycoproteins (GP), ECM-affiliated proteins, and other function-related ECM regulators. Fig. 4 shows that selective ECM degradation impacts not only the ECM’s morphology and viscoelastic properties but also its molecular composition, thus adding complexity to the observed effect. For example, our data confirmed the previously reported MMP-13 specificity for the GP tenascin-C (41), and the PG aggrecan (42) and showed much broader specificity of MMP-13 to matrisome proteins. Other differentially released noncollagenous components include PGs such as decorin and fibromodulin and GPs such as fibulin 5, all known to be critical for cell signaling, spreading, and motility (43–45). Taken together, these data demonstrate that both collagenases effectively degrade native collagen-rich ECM in a highly selective mode which consequently releases bioactive matrisome molecules and at the same time exposes new matrix-associated epitopes and protein complexes.

Discussion

The constant remodeling of the ECM environment in healthy and diseased states continuously subjects cells to a variety of stimuli. Nevertheless, we lack an understanding of the cellular implications of these stimuli. This study shows that cell–ECM crosstalk is governed by specific and selective activity of remodeling enzymes that exert intricate effects on the ECM, thus altering its morphology and viscoelastic and biochemical properties. Specifically, our results provide mechanistic insights into how homologous collagenases selectively degrade native collagen-rich ECM.

Cells seeded onto collagen matrix may secrete their own ECM proteins and remodeling enzymes over a longer time (Fig. S1). Therefore, we conducted our experiments at the early stages of cell–ECM interactions. In addition, the control and degraded collagen-rich ECM was washed extensively to minimize the presence of degradation products and exogenously added MMPs. We found that the observed changes in cell shape, motility, and signaling are affected most by the degraded matrix morphology, biochemistry, and mechanical properties and less by the degradation products. However, we could not completely exclude the contribution of cell-derived matrix proteins and proteases to the effects observed on the ECM or to the activation of other cell-produced proteases. Nevertheless, our results clearly demonstrate that homologous collagenases promote differential degradation patterns on natural ECM. Although MMP-1 and MMP-13 are structurally homologous and degrade Col I anisotropically from the C to the N terminus, we show that they have a different specificity and selectivity to natural collagen fascicles: MMP-13 exhibits broader substrate specificity than MMP-1 and produces a greater variety of degradation products. In addition, we found that ECM degradation by either MMP-1 or MMP-13 reveals distinct collagen-cleavage mechanisms, producing characteristic degradation fragments as shown by both TEM images and MS analysis (Figs. 3 and 4). The distribution analysis of Col I fragment lengths showed that each enzyme produced intrapopulation heterogeneity, confirming our MS data and indicating the existence of several cleavage sites on Col I. These significantly different cleavage patterns suggest that MMP-1 and MMP-13 access different epitopes of the assembled or partially digested collagen fibrils.

Our results are supported by recently published data indicating MMP-13’s ability to cleave a range of dissolved collagen type II peptides (46). In addition, regions of helical instability and triple-helix local dissociation recently identified in native hydrated collagen fibrils (47) may enable MMPs to access other exposed sites. These results are further supported by PCA demonstrating the distinct tryptic fragments of native MMP-1– and MMP-13–degraded Col I. Furthermore, TEM and nano-LC-ESI-MS/MS analysis provided proof that the C-telopeptides remain intact in MMP-1–degraded Col I but are cleaved by MMP-13. Previous in vitro and in silico studies suggested that the cleavage of C-telopeptides is a critical initial step in collagenolysis, enabling the MMPs to access the cleavage site (48). Remarkably, our data show that MMP-1 collagenolysis can occur efficiently without prior C-telopeptide cleavage. It is noteworthy that our experimental set-up using native collagen fascicles, a fibroblast cell line, and individual proteases may not completely mimic the complex natural action of collagenases in vivo during health and disease states. For example, excess activity of multiple interstitial collagenases could be correlated with a chronically nonhealing wound (44). Nevertheless, our results highlight another, previously unrecognized level of complexity, showing that each protease exhibits a unique proteolytic pattern within the native tissue. Although the degradation products may differ among different tissues in vivo, our mass spectrometry analysis conducted on the model we used demonstrates the general concept that homologous collagenases do not exhibit redundant enzyme activity at multiple biochemical levels. We show that the degradation of collagenous and noncollagenous proteins, such as decorin, fibromodulin, and aggrecan, which are required for the proper organization of the ECM, may also change the ECM’s spatial organization and its morphology. In addition, under physiological conditions the release or degradation of these matrisome molecules was reported to contribute to cell signaling through their interactions with cell-surface molecules (45, 49, 50). Finally, selective degradation of PGs and GPs is known to contribute to cell adhesion and activation of cell proliferation (43, 44, 51, 52).

We demonstrate that ECM degradation by MMPs improves fibroblasts’ ability to adhere to ECM, suggesting that ECM degradation leads to the exposure of adhesion sites and/or signaling molecules bound to the ECM scaffold. Indeed, some of the genes that were induced in the cells following interaction with the degraded ECM were annotated as adhesion molecules, such as protocadherins (Pcdhga3 and Pcdhga5), belonging to the cadherin family, and morphogenesis/adhesion molecules (Ddr1 and Tpm1). These genes are known to interact with a wide range of binding partners regulating cell adhesion (32). The induced transcriptional responses of genes related to proliferation (Cdk9 and Cdk14) in fibroblasts interacting with MMP-degraded ECM also support the observed induction of the ERK1/2-signaling cascade. We also examined the other main MAPKs signaling pathways (JNK, p38, and AKT) and found no significant activation under the applied experimental set-up. Furthermore, we show an increase in collagen receptor Ddr1, which regulates cellular functions such as cell spreading and polarity, in response to ECM–microenvironment alterations (53, 54). Overall, the remodeling events of collagen-rich ECM by either MMP-1 or MMP-13 contributed to fibroblast–matrix communication, suggesting that such stimulation may be involved in cell migration in normal and pathological processes. For example, MMP-13 has been shown to enhance the remodeling of 3D collagen matrix, to affect cell morphology, and to promote proliferation of dermal fibroblasts in both in vitro and in vivo models of wound healing (44, 55).

Finally, our results confirm that ECM degradation by both collagenases is accompanied by a significant loss of mechanical rigidity on the macroscale level, resulting from the degradation of the outer layers of each fascicle. Both MMP-1 and MMP-13 bring about ECM softening, with MMP-1 having a stronger softening effect than MMP-13. Softening of polyacrylamide substrates and other artificial matrices is known to reduce the spread of fibroblasts, decrease cell motility, and induce cell rounding (56, 57). Remarkably, we observed that softening of native collagen-rich ECM by individual collagenases resulted in differential cell behavior. This finding implies that the biomechanical properties of native ECM constitute only one of the parameters impacting cell motility, shape, and spreading. Our results demonstrate the importance of a combinatorial effect that includes the integration of all the events driven by MMP degradation, namely ECM morphology, molecular composition, and biomechanics, that govern cell behavior. These observations suggest that the concerted action of several collagenases, including MMP-8 and MMP-14, among other proteases detected in various pathologies, will manifest the severity of the disease by interfering with the tissue and ECM integrity in a combined or synergistic manner. Such concerted proteolytic action also can contribute to the activation of other proteases and their tissue inhibitors in vivo.

In conclusion, our results demonstrate the distinct roles of ECM-remodeling enzymes in generating specific ECM properties that affect cells and determine their fate. Our integrated experimental approach identified the specific changes in morphology, biomechanics, and molecular composition that occur in the native ECM during degradation reactions. Our study reveals the exquisite specificity and selectivity of the enzymatic activity of two structurally homologous collagenases in the context of their natural microenvironment. Collectively, our results highlight the importance of selective ECM remodeling.

Materials and Methods

Details of reagents, antibodies, detailed methods for ECM preparation, expression, and purification, activation and enzyme activity assays for MMP-1 and MMP-13, cell culture, RNA isolation, qPCR, Western blot analysis, imaging studies, proteomic analysis by mass spectrometry, rheology, and statistical analysis used in this study can be found in SI Materials and Methods.

SI Materials and Methods

Reagents and Antibodies.

All analytical-grade reagents were purchased from Sigma-Aldrich unless otherwise mentioned. Purified deionized water was prepared using a Milli-Q water-purification system. Polyclonal anti-total ERK1/2 (catalog no. M5670) and phosphorylated ERK1/2 (cat. No M8159) antibodies were purchased from Sigma-Aldrich. Monoclonal MMP-1 antibody was purchased from Thermo Fischer Scientific (catalog no. MA-515872). Monoclonal MMP-13 antibody was purchased from Invitrogen (catalog no. 701287). Secondary HRP-conjugated antibodies (both anti-rabbit and mouse) were purchased from Jackson ImmunoResearch (catalog nos.111-001-003 and 115-001-003, respectively).

Fascicle-Derived ECM Sample Preparation.

Fascicle-derived ECM was prepared from the tails of adult (6-mo-old) Norwegian rats. Specifically, rat tails were dissected, and tendon fascicles (diameter ∼0.6 mm) were gently extracted and were washed extensively in TNC buffer [50 mM Tris (pH 7.4), 150 mM NaCl, 10 mM CaCl₂, 0.02% NaN₃] to remove the macroscopic tissue debris and excess proteases. The samples were then flash-frozen and kept at −80 °C until processed. Digested ECM samples were prepared by incubation of the fascicles in 500 nM MMP-1 or MMP-13 in TNC buffer at 30 °C for 24 h.

Reaction conditions were monitored based on the published protocols of Chung et al. (58) and silver-stained gel-based analysis of degradation products and TEM. Specifically, we have optimized the reaction conditions by measuring fascicle degradation using the conditions listed in the table below:

Temp, °C	MMP-1, nM	MMP-13, nM	Time, h
25	800	800	0, 2, 6, 24
30	500	500	0, 2, 6, 24
35	100	100	0, 2, 6, 24
37	10	10	0, 2, 6, 24
45	5	5	0, 2, 6, 24

Degradation products could be detected at the desired high resolution by TEM at 30 °C after 24 h. The estimated degraded fraction of fascicles is ∼13% within the first 24 h at 30 °C using 500 nM of MMP-1 or MMP-13.Under these conditions we could detect full degradation of fascicles (0.6 mm in diameter and 7 mm in length) over a period of 8 d. Reaction was stopped by the addition of 20 mM EDTA (pH 8.0). The ECM samples then were gently washed with double-deionized water, followed by at least three washings in a suitable buffer. Untreated fascicles were stable for at least 20 d at 30 °C.

MMP Preparation and Activation and Enzymatic Assays.

MMP-1 preparation.

The human proMMP-1 was cloned in the pET3a expression vector. Bacteria were grown in LB sterile medium [1 L containing 10 g Bacto-tryptone, 5 g yeast extract, 10 g NaCl (pH 7.5)] with 150 µg/mL of ampicillin at 37 °C. Protein expression was induced with 0.4 mM isopropyl-β-d-thiogalactoside at an OD₆₀₀ = 0.6, and growth was allowed to continue for a further 4 h. Following expression, the enzyme accumulated in the fraction of inclusion bodies. Importantly, all steps and refolding of proMMP-1 were performed at 4 °C unless otherwise noted. The cells from 1 L of the culture (∼16 g) were then harvested by centrifugation (3,500 × g; Sorvall LYNX4000 centrifuge, 15 min) and were resuspended in 100 mL of lysis buffer [50 mM Tris (pH 8.5), 0.1 M NaCl, 5 mM β-mercaptoethanol, 2 mM EDTA, 0.1% Brij-35 mM containing 1 tablet of Complete (EDTA-free) protease mixture (Boehringer)]. The cells then were passed through a hand homogenizer and, after the addition of ∼10 mg lysozyme, were stirred for 10–20 min at 4 °C. The suspension was then sonicated (six cycles of 10 s on and 20 s off at 65% of Virsonic 60 power amplitude) and centrifuged at 27,000 × g (Sorvall LYNX4000) for 40 min. The pellet was further suspended in 100 mL of buffer containing 50 mM Tris (pH 8.0), 2 M NaCl, 5 mM β-mercaptoethanol, 2 mM EDTA, 0.1% Brij-35, 100 mM MgCl₂ in the presence of 10–20 µL of 10 mg/mL of DNase with 100 mM MgCl₂, sonicated as described above until the sample lost its viscosity, and collected as before. After the centrifugation at 27,000 × g for 40 min, the washing procedure was repeated, and the pellet was suspended in 100 mL buffer containing 50 mM Tris (pH 8.0) and 5 mM β-mercaptoethanol, passed through a hand homogenizer, and centrifuged at 27,000 × g for 40 min. The pellet, containing inclusion bodies was then solubilized in 25 mL of denaturation buffer [50 mM Tris (pH 8.0), 20 mM DTT, 50 mM ZnCl₂, 1 mM acetohydroxamic acid (AHA), 8 M urea], stirred overnight at room temperature, and filtered (pore size 0.2 µm). The urea-extract of protein was purified further on a Hi-Trap monoQ (GE Healthcare) 5-mL column in an FPLC ACTA purification system, using a gradient of 500 mM NaCl/25 min [buffer A: 6 M urea, 20 mM Tris (pH 8.0); buffer B: 6 M urea, 20 mM Tris (pH 8.0), 1 M NaCl]. Fractions containing MMP-1 were diluted to 75 µg/mL at room temperature using buffer [20 mM Tris (pH 8.0), 20 mM cystamine, 6 M urea] and then were dialyzed against 5–8 L of 50 mM Tris (pH 8.0), 2 mM AHA, 1 mM hydroxyethyl sulfate, 4 M urea, 5 mM CaCl₂, 0.1 mM ZnCl₂, 300 mM NaCl, 5 mM β-mercaptoethanol, and 4 M urea at 4 °C overnight with stirring. The next steps of refolding were done against 2 M urea, 50 mM Tris (pH 8.0), 10 mM CaCl_2, 0.1 mM ZnCl_2, 300 mM NaCl, 2 mM AHA overnight, with stirring, at 4 °C and 50 mM Tris (pH 8.0), 10 mM CaCl₂, 0.1 mM ZnCl₂, 300 mM NaCl, 2 mM AHA.

The renatured proteins were then filtered through a membrane with a 0.2-µm cutoff, concentrated to ∼10 mL by Amicon cells (Millipore) with a 10 molecular-weight-cut-off polyethersulfone membrane, and purified by size-exclusion chromatography using a Superdex 75 26/60 (GE Healthcare) pre-equilibrated with 50 mM Tris (pH 8.0), 300 mM NaCl, 10 mM CaCl₂. The fraction eluted at 130–155 mL of the SEC column was concentrated to ∼3–5 μM and was stored at −80 °C in TNC buffer with 10% (vol/vol) glycerol.

MMP-13 preparation.

The human proMMP-13 was cloned in the pCEP4 expression vector. HEK293 cells expressing Epstein–Barr nuclear antigen (EBNA) contain a pCEP4 expression plasmid with FLAG-proMMP-13. The cells were initially grown in DMEM supplemented with 10% (wt/vol) FBS with penicillin/streptomycin. Once the cells looked viable and begun to divide, Geneticin (Sigma G418) was added to adjust the concentration of 250 µg/L (for EBNA-1–expressing cells). The cells were selected for Hygromycin B resistance in medium containing 1,000 µg/mL Hygromycin B. Once the cells were growing well, they were passaged twice a week. When cells were confluent on 15-cm dishes, the medium was replaced with DMEM containing penicillin/streptomycin and 0.2% lactalbumin enzymatic hydrolysate (LEH; basically amino acids; Sigma L9010). The medium was collected once a week, centrifuged to get rid of cell debris, and frozen at −20 °C. The yield for Flag-tagged proMMP-13 from 1 L of medium was about 0.7 mg. A hand-made 2- to 3-mL column with the resin flag was equilibrated with TNC buffer [50 mM Tris (pH 7.5), 150 mM NaCl, 10 mM CaCl₂], and the collected medium was loaded with flow rate of 1–1.5 mL/min. The column then was washed by TNC buffer, followed by 50 mM Tris (pH 7.5), 1 M NaCl, 10 mM CaCl₂, and washed again by TNC. The protein was eluted with 3 × 5 mL Flag peptide (0.2 mg/mL in TNC). The eluted solution was concentrated to 2–5 mL and was loaded on Superdex 200 16/60 gel-filtration column (GE Healthcare) in TNC buffer. The higher-molecular-weight peak shoulder on the main peak corresponds to the proMMP-13–TIMP-1 complex. ProMMP-13 was eluted at 72–75 mL of column volume and was stored at −80 °C in TNC with 10% glycerol.

Activation of proMMPs.

MMP-1 and MMP-13 were activated with 1 mM APMA (4-aminophenylmercuric acetate) in TNC buffer [50 mM Tris⋅HCl (pH 7.5), 150 mM NaCl, 10 mM CaCl₂, 0.02% NaN₃] at 37 °C for 60 min, and enzymatic activity was tested (58).

MMP enzymatic assays.

Enzymatic assay with a fluorogenic peptide.

The enzymatic activity of MMP-1 and MMP-13 was measured at 37 °C by monitoring the hydrolysis of the fluorogenic peptide Mca-Pro-Leu-Gly-Leu-Dpa-Ala-Arg-NH₂ at λex = 340 nm and λem = 390 nm as previously described (59). The enzymatic reaction was initiated by the addition of different concentrations of fluorogenic peptide (0–100 µM). Fluorescence was recorded immediately and continuously for 30 min. Initial reaction rates were measured, and V_max and K_m were calculated. For MMP-1, the V_max = 27 ± 2 relative fluorescence units (RFU)/s, and K_m = 24.4 ± 3.0 µM. For MMP-13, the V_max = 26 ± 4 RFU/s, and K_m = 24.3 ± 3.2 µM.

Bicinchoninic acid assay.

To quantify and compare the amounts of degradation products released during ECM degradation reactions by MMP-1 and MMP-13, the supernatants were subjected to protein quantification with the bicinchoninic acid (BCA) protein assay kit (ab102536; Abcam). Six hundred to eight hundred micrograms per milliliter of total protein were detected in the supernatants of samples treated by MMPs in contrast to 0.03 µg/mL in control samples.

Cell Culture.

Rat-1 immortalized fibroblasts were acquired from ATCC (CRL 2210) and were cultured in DMEM (Invitrogen) supplemented with 2 mM l-glutamine, 1% penicillin/streptomycin (Invitrogen), and 10% FBS. Cells were maintained at 37 °C in a humidified atmosphere of 95% air and 5% CO₂. Rat-1 cells were detached using 0.25% trypsin/EDTA (Life Technologies).

Time-Lapse Video Microscopy.

ECM samples (natural or degraded by MMPs) were prepared as described (SI Materials and Methods, Fascicle-Derived ECM Sample Preparation), and the reaction was stopped by the addition of 20 mM EDTA. ECM then was gently washed with double-distilled water and placed on a eight-well flat-bottomed µ-Slide (ibidi, GmbH). ECM samples were washed three times with sterile PBS and twice with DMEM. Rat-1 fibroblasts (1.5 × 10⁵ cells/mL) then were seeded on the wells with degraded and natural ECM in serum-starving, phenol-free medium (DMEM supplemented with 2 mM l-glutamine, 1% penicillin/streptomycin, and 0.1% FBS). Each slide was placed in the stage incubator on a DeltaVision Core microscope with phase-contrast optics at a magnification of 60× at 37 °C and 5% CO₂. Images were captured with a CoolSNAP HQ2 CCD camera every 5 min for periods of up to 4 h. Data acquisition and movie assembly were performed using softWoRx for Linux; movies were imported into QuickTime format (version 10.0, Apple) for further analysis using Photoshop CS4 (Adobe). Cell velocities and the cell axial ratio were quantified using ImageJ (P < 0.05, Student’s t test, n = 50 cells per treatment).

SEM.

ECM samples of 1-cm length were prepared as described (SI Materials and Methods, Fascicle-Derived ECM Sample Preparation) and were washed as described (SI Materials and Methods, Time-Lapse Video Microscopy). Rat-1 fibroblasts (1.5 × 10⁴ cells/mL) were seeded in serum-starving, phenol-free medium (DMEM supplemented with 2 mM l-glutamine, 1% penicillin/streptomycin, and 0.1% FBS) in each well of a Corning flat-bottomed 24-well culture plate containing coverslips with natural and degraded ECM samples and were incubated for 4 h at 37 °C and 5% CO₂. At the end of the incubation period, samples were fixed in a 0.1-M cacodylate buffer solution (pH 7.4) containing 2.5% paraformaldehyde and 2.5% glutaraldehyde (pH 7.2) for 60 min at room temperature and were washed three times in the same buffer. The cells were postfixed in 1% osmium tetroxide in the cacodylate buffer for 1 h and were washed three times in cacodylate buffer. The samples then were stained with 4% (wt/vol) sodium silicotungstate (pH 7.0) for 45 min and were dehydrated through an ascending series of ethanol concentrations up to 100% ethanol. Next, the samples were dried in a critical point dryer and gold-sputtered for imaging. To observe ECM topographies, the process of postfixation with osmium tetroxide was excluded. The samples were observed under a Zeiss FEG Ultra55 SEM operating at 2 kV. Image brightness and contrast levels were adjusted using Photoshop CS4 (Adobe). Cell axial ratios were quantified using ImageJ (P < 0.05, Student’s t test).

Cell Extraction and Western Blotting.

ECM samples of 1-cm length were prepared and washed as described (SI Materials and Methods, Fascicle-Derived ECM Sample Preparation and SI Materials and Methods, Time-Lapse Video Microscopy). Rat-1 fibroblasts were grown as described (SI Materials and Methods, Cell Culture) and were serum starved (0.1% FBS, 16 h) before further analysis. Cells were seeded (7.5 × 10⁴ cells/mL) on wells in a Corning flat-bottomed 24-well culture plate containing either natural or degraded ECM samples and were incubated for 5, 30, 60, 120, or 240 min at 37 °C and 5% CO₂ in a serum-starving medium (DMEM supplemented with 2 mM l-glutamine, 1% penicillin/streptomycin, and 0.1% FBS). At the end of the incubation period, the ECM samples were removed carefully, washed with PBS, and then incubated in radioimmunoprecipitation assay (RIPA) buffer [20 mM Tris (pH 7.4), 137 mM NaCl, 10% glycerol, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 2 mM EDTA, 1 mM PMSF] to produce cell lysates. Using this procedure, we could isolate the cell population that directly adhered to the natural or degraded ECM. Cell lysates then were analyzed by Western blotting using the indicated antibodies (60). Every experiment was performed three times in duplicate to test for reproducibility and to obtain statistically significant data. Quantifications of Western blot experiments were performed using ImageJ. Blots were scanned, and band densities were measured and quantified. The calculated ERK1/2 activity was measured as the ratio of band densities between pERK and ERK1/2 (pERK/ERK1/2).

Differential Gene-Expression Response Analysis.

The Gene Expression Omnibus accession number for the RNA-sequencing (RNA-seq) dataset is GSE79749.

Sample preparation.

The wells of flat-bottomed 96 Nunc plates were completely covered with ECM. The deposition and organization of the collagen fascicles onto the plastic surfaces were not affected by the type of well. Collagenases were added to the wells, and the degraded ECMs were prepared as described (SI Materials and Methods, Fascicle-Derived ECM Sample Preparation). At the end of the degradation reaction, all wells were washed as described (SI Materials and Methods, Fascicle-Derived ECM Sample Preparation and SI Materials and Methods, Time-Lapse Video Microscopy). Rat-1 fibroblasts in a serum-starving medium (DMEM supplemented with 2 mM l-glutamine, 1% penicillin/streptomycin, and 0.1% FBS) were seeded (2.5 × 10⁴ cells per well) in duplicate and were incubated for 4 h at 37 °C and 5% CO₂. The cells that adhered to ECM were sent for whole-transcriptome mRNA profiling.

RNA isolation.

Cells that adhered to native or degraded ECM were directly lysed in the presence of QIAzol, and total RNA was extracted with the miRNeasy Mini Kit (Qiagen). The RNA integrity number (RIN) was determined using the TapeStation System (Agilent Technologies). RNA quantity was determined by the Qubit Fluorometric Quantitation kit (Life Technologies).

Preparation of RNA-seq libraries.

For preparation of RNA-seq libraries, total RNA was fragmented into an average size of 300 nt by chemical heat (95 °C) treatment for 4.5 min (NEBNext Magnesium RNA Fragmentation Module; New England Biolabs). The 3′ polyadenylated fragments were enriched by selection on poly dT beads (Dynabeads Invitrogen). Strand-specific cDNA was synthesized using a poly T-VN oligo (18 T) and AffinityScript reverse transcriptase enzyme (Agilent). dsDNA was obtained using the Second Strand Synthesis kit (New England Biolabs). DNA ends were repaired using T4 polynucleotide kinase and T4 polymerase (NEBNext). After the addition of an adenine base residue to the 5′ end using Klenow enzyme (NEBNext), a barcode Illumina-compatible adaptor (IDT) was ligated to each fragment. The washed DNA fragment was amplified by PCR (12 cycles) using primers (IDT) specific to the ligated adaptors. The quality of each library was analyzed by TapeStation (Agilent).

Preprocessing of RNA-seq data.

All reads were aligned to the rat reference genome (Rat RN5) using the TopHat aligner (61). Raw gene expression levels were calculated using the ESAT program (garberlab.umassmed.edu/software/esat/). ESAT takes as input a transcriptome annotation set (we used RefSeq annotations downloaded from the University of California, Santa Cruz genome browser) and uses a scanning window approach to assign the most enriched peak to each annotation. This procedure is done for every isoform, and the ends are collapsed for the genes. We use the collapsed gene counts for our analysis. Based on the principles of the protocol, raw read counts can be used directly for gene expression, because gene length bias is eliminated when sequencing fixed-length fragments at the gene end. Normalization was done using DESeq (62) based on the negative binomial distribution and a local regression model. For the data table used for the heat map, we applied a log₂ transformation, floor to 3, and subtracted each entry by the average of control sample genes. Top changing genes were clustered by k-means (n = 15). Significant change was considered as more than 50%.

Enrichment analysis of biological functions and pathways.

For pathways and functional analysis we compared cellular pathways using the cbl-gorilla.cs.technion.ac.il/ database. Function and pathway enrichments in each profile were calculated using a Wilcoxon test P value.

qPCR run and analysis.

RNA was isolated from cultured cells using the miRNeasy extraction kit (Qiagen) according to the manufacturer’s instructions. cDNA was obtained with cDNA Reverse Transcription Kit (Applied Biosystems). qPCR was performed using an ABI 7300 instrument (Applied Biosystems). Values were normalized to GAPDH control. Each RNA sample was run in triplicate, and results are a mean of two or three separate runs. Data are presented as mean fold changes using the 2^−∆∆CT method on the mean of all six measurements (two runs in triplicate). Namely, the height of columns on the graph corresponds to the 2^−∆∆CT (Dataset S1). The SD of the mean was calculated for the original ∆CT data. Results were statistically analyzed on original data using Student’s t test in Microsoft Office Excel or Prism.

A list of primers used for qPCR assay is presented in Dataset S3.

TEM.

Sample preparation.

The samples were prepared, and reactions were stopped as described (SI Materials and Methods, Fascicle-Derived ECM Sample Preparation). Because fascicles did not degrade in the absence of MMPs, there were no individual fibrils in samples untreated by collagenases. To obtain individual fibrils in control samples, the collagen fibrils were gently peeled from fascicles before incubation. The fascicles were kept hydrated. Fibrils prepared in this manner were used as controls.

Cryo-TEM microscopy.

Supernatants from degradation experiments (5 μL) were applied to glow-discharged copper TEM grids coated with lacey carbon (SPI Supplies). The samples were blotted and plunged into liquid ethane using a Leica EM-GP automated plunger. Grids were stored in liquid nitrogen, and images were taken in a low-dose mode at −178 °C using a Gatan 626 cryo-transfer holder under a Tecnai T12 electron microscope at 120 kV or under a Tecnai F20 microscope at 200 kV. Images were recorded on either a TVIPS F224 camera or a Gatan US4000 camera. Images were band-pass filtered for figure preparation.

Negative staining.

Supernatants from degradation experiments (5 µL) were deposited on glow-discharged, carbon-coated grids and were stained with 4% sodium silicotungstate (pH 7.0) for 30 s. The samples were then observed under a Tecnai T12 TEM (FEI) operated at 120 kV. Images were recorded with a MegaView III CCD camera (SIS), or a Tietz TVIPS F224 camera.

Proteomic Analysis.

Two approaches were taken to analyze degraded products released by MMP-1 or MMP-13. In the first approach, the supernatants were separated by SDS/PAGE, the lanes with the bands of degraded products were cut into 21 horizontal slices, and each of 21 slices was subjected to in-gel tryptic digestion followed by LC-MS/MS analysis. In the second approach, samples were subjected to in-solution digestion and ion-intensity–based label-free quantification. The mass spectrometry-based proteomics data have been deposited with the ProteomeXchange Consortium via the PRIDE (63) partner repository with the dataset identifier PXD003796 and PXD003553.

Sample preparation.

ECM preparation.

The fascicles were prepared and treated with MMPs for 24 h at 30 °C as described (SI Materials and Methods, Fascicle-Derived ECM Sample Preparation). The supernatants were used for MS-based proteomics. The total amount of degraded products was determined by BCA assay.

SDS/PAGE gels.

For silver-stained gels, 5 μL of sample reduced buffer (4×) was added immediately to 15 μL of the supernatants containing 20 mM EDTA and was boiled for 3 min at 90 °C. The samples then were loaded on a 12% gel of 0.7-mm thickness. For Coomassie-stained gels, 30 μL of samples prepared exactly as described for silver-stained gels were loaded on the 12% gel (1.5-mm thick). PageRuler unstained protein ladder (Fermentas International, Inc.) was used as a standard for molecular weight. Protein bands from the silver-stained gel (21 slices, 1.5 mm high) (Fig. S8) or the 1-cm lane from the Coomassie-stained gels were excised from gel and destained using multiple washings with 50% (vol/vol) acetonitrile in 50 mM ammonium bicarbonate. The protein bands were subsequently reduced, alkylated, and in-gel digested with bovine trypsin (sequencing grade; Roche Diagnostics) at a concentration of 12.5 ng/µL in 50 mM ammonium bicarbonate at 37 °C, as described (64). The peptide mixtures were extracted with 80% (vol/vol) CH₃CN and 1% CF₃COOH, and the organic solvent was evaporated in a vacuum centrifuge. The resulting peptide mixtures were reconstituted in 80% formic acid and were diluted immediately 1:10 with Milli-Q water before the analysis by online reversed-phase nano-LC-ESI-MS/MS.

MS from solutions.

Immediately after the enzymatic reaction was stopped, the supernatants were transformed into separate tubes and brought for MS analysis. The total protein concentration of the samples was adjusted so that an equal amount of protein was analyzed by LC-MS/MS for all samples. Proteins first were reduced using DTT (Sigma Aldrich) to a final concentration of 5 mM and were incubated for 30 min at 60 °C, followed by alkylation with 10 mM iodoacetamide (Sigma Aldrich) for 30 min in the dark at 21 °C. Proteins then were digested using trypsin (Promega) at a trypsin/protein ratio of 1:50 (wt/wt) for 16 h at 37 °C. Digestions were stopped by the addition of formic acid to a concentration of 1%. The samples were stored in aliquots at −80 °C.

Instrumentation.

Nano-LC-ESI-MS/MS of in-gel digested samples.

Peptide mixtures were separated by online reversed-phase nanoscale capillary LC and were analyzed by ESI-MS/MS. For LC-MS/MS, the samples were injected onto an in-house–made 15-cm reversed-phase spraying fused-silica capillary column (i.d. 75 µm) packed with 3 µm ReproSil-Pur C18A18 medium (Dr. Maisch GmbH,), using an UltiMate 3000 Capillary/nano LC System consisting of a FamosTM Micro Autosampler and a SwitchosTM Micro Column Switching Module (LC Packings, Dionex). The flow rate through the column was 250 nL/min. An acetonitrile (ACN) gradient was used with a mobile phase containing 0.1% and 0.2% formic acid in Milli-Q water in buffers A and B, respectively. The injection volume was 5 µL. The peptides were separated with 50-min gradients from 5 to 65% ACN. The LC setup was connected to the LTQ Orbitrap mass spectrometer (Thermo Fisher Scientific) equipped with a nano-ESI source (Thermo Fisher Scientific). In the nano-ESI, the end of the capillary from the nano-LC column was connected to the emitter with pico-tip silica tubing, i.d. 20 µm (New Objective) by a stainless steel union, with a PEEK sleeve for coupling the nanospray to the on-line nano-LC. The voltage applied to the union to produce an electrospray was 2.4 kV. Helium was introduced as a collision gas at a pressure of 3 psi. The LTQ Orbitrap mass spectrometer was operated in the data-dependent mode with the resolution set to a value of 60,000. Up to seven of the most intense ions per scan were fragmented and analyzed in the linear trap. For the analysis of tryptic peptides, survey scans were recorded in the FT mode followed by data-dependent collision-induced dissociation (CID) of the seven most intense ions in the linear ion trap (LTQ).

LC-MS/MS analysis of the tryptic peptides generated by in-solution digestion.

Ultra liquid chromatography/MS-grade solvents were used for all chromatographic steps. Each sample was loaded using splitless nano-ultra performance liquid chromatography (UPLC) (10 kpsi; nanoAcquity; Waters). The buffers used were H₂O + 0.1% formic acid (A) and ACN + 0.1% formic acid (B). Desalting of samples was performed online using a reversed-phase C18 trapping column (180-mm i.d., 20-mm length, 5-mm particle size) (Waters). The peptides were separated using a C18 T3 HSS nano-column (75-mm i.d., 150-mm length, 1.8-mm particle size) (Waters) at 0.4 μL/min. The mobile phase consisted of H₂O + 0.1% formic acid (A) and ACN + 0.1% formic acid (B). The following gradient was used to elute the peptides: 3–30% B in 50 min, 30–95% B in 10 min, hold for 7 min, and back to initial conditions. The nano-UPLC system was coupled online through a nano-ESI emitter (7-cm length, 10-mm tip; New Objective) to a quadrupole ion mobility time-of-flight (Q-IM-ToF) mass spectrometer (Synapt G2 HDMS; Waters) tuned to >20,000 mass resolution for both MS and MS/MS (FWHM). Data were acquired using MassLynx version 4.1 in an MSe Data Viewer (Waters). In low-energy (MS) scans, the collision energy was set to 5 eV and was ramped from 17 to 40 eV for high-energy (MS/MS) scans. For both scans, the mass range was set to 50–1,990 Da with a scan time set to 1 s per scan. A reference compound (Glu-1-Fibrinopeptide B; Sigma) was infused continuously for external calibration using a LockSpray ion source and was scanned every 30 s.

Data processing, searching, and analysis.

In-gel–digested samples.

The spectra acquired from Orbitrap-XL were submitted to the in-house MASCOT server (version 2.4, Matrix Science) (65) and were searched against the UniProt and National Center for Biotechnology Information databases. Search parameters included fixed modification of 57.02146 Da (carboxyamidomethylation) on Cys and variable modifications of 15.99491 Da (oxidation) on Met, 0.984016 Da (deamidation) on Asn and Gln (Q/N), and hydroxylation of Pro. The search parameters were as follows: maximum 2 missed cleavages; initial precursor ion mass tolerance, 10 ppm; fragment ion mass tolerance, 0.6 Da. Half-trypsin cleavage was allowed from either end to detect collagenase-cleaved peptides. The identity of the peptides was determined from the detected CID products by Mascot software and confirmed by manual inspection of the fragmentation series. Relative quantitation of the peptides revealing specific MMP-1 and/or MMP-13 cleavage sites was conducted using the Scaffold software (version Scaffold 3.6.3; Proteome Software Inc.). To validate the datasets generated by MS, database search files generated by Mascot were imported into Scaffold and were analyzed further from within Scaffold using the spectral quantitative value display option with the following filter settings: minimum protein, 99%; minimum no. of peptides, 2; minimum peptide, 95%. All 21 Mascot outputs derived from searches performed on the 21 horizontal gel slices of each biological replicate were imported into Scaffold and were combined, and the number of assigned peptides and spectra in each biological replicate was used for protein identification and quantification. The integrated PeptideProphet (66) and ProteinProphet algorithms (67) were used to control the false-discovery rate; the probabilities were set to a minimum of 95% and 99%, respectively, and at least two uniquely matched peptides per protein were required for confident protein identification. Remarkably, it was shown that trypsin is an exceedingly specific protease, and the existence of semitryptic peptides was described as a result of protein degradation or decomposition of peptides at labile bonds before tandem MS (40). Because silver-stained gel did not show the degradation of natural ECM, the semitryptic peptides, defined by MS, were correlated with the proteolytic activity of MMPs, allowing the estimation of fibrillary Col I cleavage sites. The cleavage sites from five experiments provided from silver and Coomassie blue stains were analyzed. The list of proteolytic cleavage sites was divided into those reproducible in all experiments and those that were randomly detected.

Analysis of the tryptic peptides generated by in-solution digestion.

Raw data from the mass spectrometer were imported into the Rosetta Elucidators System, version 3.3 (Rosetta Biosoftware). Elucidator was used for alignment of raw MS1 data in retention time and m/z dimensions. Aligned features were extracted, and quantitative measurements were obtained by integration of 3D volumes (time, m/z, intensity) of each feature, as detected in the MS1 scans. In parallel, database searching was carried out using ProteinLynx Global Server version 2.5 with the Ion Accounting algorithm described by Li et al. (68). Data were searched against the rat UniProt database (version 2011_05), appended with the sequences of MMP-1 and MMP-13. Trypsin was set as the protease. One missed cleavage was allowed, and fixed modification was set to carbamidomethylation of cysteines. Variable modification included oxidation of methionine. The criteria for protein identification were set to a minimum of three fragments per peptide, five fragments per protein, a minimal peptide sequence of six amino acids, and a minimum of two peptides per protein. Data also were searched against the randomized version of each database, and the maximum false-identification rate was calculated to be less than 1% at a score cutoff of 6.5. This approach for setting the minimum identification score (termed “PeptideProphet”) is based on reports by Keller and coworkers (66). Additionally, we set the criteria so that peptides had to be detected in at least two of three replicates and in 67% of the samples in any one of the groups. Resulting peptide mixtures were compared based on peak intensities across all samples after alignment of retention time and feature extraction. In all experiments a Student’s t test was used to evaluate the statistical significance of differential changes between the groups of MMP-1, MMP-13, and the control. P values were corrected for multiple comparisons testing by the Benjamini–Hochberg’s Q value method (69). A significance threshold of Q value of ≤0.05 was considered for differences in Col I degradation by MMPs.

Rheological Characterization.

Natural and MMP-1– or MMP-13–degraded ECM‎ was prepared and washed as described (SI Materials and Methods, Fascicle-Derived ECM Sample Preparation). ECM then was assembled densely on the lower plate of the rheometer, ensuring that it covered the entire surface of the plate. Storage (G′) and loss (G′′) moduli were measured using a Thermo Scientific rheometer in a plate–plate (P20 Ti L) configuration using the HAAKE MARS system at 25 ± 0.1 °C (working gap of 0.3 ± 0.05 mm). Dynamic frequency sweep analysis was conducted to measure the frequency-dependent G′ and G′′ moduli of various ECMs in the range of 1–100 Hz. Excess water from intact or degraded ECMs was removed carefully using KimWipe tissue papers.