| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 40, 38765-38771, October 3, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Department of Pharmacology, University of Minnesota, Minneapolis, Minnesota 55455
Received for publication, June 23, 2003 , and in revised form, July 22, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The archetypal MT1-MMP is by far the most studied and is widely considered a model system for the elucidation of MT-MMP functions (6, 20). The first function described for MT1-MMP is its ability to activate pro-MMP-2, an enzyme implicated in many physiological and pathological conditions such as tumor invasion and metastasis (19). Subsequently, similar activity has been confirmed for the other type I transmembrane MT-MMPs, i.e. MT2-, MT3-, and MT5-MMPs (8-10, 21). Interestingly, MT1-MMP knockout mice exhibit far more severe developmental defects than MMP-2 null mice, suggesting that MT1-MMP participates in other biological functions in addition to the activation of MMP-2 (13-15, 22). Indeed, MT1-MMP has been previously shown to have intrinsic proteolytic activities against extracellular components, including type I collagen (13, 22). The apparent failure of MT1-MMP null mice to develop normally has been linked to deficiencies in collagen metabolism in bone and the skeleton (14, 15), confirming that type I collagen is an MT1-MMP substrate in vivo. In addition to collagen, fibrin has been identified as a likely substrate for MT1-MMP during angiogenesis (23). Although the range of substrates for MT-MMPs may continue to expand, the structural basis of MT-MMP substrate specificity remains virtually unknown.
The fact that MT-MMPs are anchored on the plasma membrane represents a formidable challenge to investigate their structure/function relationship at the cellular level. Consequently, most of the characterizations for MT-MMP activity have been performed mostly with purified enzymes in well buffered reaction conditions (13, 22), perhaps reflecting the function of their catalytic domains or proteolytic potentials. Recently, one cellular approach has been adapted to analyze the structure/function of MT-MMPs. By substituting the pexin-like domain of MT4-MMP into the analogous position in MT1-MMP, Itoh et al. (24) concluded that the pexin-like domain of MT1-MMP is required for its ability to mediate pro-MMP-2 activation. Because MT4-MMP is not known to activate pro-MMP-2, this exchange experiment proves that the pexin-like domain of MT1-MMP plays a critical role in specifying its substrate specificity (24), demonstrating the utility of domain exchanging in analyzing the structure/function relationships among closely related members of multigene families such as the MT-MMP gene family.
In this study, we constructed several chimeras to define the contributions of various domains in MT1-MMP and MT3-MMP to pro-MMP-2 activation and type I collagen degradation at the cellular level. Despite the well recognized contribution of the pexin-like domains to MMP substrate specificity (24), our data suggest that the catalytic domains of MT1-MMP and MT3-MMP play a dominant role in defining their activities for biological substrates. Furthermore, we demonstrate that the cytoplasmic domain of MT1-MMP behaves as a negative regulator even when substituted into MT3-MMP. These results suggest that different domains of MT-MMPs could be targeted for potential drug developments.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Expression Constructs—Expression vectors carrying wild-type MT1-MMP (WT19), its cytoplasmic truncation mutant (
C16), and wild-type MT3-MMP have been described previously (25, 26). Chimeras were generated using overlapping primers designed for recombining various domains of MT1-MMP and MT3-MMP through a PCR strategy similar to the one described for MMP-11 (27). The MT1-MMP/MT3-MMP exchange constructs (MT1/MT3-1, -2, -3, -4, -5, -6, and -C1) were made with two or three MT1-MMP and MT3-MMP PCR fragments as illustrated in Fig. 1. These constructs were cloned into the pCR3.1 expression vector and characterized as described previously (10, 13, 27).
|
Cell Transfection and Generation of Stable Cell Lines—Plasmids pCR3.1-MT1WT19, pCR3.1-MT1C16, and pCR3.1-MT3WT9 and MT1-MMP/MT3-MMP exchange constructs (MT1/MT3-1, -2, -3, -4, -5, -6, and -C1) were transfected into MDCK cells by LipofectAMINE (Invitrogen), and stable clones were selected in the presence of G418 as described (10, 11). The stable clones were screened by Western analysis of the cell lysates with anti-MT1-MMP antibodies 502 and 879 and anti-MT3-MMP antibody and by zymographic analysis of pro-MMP-2 activation. Ten representative clones with similar levels of protein expression were selected for further studies: MT1-19-11, MT1/MT3-1-7-8, MT1/MT3-2-9-5, MT1/MT3-3-4-1, MT1/MT3-4-14-2, MT1/MT3-5-23-10, MT1/MT3-6-17-7, MT1/MT3-C1-5-5, MT3-9(FF4-7), and MT1C16-17.
Zymography and Western Blotting—The basic protocols for these two assays have been described previously (10, 11). Briefly, cells were allowed to grow to confluence in 6-well plates, washed three times with phosphate-buffered saline, and allowed to incubate in the presence of Dulbecco's modified Eagle's medium (DMEM; 1 ml/well) with 5% fetal bovine serum (a source of pro-MMP-2). The synthetic MMP inhibitor BB94 (50 µM) was included in the medium as needed. After 24 h of incubation, the medium was harvested and cleared by centrifugation and analyzed on SDS-polyacrylamide gel impregnated with gelatin (1 mg/ml) as described (10, 11). The gels were incubated at 37 °C for 12 h, stained with Coomassie Blue, and destained; and images were captured by scanning. The active species of MMP-2 and its total activity were quantified by densitometry using a Stratagene Eagle-Eye system. For Western blotting, cells grown in 6-well plates were lysed in 250 µl of radioimmune precipitation assay buffer (50 mM Tris (pH 7.5), 150 mM NaCl, 0.25% sodium deoxycholate, 0.1% Nonidet P-40, 10 µM leupeptin, 0.1 µM 5-p-amidinophenylmethanesulfonyl fluoride, and 1 µM aprotinin) with 10 mM of EDTA. The lysates were centrifuged at 14,000 x g for 20 min to remove the cell debris and analyzed by Western blotting with anti-MT1-MMP antibody 502 or 879 or anti-MT3-MMP antibody and developed as described (25, 26).
Invasion of Type I Collagen by MDCK Cells Transfected with MT1-MMP, MT3-MMP, and Their Derivatives—Type I collagen (0.5 ml at 2.0 mg/ml; Collaborative Research, Bedford, MA) was added to each well of a 24-well plate and incubated at 37 °C for 2 h to form a gel. Cells (2.4 x 103) were mixed with fresh medium containing 90% DMEM and 10% fetal bovine serum and added to the top of the gel with care. The medium was renewed every 2 days. BB94 (50 µM) was added as indicated. After 7 days, the cells that had invaded the collagen were photographed (magnification x200) with a video camera attached to a Nikon microscope at the University of Minnesota Bioimaging Processing Facility as described (25). The diameter of each cell cyst was measured, calculated, and presented as described (25).
Growth of MDCK Cells and Derivatives in Three-dimensional Collagen Gel—Cells (1.2 x 103) were mixed with collagen solutions (1.5 ml at 2.5 mg/ml; Collaborative Research), which were allowed to gel at 37 °C in 6-well plates to give rise to three-dimensional collagen matrices. Fresh medium containing 90% DMEM and 10% fetal bovine serum was added to the wells and changed every 2 days. BB94 (10 mM) was added as needed. After 11 days, the MDCK cells and derivatives on the three-dimensional collagen gel were photographed (magnification x100) with a video camera and analyzed as described above.
Multivariant Analysis: Hierarchical Clustering, K-means Clustering, and Principal Component Analysis—The percentages of MMP-2 activation from transient transfection or in the stable cell line and the average diameters of cell cysts in the collagen assay were analyzed by clustering principles. Because the scales and S.E. values are different among these measurements, the original data were transformed to standardized data (x' = (x - xbar)/S.D.) before clustering analysis. All of the clustering analyses were done with J-Express Version 2.1 (Molmine Bioinformatics Software Solutions). Euclidean was chosen as the distance measure for both hierarchical and K-means clustering. For hierarchical clustering, all four cluster similarity methods available in J-Express were used, including single linkage, unweighted average linkage (UPGMA), weighted average linkage (WPGMA), and complete linkage. For K-means clustering, the number of clusters was designed as 4, and the maximum iteration number was chosen as 200. The initialization method was the most widely used one: random. For principal component analysis, the repeated information among different characters was removed from analysis. The high dimensional data were transformed into lower dimensional data when the most variance information was retained. The lower dimensional data became obvious for visualization.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Despite their structural similarities, MT1-MMP and MT3-MMP differ functionally in at least two areas: 1) activation of pro-MMP-2 into distinct patterns and 2) degradation of type I collagen efficiency. We hypothesize that the functional differences between the MT-MMPs are determined by both the catalytic and pexin-like domains. In an effort to address the structural basis of these functional differences, we constructed chimeras by shuffling domains between these two genes: (a) exchanging the ectoenzymes at the end of the pexin-like domain, (b) exchanging the hinge/pexin-like domains, and (c) exchanging the pro-catalytic domains. As illustrated in Fig. 1B, these constructs were made by a two-step PCR strategy and confirmed by sequencing as described (27). Specifically, the constructs contain the following detailed sequences from MT1-MMP and MT3-MMP: MT1/MT3-1 with MT1-MMP-(1-508) and MT3-MMP-(533-607); MT1/MT3-2 with MT3-MMP-(1-532) and MT1-MMP-(509-582); MT1/MT3-3 with MT1-MMP-(1-284), MT3-MMP-(392-532), and MT1-MMP-(509-582); MT1/MT3-4 with MT3-MMP-(1-291), MT1-MMP-(284-532), and MT3-MMP-(533-607); MT1/MT3-5 with MT1-MMP-(1-284) and MT3-MMP-(392-607); MT1/MT3-6 with MT3-MMP-(1-291) and MT1-MMP-(284-582); MT1/MT3-C1 with MT3-MMP-(1-587) and MT1-MMP-(563-582); and MT1C with MT1-MMP-(1-562).
Segregation of Pro-MMP-2 Activation Pattern with Only the Catalytic Domain—Upon confirmation of all constructs to be error-free and configured as designed, we transiently transfected these constructs into MDCK cells and monitored their expression by Western blotting and their ability to activate pro-MMP-2 by zymography. The constructs expressed protein species as expected as judged by Western blotting using two anti-MT1-MMP antibodies (Fig. 2, C and D) and one anti-MT3-MMP antibody (Fig. 2E). In addition to the specific protein species, there were nonspecific bands cross-reacting with anti-MT1-MMP antibody 879 and anti-MT3-MMP antibody (Fig. 2, D and E, lane 1). Some of the specific species may represent different glycosylation forms as described previously for MT1-MMP and MT3-MMP (e.g. species in Fig. 2E, lanes 4, 5, 9, and 10) (25). The conditioned medium was analyzed for pro-MMP-2 activation. As shown in Fig. 2A, MT1-MMP and MT3-MMP activated pro-MMP-2 with different patterns (lane 2 versus lane 10). Specifically, MT1-MMP converted more MMP-2 into the final active species compared with MT3-MMP such that the ratio between active and intermediate MMP-2 converted by MT1-MMP was significantly greater than that converted by MT3-MMP (Fig. 2A, lane 2 versus lane 10). Given the purported role of the pexin domains in mediating pro-MMP-2 activation (24), we anticipated that the observed difference in pro-MMP-2 activation mediated by MT1-MMP and MT3-MMP would segregate with their respective pexin domains. Surprisingly, when all chimeras were compared for their pro-MMP-2 activation (Fig. 2A, lanes 2-11), a trend emerged to suggest that the patterns of pro-MMP-2 activation segregate only with the catalytic domain of each MT-MMP. The rest of the domains appear to play a very small role in determining the ratio between active and intermediate MMP-2 (Fig. 2A, lanes 2-11). Because transient transfection may generate an abnormally high level of protein for each construct, the observed results in Fig. 2 may not be reflective of the activities under more stable conditions such as those in vivo. Therefore, it is imperative to validate the findings in stable cell lines as shown in Fig. 2.
|
We generated stable cell lines from each construct as described previously (10). Briefly, we picked 24 clones from each construct and screened these clones for pro-MMP-2 activation by zymography and MT1-MMP/MT3-MMP protein expression by Western blotting as described in the legend to Fig. 2. On average, we obtained at least six stable clones expressing varying amounts of recombinant protein on Western blots that mediated corresponding levels of pro-MMP-2 activation (data not shown). To analyze these constructs in the same experimental settings, we selected one representative clone from each construct that expressed similar levels of recombinant proteins upon Western blotting. As shown in Fig. 3 (C-E), we determined the amounts of the recombinant proteins in those selected stable cell lines. The stable lines harboring MT1-MMP, MT1/MT3-1 to MT1/MT3-6, MT1/MT3-C1, and MT3-MMP appeared to express the recombinant proteins at comparable levels (Fig. 3, C-E, lanes 2-10). The only exception is the one expressing MT1C, with about twice as much protein as the rest of the constructs (Fig. 3C, lane 11 versus lanes 2, 3, 5, 7). As shown in Fig. 2 (C-E), there were multiple species observed, presumably representing different forms of glycosylation for both pro and active species for these constructs.
|
The pro-MMP-2 activation profiles of these stable lines recapitulate those from transient transfections (Fig. 3A versus Fig. 2A). As expected, there was a difference in pro-MMP-2 activation between transiently and stably transfected cells. The same constructs expressed in stable lines appeared to be less efficient in converting the intermediate form into the active species compared with those expressed transiently (Fig. 3A versus Fig. 2A). However, the stable lines were more active in converting pro-MMP-2 into the intermediate form (Fig. 3A versus Fig. 2A), perhaps reflecting the percentage of cells expressing the recombinant constructs (100% in stable lines versus <10% in transient transfections). Despite this difference, segregation of the pro-MMP-2 activation patterns with the catalytic domains was apparent among these constructs (Fig. 3A, lanes 2, 3, 5, 7, and 11 versus lanes 4, 6, 8, 9, and 10), i.e. the constructs with the MT1-MMP catalytic domain were capable of generating more active species than the ones containing the MT3-MMP catalytic domain.
Catalytic and Pexin Domains of MT1-MMP Cooperate to Mediate Strong Invasiveness into Type I Collagen Gel—Given the surprising finding that the catalytic domains play a dominant role in mediating pro-MMP-2 activation between MT1-MMP and MT3-MMP, one would predict a similar segregation for their other functions, i.e. MT1-MMP- or MT3-MMP-mediated invasion into type I collagen gel. It has been reported that the MT-MMPs can mediate cell invasiveness into type I collagen, a phenotype mimicking processes such as angiogenesis, tumor invasion, and metastasis (23, 25). Thus, we tested the stable clones described above for their invasiveness into type I collagen gel as described (23, 25). Invasiveness was scored by measuring the diameters of cysts invading type I collagen, and multiple measurements were taken from each assay and averaged and are presented in Fig. 4. MT1-MMP was about five times as efficient as MT3-MMP in promoting MDCK cell invasion into type I collagen matrix (Fig. 4, group 2 versus group 10). The three cell lines from constructs containing the intact catalytic/pexin domains exhibited the strongest invasiveness into type I collagen (Fig. 4, groups 2, 3, and 11), even though MT1/MT3-1 contains the stem/transmembrane/cytoplasmic domains of MT3-MMP and MT1C lacks the 20-residue cytoplasmic domain (Fig. 1B), domains thought to play critical roles in MT1-MMP function. These data suggest that both the catalytic and pexin domains of MT1-MMP cooperate to mediate strong invasiveness into type I collagen. The C-terminal domains of MT3-MMP can functionally substitute for those of MT1-MMP without any loss of activity (Fig. 4, group 2 versus group 3). The catalytic domain of MT1-MMP, when substituted into MT3-MMP, appeared to enhance the invasiveness of the transfected cells (Fig. 4, group 7 versus group 10). On the other hand, the pexin domain of MT1-MMP not only failed to generate any enhancement of invasiveness when substituted into M3-MMP, but also led to a decrease in invasiveness (Fig. 4, group 6 versus group 10). Conversely, the catalytic and pexin domains of MT3-MMP significantly reduced the invasiveness of the cells expressing the resulting exchange constructs as expected (Fig. 4, groups 5 and 8 versus group 2). Surprisingly, the cytoplasmic domain of MT1-MMP appeared to reduce the invasiveness of cells expressing the exchange construct MT1/MT3-C1 (Fig. 4, group 9 versus group 10), suggesting that it encodes a negative signal for MT-MMP function as suggested previously (26).
|
Contributions of Catalytic, Pexin, and Cytoplasmic Domains to MT-MMP-mediated Growth of Transfected MDCK Cells in Type I Collagen Matrix—The collagen invasion assay described in the legend to Fig. 4 involves at least two distinct processes: invasion into and subsequent growth in type I collagen gel. Although the data presented in Fig. 4 demonstrate a close coupling of the catalytic and pexin domains in mediating the invasiveness of the transfected cells into type I collagen gels, it is difficult to discern any functional difference among the exchange constructs separating the catalytic and pexin domains. To this end, we embedded the cells in type I collagen gels to measure their growth in the gels as the collagenolytic index of the transfected constructs as described previously (25). As expected, the cells grew as cysts in type I collagen by clearing the surrounding type I collagen gels in a process inhibited completely by BB94, a synthetic inhibitor (data not shown). The sizes of the cysts were measured, averaged, and analyzed and are presented in Fig. 5. The data are in general agreement with those in Fig. 4. First of all, MT1-MMP was much stronger (about two times) than MT3-MMP in enhancing cell growth in type I collagen (Fig. 5, group 2 versus group 10). Second, the constructs with MT1-MMP catalytic and pexin domains together consistently expressed strong collagenolytic activities (Fig. 5, groups 2, 3, and 11). Third, MT1/MT3-2 and MT1/MT3-C1, both with the cytoplasmic domain of MT1-MMP, appeared to be <50% as active as wild-type MT3-MMP (Fig. 5, groups 4 and 9 versus group 10), suggesting that the MT1-MMP cytoplasmic domain is a negative regulator not only for MT1-MMP (group 2 versus group 11), but also for MT3-MMP. Finally, whereas substituting the MT3-MMP pexin or catalytic domain into MT1-MMP reduced its collagenolytic activity (Fig. 5, groups 5 and 8 versus group 2), the reciprocal substitution of the MT1-MMP pexin or catalytic domain into MT3-MMP failed to enhance its collagenolytic activity (groups 6 and 7 versus group 10). These results suggest that there is no single domain in MT1-MMP that determines its collagenolytic activity or its ability to enhance the growth of MDCK cells in type I collagen.
|
Hierarchical Clustering Analysis of Chimeric Constructs—Given the complex nature of the data presented in Figs. 2, 3, 4, 5, we adopted a new approach to analyze large data sets with multiple variants. Referred to as hierarchical clustering for the analysis of microarray data sets, this model is designed to identify relatedness and to establish orders among different entities. We first quantified the results in Figs. 2 and 3 and then input those data along with those from Fig. 4 and 5 for hierarchical clustering analysis. Two additional statistical analyses, i.e. K-means clustering and principal component analysis, yielded virtually identical results (data not shown). As shown in Fig. 6, the relative relatedness of the constructs becomes apparent. Despite the differences in the four different assays and the likely mechanisms underlying these assays, the hierarchical clustering results in Fig. 6 demonstrate a remarkable degree of agreement in their relative activities and relatedness. In addition to the conclusions drawn above, this new analysis reveals several additional insights. First, MT1-MMP and MT3-MMP are not the opposite ends of a spectrum of activities, as we initially expected. Rather, MT1-MMP and MT3-MMP are positioned in the lower middle part of the spectrum (ranked 6 and 8 out of 10, respectively). Second, two constructs, MT1C and MT1/MT3-1, rank 9 and 10 in the clustering scale to show improved activity over wild-type MT1-MMP (ranked 8), whereas only one construct, MT1/MT3-5, exhibited higher activity than wild-type MT3-MMP. On the other hand, four constructs, MT1/MT3-C1, -2, -4, and -6 (ranked 1-4), had detrimental effects when analogous domains were exchanged (Fig. 6). These observations once again confirm that the MT1-MMP cytoplasmic domain plays a negative role in regulating MT-MMP function and that the MT1-MMP catalytic domain behaves in a dominant fashion in mediating pro-MMP-2 activation and cell invasion. Therefore, this hierarchical analysis should be a useful tool in confirming previous conclusions as well as identifying previously obfuscated functions.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To our knowledge, this report may be the first attempt to dissect the functional domains of MT-MMPs through comprehensive domain exchanges. Domain exchanges coupled with mutagenesis were employed successfully in analyzing the contribution of catalytic and pexin-like domains to collagenolysis for MMP-1, -3, and -8 (32-34). Remarkably, further fine mapping by mutagenesis identified a short segment within the catalytic domain of MMP-1 (RWTNNFREY) that is required for the triple helicase activity of MMP-1 (32). This degree of specificity may be achievable for the MT-MMPs in light of results presented in this report. Here, we chose two closely related MT-MMPs, MT1-MMP and MT3-MMP, to define their structure/function relationship and yielded readily interpretable results as described above. Furthermore, we expect that a similar strategy may be employed to further define the structure/function interplays among the MT-MMPs in interacting with or cleaving novel proteins as described recently (16, 35-38). Alternatively, these chimeras may be useful in identifying the domains targeted by cellular regulation such as type I collagen-mediated induction of pro-MMP-2 activation in tumor cells (17, 39). Inherently, the resolution of the domain exchange strategy is dependent upon the fine structural domains defined through sequence alignments and thus could be quite limited in utility. Point mutagenesis guided by data from domain exchange experiments may allow more refined definition of residues involved in substrate specificity and activity of MT-MMPs. Unfortunately, little has been attempted in this area in a comprehensive scale, although valuable information regarding the roles of the furin cleavage site, a conserved Cys residue in the cytoplasmic domain and the catalytic Asp residue, has been delineated in recent years (13, 40, 41). Recent developments in the x-ray crystallographic structure of MMPs may allow computation-based simulation of three-dimensional structures for the MT-MMPs and provide more rational design for mutagenesis and domain exchange experiments (35, 42, 43). Given the domain cooperation between catalytic and pexin domains suggested by our data, we suggest that future work be devoted to understanding the interactions between the catalytic and pexin domains of individual MMPs, which may ultimately reveal the structural codes for substrate specificity.
|
FOOTNOTES |
|---|
Career Investigator of the American Lung Association. To whom correspondence should be addressed: Dept. of Pharmacology, University of Minnesota, 6-120 Jackson Hall, 321 Church St. S. E., Minneapolis, MN 55455. Tel.: 612-626-1468; Fax: 612-624-3952/612-625-8408; E-mail: peixx003{at}tc.umn.edu.
1 The abbreviations used are: ECM, extracellular matrix; MMP, matrix metalloproteinase; MT-MMP, membrane-type matrix metalloproteinase; MDCK, Madin-Darby canine kidney; DMEM, Dulbecco's modified Eagle's medium.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  X.-Y. Li, I. Ota, I. Yana, F. Sabeh, and S. J. Weiss Molecular Dissection of the Structural Machinery Underlying the Tissue-invasive Activity of Membrane Type-1 Matrix Metalloproteinase Mol. Biol. Cell, August 1, 2008; 19(8): 3221 - 3233. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Schneider, G. K. Sukhova, M. Aikawa, J. Canner, N. Gerdes, S.-M. T. Tang, G.-P. Shi, S. S. Apte, and P. Libby Matrix Metalloproteinase-14 Deficiency in Bone Marrow-Derived Cells Promotes Collagen Accumulation in Mouse Atherosclerotic Plaques Circulation, February 19, 2008; 117(7): 931 - 939. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Morrison and C. M. Overall TIMP Independence of Matrix Metalloproteinase (MMP)-2 Activation by Membrane Type 2 (MT2)-MMP Is Determined by Contributions of Both the MT2-MMP Catalytic and Hemopexin C Domains J. Biol. Chem., September 8, 2006; 281(36): 26528 - 26539. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. G. Remacle, D. V. Rozanov, P. C. Baciu, A. V. Chekanov, V. S. Golubkov, and A. Y. Strongin The transmembrane domain is essential for the microtubular trafficking of membrane type-1 matrix metalloproteinase (MT1-MMP) J. Cell Sci., November 1, 2005; 118(21): 4975 - 4984. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Wang, J. Nie, and D. Pei The Hemopexin Domain of Membrane-type Matrix Metalloproteinase-1 (MT1-MMP) Is Not Required for Its Activation of proMMP2 on Cell Surface but Is Essential for MT1-MMP-mediated Invasion in Three-dimensional Type I Collagen J. Biol. Chem., December 3, 2004; 279(49): 51148 - 51155. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Wang, X. Wang, and D. Pei Mint-3 Regulates the Retrieval of the Internalized Membrane-type Matrix Metalloproteinase, MT5-MMP, to the Plasma Membrane by Binding to Its Carboxyl End Motif EWV J. Biol. Chem., May 7, 2004; 279(19): 20461 - 20470. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Cao, P. Kozarekar, M. Pavlaki, C. Chiarelli, W. F. Bahou, and S. Zucker Distinct Roles for the Catalytic and Hemopexin Domains of Membrane Type 1-Matrix Metalloproteinase in Substrate Degradation and Cell Migration J. Biol. Chem., April 2, 2004; 279(14): 14129 - 14139. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Zhao, M. M. Bernardo, P. Osenkowski, A. Sohail, D. Pei, H. Nagase, M. Kashiwagi, P. D. Soloway, Y. A. DeClerck, and R. Fridman Differential Inhibition of Membrane Type 3 (MT3)-Matrix Metalloproteinase (MMP) and MT1-MMP by Tissue Inhibitor of Metalloproteinase (TIMP)-2 and TIMP-3 Regulates Pro-MMP-2 Activation J. Biol. Chem., March 5, 2004; 279(10): 8592 - 8601. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Wang, D. Ma, J. Keski-Oja, and D. Pei Co-recycling of MT1-MMP and MT3-MMP through the Trans-Golgi Network: IDENTIFICATION OF DKV582 AS A RECYCLING SIGNAL J. Biol. Chem., March 5, 2004; 279(10): 9331 - 9336. [Abstract] [Full Text] [PDF]
|
|
| ||||||
MT1502; C) or 879 (
(xik - xjk)2), where i is the ith row (one gene type), j is the jth row (the other gene type), and k starts from the first column to the last column (using all four parameters in this study: gel A activation in transient transfection; gel A activation by stable transfection; invasion (diameters); and collagen growth assay (diameters)). The green spots show that the values are below the average of all of the constructs, whereas the red spots show that the values are above the average of all of constructs. The brighter a green square, the smaller the number of the character; the brighter a red square, the larger the number of the character. The minimum number (-45.906) and the maximum number (45.906) are the standardized data. Four cluster similarity methods were applied: single linkage, unweighted average linkage (UPGMA), weighted average linkage (WPGMA), and complete linkage. The four panels gave same clustering results as shown here. When we separated the constructs into five groups based on these dynamic cluster diagrams, the five clusters (shaded) are as follows: (a) cluster 1 with the lowest values (ranked 1-4), including vector, MT1/MT3-C1, -2, -4, and -6; (b) cluster 2 (ranked 5 and 6), including MT1/MT3-3 and MT3-MMP; (c) cluster 3 (ranked 7), including MT1/MT3-5; (d) cluster 4 (ranked 8 and 9), including MT1-MMP and MT1/MT3-1; and (e) cluster 5 with the highest values (ranked 10), including MT1