Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Tissue inhibitor of metalloproteinases-4 and progelatinase A activation : the role of the C-terminal… Kai, Heidi S.-T. 2002

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2002-0123.pdf [ 10.44MB ]
Metadata
JSON: 831-1.0090224.json
JSON-LD: 831-1.0090224-ld.json
RDF/XML (Pretty): 831-1.0090224-rdf.xml
RDF/JSON: 831-1.0090224-rdf.json
Turtle: 831-1.0090224-turtle.txt
N-Triples: 831-1.0090224-rdf-ntriples.txt
Original Record: 831-1.0090224-source.json
Full Text
831-1.0090224-fulltext.txt
Citation
831-1.0090224.ris

Full Text

Tissue Inhibitor of Metalloproteinases-4 and Progelatinase A Activation: The Role of the C-terminal Tail in Mediating Activation By Heidi S. -T. Kai B.Sc, The University of British Columbia, 1998 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Oral Biological and Medical Sciences We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 2002 © Heidi S. -T. Kai, 2002 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT The activation of gelatinase A (matrix metalloproteinase-2) occurs on the cell surface via a unique mechanism that requires a ternary complex of membrane-type 1 matrix metalloproteinase (MT1-MMP) and tissue inhibitor of metalloproteinases (TIMP)-2, which acts as a cell surface receptor for progelatinase A. TIMP-2 binds to both the MT1-MMP and the hemopexin C-domain of progelatinase A thus linking progelatinase A to the cell surface where a second MT1-MMP can begin the activation process by cleaving the propeptide of progelatinase A. Recently, our lab has shown that TIMP-4, which shares a high degree of homology with TIMP-2, can bind and inhibit MT1-MMP and independently bind the hemopexin C-domain of progelatinase A in a manner similar to TIMP-2 but is unable to support the activation of progelatinase A. TIMPs consist of two domains: the N-terminal inhibitory domain and the C-terminal domain. Additionally, TIMPs-2, -3, and -4 contain nine residues at the C-terminus known as the C-terminal tail. TIMPs-2 and -4 are hypothesized to bind to progelatinase A through their non-inhibitory C-domains. To investigate the differences in the interactions of TIMP-2 and TIMP-4 with progelatinase A, the C-terminal domains of TIMPs-2 and -4 were each cloned as a C-terminal fusion partner to horse heart myoglobin. The anionic C-terminal tail of TIMP-2 is hypothesized to bind a cationic site on the hemopexin C-domain of progelatinase A during activation. Therefore, to investigate the aspects of TIMP-4 that renders it deficient in the activation of progelatinase A, a set of mutations were introduced into the C-terminal tail of TIMP-2 and -4 and the activities of 8 fusion proteins were investigated. Myoglobin-C-TIMP-2 (MbcT2) bound progelatinase A with reduced affinity relative to full length TIMP-2, whereas deletion of the tail decreased the affinity further. In contrast, deletion of the C-terminal tail of C-TIMP-4 did not alter the binding of Mb-C-TIMP-4 (MbcT4) to progelatinase A and both proteins bound with only slightly less affinity than full length TIMP-4. The similarity of the binding affinities of TIMP-4 and MbcT4 to progelatinase A indicates that the N-terminal domain and C-terminal tail of TIMP-4 is unlikely to contribute to its interaction with progelatinase A. Mutations that removed the negative charges in the C-terminal tail of MbcT2 (E192V/D193Q) also resulted in a reduction in binding, as did a substitution of the C-terminal tail of MbcT2 with that of TIMP-4. Conversely, addition of negative charges to the C-terminal tail of MbcT4 (V193E/Q194D) resulted in an increased affinity for progelatinase A, as did a substitution of its tail with that of TIMP-2. The overall analysis of these fusion protein constructs show that the C-terminal tail of TIMP-2 is integral to its ability to support the activation of progelatinase A, specifically the negatively charged residues Glu192 and Asp193. The data also indicate that the reason TIMP-4 cannot support the activation of progelatinase A is because of the lack of these negatively charged residues in the TIMP-4 C-terminal tail. TABLE OF CONTENTS Abstract ii List of Figures vi List of Tables vii List of Abreviations viii Acknowledgments x Chapter 1 - Introduction 1 1.1 General overview 1 1.2 Matrix Metalloproteinases 4 1.2.1 Structure of MMPs 4 1.2.2 Regulation of MMPs 6 1.2.3 GelatinaseA 11 1.3 Tissue Inhibitor of Metalloproteinases 15 1.3.1 Structure of TIMPs 15 1.3.2 Mechanism of inhibition 18 1.3.3 TIMP-1 19 1.3.4 TIMP-2 21 1.3.5 TIMP-3 26 1.3.6 TIMP-4 29 1.4 Aims of research, hypothesis, and experimental approach 34 Chapter 2 - Materials and Methods 40 2.1 Cloning of pGYMC-MbcTIMP-4 40 2.2 Creation of the MbcTIMP mutant constructs 41 2.2.1 Mutagenesis of pGYMC-MbcTIMP-2 42 2.2.2 Mutagenesis of pGYMC-MbcTIMP-4 43 2.3 Screening of E. coli cell lines for protein expression 44 2.4 Stability testing of MbcTIMP-4 45 2.5 Expression, refolding and purification of recombinant proteins 45 2.6 Analysis of the MbcTIMP recombinant proteins 48 2.6.1 SDS-PAGE analysis 48 2.6.2 Western blotting analysis 48 2.6.3 Mass spectrometry analysis 49 2.6.4 Soret peak analysis 50 2.7 Microwell solid phase binding assay 50 2.8 Affinity columns 51 2.8.1 Gelatin Sepharose mini columns 52 2.8.2 Affi-gel Coupled mini columns 52 2.9 Velocity sedimentation 53 2.10 Glutaraldehyde cross-linking 54 2.10.1 Chemical cross-linking 54 2.10.2 Analysis by zymography 54 2.11 Enzyme capture assay 55 iv Chapter 3 - Results 56 3.1 Recombinant protein expression and characterization 56 3.2 Microwell Solid Phase Binding Assay 65 3.3 Affinity Columns 67 3.4 Velocity Sedimentation 67 3.5 Chemical Cross-linking 70 3.6 Enzyme Capture Assay 73 3.6.1 Visualization by Fluorimetry Assay 73 3.6.2 Visualization by Zymography 75 Chapter 4 - Discussion 77 4.1 Perspective 77 4.2 Recombinant protein expression, purification and characterization 78 4.3 Protein interaction studies 83 4.3.1 Microwell solid phase binding assay 83 4.3.2 Affinity columns 84 4.3.3 Velocity sedimentation 85 4.3.4 Chemical cross-linking 87 4.3.5 Enzyme Capture Assay 89 Chapter 5 - Conclusion 95 Chapter 6 - References 98 v LIST OF FIGURES Figure 1: Schematic representation of the different MMPs and their domain structure . 7 Figure 2: The crystal structure of progelatinase A 13 Figure 3: Sequence alignment of human TIMPs 16 Figure 4: Crystal Structure of TIMP-2 17 Figure 5: Crystal structure of the gelatinase A hemopexin C-domain in stereo...24 Figure 6: Diagrammatic representation of the ternary complex mechanism of the cell surface activation of progelatinase A 25 Figure 7: Schematic representation of the MbcTIMP recombinant proteins 37 Figure 8: Flow diagram of the purification scheme for MbcTIMPs 46 Figure 9: Mini expression of MbcT4 in BL21 57 Figure 10: Purification of the mutant MbcT4+q" 58 Figure 11: Spectrophotometric properties of the Mb-fusion proteins 60 Figure 12: SDS-PAGE and Western blotting analysis of MbcTIMP recombinant proteins. Panel A 63 Figure 12: SDS-PAGE and Western blotting analysis of MbcTIMP recombinant proteins. Panels B and C 64 Figure 13: Comparison of the binding of MbcT4 and MbcT4T to gelatinase A hemopexin C domain 66 Figure 14: Determination of protein interactions using velocity sedimentation Panels A-C... 68 Figure 14: Determination of protein interactions using velocity sedimentation Panels D-F 69 Figure 15: Determination of protein interactions using glutaraldehyde cross-linking. Panels A-C : 71 Figure 15: Determination of protein interactions using glutaraldehyde cross-linking. Panels D-G 72 Figure 16: Visualization of enzyme capture assay by fluorimetry 74 Figure 17: Visualization of enzyme capture assay by zymography 76 Figure 18: Hypothetical binding site for the C-terminal tail of TIMP-2 on the hemopexin C domain of gelatinase A 94 vi LIST OF TABLES Table 1: Members of the matrix metalloproteinase family 5 Table 2: Protein yields from purification process 61 Table 3: Molecular mass determination of horse heart myoglobin and myoglobin C-TIMP fusion proteins by electrospray ionization mass spectrometry 62 vii LIST OF ABBREVIATIONS ADAM ADAM-TS AP AP-1 APMA BSA CA-MMP cDNA Con A C-terminal C-TIMP DMSO DTT ECL ECM ELISA EPA FAB hhMb HRP IL-1 Kd MbcTIMP MCP MMP A disintegrin and a metalloproteinase A disintegrin and metalloprotease with thrombospondin type I modules Alkaline phosphatase Activating protein 1 4-aminophenylmercuric acetate Bovine serum albumin Cysteine array matrix metalloproteinase Complementary deoxyribonucleic acid Concanavalin A Carboxy terminal The carboxy terminal domain of a tissue inhibitor of metalloproteinases Dimethyl sulfoxide Dithiothreitol Enhanced chemiluminescence Extracellular matrix Enzyme-linked immunosorbance assay Erythroid potentiating activity Fluorimetry assay buffer Horse heart myoglobin Horseradish peroxidase Interleukin 1 Dissociation constant Fusion protein consisting of a myoglobin at the amino terminus of the molecule and the carboxy terminal domain of a tissue inhibitor of metalloproteinases at the carboxy terminus Monocyte chemoattractant protein Matrix metalloproteinase MMPI Matrix metalloproteinase inhibitors (synthetic) mRNA Messenger ribonucleic acid MT-MMP Membrane type matrix metalloproteinase N-terminal Amino terminal PAGE Polyacrylamide gel electrophoresis PBS Phosphate buffered saline PCPE Procollagen C-terminal proteinase enhancer protein PCR Polymerase chain reaction PDB Id Protein data bank identification number PEA3 Polyoma enhancer activator SDF Stromal cell-derived factor SDS Sodium dodecyl sulphate SV-40 Simian virus 40 TBS Tris buffered saline TGF-P Transforming growth factor P TIMP Tissue inhibitor of metalloproteinase TNF-cc Tumour necrosis factor a ACKNOWLEDGEMENTS There are a number of people whom I wish to acknowledge for guiding me through the often times bumpy road that my masters experience became. First I'd like to thank my supervisor Dr. Chris Overall for giving me this project and allowing me to work in his laboratory. In my time at the Overall lab, I've increased my knowledge of science and gained many friends and fond memories. I'd also like to thank Dr. Clive Roberts for all his advice and for teaching me the technique of immunohistochemistry. Many thanks to Dr. Georgina Butler for the generous donation of copious amounts of progelatinase A and for her advice and help in planning experiments. In addition, I'd like to acknowledge Dr. Charlotte Morrison for the use of her TIMP-2 and Dr. Heather Bigg and Andrea Connor for the use of their TIMP-4. I'd also like to thank Andrea for the wonderful job she has done of keeping the lab in running order and ordering in all the reagents and equipment the lab required. For help regarding computer issues and for the construction of the pGYMC-MbMCS vector, I'd like to acknowledge Gayle Pelman. For the cloning of MbcT2 and guiding me through my first year in the lab, I'd like to thank Angela King. Two key experiments in my thesis were velocity sedimentation and the detection of enzyme capture by fluorimetry. For the ideas behind these experiments and for teaching me how to do these experiments, I must acknowledge Dr. Chris Loewen and Eric Tarn. Without the fluorimetry experiment, my thesis would have been a null results thesis, which would have been difficult to come to terms with after so many years and so many proteins purified. I would like to thank my friends and labmates for listening to me complain about when things were not working in the lab (and unfortunately, there were a lot of these times). I'd also like to thank Dr. B. K. C. Chow for giving me my first chance (and paper) in the field of biochemistry. Lastly, special thanks to Todd Moore for putting up with me and giving me ideas, advice, moral support, and pushing me to succeed and also to my friends and family, especially my parents and sister, for their support and for not asking me too many questions about things I don't know how to explain. x CHAPTER 1 - INTRODUCTION 1.1 General overview Many of the normal physiological processes undertaken by cells in multi-cellular organisms involve controlled remodelling of the extracellular matrix (ECM). Connective tissue remodelling is essential for normal development and growth. Processes such as wound healing, angiogenesis, and ovulation cannot occur without controlled remodelling of the ECM (Parks, 1999; Edwards et al., 1996; Nguyen et al., 2001; Smith et al., 1999). In addition to a structural role, the ECM also provides cells with biological information and signals about their environment. The ECM can sequester certain signalling molecules within its lattice, which are released upon breakdown. Exposure of cryptic sites through cleavage of ECM molecules during remodelling can also provide signals that are important for processes such as cellular growth, differentiation, migration, and apoptosis and can be a factor in the progression of disease (Boudreau and Jones, 1999; Aumailley and Gayraud, 1998; Streuli, 1999). The ECM is composed of a number of proteins and proteoglycans of which collagen is the most abundant. Collagen consists of three individual a-chains that have a right-handed triple helical conformation and exists in numerous forms that can be subdivided into fibril-forming collagens (types I, II, III, V, XI), and non-fibril-forming collagens such as the network-forming collagens (types IV and Vll) (Aumailley and Gayraud, 1998). 1 ECM homeostasis is achieved through the collective efforts of a number of proteinase and inhibitors families including a family of enzymes known as matrix metalloproteinases (MMPs) and their inhibitors, the tissue inhibitors of metalloproteinases (TIMPs). A delicate balance between MMPs and TIMPs aids in controlling the breakdown of the components of connective tissue. Errant expression of MMPs and/or TIMPs resulting in an imbalance can lead to diseases such as periodontitis, arthritis, and atherosclerosis (Birkedal-Hansen et al., 1993; Birkedal-Hansen, 1993; Yamanishi and Firestein, 2001; Newby era/., 1994; McDonnell et al., 1999). MMPs and TIMPs also play a role in neoplasia, especially in tumour angiogenesis and metastasis (Kleiner and Stetler-Stevenson, 1999). Indeed, many MMPs are prognostic indicators in various cancers (Murray era/., 1996; Yamamoto era/., 2001; Gohji et al., 1996; Curran and Murray, 2000); (Egeblad and Werb, 2002). Synthetic MMP inhibitors (MMPIs) are currently undergoing clinical trials for the treatment of several cancers because of the importance of MMPs in the progression of cancer. However, it is difficult to judge their effectiveness in early phase clinical trials because MMPIs are anti-proliferative and anti-angiogenic but not cytotoxic (Hoekstra et al., 2001; Egeblad and Werb, 2002). MMPIs are also under investigation for the treatment of arthritis and heart disease (Elliott and Cawston, 2001; Creemers era/., 2001). Because of the vast and varied roles of MMPs and TIMPs in normal physiological processes and in disease, it is important to study how MMPs and TIMPs interact 2 and bind to each other. This knowledge may be useful in controlling tissue remodelling and may lead to the discovery of strategies to combat diseases that involve the uncontrolled degeneration of the ECM such as rheumatoid arthritis, cancer and a number of cardiovascular diseases. MMP-2 or gelatinase A is an enzyme that is implicated in many diseases including cancer because of its ability to degrade type IV collagen, the basement membrane collagen, although recently, there has been evidence that its actions on the basement membrane causes the exposure of cryptic sites and release of growth factors and cytokines, which is in turn responsible for its effects on tumour progression (Giannelli and Antonaci, 2002). Indicative of the role of gelatinase A in neoplasia is the fact that gelatinase A deficient mice have reduced tumour angiogenesis and progression (Itoh et al., 1998). Because of the importance of gelatinase A in disease processes, it is important understand how its inactive precursor, progelatinase A, is activated on the surface of cells. It is hoped that knowledge of the activation mechanisms of this enzyme will lead to a better understanding of its role in the progression of cancer so that therapeutic agents can be designed to combat this disease. The role of MMPs and TIMPs in many processes are currently the subject of intense research but of particular of interest is the fact that TIMP-2, besides its ability to regulate MMP activity through inhibition, has also been shown to participate in the activation of progelatinase A, thus regulating the activation of MMPs (Strongin et al., 1995). It was recently shown that another TIMP, TIMP-4 is not able to promote the activation of progelatinase A although it can interact with progelatinase A in a manner similar to TIMP-2 (Bigg et al., 2001; 3 Toth et al., 2000; Hernandez-Barrantes et al., 2001). This study investigates the dichotomous relationship between progelatinase A and TIMPs-2 and -4 and why TIMP-4 cannot support the activation of progelatinase A. 1.2 Matrix Metalloproteinases Matrix metalloproteinases, also known as matrixins, are a family of endopeptidases that are collectively capable of degrading most components of the ECM. The MMPs utilize a zinc ion for catalysis and have a requirement for structural zinc and calcium ions. MMPs are members of a supeitamily of metalloproteinases, the metzincins, that are characterized by their reliance on zinc for catalysis and a conserved methionine which constitutes the "Methionine turn" which forms the end of the lower edge of the active site (Bode et al., 1999). MMPs have a conserved zinc-binding motif with the sequence HExxHxxGxxH in which the underlined histidine residues coordinate the catalytic zinc ion (Overall, 2001). 1.2.1 Structure of MMPs There are 22 human MMPs known to date. The MMPs are structurally related and have conserved regions of homology and can be subdivided into different categories including the collagenases, stomelysins, gelatinases, matrilysins, and membrane-type MMPs (MT-MMPs) (Table 1). With the exception of the matrilysins and MMP-23 (cysteine array (CA)-MMP), MMPs are organized into 4 Table 1: Members of the Matrix Metalloproteinase Family MMP Designation Common Name Sub-Family 1 Collagenase-1 Collagenase Interstitial collagenase Fibroblast collagenase Tissue collagenase 2 Gelatinase A Gelatinase 72-kDa gelatinase 72-kDa type IV collagenase 3 Sromelysin-1 Stromelysin Proteoglycanase 7 Matrilysin-1 Matrilysin PUMP-1 8 Neutrophil collagenase Collagenase Collagenase-2 9 Gelatinase B Gelatinase 92-kDa gelatinase 92kDa type IV collagenase 10 Stromelysin-2 Stromelysin 11 Stromelysin-3 Stromelysin 12 Metalloelastase Macrophage elastase 13 Collagenase-3 Collagenase 14 Membrane-Type (MT)1-MMP Membrane-Type 15 MT2-MMP Membrane-Type 16 MT3-MMP Membrane-Type 17 MT4-MMP Membrane-Type 18 Collagenase-4 Collagenase xCol4 (Xenopus, no known human homologue) 19 Formerly MMP-18 RASI-1 20 Enamelysin 21 XMMP (Xenopus) Human homologue unpublished 22 CMMP (Chicken) 23 Cysteine Array (CA)-MMP 23a Formerly MMP-21 MIFR-1 23b Formerly MMP-22 MIFR-2 Femalysin 24 MT5-MMP Membrane-Type 25 MT6-MMP Membrane-Type Leukolysin 26 Matrilysin-2 • - Matrilysin Endometase 27 Human homologue of MMP-22 28 Epilysin 5 three basic domains: the propeptide domain, which is responsible for maintaining enzyme latency; the catalytic domain; and the carboxy (C)-terminal hemopexin domain, which contains exosites for substrate specificity (Figure 1). MMPs also contain an N-terminal signal peptide. Some MMPs also have additional domains such as the C-terminal transmembrane domain and cytoplasmic tail of some MT-MMPs (Sato et al., 1994). Other MT-MMPs are known to be glycosylphosphatidyl inositol anchored to the cell membrane (Itoh era/., 1999; Kojima et al., 2000). Many of the MMPs have domains that share homology to matrix proteins. These include the fibronectin-like domains of gelatinase A and B and the a2(V) collagen chain of gelatinase B (contained within the O-linked oligosaccharide domain) (Figure 1) (Overall, 2001; Opdenakker etal., 2001). 1.2.2 Regulation of MMPs Highly controlled regulation of MMP activity is essential due to the potentially devastating effects of MMPs. There are three main mechanisms of regulation for MMPs: transcriptional, post-transcriptional, and post-translational. Together, these mechanisms tightly and differentially control the levels and activity of MMPs in tissues so that they can participate in normal physiological processes without causing harm. A breakdown in the regulatory mechanisms of MMPs can lead to serious disease and is thought to contribute greatly to tumour angiogenesis and metastasis (Egeblad and Werb, 2002). 6 Matrilysin 1 and 2 Collagenase-1, -2, and -3, Stromelysin-1, and -2, Metalloelastase, Enamelysin, and MMP-19 Stromelysin-3 Membrane-Type-1, -2^-3, and -5 MMP Membrane-Type-4 and -6 MMP Gelatinase A "mm Cysteine-Array-M M P Signal Peptide '/Z^Z^X Pro-domain N-terminal signal anchor/pro-domain Vitronectin-like insert Furin Recognition Sequence Catalytic Domain fX CO Fibronectin Type ll-like modules Catalytic Zinc O-linked Oligosaccharide Domain 3 Cysteine Array Domain Hinge Region Hemopexin C domain Cysteine/Proline-Rich IL-1 Receptor-like Domain Transmembrane Domain ••I Cytoplasmic Tail Stalk Region • • I GPI Anchor F igu re 1: S c h e m a t i c r e p r e s e n t a t i o n of t h e d i f fe ren t M M P s a n d the i r d o m a i n s t r u c t u r e . 7 Firstly, the expression of MMPs is regulated transcriptionally. Members of the MMP family are differentially and tissue specifically regulated. While some MMPs genes are inducible, others, such as MMP-2, are not readily inducible and are more constitutively expressed at a basal level. The readily inducible MMPs (MMP-1, -3, -7, -9, -10, -12, and -13) are under the influence of two main regulatory elements found within their promoter regions, the activating protein-1 (AP-1) site and the polyoma enhancer activator (PEA3) site (Curran and Murray, 2000; Sternlicht and Werb, 2001). The AP-1 element located at approximately -70 base pairs upstream of the transcriptional start site, is important for both basal transcription of the inducible MMP genes and for repression of transcription. Basal transcription and transactivation of MMPs require that AP-1 interact with other cis-acting elements such as PEA3 (Benbow and Brinckerhoff, 1997). lnterleukin-1 (IL-1), tumour necrosis factor a (TNF-a), interferon-p\ 12-0-tetradecanoylphorbol 13-acetate, glucocorticoids, retinoids, thyroid hormones, progesterone, androgens, and transforming growth factor |3 (TGF-|3) all act in some way on the AP-1 site to modulate the expression of inducible MMPs (Curran and Murray, 2000). MMPs can also be post-transcriptionally regulated via increased mRNA stability. This is the case for gelatinase A. Upon addition of TGF-(3 to human gingival fibroblasts, the abundance and stability of gelatinase A mRNA was markedly increased (Overall et al., 1991). Another post-transcriptional regulation mechanism is zymogen activation. MMPs are synthesized as inactive zymogens 8 that contain an -80 amino acid amino (N)-terminal prodomain which is responsible for maintaining enzyme latency (Figure 1). The prodomains of all MMPs except CA-MMP (Pei, 1999) contain a cysteine switch motif PRCG(N/V)D where the cysteine residue (underlined) coordinates the active site zinc, which excludes a water molecule necessary for catalysis, thus preserving enzyme latency (Overall, 2001). Most MMPs are secreted by cells in a latent, proenzyme/zymogen form, which is later converted to the active form by the cleavage of the prodomain. The exceptions are MMP-11 (stromelysin-3), the MT-MMPs and CA-MMP. Stromelysin-3 and the MT-MMPs are activated intracellular^ by furin cleavage at the furin recognition sequence before secretion or transport to the cell surface, respectively (Pei and Weiss, 1995; Sato et al., 1996). CA-MMP is N-terminally tethered to the cell surface as a zymogen by an N-terminal signal anchor and is released as a fully active enzyme following proteolytic cleavage at a proprotein convertase motif that is located between the pro and catalytic domains (see Figure 1) (Pei et al., 2000). The activation of MMPs in vitro can also be accomplished by the addition of organomercurials such as aminophenylmercuric acetate (APMA) (Stricklin et al., 1983). Most MMPs can be activated by other proteases such as serine proteinases (i.e. plasmin or trypsin) (Okada and Nakanishi, 1989; Nagase era/., 1990; Murphy et al., 1992a; Lee et al., 1996). However, gelatinase A is different in that it lacks serine protease cleavage sites in the prodomain (Okada et al., 1990). Thrombin has been shown to induce activation of progelatinase A in microvascular endothelial cells (Nguyen et al., 1999). This pathway is hypothesized to be 9 independent of the thrombin receptor and is thought to involve thrombin forming a complex with thrombomodulin. The complex then converts protein C to activated protein C, which then activates progelatinase A (Nguyen et al., 2001). MMPs can also activate other MMPs, as is the case with MMP-3 (stromelysin-1), which can activate MMP-1, -8, and -13 (collagenase-1, -2, and -3) (Murphy etal., 1987; Knauper et al., 1993; Knauper et al., 1996a). Gelatinase A and collagenase-3 can activate MMP-9 (gelatinase B) (Fridman et al., 1995; Knauper et al., 1997). MT1-MMP is also an important activator of MMPs and is capable of activating procollagenase-3 (Knauper et al., 1996b). However, the most intriguing mechanism of MMP activation involving MT1-MMP is the activation of progelatinase A (Sato et al., 1994). Activation of progelatinase is lacking in MT1-MMP deficient mice indicating that MT1-MMP is the primary activator of progelatinase A in vivo (Zhou et al., 2000). This mechanism of activation occurs at the cell surface and requires a ternary complex of MT1-MMP and one of the inhibitors of MMPs, TIMP-2, to tether progelatinase A to the cell surface in order for another TIMP-2 free MT1-MMP molecule to cleave progelatinase A. The activation of progelatinase A by MT1-MMP and these progelatinase A/TIMP-2 interactions will be discussed in detail in section 1.3.4 of this thesis. The final mechanism of MMP regulation occurs posWranslationally via MMP inhibitors. The major physiological inhibitors of MMPs are TIMPs. Currently, there are four known members of the TIMP family, numbered 1-4. TIMPs noncovalently inhibit MMPs stoichiometrically at a 1:1 molar ratio. The 10 expression of TIMPs is also tightly regulated to maintain the local balance between MMPs and TIMPs to control ECM breakdown and accumulation. Their function and importance will also be discussed in greater detail in section 1.3 of this thesis. Besides the TIMPs, other endogenous inhibitors of MMPs include the plasma protein oc2-macroglobulin (Sottrup-Jensen and Birkedal-Hansen, 1989), the noncollagenous NC1 domain of type IV collagen (Netzer et al., 1998), and a C-terminal fragment of procollagen C-terminal proteinase enhancer protein (PCPE) (Mott et al., 2000). The NC1 domain and C-terminal fragment of PCPE are protein subdomains that have structural similarity to the TIMPs. oc2-macroglobulin may be the major inhibitor of MMP in tissue fluids because it is so abundant and a2-macroglobulin/MMP complexes are irreversibly cleared by scavenger receptor-mediated endocytosis (Sternlicht and Werb, 2001; Egeblad and Werb, 2002). The inhibitory interactions between the NC1 domain or the C-terminal fragment of PCPE and MMPs appear to be of lower affinity than that of TIMPs, although they can inhibit angiogenesis and tumour growth much like TIMPs (Petitclerc etal., 2000). 1.2.3 Gelatinase A Gelatinase A (72-kDa gelatinase/type IV collagenase) is a 72 kDa proteinase that is constitutively expressed and secreted by most cell types. It contains an 80 amino acid propeptide, a 338 amino acid catalytic domain that has within it three 58-residue tandem repeats similar to the collagen-binding domain of fibronectin (the fibronectin type ll-like modules), a 28 amino acid hinge region, and a 185 11 amino acid hemopexin C. domain (Figures 1 and 2) (Collier et al., 1988; Morgunova et al., 1999). Its matrix-derived substrates include types I, III, IV, V, Vll, X, and XI collagens, gelatin, fibronectin, laminin, agrrecan, elastin, and decorin (Sternlicht and Werb, 2001; Egeblad and Werb, 2002; Imai et al., 1997). In addition to its role in matrix degradation, gelatinase A can also cleave non-structural molecules such as, pro-TGF-(3 (Yu and Stamenkovic, 2000), pro-IL-1 f3 (Schonbeck et al., 1998), and pro-TNF-a (Gearing et al., 1994), resulting in a variety of biological activities. Recently, it was shown by McQuibban et al. that gelatinase A cleaves chemokines including stromal cell-derived factor (SDF)-1 (McQuibban et al., 2001) and monocyte chemoattractant protein (MCP)-3 (McQuibban et al., 2000). Cleavage of SDF-1 and MCP-3 destroys their chemotatic ability and the cleaved MCP-3 has been shown to act as a receptor antagonist, preventing functional MCP-3 from binding to its receptors (McQuibban et al., 2000; McQuibban et al., 2001). This suggests that gelatinase A plays an important role in the regulation of inflammatory responses. C-terminally truncated gelatinase A has been shown to exhibit full proteolytic activity in the degradation of gelatin and type IV collagen (Fridman et al., 1992), thus the hemopexin C domain of gelatinase A does not appear to be required for its gelatinolytic activity. However, it appears to be necessary for its collagenolytic activity (Patterson et al., 2001), although it does not bind to native collagen (Wallon and Overall, 1997). The gelatinase A hemopexin C domain is also essential for interaction with its inhibitors, namely TIMPs (Murphy et al., 1992b; 12 Figure 2: The crystal structure of progelatinase A including the prodomain, the catalytic domain including the three fibronectin like modules, the hinge region and the hemopexin C domain as indicated. Note the cysteine of the cysteine switch shown in space-fill, coordinating the active site Zn2+. Part of the hinge region remains unresolved due to the molecular motion of that region. PBD Id: 1CK7 (Morgunova et al., 1999). 13 Ward et al., 1994). Thus the hemopexin C domain is important for the activation of progelatinase A by a ternary complex involving TIMP-2 and MT1-MMP (Strongin et al., 1995). The ability of invasive tumour cells to degrade connective tissue stroma and basement membrane is pivotal to metastasis (Liotta and Stetler-Stevenson, 1991; Stetler-Stevenson et al., 1993) and the ability of endothelial cells to degrade the basement membrane and migrate through the ECM is essential for angiogenesis (Nguyen et al., 2001). Since type IV collagen is the major basement membrane structural protein, its degradation is a key event in tumour spread and growth. Thus, understanding and studying the structure and function of the proteases involved in metastasis and angiogenesis and their inhibitors, and their regulation is essential for understanding and controlling cancer. Due to the ability of gelatinase A to degrade type IV collagen, it has been implicated in the breakdown of the basement membrane. Cell lines transfected with gelatinase A have been shown to have increased invasiveness (Kawamata etal., 1995) and metastatic tumours often have high levels of gelatinase A expression either in the tumour itself or more commonly in the surrounding stroma (Stetler-Stevenson and Yu, 2001; Egeblad and Werb, 2002). Interestingly, gelatinase A knockout mice show reduced tumour growth and metastasis (Itoh et al., 1998). A tumour cannot grow larger than 1mm3 before it requires its own blood supply for nutrients and waste removal, thus gelatinase A has also been implicated in tumour angiogenesis (Nguyen et al., 2001). The role of gelatinase A in 14 angiogenesis was established by Kato et al., who showed that gelatinase A deficient mice show reduced corneal angiogenesis (Kato et al., 2001). The hemopexin C domain of gelatinase A can block angiogenesis by displacing active gelatinase A from the integrin a v(3 3, providing further evidence of the role of gelatinase A in angiogenesis (Brooks etal., 1998; Bello etal., 2001). 1.3 Tissue Inhibitor of Metalloproteinases The tissue inhibitors of metalloproteinases are a family of naturally occurring protein inhibitors that are responsible for modulating the activity of MMPs. Currently, there are four known mammalian TIMPs. They inhibit MMPs by forming a non-covalent, stoichiometric complex. The four TIMPs are highly homologous proteins both between the different members of the family and interspecies. Human TIMPs share approximately 40-50% identity and 50-70% similarity (Figure 3) (Douglas et al., 1997). TIMPs also have conserved gene structures and are located within introns of synapsin genes, indicating that the different TIMPs are likely due to gene duplications (Brew etal., 2000). 1.3.1 Structure of TIMPs TIMPs are organized into two distinct, independently folding domains: the N-terminal domain and the C-terminal domain (Figure 4). TIMPs also contain a short leader sequence consisting of 23-29 amino acids, which is cleaved off to produce the mature protein. TIMPs share a high degree of similarity in their tertiary structure. Each TIMP contains 6 loops held together by 12 conserved 15 M L . . L A C . C . P . H P Q . Consensus #1 M X P X X R X A X X W L G L L X L L A L L R P X - - G - E A C S C S P X H P Q Q M a j o r i t y 1 1 ; 1 r-10 20 30 40 . A F E P L . S _l_ G I . L . W I _ L _ P s . T V P . . T . G A A A . T L R L A . L A T . . A - - - - D . V . . T I V G s w S L G D W . A . . . T s . . D . P G S P . P . P S V L . R , , P G L . - . . A A . hTIMP-2 hTIMP-3 hTIMP-4 . . C . . . . V I R A K V Y . I K . . Consensus #1 A F C N S D L V I R A K X V G E K E V X X G X D P Y G X - - K - L R Y E I K Q I M a j o r i t y 1 1 1 1-50 60 70 80 i i i t— 34 F . . T P . . N Q T T - L . Q - - - - - - M T hTIMP-1 37 . . . . A . V A . S . . . . D S . N . I . . N P I . R I Q hTIMP-2 34 I V . . K . L . K E . - - . F . T - - - - . V . T . . . M hTIMP-3 40 H I . H . A I S S . . V . P A S A . . - A D T E . M hTIMP-4 K M . . G T C G Consensus #1 K M F K G F E K X - - - X D V Q F I Y T P A S S S V C G V K L E V N X K K E - Y M a j o r i t y 1 1 1 r-90 100 110 120 i i i i _ 67 . . Y . . . Q A L G D A A . I R . V . . . . M E . . . . Y F H R S H N R S . E F hTIMP-1 77 P . . - - - - - . I E . . . . A P . . A . . . . S . D . G G . . . - . hTIMP-2 68 . . Y R . . T . M - - - P H . . Y . H . E . . E . L . . L - . Y Q - . hTIMP-3 79 V - - - K . . . Y . . . . F D . . L A . S Q . Q - . hTIMP-4 L . . G . . . . D G C W . . L . . . Q Y . . Consensus #1 L L A G K V X - D G K M H I T L C N F V E P W D X L S L X Q R K G L N H R Y H L M a j o r i t y 1 1 1 1 130 140 150 160 i i i i _ 107 . 1 . . . L Q - . . L L . . . T . S . . A . . N S . . . A . . R . F T K T . T V hTIMP-1 111 . I . . . A E G D . I V . . . T . . T T . K . S Q M hTIMP-2 103 . . T . R . Y - . . . . Y T G R . . Q . T . S Y . . . . hTIMP-3 115 . . T . Q . L S . . . V F . H . . . Y l . . . E D . . . V . . E S . . . H . . . hTIMP-4 . C . . C . . . . C . . . P C C L W . D G . Q . . Consensus #1 G C E - C K I T X C Y X I P C X I S S P N E C L W T D W L L E X X X X G Y Q A K M a j o r i t y 1 1 1 r— 170 180 190 200 I i i i _ 146 . . . E . T V F P . L S . . . K L Q . G T H Q . . Q G S E K . F . S R hTIMP-1 151 . . . - . . . . R . P M . . . Y . . . . D . . . . M . . V T . K N I N . H . . . hTIMP-2 142 . . N - . . . K S . . Y L . . F V T . K M . S N F G Y P . . . S . hTIMP-3 155 N . G - . Q . . T . . T V . . T . . A R K L Y . . . . Q hTIMP-4 . . . C G . C . W Consensus #1 H Y A C I K R X D G X C S W Y R G X A P P X K E F X D I X D P M a j o r i t y 210 220 230 _ i i _ 186 . h . . L P . E P . L . T . Q S L R S Q I A hTIMP-1 190 F F S . . S . A . . . . A . . . K Q . . L . . E . . hTIMP-2 181 R Q K G . Y W . . . D . S I I N A T . . hTIMP-3 194 . . V . M . H V . . T . . . . . . H L . L R . . . V . . V Q . hTIMP-4 Figure 3: Sequence alignment of human TIMPs. Sequences of TIMP-1 (Accession number P01033), TIMP-2 (P16035), TIMP-3 (P35625), and TIMP-4 (Q99727) were obtained from the National Center for Biotechnological Information protein data bank and aligned using Megalign from DNAStar. Residues that match the consensus are shown as When all residues match the consensus, the residue of the consensus is shown, otherwise "." is shown. Note the 9 extra residues present in TIMPs-2, -3, and -4, C-terminal tail (underlined). 16 Figure 4: Crystal Structure of TIMP-2. The N-terminal and C-terminal domains are as indicated. The N-terminus is marked by residue Cys1. The C-terminal tail is not visible due to its disordered nature in the crystal. The approximate location of the C-terminal tail is indicated by the arrow. PDB Id: 1BR9 (Tuuttila et al., 1998). 17 cysteine residues that form disulphide bonds (Williamson et al., 1990). The disulphides are essential for the function of TIMPs as reduction and alkylation of the inhibitors abrogate their inhibitory activity (Cawston et al., 1981). Each domain of the TIMPs contains 3 loops with the N-terminal domain consisting of loops 1-3 and the C-terminal domain consisting of loops 4-6 (Williamson et al., 1990). The N-terminal domain of the TIMPs is highly conserved, especially the first 22 amino acids, and this domain is responsible for the inhibitory effect of TIMPs (Willenbrock and Murphy, 1994). The C-terminal regions of the TIMPs are much more diverse and may have a role in the binding efficiencies and specificities of the TIMPs to different MMPs as well as the other effects of TIMPs. The divergence of the C-terminal domain of TIMPs in conjunction with the differences in their expression has led to the hypothesis that TIMPs have differing biological roles (Douglas etal., 1997). 1.3.2 Mechanism of inhibition The mechanism of the inhibition of MMPs by TIMPs involves the first five residues of the N-terminus of TIMPs binding to MMPs in a substrate-like manner. The first residue, Cys1, coordinates to the active site zinc of MMPs while the hydroxyl group of the side chain of Ser2/Thr2 hydrogen bonds to the catalytic Glu of MMPs, thereby displacing a water molecule that is involved in catalysis (Williamson et al., 1994; Gomis-Ruth et al., 1997; Fernandez-Catalan et al., 1998; Stratmann et al., 2001a; Butler et al., 1999b). The C-terminal domain of TIMPs is not required for inhibition to occur, as C-terminally truncated TIMP 18 mutants (N-TIMPs) can efficiently inhibit MMPs (DeClerck et al., 1993; Willenbrock and Murphy, 1994). However, the C-terminal domain may provide extra contact residues because the rate of inhibition by full length TIMPs is greater than that of the N-TIMPs (Willenbrock and Murphy, 1994; Stratmann et al., 2001a; Stratmann etal., 2001b). It is likely that the C-terminal domain of the TIMPs contacts the hemopexin C domain of MMPs in the inhibitory complex as C-terminally truncated MMPs and matrilysin, which have no hemopexin C domain, bind TIMPs with greatly reduced affinity (Bigg et al., 1994; Baragi et al., 1994). 1.3.3 TIMP-1 TIMP-1 is a 184 amino acid glycoprotein that was discovered in the mid 1970's as a very stable protein that inhibited collagenase activity (Bauer et al., 1975) and was subsequently partially characterized by Stricklin and Welgus in 1983 (Stricklin and Welgus, 1983). Its molecular weight ranges from 28.5-30 kDa depending on its glycosylation state. TIMP-1 contains 2 separate N-glycosylation sites; one in loop 1 and another in loop 2. It has been shown that the inhibitory activity of TIMP-1 is unaffected by the loss of its carbohydrate moieties as mutation of the glycosylation sites have no effect on its ability to inhibit MMPs (Caterina etal., 1998). TIMP-1 was not fully cloned until 1985, when Docherty et al. discovered that it was identical to erythroid potentiating activity (EPA) cloned by Gasson et al. earlier in 1985 (Docherty et al., 1985; Gasson et al., 1985). Therefore from the time it was first cloned, it was known that TIMP-1 was a 19 multifunctional molecule that could both inhibit MMPs and act as EPA. The EPA activity of TIMP-1 was later shown to be separate from that of its MMP inhibitory activity, as destruction of its inhibitory activity through mutation did not affect its erythroid potentiating activity (Chesler et al., 1995). TIMP-1 has been shown to have growth factor effects (Hayakawa et al., 1992) and anti-apoptotic effects that are independent of its inhibitory ability (Guedez et al., 1998). TIMP-1 is a soluble protein secreted by many cell types, whose expression is regulated by a variety of factors such as phorbol esters, growth factors, and cytokines. TIMP-1 is expressed in most tissues although its levels are highest in the uterus, decidua, and ovary (Waterhouse et al., 1993; Young et al., 2002), which is in agreement with its involvement in reproductive processes that require matrix remodelling, such as ovulation (Inderdeo et al., 1996; Einspanier et al., 1999; Tsafriri and Reich, 1999). TIMP-1 has also been shown to reduce tumour cell growth, metastasis, and angiogenesis (Khokha, 1994; Johnson era/., 1994). With the exception of the MT-MMPs (Will etal., 1996; Butler era/., 1997), TIMP-1 inhibits all MMPs with high affinity, as well as a member of the ADAM family (a disintegrin and a metalloproteinase) of metzincins, ADAM-10 (Amour et al., 2000). Besides being able to form an inhibitory complex with MMPs, TIMP-1 has been found complexed with progelatinase B (Wilhelm et al., 1989). Binding of TIMP-1 to progelatinase B via its C-terminal non-inhibitory domain at a site other than the catalytic site of progelatinase B is suggested since: TIMP-1 cannot bind to progelatinase B in its usual inhibitory fashion because of the interference 20 of the pro-domain and as this proenzyme-TIMP complex inhibits other MMPs (Kolkenbrock et al., 1995); The complex inhibits the dimerization of gelatinase B and its ability to form a complex with MMP-1 (Goldberg et al., 1992); It also protects progelatinase B from activation by stromelysin-1 (Ogata etal., 1995). 7.3.4 TIMP-2 TIMP-2 is a 197 amino acid 21 kDa protein (Figure 4) that was discovered as a metalloproteinase inhibitor complexed to progelatinase A and was first characterised by Stetler-Stevenson et al. who, through amino acid sequencing of overlapping peptides, recognized its homology to TIMP-1 (DeClerck et al., 1989; Goldberg et al., 1989; Stetler-Stevenson et al., 1989). TIMP-2 is a soluble protein that is widely expressed in most tissues at relatively high levels (Young et al., 2002). Unlike TIMP-1, TIMP-2 is more constitutively expressed (Overall, 1994). TIMP-2 can inhibit all known MMPs with high affinity including the MT-MMPs (Will et al., 1996; Butler et al., 1997; Vincenti, 2001). TIMP-2 is a multi-functional molecule as it also possesses growth-stimulating activity in a manner similar to TIMP-1 and has been isolated as a factor promoting cell survival and an autocrine factor that stimulates simian virus (SV) 40 transformed human fibroblasts and osteosarcoma cells (Matsumoto et al., 1993; Nemeth and Goolsby, 1993; Woessner, 1996). TIMP-2 has also been shown to inhibit the proliferation of human microvascular endothelial cells (Murphy et al., 1993). In addition, TIMP-2 can suppress apoptosis in cells and inhibit tumour progression, metastasis, and angiogenesis (Guedez et al., 1998; Gomez et al., 1997). 21 From the time TIMP-2 was discovered, its ability to bind progelatinase A was known. Goldberg et al. showed that progelatinase A purified from HRAS-transformed human bronchial epithelial cells, SV-40 transformed lung fibroblasts and skin fibroblasts was noncovalently complexed to TIMP-2 in a stoichiometric manner (Goldberg et al., 1989). Goldberg et al. also showed that the TIMP-2/progelatinase A complex can by activated by organomercurials and inhibited by additional TIMPs (Goldberg et al., 1989). This indicated that the binding of TIMP-2 to progelatinase A does not involve the catalytic domain of progelatinase A, much like the TIMP-1/progelatinase B complex. TIMP-2 does not associate with C-terminally truncated progelatinase A (Murphy et al., 1992b) and therefore the TIMP binding site is likely to be located on the hemopexin C domain of progelatinase A. Olson et al. showed through kinetic analyses that the binding of TIMP-2 to both pro and active gelatinase A was biphasic with a higher and lower affinity site and that the gelatinase A hemopexin C domain contains the higher affinity binding site (Olson etal., 1997). As the active site in the proform is unavailable for binding and as these TIMP/proMMP complexes retained MMP inhibitory activity (Kolkenbrock et al., 1991), it was hypothesised that TIMPs can bind proMMPs with their C-terminal domain. TIMP-2, unlike TIMP-1, contains 9 extra residues at the C-terminus known as the C-terminal tail. The C-terminal tail of TIMP-2 has a number of negatively charged residues which are hypothesised to contribute to the binding of TIMP-2 to the positively charged hemopexin C domain of gelatinase A. The 22 binding site of TIMP-2 on progelatinase A is, as expected, on the hemopexin C domain and has been mapped by Overall et al. It is located on the peripheral rim of the gelatinase A hemopexin C domain at the junction between hemopexin modules III and IV and involves a number of cationic residues (Figure 5) (Overall et al., 1999). Deletion of the anionic C-terminal tail of TIMP-2 was shown to slow its rate of inhibition and it was demonstrated that ionic interactions were involved (Willenbrock and Murphy, 1994). The C-domain to hemopexin C domain interaction between TIMP-2 and progelatinase A is hypothesised to be important in the activation of progelatinase A (Butler et al., 1998). With the discovery of the cell surface activator of progelatinase A, MT1-MMP (Sato et al., 1994), it was soon hypothesised that the role of TIMP-2 in the TIMP-2/progelatinase A complex may be to aid in the activation of the zymogen. Activation of progelatinase A by TIMP-2 is a seemingly paradoxical function for this inhibitor. Cell surface activation was first shown by Overall et al. who reported that concanavalin A (con A) added to human fibroblasts induced the activation of progelatinase A (Overall and Sodek, 1990). The progelatinase A activation effect of con A is hypothesized to induce upregulation and clustering of MT1-MMPs at the cell surface (Fishman et al., 1996; Gervasi et al., 1996). The activation of progelatinase A at the cell surface is hypothesized to occur via a ternary complex mechanism (Figure 6). TIMP-2 plays a significant role in the ternary complex in that it forms part of a progelatinase A receptor on the cell surface by binding and inhibiting an MT1-MMP molecule with its inhibitory domain. This receptor acts to 23 IV I III II Figure 5: Crystal structure of the gelatinase A hemopexin C-domain in stereo. Shown above are cationic residues hypothesised to participate in the binding of TIMP-2 to progelatinase A. The residues are located on blades III and IV of the hemopexin C-domain. The four blades of the four-bladed beta propellor structure of the gelatinase A hemopexin C domain are numbered as indicated. Modified from PDB Id 1CK7 (Morgunova etal., 1999). 24 MT1-MMP f U L c ^ TIMP-2 Gelatinase A Q £ J Progelatinase A Figure 6: Diagrammatic representation of the ternary complex mechanism of the cell surface activation of progelatinase A. Activation of progelatinase A requires TIMP-2 binding the hemopexin C domain of progelatinase A through its non-inhibitory C-terminal domain and the association of TIMP-2 with activator proteinases, the membrane type (MT)1-MMPs, to form a trimolecular complex. TIMP-2, together with MT1-MMP, forms a receptor for progelatinase A that brings progelatinase A to the cell surface where it is cleaved by another free MT1-MMP molecule. 25 recruit progelatinase A to the cell surface through the interaction of the C-domain of TIMP-2 to the hemopexin C domain of progelatinase A, thus forming a tri-molecular complex (Butler et al., 1998). Another free molecule of MT1-MMP can then initiate the activation process by cleaving the prodomain of gelatinase A between residues Asn37 and Leu38 (Will et al., 1996). This enables the autocatalytic processing of the remainder of the prodomain between residues Asn80 and Tyr81 to result in active enzyme (Figure 6) (Will etal., 1996). In order for progelatinase A activation to occur, TIMP-2 concentrations must be optimal. If there is too little TIMP-2, activation is not supported. If there is too much TIMP-2, activation is inhibited because TIMP-2 inhibits all the available MT1-MMP (Butler et al., 1997). MT2 and MT3-MMP can also initiate the activation of progelatinase A at the cell surface although their mechanisms of action are unknown (Butler et al., 1997; Takino era/., 1995). The activation mechanism of progelatinase A by MT2-MMP does not require TIMP-2 or TIMP-4, although TIMP-3 involvement has not been ruled out (Morrison era/., 2001). 1.3.5 TIMP-3 TIMP-3 is a 188 amino acid glycoprotein that is unique among the TIMPs in that it is a relatively insoluble and ECM associated. TIMP-3 was first discovered as a matrix bound molecule produced by cells undergoing transformation (Blenis and Hawkes, 1984). The matrix-binding component of TIMP-3 was first attributed to its C-terminal domain (Langton et al., 1998), however, it is now thought to be the due to the interaction of the N-terminal domain with the sulphated 26 glycosaminoglycans of the ECM (Yu et al., 2000). TIMP-3 also has a 9 amino acid C-terminal tail like TIMP-2, although its tail is less anionic and it contains one N-glycosylation site (Wilde et al., 1994). As its glycosylation site is located within the C-terminal domain, removal of the carbohydrate moiety of TIMP-3 does not affect inhibition. It has been further shown that both the glycosylated and unglycosylated forms of TIMP-3 can localize to the ECM, indicating that glycosylation is not required for the ECM binding properties of TIMP-3 (Langton et al., 1998). TIMP-3 has also been shown to interact with the hemopexin C domain of gelatinases. However, unlike the other TIMPs, TIMP-3 can interact with the hemopexin C domain of both gelatinases A and B (Butler et al., 1999a) although the significance of these interactions are unknown. TIMP-3 also inhibits all known MMPs with high affinity (Butler et al., 1999a; Vincenti, 2001). However, TIMP-3, like TIMP-1, also inhibits ADAMs such as TNF-a converting enzyme (TACE or ADAM-17) (Amour et al., 1998) and ADAM-10 (Parks, 1999), which implies that the structure of the catalytic domain of ADAMs is similar to that of the MMPs. Interestingly, TIMP-3 also inhibits members of another branch of the ADAM family, a disintegrin and metalloprotease with thrombospondin type I modules (ADAM-TS), specifically ADAM-TS4 (aggrecanase-1), and ADAM-TS5 (aggrecanase-2) (Hashimoto etal., 2001; Kashiwagi et al., 2001). The aggrecanases are enzymes that have been shown to cleave aggrecan at sites that result in fragments that are seen in 27 various forms of arthritis and thus a reduction in the expression of TIMP-3 may be involved in the progression of arthritis. TIMP-3 is found in various tissues with thymus, heart, kidney and ovary having the highest levels (Young et al., 2002). TIMP-3 is also expressed in the Bruch's membrane of the eye where mutations in TIMP-3 are known to cause Sorsby's fundus dystrophy (Weber et al., 1994). Like TIMPs-1 and -2, TIMP-3 has also been shown to inhibit tumour growth, metastasis and angiogenesis although there have been reports of TIMP-3 enhancing tumorigenicity and invasiveness as well (Anand-Apte etal., 1996; Baker etal., 1999; Anand-Apte etal., 1997; Sun et al., 1996). However, unlike TIMPs-1 and -2, TIMP-3 has the ability to cause apoptosis in cells (Ahonen et al., 1998; Baker et al., 1998). Although synthetic MMP inhibitors cannot cause apoptosis at physiologically relevant concentrations, this function of TIMP-3 is attributed to its inhibitory activity as mutation of its N-terminal cysteine residue abrogates the effect (Bond et al., 2000). It is thought that the apoptotic effect of TIMP-3 is due to its ability to stabilize TNF-oc receptors on cell surfaces (Smith et al., 1997). TIMP-3 also stimulates detachment of transformed cells from the ECM and appears to accelerate morphological changes associated with cell transformation (Yang and Hawkes, 1992). In addition, mutations in TIMP-3 are responsible for an autosomal dominant retinal degenerative disease called Sorsby's fundus dystrophy, which leads to blindness (Weber et al., 1994). Mutations in TIMP-3 that are responsible for Sorsby's fundus dystrophy usually result in an unpaired 28 cysteine in the C-terminal domain, giving rise to TIMP-3 dimers that can inhibit MMPs and bind to matrix components but have impaired turnover and therefore accumulate in the eye causing degeneration (Langton et al., 2000; Sternlicht and Werb, 2001). Because of the role of TIMP-3 in a genetic disease, it was suggested that the other TIMPs could have roles in genetic diseases as well, specifically, TIMP-4 and dilated cardiomyopathy as TIMP-4 is highly expressed in the heart, although no known genetic defects in the TIMP-4 gene have yet been revealed in affected individuals (Olson etal., 1998). 1.3.6 TIMP-4 In 1996, the newest member of the TIMP family, TIMP-4, was described. TIMP-4 was the first TIMP cloned by analysis of expressed sequence tags. A partial clone was first identified from a human brain library and the full-length cDNA was later isolated from a human heart cDNA library (Greene et al., 1996). TIMP-4 is a 195-aa, 22 kDa protein. Like TIMP-2, human TIMP-4 is not glycosylated, although rabbit TIMP-4 is (Riley etal., 1999a; Devlieger et al., 2000; Douglas et al., 1997). TIMP-4 has the ability to inhibit all known MMPs, much like the other TIMPs (Vincenti, 2001). However, it is a much more potent inhibitor of gelatinase A and matrilysin-1 than of the other MMPs (Liu et al., 1997). TIMP-4 shares a high level of sequence similarity with TIMP-2 and TIMP-3. TIMP-4 is only -37% identical and -57% similar to TIMP-1 but it is -50% identical and -69% similar to TIMP-2 and -50% identical and -68% similar to TIMP-3 (Douglas et al., 1997; Greene etal., 1996). Because of the high homology between TIMP-2 and TIMP-29 4, it has been suggested that they inhibit MMPs by similar mechanisms. Studies show that TIMP-4 is very similar to TIMP-2 in its MMP inhibitory profile. TIMP-4, like TIMP-2 is a much more potent inhibitor of gelatinase A than gelatinase B (Liu et al., 1997). Although TIMP-2 and TIMP-4 both bind gelatinase A strongly, TIMP-4 has been shown to be a less potent inhibitor of gelatinase A than TIMP-2 (Bigg et al., 2001). However, TIMP-2, like gelatinase A, is expressed in a wide range of tissues, which is not the case with TIMP-4 (Goldberg etal., 1989). TIMP-4 transcripts are expressed at high levels in the heart, blood vessels, platelets, and brain, at lesser levels in the thymus and testis, and at extremely low or non-existent levels in all other tissues (Greene et al., 1996; Frolichsthal-Schoeller et al., 1999; Vaalamo et al., 1999; Young et al., 2002). In the adult mouse, TIMP-4 is expressed in the brain, heart, ovary and skeletal muscle (Leco etal., 1997). Because of its limited expression, it has been suggested that TIMP-4 is a tissue specific inhibitor of MMPs and may function as part of an acute response to tissue remodeling (Greene era/., 1996; Leco et al., 1997). However, in rat, TIMP-4 appears to have a wider expression profile which includes retina, smooth and skeletal muscle, skin, pancreas, heart, brain, lung, kidney, and testis (Wu and Moses, 1998). The aberrant expression of TIMP-4 is thought to be a factor in a number of diseased states. Although TIMP-4 is not normally expressed in the lung, it has been found in interstitial macrophages, plasma cells, and alveolar epithelial cells 30 of the lungs of patients with idiopathic pulmonary fibrosis (Selman et al., 2000). TIMP-4, like the other TIMPs, has been found to inhibit tumour angiogenesis and growth. It has also been found to initiate apoptosis of transformed cardiac fibroblasts and inhibit the migration and invasion of cardiac fibroblasts in a Boyden chamber assay (Tummalapalli et al., 2001). Also, the addition of recombinant human TIMP-4 and overexpression of TIMP-4 has been shown to alter the invasive potential of human breast cancer cells using a Matrigel assay (Liu et al., 1997; Wang et al., 1997). Cells transfected with TIMP-4 and orthotopically injected into nude mice exhibited significant inhibition of tumour growth by 4-10 fold in primary tumour volumes and a significant decrease was seen in axillary lymph node and lung metastases as compared to non-transfected cells (Wang et al., 1997). TIMP-4 expression has not been detected in most breast cancer cell lines. This low to nonexistent level of expression could be due to down-regulation of the gene during malignant progression (Wang et al., 1997). However, TIMP-4 appears to be expressed in the stromal cells. It has been suggested that stromal expression of TIMP-4 by fibroblasts surrounding breast carcinomas could be a host response to attempt to balance the local tissue degradation due to tumour cell invasion. Availability of excess TIMP-4 relative to MMP could create a microenvironment in the tumoral-stromal interface where subsequent tumour cell invasion can be inhibited by TIMP-4 (Leco et al., 1997). TIMP-4 also appears to be expressed mainly in the stroma in non-small cell lung carcinoma (Thomas et al., 2000). In clear cell renal cell carcinomas, a downregulation of TIMP-4 expression is attributable to the chromosomal loss of 31 3p, which contains the loci for TIMP-4 (Hagemann et al., 2001). Although TIMP-4 can inhibit tumour growth in vitro, there are conflicting views of its actions in vivo. Delivery of the TIMP-4 gene inhibits Wilm's tumour growth yet has been shown to stimulate mammary tumorigenesis (Celiker et al., 2001; Jiang et al., 2001). The high level of expression of TIMP-4 in the brain suggests that it has a potential role in this organ. TIMP-4 is expressed in both human embryonic central nervous system stem cells and mature and differentiated central nervous system cells (Frolichsthal-Schoeller et al., 1999). In addition, TIMP-4 transcripts have been found to increase during the development of the rat brain and are found in high levels in the embryonic dorsal root ganglia, indicating a role for TIMP-4 in the developing central nevous system. In the adult rat brain, TIMP-4 mRNA is found primarily in the cerebellum; particularly in the cerebellar Purkinje cells (Fager and Jaworski, 2000). TIMP-4 also appears to play a role in the progression of gliomas. TIMP-4 levels are elevated in low and middle grade gliomas and downregulated in high grade gliomas with respect to the lower grade gliomas indicating a negative correlation with glioma malignancy (Groft et al., 2001). A depression of TIMP-4 levels is also thought to have a role in the wound healing of ligaments, obesity, and the degradation of ECM at the periprosthetic interface of artificial hips (Reno et al., 1998; Maquoi et al., 2002; Sasaki et al., 2001). In 32 addition, TIMP-4 is expressed during certain reproductive processes: TIMP-4 is found in human milk with the maximum at 7 days and is thought to be involved in the protective actions of milk (Cheung et al., 2001); TIMP-4 has also been found in the Sertoli cells and gonocytes of human fetal testis and the granulosa and theca cells of the follicle in equine ovaries (Robinson et al., 2001; Riley et al., 2001) ; During the second trimester of pregnancy, low amounts of TIMP-4 protein are present in extraembryonic coelomic fluid and in amniotic fluid (Riley et al., 1999a); In addition, it has been shown by the use of cDNA arrays that TIMP-4 levels are increased in pre-term labour without chorioamnionitis (Marvin et al., 2002) . A decrease in the secretion of TIMPs, including TIMP-4, from the chorion laeve of the fetal membranes and the deciduas parietalis into the amniotic fluid is apparent during active labour. This suggests that decreased levels of TIMP-4 play a role in the detachment of fetal tissues from the uterus during childbirth, allowing proteases to function (Riley et al., 1999b). Of particular interest is the fact that the highest level of TIMP-4 expression is found in the heart where human cancer metastasis rarely occurs (Greene et al., 1996). The expression of TIMP-4 in the heart is thought to be regulated by cytokines as TNF-a and IL-1p\ which transiently increase TIMP-4 protein levels (Li etal., 1999). Dollery etal. found that vascular injury produced elevated TIMP-4 levels in rats and that TIMP-4 reduces the invasion of rat vascular smooth muscle cells in vitro (Dollery et al., 1999). TIMP-4 levels have also been shown to be decreased by over 50% in the left ventricle at 1 and 8 weeks following 33 myocardial infarction in rats (Peterson et al., 2000). However, because of its high levels of expression in the normal heart, it has been speculated that absence of this expression plays a role in idiopathic dilated cardiomyopathy (Douglas etal., 1997; Gunja-Smith etal., 1996; Olson etal., 1998). The amount of TIMP-4 is significantly reduced in idiopathic dilated cardiomyopathy (Li et al., 1998). The TIMP-4 gene was mapped to chromosome 3p25, a region that contains a locus for dilated cardiomyopathy, providing further evidence for the role of TIMP-4 in heart failure (Olson et al., 1998). TIMP-4 levels are also decreased in rats with spontaneously hypertensive heart failure (Li et al., 2000) and spontaneously hypertensive heart failure rats treated with the drug ramipril had increased TIMP-4 levels and decreased the mortality rate (Seeland et al., 2002), again indicating a role for TIMP-4 in heart failure. TIMP-4 is also suspected to play a role in the pathology of atherosclerosis (Douglas et al., 1997). 1.4 Aims of research, hypothesis, and experimental approach Understanding of the TIMP family is constantly increasing due to their apparent importance in the prevention of tumour angiogenesis and metastasis. Although much is known about TIMP-2 and its role in the remodelling of tissues, TIMP-4, due to its relatively recent discovery, is less well understood. Because of the high homology between TIMP-4 and TIMP-2, much of what is true for TIMP-2 was assumed to also apply to TIMP-4. Indeed, there appears to be some functional redundancy in the actions of TIMP-2 and TIMP-4. In addition to its 34 strong inhibitory effect on gelatinase A, it was suggested that TIMP-4 could also play a role in the activation of progelatinase A like TIMP-2 (Bigg et al., 1997). TIMP-4 was reported to bind to the gelatinase A hemopexin C domain although it has a lower binding affinity (Kd of 1.7 x 10"7) than TIMP-2 {Kd of 6.6 x 10"8) (Bigg et al., 1997). TIMP-4, like TIMPs-2 and -3 has a C-terminal tail and TIMP-4 was shown to interact with progelatinase A at or near to the TIMP-2 binding site based on the competitive inhibitory effect that TIMP-2 has on TIMP-4 in the binding of the C-domain of progelatinase A (Bigg et al., 1997). It was suggested that TIMP-4 could either compete with TIMP-2 to prevent activation or it could participate in activation itself. However, this has since been refuted by further work from Bigg et al. who showed that although TIMP-4 can bind progelatinase A (Bigg et al., 1997) and inhibit MT1-MMP (Bigg et al., 2001), it cannot form a tri-molecular complex and promote the activation of progelatinase A (Bigg et al., 2001; Toth et al., 2000; Hernandez-Barrantes et al., 2001). It was also shown that TIMP-4 cannot compete TIMP-2 from the gelatinase A hemopexin C domain but TIMP-2 can compete off TIMP-4 that is bound to the gelatinase A hemopexin domain (Bigg et al., 2001). However, TIMP-4 can inhibit progelatinase A activation at higher concentrations (Bigg et al., 2001; Toth et al., 2000). This inhibition of progelatinase A activation is assumed to be due to the ability of TIMP-4 to inhibit MT1-MMP, similar to the effect of excess TIMP-2 on the cell surface activation of progelatinase A. 35 Thus the aim of this thesis is to identify the aspects of TIMP-4 that precludes its participation in the activation of progelatinase A as opposed to TIMP-2, which can support the activation of progelatinase A. Since the anionic C-terminal tail of TIMP-2 is hypothesized to play a role in its association with the cationic hemopexin C domain of gelatinase A, it is interesting to note that although TIMP-4 has a C-terminal tail like that of TIMP-2 and -3, its tail is less anionic than that of TIMP-2 (net negative charge of 1 as opposed to 4). Therefore my hypothesis was that TIMP-4 cannot support activation because of the less negatively charged nature of its C-terminal tail and that the negatively charged C-terminal tail of TIMP-2 allows it to bind to the hemopexin C domain of gelatinase A in a manner that supports activation. In an effort to better understand the role that TIMP-4 plays in the regulation of gelatinase A and to elucidate the molecular determinants of gelatinase A activation, the C-terminal domain of TIMP-4 (C-TIMP-4) was cloned, expressed, and purified. The C-terminal domain of TIMP-2 (C-TIMP-2) was previously cloned in our lab. Because of the small size of the C-TIMPs (~8kDa), they were cloned into a vector containing horse heart myoglobin (hhMb) cDNA and expressed as a fusion protein (MbcTIMP) on the C-terminal side of hhMb separated by a glycine linker (Figure 7). Since the hhMb contains a heme group, it has a distinctive red colour, which makes it relatively easy to track both spectrophotometrically at 409 nm (Soret band) and visually. This also allows for 36 hhMb C Domain Tail hhMb MbcT2 X////////A E F L D I E D P MbcT2AT - f MbcT2Aq" Y / / / / / / / / \ //' \Q E F L D I PWIfp MbcT2^T4 V / / / / / / / A "ft K E F V D I V Q P MbcT4 K E F V D I V Q P MbcT4AT \ / / / / / / / / , MbcT4 + q- r////////, K E F V D I E D MbcT4-^T2 Y / / / / / / / Z Q E F L D I E D P Figure 7: Schematic representation of the MbcTIMP recombinant proteins. The hatched section represents hhMb. The white sections represent portions of TIMP-2. The black sections represent portions of the proteins that came from TIMP-4. The amino acid sequences shown represent the C-terminal tail and the tail mutations performed. (Not to scale) 37 the accurate quantification of the concentration of the MbcTIMPs as the soret absorbance is concentration dependent. It was hypothesized that the negative charges in the C-terminal tail of TIMP-2 allow it to dock to the hemopexin C domain of progelatinase A, providing stronger and faster interactions than TIMP-4. The amino acid sequences of the C-terminal tails of TIMP-2 (QEFLDIEDP) and TIMP-4 (KEFVDIVQP) only differ by four residues (underlined). Thus 6 mutations involving the C-terminal tails of TIMPs-2 and -4 were generated to determine the binding contribution of the differing residues. To determine the contribution of the C-terminal tail to the binding of the gelatinase A hemopexin C domain, mutations were made to remove the C-terminal fails of both TIMPs-2 and -4 resulting in the mutants MbcT2AT and MbcT4AT respectively. To determine the contribution to binding of the negative residues Glu192 and Asp193 within the C-terminal tail of TIMP-2, mutations were also made to remove these negative charges, making it more like the TIMP-4 tail (MbcT2Aq). In addition, mutations were introduced into the C-terminal tail of MbcT4 to add in the charged residues contained within the tail of TIMP-2 (MbcT4+q"). The C-terminal tails of MbcT2 and MbcT4 were also substituted with one another (MbcT2->T4 and MbcT4->T2) to determine if the chimeras would show altered binding to gelatinase A hemopexin C domain or progelatinase A (Figure 6). It was hypothesised that these recombinant proteins would be useful tools in determining why TIMP-4 cannot participate in the activation of progelatinase A by MT1-MMP but TIMP-2 can. This thesis presents 38 experiments which reveal that the fusion C-TIMP proteins can bind progelatinase A. I also present data demonstrating the differences between C-TIMP-2 and C-TIMP-4 in the binding of progelatinase A in the context of progelatinase A activation through the use of various binding assays. 3 9 CHAPTER 2 - MATERIALS AND METHODS 2.1 Cloning of pG YMC-MbcTIMP-4 Primers for polymerase chain reaction amplification (PCR) were designed to correspond to the 5' and 3' sequences of the C-terminal domain of hTIMP-4. The 5' primer, 5'cTIMP4 (5'-CGGGGGGCTAGCGGCTGCCAAATCACC-3') and 3' primer, 3'cTIMP4 (5'-GTCAAGCTTCTAGGGCTGAACGATGTC-3') contain Nhe I and Hind III restriction sites, respectively, for directional cloning (underlined). The C-terminal coding region of hTIMP-4 (from Gly128 to Pro195) was amplified from the full length cDNA located on pBS II SK-hTIMP-4 (obtained from Dr. Y. E. Shi, Department of Pediatrics, Long Island Jewish Medical Center, The Long Campus for the Albert Einstein College of Medicine, New Hyde Park, New York) (Greene et al., 1996). The following were the conditions under which the PCR was performed: 50 pmol of primer 5TIMP4, 50 pmol of 3TIMP4, 1 ng of pBS II SK-hTIMP-4, 43.5 pi of Reaction Mix (20 pi of 10x Mg-free PCR buffer (500 mM KCI, 100 mM Tris-HCI pH 9.0 at 25 °C, 1.0% Triton X-100), 200 pM of dNTP mix (25% (v/v) each of dATP, dCTP, dGTP, dTTP), nanopure dH 2 0 to 43.5 pi), 12 pi of 25 mM MgCI2, and 2 units of Taq polymerase in a 200 pi volume followed by the addition of 200 pi of mineral oil. The PCR was performed on a Cetus DNA amplification machine (Perkin-Elmer) and the reaction times were as follows: a 5 min denaturation period at 94 °C followed by 30 cycles of 94 °C for 40 s, 58 °C for 90 s, 72 °C for 60 s followed by a 10 min incubation period at 72 °C. C-TIMP-4 PCR product was recovered on a 1% agarose gel and purified by QIAEX II 40 Agarose Gel Extraction Kit. The gel-purified C-TIMP-4 was ligated into the expression vector pGYMC-MbMCS (Pelman and Overall unpublished) using the Nhe I and Hind III sites. The ligation reaction was performed using a 1:3 vector to insert ratio and 1 unit of T4 ligase (Gibco-BRL) at 16 °C for 16 h and 3 uJ of the resulting ligation reaction was transformed into sub-cloning efficiency Escherichia coli (E. Coli) DH5a (Gibco-BRL) by heat shock and plated onto LB-agar plates containing 100 u,g/ml ampicillin. The plates were incubated at 37 °C for 16 h. Colonies were selected for growth in liquid media + 100 u,g/ml ampicillin and plasmid DNA was isolated using the QIAprep Spin Miniprep kit from Qiagen. Potential clones were screened by restriction enzyme digestion and the resulting positives were sent to the Nucleic Acids-Protein Service (NAPS) Unit at the University of British Columbia for automated sequencing (500 ng pGYMC-MbcTIMP-4 and 3.2 pmol primers M13R or M13F-17 per reaction). The sequencing reaction was analyzed on an Applied Biosystems (ABI) Prism 377 using ABI's AmpliTaq FS DyeDeoxy™ Terminator Cycle Sequencing chemistry. 2.2 Creation of the MbcTIMP mutant constructs Six additional pGYMC-MbcTIMP constructs were created using the QuikChange™ Site-Directed Mutagenesis Kit from Stratagene. The mutagenesis reactions were performed according to the protocol provided in the kit and the resulting clones were sequenced using the Perkin Elmer Applied Biosystems Prism Big Dye Terminator Cycle Sequencing Ready Reaction Kit and the purified reaction products were analyzed by automated sequencing at the NAPS 41 sequencing unit. The sequencing data was analyzed using Sequencher 4.1.2 from Gene Codes Corporation, which confirmed the fidelity and accuracy of the constructs. 2.2.1 Mutagenesis of pGYMC-MbcTIMP-2 Primers were designed to create three constructs based on the C-TIMP-2 construct that was previously cloned in our laboratory, pGYMC-MbcTIMP-2 (King and Overall unpublished). To create the mutant lacking a C-terminal tail (MbcT2AT), the following primers were designed to introduce a stop codon nine residues before the end of MbcTIMP-2 at position 186 (Gln186Stop) based on the numbering of full-length TIMP-2 starting from Cys1 (mutated base(s) in bold and italics): CT2ST1 5' - G CG CG G CG CCCCCCAAG TAGG AGTTTCTCG ACATC - 3' GT2ST2 5' - GATGTCGAGAAACTCCTdCTTGGGGGGCGCCGCGC - 3'. Primers were also designed to create the mutant lacking key charged residues located near the end of the anionically charged tail of MbcTIMP-2 (Glu192Val/Asp193Gln), MbcT2Aq": 5'2T4#1 5' - G G AGTTTCTCG AC ATCG TG CA GCCATAACTG C AG G C ATG - 3' 3'2T4#1 5' - CATGCCTGCAGTTATGGCTGC4CGATGTCGAGAAACTCC - 3'. To mutate the MbcTIMP-2 tail to resemble the tail of TIMP-4, primers with two additional mutations (Gln186Lys/Leu189Val) were designed and introduced to the MbcT2Aq" template to make MbcT2—>T4: 5'2T4#2 5' - CGCCCCCCAAG>1AGGAGTTTGTCGACATCGTG - 3' 3'2T4#2 5' - CACGATGTCGACAAACTCCTTCTTGGGGGGCG - 3'. 42 2.2.2 Mutagenesis of pGYMC-MbcTIMP-4 In addition to the three mutants based on the pGYMC-MbctlMP-2 construct, primers were designed to create an additional three constructs based on pGYMC-MbcTIMP-4. To create the mutant lacking a C-terminal tail (MbcT4AT), the following primers were designed to introduce a stop codon nine residues before the end of MbcTIMP-4 at position 187 (Lys187Stop) based on the numbering of full-length TIMP-4 starting from Cys1 (mutated base(s) in bold and italics): CT4ST1 5' - GCCACCTGCGTCTCAGG TAGGAGTTTGTTGACATC - 3' CT4ST2 5' - GATGTCAACAAACTCCTi4CCTGAGAGGCAGGTGGC - 3'. Primers were also designed to create the mutant with the negatively charged residues in the C-terminal tail of MbcTIMP-4 (Val193Glu/Gln194Asp), MbcT4+q" that mimics the charged residues located in the C-terminal tail of TIMP-2: 5'4T2#1 5' - GAAGGAGTTTGTTGACATCGAGGACCCCTAGAAGCTTGGCACTG - 3' 3'4T2#1 5' - C AG TG CC AAG CTTCTAG G G GT CC7CG ATGTC A AC A AACTCCTTC - 3'. To mutate the MbcTIMP-4 C-terminal tail to resemble the C-Terminal tail of TIMP-2, primers with two additional mutations (Lys187GlnA/al190Leu) were designed and introduced to the MbcT4+q" template to make MbcT4->T2: 5'4T2#2 5' - CCTGCCTCTCAGGCAGGAGTTTCTTGACATCGAGG - 3' 3'4T2#2 5' - CCTCG ATGTC AA GAAACTCCT GCCTG AG AG GC AG G - 3'. 43 2.3 Screening of E. coli cell lines for protein expression pGYMC-MbcTIMP-4 was transformed into the following E. coli cell lines in order to determine the most suitable cell line for expression: BL21, BL21 Gold, BL21 + pTrix, BL21 + pGroesI, BL21 + pMagic, DH5cc, DH5ccF, JM101, Le392F, R1360, RZ1032, TB1, and XLI (blue). 5 mL cultures of Superbroth (1% (w/v) trypticase peptone, 0.8% (w/v) yeast extract, 0.5% NaCl, 0.001% (v/v) glycerol, pH 7.4) +100 pg/ml ampicillin were inoculated with each transformed cell line. 5 ml control cultures of superbroth without ampicillin were inoculated with each cell line (non-transformed). The cultures were grown for 16 h at 37 °C. Each of the resulting bacterial cultures were split into two tubes and the bacterial cell pellet was collected by centrifugation at 13,000 x g for 2 min in microfuge tubes. The cell pellets for all tubes were resuspended in 250 pi of 1 x NET (100 mM NaCl, 1 mM EDTA, 20 mM Tris pH 8) and sonicated on ice. 250 pi of 4 x SDS-PAGE sample buffer and 500 pi of ddH 20 were added to one of the two tubes of lysed cells from each culture. The other tubes of lysed cells were collected by centrifugation at 13,000 x g for 2 min and the supernatant (soluble protein fraction) transfered into a new microfuge tube containing in 250 pi of ddH 20 and 250 pi of 4 x SDS-PAGE sample buffer. The remaining protein pellets (insoluble inclusion body fractions) were resuspended in 750 pi of ddH 20 and 250 pi of 4 x SDS-PAGE sample buffer. The samples were analysed on 15% SDS-PAGE gels. 44 2.4 Stability testing of MbcTIMP-4 To determine suitable storage and working buffer conditions for MbcTIMP proteins, stability testing was performed according to a previously published protocol for optimization of solvent conditions for NMR spectroscopy of proteins (Lepre and Moore, 1998). Partially purified MbcTIMP-4 was used to screen buffer conditions. The following buffer conditions were tested: 100 mM Sodium Phosphate at pH 5.5, 6.0, 6.5, 7.0, 7.5, and 8.0; 100 mM Tris at pH 7.0, 7.5, 8.0, 8.5, and 9.0; 100 mM HEPES at pH 6.5, 7.0, 7.5, 8.0, and 8.5, 100 mM Sodium Carbonate at pH 9.0, 9.5, 10.0, and 10.5; and Phosphate buffered saline (PBS -137 mM NaCI, 2.68 mM KCI, 4.29 mM Na 2HP0 4»7H 20, 1.47 mM KH 2P0 4, pH 7.4). ' 2.5 Expression, refolding and purification of recombinant proteins All MbcTIMP recombinant proteins were expressed in the E. coli cell line BL21 Gold (see figure 8 for flow diagram of purification scheme). Transformed BL21 Gold cells were grown in 5 ml seed cultures of superbroth + 100 u,g/ml ampicillin for 16 h at 37 °C. Four 2 L flasks containing 700 ml of superbroth + 100 u,g/ml ampicillin were inoculated with 700 uJ of seed culture. The cultures were grown for 24 h at 37 °C and the bacterial cells collected by centrifugation (6000 x g for 15 min) in a Sorvall RC-5B refrigerated superspeed centrifuge. The cell pellet was washed once with 300 ml of 1 x NET and then resuspended in 250 ml of lysis buffer (20 mM Tris, 0.15 M NaCI, 5 (ig/ml DNase, 0.1 mg/ml Lysozyme, 10 mM Benzamidine, 2 mM MgCI2 pH 8). The cells were lysed for 16 h at 4 °C and 45 2.8 L overnight culture (E. coli BL21 Gold, grown 24 hrs). Media (Discarded) p i ( C e l 1 P e l l e t ) Y Resuspend in Lysis Buffer for 16 h at 4 °C. Sonicate. 15 000 x g for 15 min. Soluble protein P 2 (Inclusion Bodies) Y Resuspend in Solubilization Buffer for 16 h at 4 °C. I 19,000 x g for 1 h. Add 5mg of heme/1 OOmL of solubilized protein < Soluble Protein Y Refold protein (dialyse into 0.1 M Bicarbonate buffer pH 10) Y Y Y Dialyse into 10mM Tris pH 8 • Benzamidine Sepharose ^ n D o u n c * ^ DEAE Sepharose Unbound Unbound Wash with equilibration b u f f e r - ^ — • Nickel Chelate Sepharose J^W1 Wash with eq buffer + 1M NaCl, | W 2 Wash with 10mM, Tris pH 7 J^W3 Elute 0-500mM Imidazole gradient in 10mM Tris, pH 7 I Dialyse into 10mM Tris 1 pH unit below pi • C M Sepharose I Wash with 10mM Tris, pH 7 J^W1 Dialyse into 50mM Tris, 0.15M NaCl, pH 8 4 P°°l Elute 0-500mM NaCl in 10mM Tris Quantify protein Aliquot • Flash freeze protein Figure 8: Flow diagram of the purification scheme for MbcTIMPs. 46 then sonicated on ice to ensure complete cell lysis and to shear the remaining DNA. The inclusion bodies were harvested by centrifugation for 1 h at 19,000 x g. The inclusion body pellet was solubilized in 125 ml of solubilization buffer (20 mM Tris, 0.15 M NaCI, 6 M Guanidine hydrchloride, pH 8) for 16 h at 4 °C. The solubilized protein was then separated from the insoluble protein by centrifugation (19,000 x g for 1 h) and the supernatant was decanted off and the volume measured. 5 mg/100 ml of Hemin (Sigma H-5533) was added to the solubilized protein for the reconstitution of myoglobin during refolding and the volume of the solubilized protein was made up to 250 ml with refolding buffer (55 mM Na 2C0 3, 45 mM NaHC02, 0.02% (w/v) NaN3, pH 10) before dialysis into refolding buffer and then 10 mM Tris at the pi of the myoglobin fusion protein. The refolded protein was applied to Benzamidine Sepharose resin to remove serine proteases and the unbound fraction was then applied to 30 ml of equilibrated DEAE Sepharose. The NaCI concentration in the unbound fraction from the DEAE column was slowly raised to 0.5 M and then applied to a 30 ml Ni-Sepharose column, washed with 10 bed volumes of equilibration buffer (W1), 10 bed volumes of equilibration + 1 M NaCI (W2), and 10 bed volumes 10 mM Tris pH 7 (W3), before elution by a 0-500 mM imidazole gradient in 10 mM Tris pH 7. Elution fractions were pooled and dialysed into 10 mM Tris, 1 pH unit below the pi of the myoglobin fusion protein in no salt conditions and then applied to a 10 mL CM-Sepharose column to remove the lower molecular weight contaminating proteins. The column was washed with 10 bed volumes of 10 mM Tris at the equilibration pH (W4) and the recombinant protein was eluted using a 47 0-500 mM NaCl gradient in 10 mM Tris and dialysed into Tris storage buffer (50 mM Tris, 0.15 M NaCl, pH 8) for storage at -80 °C until use. If further purification was required, elution fractions from the CM-Sepharose column were pooled and dialysed into 10 mM Tris, 1 pH unit above the pi of the myoglobin fusion protein in no salt conditions and then applied to a DEAE-Sepharose column. The column was washed with 10 bed volumes of 10 mM Tris at the equilibration pH and the purified protein was eluted using a 0-500 mM NaCl gradient and then dialysed into Tris storage buffer. 2.6 Analysis of the MbcTIMP recombinant proteins 2.6.1 SDS-PAGE analysis 0.1 pg of each MbcTIMP recombinant protein, TIMP-2 and -4 standards, and a recombinant myoglobin standard were electrophoresed on 15% polyacrylamide separating gels using the Laemmli denaturing SDS discontinuous gel electrophoresis method (Laemmli, 1970). Protein bands were visualized by silver staining. 2.6.2 Western blotting analysis Proteins were prepared and electrophoresed as above and the gels soaked in Western blot semi-dry transfer buffer (39 mM glycine, 48 mM Tris, 0.0375% SDS, 20% methanol) for 10 min prior to transferring onto Immobilon-P membranes (Millipore) activated according to the manufacturer's protocol. The transfer was performed on a Pharmacia Biotech Multiphor II apparatus at 100 mA per gel for 1 48 h. The membranes were then blocked with 5% skim milk powder in 50 mM Tris, 0.15 M NaCI, pH 8 for 16 h at 4 °C. Membranes were washed with 50 mM Tris, 0.15 M NaCI, + 0.05% (v/v) Tween between blocking, application of primary and secondary antibodies, and development. Peptide antibodies previously available in our laboratory and raised in rabbits to the C-terminal tail of TIMP-2 and the C-terminal tail of TIMP-4 (GGLRKEFVDIVQ-NH? where the underlined residues differ from the natural sequence) as described by Overall et al., (Overall et al., 2000), and an antibody raised to recombinant hhMb were used to detect the proteins of interest. Secondary antibody was purchased from Bio-Rad (EIA Grade Affinity Purified Goat Anti-Rabbit IgG (H+L)-HRP Conjugate). Development of membranes was performed using ECL™ Western blotting detection reagents from Amersham Pharmacia Biotech according to their instructions. 2.6.3 Mass spectrometry analysis Electrospray ionization mass spectrometry was performed on the MbcTIMP recombinant proteins and hhMb to further assess the fidelity of the expressed and purified proteins. 10 ng of each recombinant protein in storage buffer was analysed on a SCIEX API 300 (Perkin-Elmer) in single quadrupole mode. All samples were found to contain both N-terminal processed (minus methionine) and unprocessed forms of the expected proteins. 49 2.6.4 Soret peak analysis To determine if the MbcTIMPs were fully reconstituted by the addition of heme during refolding, soret analysis was performed on a MbcTIMP ± the addition of 1 molar equivalent excess heme. Analysis was performed on a Varian Cary spectrophotometer. The spectrophotometric absorbance reading at 409 nm was used to determine the concentration of the MbcTIMPs. In order to determine the concentration of the recombinant MbcTIMPs, first spectrophotometric readings were taken of known amounts of serially diluted hhMb at 409 nm and a linear standard curve was constructed. This curve was then used to calculate the approximate number of moles within each sample of MbcTIMP. 2.7 Microwell solid phase binding assay In order to test the binding interactions between the C-TIMP fusion proteins and the gelatinase A hemopexin C domain, a solid phase assay, enzyme-linked immunosorbent assay (ELISA) was used. 0.5 pg of TIMP-2 (cloned, expressed and purified from Chinese hamster ovary cells by Dr. C. J. Morrison) (Overall et al., 2000; Bigg et al., 2001), TIMP-4 (cloned, expressed and purified from baby hamster kidney cells by Dr. H. F. Bigg) (Bigg etal., 2001), MbcTIMP, or hhMb in 100 pi of Voller's buffer (15 mM Na 2C0 3, 35 mM NaHC0 3 l 0.02% (w/v) NaN3, pH 9.6) was immobilized onto 96-well plates for 16 h at 4 °C. The wells were washed four times with -300 pi of Tris buffered saline (TBS - 20 mM Tris, 0.15 M NaCl, pH 7.4) + 0.05% (v/v) Tween using a Dynex Technologies Ultrawash Plus 96 well plate washer. To block the remaining binding sites on the wells, 200 pi of 50 blocking buffer (2.5% BSA (w/v) in TBS) was added to each well and incubated at 25 °C for 1 h. After blocking, the wells were washed as above prior to adding serial dilutions of gelatinase A hemopexin C domain (previously cloned in our laboratory by U. M. Wallon. Expressed and purified by A. E. King) (Wallon and Overall, 1997), which was incubated for 1 h at 25 °C. The excess gelatinase A hemopexin C domain was washed off as above and replaced with an affinity-purified antibody raised to a 6 histidine tag (anti-His6 antibody) located at the N-terminus of the recombinant gelatinase A hemopexin C domain. After incubation for 1 h at 25 °C, the anti-His6 antibody was washed off as above. Diluted secondary antibody (1/1000 in TBS + 0.25% (w/v) BSA + 0.05% (v/v) Tween), purchased from Bio-Rad (EIA Grade Affinity Purified Goat Anti-Rabbit IgG (H+L)-AP Conjugate), was then added and incubated for 1 h at 25 °C. The secondary antibody was then washed off as above and detection by alkaline phosphatase levels was quantified by p-nitrophenyl phosphate substrate (Sigma) in an ELISA plate reader (Molecular Devices) at 405 nm. 2.8 Affinity columns To test for an interaction between gelatinase A and the MbcTIMPs, various mini-columns such as gelatin Sepharose and gelatinase A hemopexin C domain coupled affi-gel were attempted. 51 2.8.1 Gelatin Sepharose mini columns 100 pi of gelatin Sepharose resin (Pharmacia) was equilibrated with equilibration buffer (50 mM Tris, 0.15 M NaCl, 0.05% (v/v) Brij-35, pH 8) and then blocked with 2.5% (w/v) BSA. The resin was then washed with 2 x 1 ml of equilibration buffer. 2 pg of progelatinase A (cloned, expressed and purified from TIMP-2 -/-cells by Dr. G. S. Butler) (Bigg et al., 2001) was incubated for 1 h at 25 °C with 2 pg TIMP-2, -4, MbcT2, MbcT4, MbcT2AT, MbcT4AT, or hhMb prior to addition to the resin. The resin was washed with equilibration buffer and subsequently eluted with equilibration buffer + 10% DMSO and then 4 x SDS-PAGE sample buffer. The resulting fractions were analysed by SDS-PAGE. 2.8.2 Affi-gel Coupled mini columns Gelatinase A hemopexin C domain (expressed and purified by I. Mak) (Wallon and Overall, 1997), MT1-MMP hemopexin C domain (previously cloned, expressed and purified in our laboratory by E. Tarn) (Overall et al., 2000), BSA and a PBS control were coupled to Affigel-10 (Bio-Rad) according to the protocol provided by the manufacturer (-0.685 mg to 3ml of resin). The resins were blocked with ethanolamine as described in the protocol and then equilibrated with equilibration buffer + 0.02% (w/v) sodium azide. Resin was further blocked with 2.5% BSA and washed with equilibration buffer. 0.65 nmoles of TIMP-2, MbcT2, MbcT4, or hhMb was loaded onto 100 pi of resin, twice, and allowed to incubate for 1 h at 25 °C. The resin was then washed with 1 ml of equilibration buffer, 1 ml of equilibration buffer + 1 M NaCl, and equilibration buffer + 10% DMSO and 52 then eluted with equilibration buffer + 0.1 M EDTA and 4 x SDS-PAGE sample buffer. The column fractions were analysed by SDS-PAGE. 2.9 Velocity sedimentation A 5-20% (w/w) sucrose gradient in 50 mM Tris, 0.15 M NaCI, 0.05% Brij-35, pH 8 was formed by layering 0.5 ml of 5%, 10%, and 15% sucrose solutions, on top of 0.5 ml of 20% sucrose solution in a Beckman Ultra-Clear centrifuge tube and letting stand for 1 h at room temperature. The gradients were then placed on ice for a further 30 min before layering 100 uJ of sample. Samples consisted of 8.5 |ig of the catalytically inactive mutant progelatinase A E375A (pE375A) (Crabbe et al., 1994) expressed in TIMP-2 -/- cells and purified by Dr. G. S. Butler (McQuibban etal., 2001) was incubated with 4.7 ug TIMP-2, 4.7 \ig TIMP-4, 5 ng MbcT2, or 4.92 u.g MbcT4 for 7 h at 4 °C, each protein on its own or pE375A from Chinese hamster ovary cells that is secreted pre-complexed with TIMP-2 (expressed and purified by Dr. G. S. Butler) (Bigg era/., 2001). The appropriate marker proteins (1 u,g each hhMb, carbonic anhydrase, ovalbumin, BSA, phosphorylase B with carbonic anhydrase being omitted for samples containing C-TIMP and BSA being omitted for samples containing progelatinase A due to similarity in size) were added to the sample prior to layering on the gradient. The tubes were then spun at 50,000 rpm for 16 h at 4 °C in a Beckman Ultracentrifuge. Fractions were collected and analysed by SDS-PAGE. 53 2.10 Glutaraldehyde cross-linking 2.10.1 Chemical cross-linking TIMP-2, TIMP-4, MbcT2, and MbcT4 were incubated with or without gelatinase A hemopexin C domain (expressed and purified by myself) (Wallon and Overall, 1997) at 6:1, 4:1, 3:1, and 2:1 molar ratios of TIMP:gelatinase A hemopexin C domain for 1 h at room temperature. Glutaraldehyde was then added to 0.5% (v/v) for 30 min at room temperature to cross-link interacting proteins. The reactions were stopped by addition of an equal volume of 2 x SDS-PAGE sample buffer. The interactions were analysed by SDS-PAGE and Western blotting (for TIMP-4 only) using the anti-TIMP-4 tail antibodies. MbcT2, MbcT4, MbcT2AT, MbcT4AT, and hhMb were also incubated with or without progelatinase A at 2.5:1 molar ratios of MbcTIMP or hhMb:progelatinase A and cross-linked as above. The reactions were stopped with equal volumes of 2 x non-reducing SDS-PAGE sample buffer and analysed by SDS-PAGE and zymography. 2.10.2 Analysis by zymography Samples were run at 20 mA/gel on 10% SDS-PAGE gels containing 0.025% gelatin. The gels were then incubated twice in 2.5% (v/v) Triton X-100 for 15 min each and then rinsed 3 x with dH 2 0 before incubation in zymography assay buffer (20 mM Tris, 0.15 M NaCl, 5 mM CaCI2, pH 8) for 5 h at 37 °C. The gels were then stained with Coomassie R-250. 54 2.11 Enzyme capture assay In order to test the binding interactions between the C-TIMP fusion proteins and full-length progelatinase A, an enzyme capture approach was used. 0.5 pg of TIMP-2, TIMP-4, MbcTIMP, or hhMb in 100 pi of Voller's buffer was immobilized onto 96-well fluorimetry plates for 16 h at 4 °C. The wells were washed 3 x with 200 pi of TBS. To block the remaining binding sites on the wells, 200 pi of blocking buffer (1% BSA (w/v) in TBS) was added to each well and incubated at 25 °C for 1 h. After blocking, the wells were washed as above prior to the addition of serial dilutions or progelatinase A or 250 pg or 125 pg of progelatinase A in fluorimetry assay buffer (FAB - 100 mM Tris, 10 mM CaCI2, 100 mM NaCl, 0.05% (v/v) Brij-35, pH 7.5), which was incubated for 2 h at 25 °C. The excess progelatinase A was washed off as above and the progelatinase A was activated by adding 180 pi of 2 mM 4-aminophenylmercuric acetate (APMA) and incubating for 1 h at 25 °C. 20 pi of fluorescence substrate QF-24 ((7-methoxycoumarin-4-yl)acetyl-Pro-Leu-Gly-Leu-[3-(2,4-dinitrophenyl)-L-2,3-diaminoproprionyl]-Ala-Arg-NH2) (Knight etal., 1992) was added to each well and the reaction was monitored on a Molecular Devices Fmax fluorescence plate reader for 4 h at 37 °C. The progelatinase A that was bound to the plate was also eluted off with 4 x non-reducing sample buffer and analyzed by zymography as above. 55 CHAPTER 3 - RESULTS 3.1 Recombinant protein expression and characterization C-TIMP-4 was cloned as a fusion protein C-terminal to hhMb for protein interaction studies with progelatinase A. C-TIMP-2 was previously cloned in our laboratory. Six mutants were created based on the MbcT2 and MbcT4 constructs (Figure 7). These mutants, like MbcT2 and MbcT4, were sequenced and found to be correct and containing the desired mutation(s). Mini-expression showed that the E. Coli strain BL21 was suitable for expression and that the MbcTIMPs are expressed and accumulated in insoluble inclusion bodies (Figure 9). Initial purification attempts revealed that the MbcTIMPs did not contain heme upon refolding and required reconstitution and that the MbcTIMPs were unstable and tended to precipitate in phosphate buffers and at lower pH. Stability testing was performed to determine suitable buffer conditions for the purification and storage of MbcTIMPs. MbcTIMPs were determined to be stable in Tris and carbonate-based buffers at neutral to high pH (data not shown). Typically, purification of the recombinant MbcTIMPs involved harvesting and lysing the MbcTIMP expressing BL21 cells, recovery of the insoluble inclusion bodies, solubilization of the inclusion bodies, reconstitution of heme during refolding and a series of chromatography columns. Figure 10 is an example of a purification scheme, in this case for MbcT4+q" (see figure 7 for schematic of this mutant). To remove proteases that may degrade the MbcTIMPs, benzamidine Sepharose was used. DEAE Sepharose, an anion exchange resin, was used to remove 56 pGYMC-MbcTIMP-4 - + - + - + 29kDa 18kDa 29kDa 18kDa a-cT4 Tail Figure 9: Mini expression of MbcT4 in BL21 revealed that the protein is expressed and accumulated in insoluble inclusion bodies. Transformation of pGYMC-MbcTIMP-4 vector into E. Coli BL21 cells resulted in an extra band at an apparent molecular weight of approximately 26kDa as indicated by the arrows, which is in agreement with the predicted molecular weight of MbcT4. The BL21 cells were harvested, lysed, and centrifuged to separate the soluble (supernatant) and insoluble (pellet) protein fractions. Western blot analysis with the a-C-TIMP-4 antibody indicates that this protein is likely MbcT4. 57 A B Elution (Imidazole) A Elution (Imidazole) ^ Elution (Imidazole) A M ] Ar *•* **** *•-*" <-MbcT4+q" Pooled (B3) Pooled (B3) Elution (NaCl) Elution (NaCl) v<-MbcT4+q" Pooled, dialysed and frozen Elution (NaCl) ^ Figure 10: Purification of the mutant MbcT4+q". MWt denotes molecular weight, B1 denotes before Benzamidine and DEAE Sepharose columns, B2 denotes before Ni 2 + chelate column, W1-W4 are as indicated in section 2.5 of chapter 2 (also see figure 8 for the detailed purification scheme). Pooled fractions are as indicated. Panel A: SDS-PAGE analysis of the DEAE Sepharose column fractions. The U1 fraction was loaded onto the nickel chelate column shown in panel B. The pooled fractions from the nickel chelate column were dialysed and further purified on a CM Sepharose column as shown in panel C. 58 many contaminating proteins (Figure 10 A). Ni chelate Sepharose was used to exploit the metal chelate binding properties of the heme (Figure 10 B), and the cation exchange, CM Sepharose column was used for further purification and was useful in removing the lower molecular weight contaminanting proteins (Figure 10 C). The majority of the MbcTIMPs were purified to homogeneity after the CM Sepharose column. The recombinant proteins could be tracked visually due to their red colour, indicating that reconstitution was successful (Figure 11 A). Heme reconstitution studies showed that the fusion proteins were fully reconstituted and no additional heme was required (Figure 11 B). The concentrations of the recombinant proteins were calculated from a standard curve based on a serial dilution of a known amount of hhMb. The extinction coeffcient for the soret absorbance of hhMb was calculated to be 149,415 M~1crrf1 (see table 2 for protein concentrations and yields). The fidelity of gene expression and identity of the proteins was confirmed by electrospray ionisation mass spectrometry. As shown in table 3, all MbcTIMPs tested contained both N-terminal methionine processed and non-N-terminal methionine processed forms. The wild-type and mutant recombinant C-TIMPs were purified to homogeneity as shown in figure 12. To confirm the presence of the mutations in the expressed proteins, Western blot analysis was performed as shown in panels B and C of figure 12. The anti-hhMb antibody recognized all Mb-fusion proteins. The anti-C-TIMP-2 tail (a-cT2 tail) antibody recognized TIMP-2, MbcT2, MbcT4+q", and MbcT4-»T2 while the anti-C-TIMP-4 tail (cc-cT4 tail) antibody recognized only 59 A o.o -I , , 1 1 200 300 400 500 600 Wavelength (nm) Figure 11: Spectrophotometric properties of the Mb-fusion proteins. Panel A: MbcT4+q", when reconstituted with heme, can be visually tracked during purification due to red colouring of the protein. Panel B: Spectrophotometric analysis is performed on 1.32 nmoles of reconstituted, heme containing MbcT2AT after purification (thin line - before additional heme). Shown above is the spectrum from 200-600nm with the soret peak absorbance at 409nm. The peak soret absorbance does not change significantly upon the addition of an additional 1 molar equivalent of heme to the already reconstituted MbcT2AT with a larger increase in the absorbance around 380nm indicating the presence of excess heme (thick line - after additional heme). This indicates that the reconstitution during refolding process was successful and that no further reconstitution of the MbcTIMPs is necessary. 60 Table 2: Protein yields from purification process Protein Absorbance Absorbance Highest Approximate at 280 nm1 at 409 nm1 concentration pool overall yield (mg/ml)2 (mg) MbcT2 1.43 4.92 0.841 6 MbcT4 1.44 2.87 0.492 3 MbcT2AT 0.3 0.99 0.162 1 MbcT4AT 6.0 17.1 2.82 17 MbcT2Aq" 1.05 3.66 0.625 8 MbcT4+q" 9.74 29.6 5.09 93 MbcT2^T4 1.23 3.58 0.611 8 MbcT4->T2 1.53 4.01 0.689 12 1 Absorbance of the highest concentration pool after multiplying with the dilution factor. 2 Calculated from the soret absorbance (409 nm) using the extinction coefficient 149,415 M"'cm"1. 61 Table 3: Molecular mass determination of horse heart myoglobin and myoglobin C-TIMP fusion proteins by electrospray ionization mass spectrometry Protein Measured mass A Mass from predicted2 ~ + N-terminal1 A N-terminal + N-terminal A N-terminal methionine methionine methionine methionine DA DA hhMb 17086 16954 +2 +2 MbcT2 25534 25400 +5 +3 MbcT4 25658 25526 +4 +4 MbcT2AT 24448 24312 +7 +2 MbcT4AT 24598 24472 +1 +6 MbcT2Aq" 25520 25388 +8 +8 MbcT4+q" 25672 25542 +1 +3 MbcT2^T4 25502 25372 +4 +6 MbcT4^T2 25688 25556 +3 +3 1 All the proteins measured registered peaks for ± N-terminal methionine although to varying degrees. For hhMb, MbcT4AT, MbcT2Aq", MbcT4+q", and MbcT4-»T2, the + Met peak is greater than the A Met peak. The A Met peak is substantially larger than the + Met peak for MbcT2, MbcT2AT, and MbcT2->T4. The + Met and A Met peaks are approximately the same size for MbcT4. 2 Predicted mass was determined from the amino acid sequence minus 2 Da per disulphide bond. 62 + DTT - DTT *o o> <y \* «9 o > ^ 29 kDa 18.4 kDa 14.2 kDa Mass -25400 -25526 Figure 12: SDS-PAGE and Western blotting analysis of MbcTIMP recombinant proteins as compared to full length TIMPs-2 and -4 and hhMb. Samples were analysed ± reduction with 65 mM DTT. Panel A: Wild-type recombinant C-TIMPs were purified to homogeneity. Shown are the masses of MbcT2 and MbcT4 from mass spectrometry. The full-length TIMPs have an apparent molecular weight of approximately 22 kDa and hhMb has an apparent molecular weight of 18 kDa. Continued on next page. 63 B + DTT _QII_ 1? 29 kDa 18 kDa Silver Stain a-cT2-Tail a-cT4-Tail a-hhMb + DTT J J I L 29 kDa 18 kDa Silver Stain a-cT2-Tail •cT4-Tail N||:: a-hhMb Figure 12: SDS-PAGE and Western blotting analysis of MbcTIMP recombinant proteins as compared to full length TIMPs-2 and -4 and hhMb. Panel B: Wild-type recombinant MbcT2 as compared to mutants based on the C-TIMP-2 construct. Panel C: Wild-type recombinant MbcT4 as compared to mutants based on the C-TIMP-4 construct. 64 TIMP-4, MbcT4, and MbcT2->T4. The lack of recognition for MbcT2Aq" by the cc-cT2 tail antibody and the ability of the a-cT2 tail antibody to recognise MbcT4+q" indicates that the a-cT2 tail antibody relies on the negatively charged residues Glu192 and Asp193 for specificity. The lack of recognition for MbcT4+q" by the a-cT4 tail antibody indicates that the epitope for the antibody includes the residues Val193 and Gln194. However, because the cc-cT4 tail antibody also does not recognise MbcT2Aq", this indicates that the epitope for the oc-cT4 tail antibody includes more of the C-terminal tail of TIMP-4. There was no difference in apparent molecular weight or antibody recognition ± reduction with the fusion proteins. The lack of dimers or multimers in the absence of DTT (Figure 12 A) shows that the majority of the protein is in the monomeric form, indicating that the proteins are correctly folded. 3.2 Microwell Solid Phase Binding Assay To determine whether the MbcTIMPs can bind gelatinase A hemopexin C domain, enzyme-linked immunosorbent assays (ELISAs) were performed. All MbcTIMPs tested could bind gelatinase A hemopexin C domain, however, the binding affinities varied with coating protein and were inconsistent although TIMP-2 consistently bound to gelatinase A hemopexin C domain with much greater affinity than any other protein tested (data not shown). As well, MbcT4 and MbcT4AT were shown to have similar binding affinities to gelatinase A hemopexin C domain (Figure 13). 65 m O cu o c ca .a i_ o to < 1.0 0.8 0.6 h-0.4 h 0.2 h 0.0 10"' O M b c T 4 • MbcT4AT T h h M b m i . 1 0 " 10"1 10" 10"1 0 1 0 9 1 0 s 10"7 1 0 6 1 0 s 10"4 Protein Concentration (Molar) Figure 13: Comparison of the binding of MbcT4 and MbcT4AT to gelatinase A hemopexin C domain. Both MbcT4 and MbcT4AT bind with similar affinities (K d approximately 4 x 10"6) to gelatinase A hemopexin C domain. 66 3.3 Affinity Columns Another method of testing for protein interactions is through the use of mini-affinity columns. Two types of columns were used: gelatin Sepharose and affi-gel. Gelatin Sepharose exploits the fact that progelatinase A can bind gelatin through its collagen binding domain. When TIMP-2 is pre-incubated with progelatinase A, a small amount of TIMP-2 co-eluted with progelatinase A using 10% DMSO (data not shown). However, TIMP-4, MbcT4, and MbcT2 did not co-elute with progelatinase A (data not shown). TIMP-2 could bind gelatinase A hemopexin C domain coupled affi-gel resin and not the blocked control resin while hhMb could not bind either (data not shown). However, MbcT2 and MbcT4 bound both the hemopexin C domain coupled resin and the control resin (data not shown). Addition of detergent, blocking proteins, or coupling of a different protein to the control resin did not abrogate the interaction between the MbcTIMPs and the control resin, thus no useful data was obtained from this method. 3.4 Velocity Sedimentation Velocity sedimentation was employed to determine if the wild-type C-TIMP fusion proteins, MbcT2 and MbcT4 could interact with progelatinase A. The predicted masses of the complexes are approximately 97kDa. Pre-complexed TIMP-2 and a catalytically inactive mutant of progelatinase A, E375A (pE375A) (Crabbe etal., 1994) were found to co-elute (Figure 14 A). Figure 14 B shows the SDS-PAGE analysis of the elution fractions from velocity sedimentation of TIMP-2 free pE375 67 A B 0 2 4 6 8 10 12 14 16 18 - pE375A Fraction Number -TIMP-2 Fraction Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 97 kDa-66 kDa-45 kDa-29 kDa-ISffil: -proMMP-2 -Ovalbumin -TIMP-2 -hhMb Figure 14: Determinat ion of protein interactions using velocity sedimentat ion. 8.5 p.g of progelat inase A E375A (pE375A) was incubated with 4.7 pg T IMP-2 or T IMP-4 or 5 ng of MbcT2 or MbcT4 for 7 hours. Samples plus markers were then layered on a 5 -20% sucrose gradient and centr i fuged for 16 hours. Densi tometry analysis was per formed on the silver stained S D S - P A G E analyses (representat ive gel shown in panel B) and plotted (panels A, C-F). The arrows indicate the peak fraction of the protein marker indicated. The T IMP or MbcTIMP standard represents the elution profile of the T I M P or MbcTIMP when not complexed with pE375A. Cont inued on the fol lowing page. 68 D -a- - pE375A —A TIMP-4 • TIMP-4 Standard 6 8 10 12 14 16 Fraction Num ber -B- - DE375A - A — MbcT2 2 4 6 8 10 12 14 16 Fraction Num ber MbcT2 Standard 0 2 4 6 8 10 12 14 16 18 Fraction Number - -B- - DE375A — * — MbcT4 MbcT4 Standard Figure 14: Determinat ion of protein interactions using velocity sedimentat ion. Panel A: Pre-complexed pE375A and TIMP-2 co-elute. Panel B: S D S - P A G E analysis of the elution fract ions of pE375A incubated with T IMP-2 (densi tometry presented in panel C). In panel C, and D, TIMP-2- f ree pE375A was complexed with T IMP-2 and T IMP-4 respectively. Panel E and F show the results of pE375A incubat ion with MbcT2 and MbcT4 respectively. 69 complexed with exogenous TIMP-2 (see figure 14 C for the densitometry analysis of this gel). TIMP-2 or TIMP-4 incubated with TIMP-2 free pE375A eluted in earlier fractions as compared to the TIMP-2 or TIMP-4 standards as shown in figure 14 C and D respectively. When MbcT2 or MbcT4 were incubated with pE375A, there was a shift in the peak fraction as compared to the MbcT2 or MbcT4 standards indicating a binding interaction between the MbcTIMPs and pE375A (Figure 14 E and F respectively). 3.5 Chemical Cross-linking To confirm an interaction between the MbcTIMPs and progelatinase A, glutaraldehyde cross-linking was performed. In figure 15 A, different concentrations of gelatinase A hemopexin C domain were incubated with TIMP-2, a known interactor, to determine protein concentrations that would give detectable levels of the TIMP-2/gelatinase A hemopexin C domain complex. TIMP-4 was also shown to interact with gelatinase A hemopexin C domain in figure 15 B. An increasing level of gelatinase A hemopexin C domain gave rise to an increase in the TIMP-4/gelatinase A hemopexin C domain complex as seen in figure 15 B that was identified by Western blotting using the a-C-TIMP-4 tail antibody. The interaction between MbcT2 and progelatinase A was confirmed by the presence of the MbcT2/progelatinase A complex band in both a silver stained gel and zymogram as indicated in figure 15 C. A band representing the MbcT4/progelatinase A complex was also present when MbcT4 was cross-linked 70 MMP-2 HexCD - + ++ + + + + + + + ++ +++ TIMP-2 + + + + + - - -0.5% G A + + + + - + + + • H C * ™ | r * 66kDa > B HexCD homodimer 4 5 k D a > tm mmwm < TIMP-2/HexCD complex <• TIMP-2 homodimer 29 kDa > tat ^•B< HexCD monomer •< TIMP-2 monomer MMP-2 HexCD - + ++ ++ TIMP-4 + + + + 0.5% GA + + + 45 k D a m t k i 4bKua ^ W — ^ TIMP-4 homodimer 29 kDa > .» HexCD monomer • W -< TIMP-4 monomer 18 kDa -66 kDa > 45 kDa > 29 kDa > 18 kDa > ! • TIMP-4/HexCD complex • TIMP-4 homodimer TIMP-4 monomer a-C-TIMP-4tail proMMP-2 MbcT2 0.5% GA 97 kDa > — 66 kDa > " " * 45 kDa > « • 29 kDa > « # - - + + + + + + + + - -- + + - + -II -MMP-2/MbcT2 .proMMP-2 -active MMP-2 -MbcT2 homodimer MbcT2 monomer MMP-2/MbcT2 proMMP-2 active MMP-2 Figure 15: Determination of protein interactions using glutaraldehyde cross-linking. Proteins were incubated together for 1 h prior to the addition of glutaraldehyde (GA) to 0.5%. Reactions were incubated for a further 30 min before stopping with equal volume 2 x SDS-PAGE sample buffer. For gelatinase A (MMP-2) hemopexin C domain (HexCD) and progelatinase A (proMMP-2), (+) denotes % molar equivalent, (++) denotes 1/2 molar equivalent, (+++) denotes 1 molar equivalent versus TIMP, MbcTIMP, or hhMb. Continued on the following page. 71 D proMMP-2 - - + + MbcT4 + + + + 0.5% GA - + + -9 7 k D a > — - H < ^ M M P - 2 / M b c T 4 66 kDa > * » * * ~ proMMP-2_ m ™ act ive M M P - 2 45 kDa > fP m M ^ = = - M b c T 4 h o m o d i m e r 29 k D a - - M b c T 4 h o m o d i m e r M M P - 2 / M b c T 4 p r o M M P - 2 act ive MMP-2 proMMP-2 MbcT2AT 0.5% GA „ 97 kDa > 66 kDa > 45 kDa >jB 29 kDa > « P VHP :!HP M M P - 2 / M b c T 2 A T p r o M M P - 2 act ive MMP-2 M b c T 2 A T homod imer proMMP-2 MbcT4AT 0.5% GA „ 97 k D a -66 k D a --> 1 45 k D a -— > # * 29 k D a -- - + + + + + + + + - -- + + - + -M M P - 2 / M b c T 2 A T p r o M M P - 2 act ive M M P - 2 - M M P - 2 / M b c T 4 A T H ^ p roMMP-2 proMMP-2 hhMb 0.5% GA 97 k D a — > 66 kDa > act ive MMP-2 MbcT4AT homod imer 4 5 k D a -29 k D a -M M P - 2 / M b c T 4 A T p roMMP-2 act ive MMP-2 - + + + + + + -+ + - + Si 9 Figure 15: Determinat ion of protein interactions using glutaraldehyde cross-l inking. Panel A: Three different concentrat ions of HexCD were incubated with T IMP-2 to determine the appropriate concentrat ion to use in this exper iment. Panel B: Two different concentrat ions of TIMP-4 were incubated with HexCD. The TIMP -4 /HexCD complex and the TIMP-4 homodimer were visual ized by Wes te rn blotting using the cc-C-TIMP-4 tail ant ibody. Panels C-F: MbcTIMPs were incubated at 1A molar equivalent to p roMMP-2 A. Posit ion of the MMP-2 /MbcTIMP complex on the silver stain (upper) and the zymogram (lower) are indicated. Panel G: Addit ion of g lutaraldehyde to pro-incubated proMMP-2 and hhMb does not give rise to p roMMP-2 /hhMb complexes. 72 to progelatinase A as indicated in figure 15 D. Figure 15 E and F show that the tail-less mutants MbcT2AT and MbcT4AT bound progelatinase A respectively. In contrast, hhMb did not interact with progelatinase A and therefore no hhMb/progelatinase A complex bands are seen in either the silver stain or through zymography (Figure 15 G). 3.6 Enzyme Capture Assay-To quantitate the binding interaction between progelatinase A and the MbcTIMPs, an enzyme capture experiment was performed. The results of the assay were visualized by two methods: fluorimetry and zymography. 3.6.1 Visualization by Fluorimetry Assay Progelatinase A captured on a fluorimetry plate by binding to coated TIMPs/MbcTIMPs or control proteins was APMA activated and subjected to a fluorimetry assay. Bound progelatinase A was quantified by measuring the rate of cleavage of a quenched fluorescent peptide substrate (Figure 16) (Knight et al., 1992). The results are normalized to the amount of enzyme captured by TIMP-2. Figure 16 demonstrates that TIMP-2 can capture higher amounts of progelatinase A than any other proteins tested. The amount of enzyme captured by MbcT2 was reduced by the loss or mutation of the C-terminal tail. MbcT4 did not capture as much enzyme as MbcT2 and loss of its C-terminal tail did not affect its ability to capture progelatinase A. The introduction of charges present 73 T J Q) N E i_ o c c: 0.75 1.25 1.00 A 0) o c (D o (/> V— o LL < 0.00 0.50 0.25 x"V A ' x> _> xtx A ' x<V *p ^> Coating Protein c r Figure 16: Visualization of enzyme capture assay by fluorimetry. Progelatinase A captured by the coating protein on a fluorimetry plate was APMA activated and detected by cleavage of a fluorogenic substrate. The values are expressed in fold of the amount of progelatinase A captured by TIMP-2. 74 in the C-terminal tail of TIMP-2 in MbcT4+q" and MbcT4->T2 increased ability to capture progelatinase A when compared to MbcT4. 3.6.2 Visualization by Zymography Progelatinase A bound to TIMP-2, TIMP-4, the MbcTIMPs, or hhMb coated on a fluorimetry plate was visualized using zymography, a sensitive method for the detection of gelatinase A bands by degradation of gelatin contained within a polyacrylamide gel by enzyme present in the protein band. TIMP-2 captured more progelatinase A than TIMP-4 as shown in figure 17 A. In contrast, hhMb could not capture any progelatinase A (Figure 17 A). MbcT2 captured less progelatinase A than TIMP-2 as shown in figure 17 B. MbcT4 and MbcT4AT captured similar but low amounts of progelatinase A. Again, figure 17 B shows that the loss of the negative charges or mutation of the C-terminal tail of MbcT2 reduced the amount of progelatinase A captured. Introduction of the negative charges in the C-terminal tail of TIMP-2 in MbcT4+q_ and MbcT4->T2 captured amounts of progelatinase A similar to MbcT2. 75 A proMMP-2 (ng) ^ <& 1? N v <bv V N ° Co TIMP-2 TIMP-4 hhMb B proMMP-2 (ng) ^ o? \V ( b v V N° A- Q> MbcT2 MbcT4 MbcT2AT MbcT4AT MbcT2Aq" MbcT4+q" MbcT2-^T4 M b c T 4 — T 2 Figure 17: Visualization of enzyme capture assay by zymography. Serial dilutions of progelatinase A were overlaid on the coating protein on a fluorimetry plate. After activation and fluorimetry assay, the proteins were eluted with 4 x non-reducing SDS-PAGE sample buffer and analysed by zymography. In panel A, progelatinase A eluted from wells coated with full length TIMPs-2 and -4 and hhMb are analysed as positive and negative controls. Panel B shows the progelatinase A captured when the MbcTIMPs are the coating proteins. 76 CHAPTER 4 - DISCUSSION 4.1 Perspective Because of the importance of type IV collagen degradation in tumour metastasis and angiogenesis, gelatinase A, a type IV collagenase, has been the subject of much research. Especially of interest is the unique ability of progelatinase A to be activated at the cell surface. This cell surface activation is paradoxical as it involves an inhibitor of MMPs, TIMP-2, effective over a narrow concentration range. Since TIMP-4 is highly homologous to TIMP-2 and can bind to both progelatinase A and MT1-MMP, it was proposed that TIMP-4 would also support the activation of progelatinase A (Bigg et al., 1997). However in 2001, Bigg et al. demonstrated that TIMP-4, unlike TIMP-2, cannot participate in the activation of progelatinase A, but can inhibit activation by MT1-MMP at higher concentrations (Bigg et al., 2001). It was also demonstrated that TIMP-4 could inhibit progelatinase A activation by MT2-MMP (Morrison era/., 2001). This has led to the work in this thesis, which addresses the issue of why TIMP-2 can participate in the activation of progelatinase A, but TIMP-4 cannot. TIMP-2 is hypothesized to bind progelatinase A hemopexin C domain through its C-terminal domain (Overall et al., 2000), utilizing an anionic C-terminal tail to aid in binding to the cationic residues of the hemopexin modules III and IV (Overall et al., 1999). As TIMP-4 does not possess these negatively charged residues in its C-terminal tail, we hypothesised that these negative charges may play an important role in promoting the interaction between TIMP-2 with progelatinase A. By cloning the C-terminal domain of TIMPs-2 and -4 and generating C-terminal tail mutants we 77 investigated the essential components that enable TIMP-2 to play a role in the activation of progelatinase A, which are absent in TIMP-4. 4.2 Recombinant protein expression, purification and characterization In order to dissect the protein interactions that occur in the activation of progelatinase A, the C-domains of TIMPs-2 and -4 were cloned as fusion proteins C-terminal to horse heart myoglobin (hhMb) and expressed in E. coli. It has never been directly shown that the C-terminal domains of TIMPs interact with progelatinase A through its hemopexin C domain, as C-TIMPs have not previously been expressed independently. It was assumed that since the complexed TIMP-2 was inhibitory, the N-terminal domain didn't interact and therefore the C-terminal domain was interacting with progelatinase A (Kolkenbrock et al., 1991). It was demonstrated by Butler et al., in 1998 that the catalytic domain of progelatinase A was unable to bind to full-length TIMP-2 (Butler et al., 1998), indicating that the hemopexin C domain is required for the TIMP-2/progelatinase A interaction to occur. Yeast-two-hybrid analysis later confirmed that the gelatinase A hemopexin C domain is sufficient to interact with full-length TIMP-2 (Overall et al., 2000; McQuibban et al., 2000). The cell line BL21 DE3 Gold was used to express the fusion proteins because it was found to have good expression levels. We predicted that since hhMb is soluble when expressed in E. coli (Tang et al., 1994) and easily purified by metal chelate chromatography that the recombinant fusion proteins would be soluble and easy to purify. However, this was not the case as the myoglobin C-TIMPs 78 (MbcTIMPs) when expressed, were located in inclusion bodies (Figure 9). This posed a significant problem, as upon denaturation, solubilization and refolding, the heme contained within the hhMb was lost. This made the proteins difficult to purify and decreased their stability. Besides the problems encountered with the loss of heme, it was soon discovered that the MbcTIMPs were unstable in phosphate-based buffers and at low pH. Exposure to low pH resulted in precipitation of the recombinant proteins as did freeze thawing in phosphate buffers. I therefore scanned for suitable storage and purification buffers. This was accomplished by the protocol established by Lepre and Moore for optimisation of solvent conditions for NMR spectroscopy of proteins (Lepre and Moore, 1998). It was discovered that the MbcTIMPs are relatively stable in tris-based buffers and very stable in carbonate-based buffers. Unreconstituted MbcTIMPs were less stable than MbcTIMPs containing heme and tended to precipitate during dialysis. In addition, the benefits of visually tracking the recombinant protein and using spectrophotometry to quantify the fusion proteins was lost as the heme is responsible for the spectrophotometric absorbance of hhMb at 409 nm. However, heme can be reconstituted into the fusion protein during refolding by addition to the solubilized protein solution. Heme reconstitution is normally accomplished by adding a 1:1 molar ratio of heme to myoglobin (Paulson et al., 1979). However, as the concentration of MbcTIMP in the protein preparation 79 was unknown, an excess amount of heme was typically added. Reconstituted MbcTIMP binds metal chelate columns and can be visually tracked (Figure 11 A). The red colour of the protein solution indicates that the reconstitution process was successful. However, the excess heme made purification difficult because heme is known to bind to the surface of other proteins and therefore purification of the MbcTIMPs required several steps (Figure 10). Benzamidine Sepharose resin was first added to the refolded protein solution to remove serine proteases that degrade MbcTIMPs. Because the pi of the MbcTIMPs is higher than many proteins, these proteins would be negatively charged at the pi of the MbcTIMPs and therefore would bind DEAE Sepharose while the MbcTIMPs would not and therefore many contaminating proteins were removed (Figure 10 A). The unbound fraction from the DEAE Sepharose, anion-exchange column was then applied to a Ni 2 + chelate column to exploit the metal chelate binding properties of myoglobin (Figure 10 B). The pooled fractions from the Ni 2 + chelate column were dialysed to one pH unit below the pi of the recombinant proteins and then applied to a CM-Sepharose, cation-exchange column. This would allow the MbcTIMPs to bind to the resin and was effective in removing the lower molecular weight contaminating proteins (Figure 10 C). The MbcTIMPs were also subjected to soret analysis to determine if the heme reconstitution was complete. The addition of additional heme to the purified proteins did not significantly alter the soret absorbance while increasing the absorbance at ~380nm where unincorporated heme absorbs (Figure 11 B), 80 indicating that the reconstitution was complete. Soret analysis also provided an excellent way to determine the concentration of the MbcTIMPs. As the absorbance of heme containing hhMb at 409 nm is concentration dependent, a standard curve was constructed using known amounts of hhMb. This allowed for the calculation of the molar concentration of the MbcTIMPs based on their soret absorbance, using the molar extinction coefficient of 149,415 M"1cm"1. The approximate overall yields of protein purification are presented in table 2. In addition to soret analysis, the purified recombinant proteins were subjected to electrospray ionization mass spectrometry to confirm the fidelity of gene expression, N-terminal methionine processing, and identity of the MbcTIMP proteins (table 3). The results of the mass spectrometry showed that all the MbcTIMPs and hhMb had both N-terminally processed (-met) and non N-terminally processed (+met) species and that the masses were within the limits of experimental error. The recombinant proteins were also visualized by SDS-PAGE followed by silver staining and Western blotting with the a-Mb, a-cT2 tail and a-cT4 tail antibodies, which were previously available in our laboratory (Figure 12). The a-cT2 tail and a-cT4 tail antibodies were previously described (Bigg et al., 2001). The apparent molecular weight of MbcT2 and MbcT4 does not change upon reduction with DTT, unlike the full-length TIMPs (Figure 12 A). This could be because the full-length TIMPs have 6 disulphides whereas the MbcTIMPs only have three disulphides, making a shift in apparent molecular weight less visible. The 81 absence of visible dimeric/multimeric bands in the fractions run without reduction indicate that the majority of the protein does not contain intermolecular disulphides. Silver staining showed that the proteins were purified to homogeneity and Western blotting with the anti-Mb antibody confirmed the presence of the hhMb fusion tag. As expected, MbcT2 and MbcT4 are recognized by the a-cT2 tail and a-cT4 tail antibodies, which are raised to the C-terminal tails of TIMPs-2 and -4 respectively and the tail-less mutants (MbcT2AT and MbcT4AT) are not recognized by either cc-C-TIMP tail antibody, thus confirming the loss of their C-terminal tails (Figure 12 B and C). It is interesting to note that both MbcT4 tail mutants (MbcT4+q" and MbcT4—>T2) are recognized by the cc-cT2 tail antibody and not the a-cT4 tail antibody (Figure 12 C), which confirmed the success of the mutations and indicated that the negatively charged residues are the major determinants of specificity for the cc-cT2 tail antibody. However, in the case of the MbcT2 tail mutants, MbcT2->T4 is recognized by the a-cT4 tail antibody whereas MbcT2Aq" is recognized by neither tail antibody (Figure 12 B). This could be because the epitope for the a-cT2 tail antibody encompasses the charged residues that are removed in the MbcT2 mutants and added in the MbcT4 mutants while the epitope for the oc-cT4 tail antibody encompasses a larger portion of the tail. The a-cT2 tail antibody does not recognize full length TIMP-2 as well as MbcT2 and this low level of detection of TIMP-2 has previously been noted by others in our laboratory. 82 4.3 Protein interaction studies The purpose of this thesis was to study the protein-protein interactions between progelatinase A or the gelatinase A hemopexin C domain and the different MbcTIMPs compared with full length TIMPs. Protein-protein interactions can be detected by many different methods including enzyme-linked immunosorbent assay (ELISA), affinity columns, and glutaraldehyde cross-linking, each of which has their own advantages and disadvantages. In addition to these methods, detection of interactions was performed by velocity sedimentation and enzyme capture assays. It was hoped that these experiments would provide insight into the ability of TIMP-2 to participate in the activation of progelatinase A. 4.3.1 Microwell solid phase binding assay An ELISA is a solid phase assay where a secondary protein bound to protein coated onto the wells on an ELISA plate is detected by antibodies. This proved to be an unreliable method of detecting interactions between- the proteins of interest. Although the MbcTIMPs were able to bind to gelatinase A hemopexin C domain, the binding affinities obtained were different if TIMPs or MbcTIMPs were coated on the plate versus if gelatinase A hemopexin C domain was coated on the plate. The presence of Brij 35 in the protein preparations also appeared to affect the results. There is also a large amount of error associated with this kind of assay as it involves the accurate pipetting of a primary protein, serial dilution of a secondary protein, and the accurate pipetting of primary antibody, secondary antibody, and substrate over a 96-well plate using a multichannel pipette. Plus, it 83 also depends on the affinity of the primary antibody for the second protein and only the second protein. In the end, the only useful information gained from these experiments was that TIMP-2 binds with greater saturation and affinity to the gelatinase A hemopexin C domain than TIMP-4 or any of the MbcTIMPs (data not shown) and that MbcT4 and MbcT4AT bind with similar affinities (Kd approximately 4 x 10"6) to gelatinase A hemopexin C domain (Figure 13). 4.3.2 Affinity columns Through the use of gelatin Sepharose mini columns, we previously showed that TIMP-4 can bind to progelatinase A (Bigg et al., 1997). This experiment was repeated and I could show binding of TIMP-2 to progelatinase A, but I could not reliably reproduce the result with TIMP-4. I was also unable to show that any of the MbcTIMPs tested bound to progelatinase A. In order to test the interaction between the hemopexin C domain and the MbcTIMPs, affinity columns using hemopexin C domain coupled to affi-gel resin were used. These columns showed that TIMP-2 could bind to the gelatinase A hemopexin C domain as opposed to PBS control resin while hhMb could not bind either of the resins. However, MbcT2 and MbcT4 bound to both the gelatinase A hemopexin C domain-coupled resin and the PBS control resin (data not shown). In an effort to reduce the non-specific interactions between the MbcTIMPs and the affi-gel resin, resin coupled to BSA and MT1-MMP hemopexin C domain were used as controls with no impact. In addition to these measures, the mini-columns were also blocked again with BSA prior to loading the MbcTIMPs onto the columns. 84 Unfortunately, the MbcTIMPs appeared to bind to the resin itself and this interaction was not reduced by the use of detergents or 1 M NaCl and the MbcTIMPs could only be eluted from the resin with 4 x SDS-PAGE sample buffer. The ability of these recombinant proteins to adhere directly to resin could be because the packing interface of the C-TIMP, which is normally closely interacting with the N-terminal domain of the TIMPs, is exposed in the fusion protein. It is likely that the residues along this interface are more hydrophobic contributing to its non-specific binding to resin. 4.3.3 Velocity sedimentation Another method of detecting protein-protein interactions is by velocity sedimentation. This method is useful for detecting interactions between proteins that have a large difference in molecular weight. Therefore to assess the protein interactions, full-length progelatinase A was used. Moreover, the hemopexin C domain, which is 29kDa, is too similar in size to the TIMPs and MbcTIMPs and therefore complexes between them would be difficult to resolve. Since the sucrose is unlikely to disrupt protein-protein interactions, the progelatinase A and TIMP or C-TIMP should remain complexed during the centrifugation and elute in the same fractions. As elution is accomplished from the bottom of the tube, the higher molecular weight proteins or protein complexes would elute in earlier fractions. A non-catalytically competent progelatinase A mutant, E375A (pE375A) made in Chinese hamster ovary cells (by Dr. G. S. Butler) is secreted as a complex with TIMP-2. This complex was used to test the ability of velocity 85 sedimentation to isolate complexed pE375A and TIMPs. pE375A was used in these experiments because progelatinase A autoactivates and autodegrades over time. When analysed by SDS-PAGE and densitometry, the pre-complexed pE375A and TIMP-2 co-elute (Figure 14 A), which indicated the suitability of this method for detection of pE375A-TIMP/C-TIMP complexes. When TIMP-2 and pE375A (expressed in TIMP-2 -/- cells) were incubated together to form a complex, there was a shift in the fraction number in which TIMP-2 eluted as compared to TIMP-2 alone (Figure 14 B and C). As an excess amount of TIMP-2 was added, not all TIMP-2 was complexed with progelatinase, resulting in the two distinct TIMP-2 peaks shown in figure 14 C. When progelatinase A and TIMP-4 were incubated together, there was also a shift in the peak elution fraction of TIMP-4 to coincide with the peak elution fractions of pE375A (Figure 14 D). However, the shift is more subtle, which likely reflects the lower affinity of TIMP-4 to pE375A compared with TIMP-2. When tested, both MbcT2 and MbcT4 peaked at an earlier fraction number when incubated with pE375A than when incubated on their own (Figure 14 E and F). This suggests that both MbcT2 and MbcT4 can bind pE375A like TIMP-2 and TIMP-4. The second, higher molecular weight peak found in the elution profiles of uncomplexed TIMP-2, TIMP-4, and MbcTIMP-4 could be because of protein aggregation resulting in homodimers. Unfortunately, the resolution of this experiment is relatively poor as the dimension of the tubes used was 2cm long with a diameter of 0.5cm. This meant that subtle differences in binding and weak binding interactions could not be detected and therefore the remaining recombinant proteins were not tested. 8 6 The resolution issue could be resolved by using a narrower but longer tube to perform this experiment but a centrifuge with a suitable rotor was not available. However, this experiment provided evidence that the recombinant C-TIMPs can bind to progelatinase A. 4.3.4 Chemical cross-linking The velocity sedimentation showed that MbcT2 and MbcT4, like TIMPs-2 and -4 can bind progelatinase A, so to confirm an interaction between the MbcTIMPs and progelatinase A or gelatinase A hemopexin C domain, glutaraldehyde cross-linking was employed. This method is only able to determine whether or not proteins interact and is unable to quantitate the interaction in solution. However, it is a useful tool for demonstrating protein interactions. To determine whether the method of glutaraldehyde cross-linking could show an interaction between TIMPs and gelatinase A and to determine the concentration of proteins to use, this method was first tested on TIMPs-2 and -4 and gelatinase A hemopexin C domain. Although glutaraldehyde cross-linking produced some TIMP homodimers, as an increasing amount of gelatinase A hemopexin C domain was added, the homodimers species lessened and the TIMP/gelatinase A hemopexin C domain complex became dominant (Figure 15 A and B). This indicates that the gelatinase A hemopexin C domain can bind to TIMP-2 and -4 and therefore the same conditions could be used to determine if the recombinant C-TIMPs could also bind gelatinase A hemopexin C domain. Progelatinase A was used in cross-linking experiments with the MbcTIMPs because the sizes of the 8 7 recombinant proteins and gelatinase A hemopexin C domain are very similar and therefore homodimers would be indistinguishable from heterodimers. Moreover, progelatinase A interactions with C-TIMPs would be more natural. Using progelatinase A had the added advantage of being able to detect the complexes by the highly sensitive technique of zymography. In figure 15 C and D, MbcT2 and MbcT4 are once again shown to interact with progelatinase A by the presence of an extra band in the silver stain and the shift in progelatinase A activity in the zymogram as indicated by the arrows. The addition of glutaraldehyde to the reactions appears to have an effect on effectiveness of the gelatinase A to cleave the gelatin embedded in the gel. This could be due to intramolecular cross-linking within the gelatinase A thereby inactivating a small proportion of the enzyme present in the samples. The tail-less mutants (MbcT2AT and MbcT4AT) were also shown to interact with progelatinase A (Figure 15 E and F) whereas hhMb did not interact (Figure 15 G). Because removal of the C-terminal tails of MbcT2 and MbcT4 does not stop the proteins from interacting with progelatinase A, it was assumed that the remaining mutants would interact with progelatinase A. This is because the tails of the remaining mutants contain only a few amino acid substitutions and are likely to interact with progelatinase A if the tail-less mutants can and therefore MbcT2Aq", MbcT4+q", MbcT2—>T4, and MbcT4-»T2 were not tested using this method. 8 8 4.3.5 Enzyme Capture Assay The glutaraldehyde cross-linking experiment demonstated that the MbcTIMPs, with or without their C-terminal tails, could interact with progelatinase A. In order to quantify these interactions, an enzyme capture method was employed. Enzyme capture is a solid phase assay much like ELISA. TIMP-2, TIMP-4, the MbcTIMPs, hhMb, and ovalbumin were coated on a high protein adsorption fluorimetry plate. Progelatinase A was added to the wells, incubated, and then the wells were washed. The progelatinase A that was bound to the coating protein was then APMA-activated and a fluorimetry assay was performed to detect gelatinase A activity. As the amount of enzyme captured by the coating protein increases, the rate of substrate cleavage increases accordingly. Therefore, the rate of cleavage of a quenched fluorescent peptide substrate was used to determine the relative amount of enzyme "captured" by the coating protein. Figure 16 shows the rates of gelatinase A captured by TIMP-2, TIMP-4, and the MbcTIMPs, while hhMb and ovalbumin served as controls. The rates are expressed relative to the rate of progelatinase A captured by TIMP-2. However, it must be noted that the activity captured by TIMP-2 and TIMP-4 most probably does not represent the total amount of enzyme captured by these proteins, as some activity would be blocked due to the inhibitory activity of the TIMPs, as the enzyme is activated by APMA, it is possible for full-length TIMPs to bind the catalytic domain through an inhibitory interaction. This inhibitory interaction between the full length TIMPs and activated gelatinase A becomes more apparent as the concentration of progelatinase incubated on the coating protein 89 is lowered. Despite this, the assay shows that TIMP-2 binds progelatinase A more avidly than TIMP-4, which is consistent with our previous data (Bigg et al., 1997). It also shows that MbcT2 binds to progelatinase A with a higher affinity than MbcT4, although much lower than TIMP-2. The lower activity seen in MbcT2 wells than in TIMP-2 wells could be because of N-terminal domain contributions to binding of progelatinase A. This would be consistent with data from our lab that shows that N-TIMP-2 can bind gelatinase A hemopexin C domain in low salt conditions, which could indicate a small role for the N-terminal domain of TIMP-2 in progelatinase A binding (Wallon and Overall, unpublished data). Another possibility is that the normally close interaction between the N-and C-terminal domains of TIMPs stabilizes the TIMP C-terminal domain. The lack of the N-terminal domain in the MbcTIMPs could destabilize C-TIMP leading to a weaker interaction between it and gelatinase A hemopexin C domain. It is also possible that the hhMb fusion domain sterically hinders the interaction between C-TIMP-2 and progelatinase A. The ability of MbcT4 to bind progelatinase A is only slightly lower than that of TIMP-4, unlike the difference between TIMP-2 and MbcT2. This is an indication that perhaps the majority of the difference between the ability of TIMP-2 and MbcT2 to bind progelatinase A is because of N-terminal contributions or C-TIMP destabilization as steric hinderance would likely affect the binding of MbcT4 to progelatinase A as well. As expected, MbcT2AT bound less progelatinase A than MbcT2 indicating a role for the C-terminal tail of TIMP-2. When the charges in the C-terminal tail of TIMP-2 were removed as in the case of MbcT2Aq", binding of progelatinase A 90 was also reduced. This suggests that the negatively charged residues in the C-terminal tail play a dominant role in the interaction between C-TIMP-2 and gelatinase A hemopexin C domain. However, removal of the C-terminal tail of C-TIMP-4 does not appear to significantly affect its ability to bind to progelatinase A. This indicates that the remainder of the C-domain of TIMP-4 is able to interact with the gelatinase A hemopexin C domain but this interaction is not enhanced by the presence of the C-terminal tail of TIMP-4. Significantly, addition of the charges found in the tail of TIMP-2 to C-TIMP-4 increases the amount of progelatinase A captured when compared to MbcT4, providing further evidence that lack of these charges are involved in binding to the hemopexin C domain of gelatinase and and that lack of these charges in the TIMP-4 C-terminal tail contribute to its inability to participate in activation. Also, substitution of the C-terminal tail of C-TIMP-4 with that of TIMP-2 increases the rate of progelatinase A captured as compared to MbcT4 which again suggests that the differences between the C-terminal tails of TIMPs-2 and -4 prevent TIMP-4 from supporting the activation of progelatinase A. As expected of the negative controls, hhMb and ovalbumin captured the lowest activity of all proteins tested. Enzyme capture can also be visualized through zymography as the proteins in the wells can be eluted with 4x SDS-PAGE sample buffer and electrophoresed on zymograms. Again, this assay shows that TIMP-2 binds progelatinase A to a greater extent than TIMP-4 while the hhMb-coated wells do not have a significant amount of gelatinase A activity (Figure 17 A). Like the results in the fluorimetry 91 assay, MbcT2 captured more progelatinase A activity than MbcT4 but less than TIMP-2 (Figure 17 A and B). MbcT2AT captures less activity than MbcT2 whereas MbcT4AT appears to have captured an amount of progelatinase A similar to MbcT4 (Figure 17 B). The activity in the MbcT2Aq_ zymogram was less than that of MbcT2 again indicating that the negative charges Glu192 and Asp193, that were removed in MbcT2Aq" play a role in the binding of TIMP-2 to progelatinase A. MbcT2->T4 was not significantly different from MbcT2Aq". Interestingly, when the corresponding residues in the C-terminal tail of TIMP-4 (Val193 and Gln194) are mutated to the negatively charged residues found in the C-terminal tail of TIMP-2, as is the case with MbcT4+q_, an increase in the amount of gelatinase A captured, is seen (Figure 17 B) which is consistent with the fluorimetry data. The activity seen in wells coated with MbcT4—>T2 is also increased as compared with MbcT4 and this result is also consistent with the result from the fluorimetry assay and is likely due to the higher affinity of the C-terminal tail of TIMP-2 to progelatinase A. From the location of the TIMP-2 binding site on the gelatinase A hemopexin C domain mapped by Overall et al. in 1999 and the data presented above, it appears that the anionic C-terminal tail of TIMP-2 is involved in the TIMP-2/progelatinase A interaction. It is likely that the negative character of the C-terminal tail leads to an interaction between it and the clusters of cationic residues that form the TIMP-2 binding site. The negative residues Glu192 and Asp193 appear to be especially important in the interaction between C-TIMP-2 92 and progelatinase A as removal of these charges reduces its ability to bind progelatinase A. As the negative charges within the C-terminal tail of TIMP-2 are localised and within close proximity to one another, cationic clusters located on or near the rim of blade III such as Lys547/Lys549/Lys550/Lys558/Arg561 and Lys566/Lys567/Lys568, (Figure 18) could serve to attract and bind the anionic TIMP-2 tail, which stabilizes the interaction between TIMP-2 and the gelatinase A hemopexin C domain. The C-domain of TIMP-2 could then associate with the cationic charges of blade IV. In addition, hydrophobic interactions are likely to be involved in the packing interface around the charged residues. The work in this thesis shows a clear role for the negative residues Glu192 and Asp 193 in the anionic tail of TIMP-2 which indicates that the lack of these charges in the C-terminal tail of TIMP-4 is partially responsible for its inability to participate in the activation of progelatinase A. 93 Figure 18: Hypothetical binding site for the C-terminal tail of TIMP-2 on the hemopexin C domain of gelatinase A. The two clusters of cationic charges hypothesized to bind to the anionic tail of TIMP-2 are coloured in dark grey (Lys547/Lys549/Lys550/Lys558/Arg561) and black (Lys566/I_ys567/Lys568). Modified from PDB Id 1CK7 (Morgunova etal., 1999). 94 CHAPTER 5 - CONCLUSION The data from the velocity sedimentation and the glutaraldehyde cross-linking experiments showed that the MbcTIMPs can interact with progelatinase A in a manner similar to full length TIMP-2 and TIMP-4. To further dissect the interaction between the C-TIMPs and progelatinase A and to quantify the contribution to binding of the C-terminal tail, an enzyme capture was performed using the MbcTIMPs as "bait". From the use of different tail mutants in the enzyme capture assay, the contribution to binding of C-terminal tails of TIMPs-2 and -4 was characterized. The data from both the fluorimetry and the zymography analyses of the enzyme capture assay indicate a role for the C-terminal tail of TIMP-2 in its binding to progelatinase A, thereby supporting the hypothesis that the C-terminal tail of TIMP-2 is involved its binding to progelatinase A. In addition, it was shown by removal of the charged residues Glu192 and Asp193 that these negative charges contribute to the binding of C-TIMP-2 to progelatinase A and that lack of these negative charges in the C-terminal tail of TIMP-4 decrease its affinity for progelatinase A as compared to TIMP-2. This lowered affinity of TIMP-4 for progelatinase A is likely responsible for its inability to form a trimolecular complex and support activation. In addition, mutation of the C-terminal tail of TIMP-4 to that of TIMP-2 increased the affinity of C-TIMP-4 to progelatinase A and this is likely to be due to the addition of the charged residues in the C-terminal tail as addition of these charges also increase the affinity of C-TIMP-4 for progelatinase A. This study has increased our understanding of the inability of the TIMP-4/MT1-MMP complex to act as a cell 95 surface receptor of progelatinase A. It has also shed light on the role of TIMP-2 in the activation of progelatinase A through elucidation of the molecular determinants involved in the participation of TIMP-2. The nature of the less negatively charged C-terminal tail of TIMP-4 does not allow it to bind to progelatinase A with as strong an affinity as TIMP-2. This weaker interaction could account for the inability of TIMP-4 to support progelatinase A activation. The C-terminal tail of TIMP-4 also contains a positively charged lysine residue at position 187, which could inhibit its binding to the positively charged hemopexin C domain. It is also possible that TIMP-4 lacks key N-terminal interactions with the gelatinase A hemopexin C domain that contribute to the TIMP-2/progelatinase A interaction. In addition, TIMP-3, like TIMP-4 can bind progelatinase A although it has not been shown to participate in the MT1-MMP mediated activation of progelatinase A (Butler et al., 1999a). This could be because the C-terminal tail of TIMP-3 has a net charge of 0, including a positively charged lysine residue at the same position as the one in the C-terminal tail of TIMP-4. TIMP-1, the only member of the TIMP family without a C-terminal tail is also the only member of the TIMP family unable to bind to progelatinase A. However, TIMP-1 does form a complex with progelatinase B although the progelatinase B/TIMP-1 interaction does not appear to mediate the activation of progelatinase B, which is activated through other pathways. It is likely that TIMP-1 binds to progelatinase B at a site similar to the TIMP-2 binding site on progelatinase A as TIMP-1 and gelatinase B are homologous to TIMP-2 and gelatinase A respectively. The interactions between C-TIMPs and progelatinase 96 A could be further investigated through cell based activation studies and fluorimetry based competition assays for additional characterization the role of the C-terminal tail, and specifically Glu 192 and Asp 193, in the activation complex. 97 CHAPTER 6 - REFERENCES Ahonen, M., Baker, A. H. and Kahari, V. M. Adenovirus-mediated gene delivery of tissue inhibitor of metalloproteinases-3 inhibits invasion and induces apoptosis in melanoma cells. Cancer Res. 58(11): 2310-5 1998 Amour, A., Slocombe, P. M., Webster, A., Butler, M., Knight, C. G., Smith, B. J., Stephens, P. E., Shelley, C, Hutton, M., Knauper, V., Docherty, A. J. and Murphy, G. TNF-alpha converting enzyme (TACE) is inhibited by TIMP-3. FEBS Lett. 435(1): 39-44. 1998 Amour, A., Knight, C. G., Webster, A., Slocombe, P. M., Stephens, P. E., Knauper, V., Docherty, A. J. and Murphy, G. The in vitro activity of ADAM-10 is inhibited by TIMP-1 and TIMP-3. FEBS Lett. 473(3): 275-9. 2000 Anand-Apte, B., Bao, L, Smith, R., Iwata, K., Olsen, B. R., Zetter, B. and Apte, S. S. A review of tissue inhibitor of metalloproteinases-3 (TIMP-3) and experimental analysis of its effect on primary tumor growth. Biochem Cell Biol. 74(6): 853-62 1996 Anand-Apte, B., Pepper, M. S., Voest, E., Montesano, R., Olsen, B., Murphy, G., Apte, S. S. and Zetter, B. Inhibition of angiogenesis by tissue inhibitor of metalloproteinase-3. Invest Ophthalmol Vis Sci. 38(5): 817-23 1997 Aumailley, M. and Gayraud, B. Structure and biological activity of the extracellular matrix. J Moi Med. 76(3-4): 253-65. 1998 Baker, A. H., Zaltsman, A. B., George, S. J. and Newby, A. C. Divergent effects of tissue inhibitor of metalloproteinase-1, -2, or -3 overexpression on rat vascular smooth muscle cell invasion, proliferation, and death in vitro. TIMP-3 promotes apoptosis. J Clin Invest. 101(6): 1478-87 1998 Baker, A. H., George, S. J., Zaltsman, A. B., Murphy, G. and Newby, A. C. Inhibition of invasion and induction of apoptotic cell death of cancer cell lines by overexpression of TIMP-3. BrJ Cancer. 79(9-10): 1347-55 1999 Baragi, V. M., Fliszar, C. J., Conroy, M. C, Ye, Q. Z., Shipley, J. M. and Welgus, H. G. Contribution of the C-terminal domain of metalloproteinases to binding by tissue inhibitor of metalloproteinases. C-terminal truncated stromelysin and matrilysin exhibit equally compromised binding affinities as compared to full-length stromelysin. J Biol Chern. 269(17): 12692-7. 1994 Bauer, E. A., Stricklin, G. P., Jeffrey, J. J. and Eisen, A. Z. Collagenase production by human skin fibroblasts. Biochem Biophys Res Commun. 64(1): 232-40 1975 98 Bello, L, Lucini, V., Carrabba, G., Giussani, C., Machluf, M . , Pluderi, M. , Nikas, D., Zhang, J., Tomei, G., Villani, R. M. , Carroll, R. S., Bikfalvi, A. and Black, P. M . Simultaneous inhibition of glioma angiogenesis, cell proliferation, and invasion by a naturally occurring fragment of human metalloproteinase-2. Cancer Res. 61(24): 8730-6. 2001 Benbow, U. and Brinckerhoff, C. E. The AP-1 site and MMP gene regulation: what is all the fuss about? Matrix Biol. 15(8-9): 519-26 1997 Bigg, H. F., Clark, I. M. and Cawston, T. E. Fragments of human fibroblast collagenase: interaction with metalloproteinase inhibitors and substrates. Biochim Biophys Acta. 1208(1): 157-65. 1994 Bigg, H. F., Shi, Y. E., Liu, Y. E., Steffensen, B. and Overall, C. M. Specific, high affinity binding of tissue inhibitor of metalloproteinases-4 (TIMP-4) to the COOH-terminal hemopexin-like domain of human gelatinase A. TIMP-4 binds progelatinase A and the COOH- terminal domain in a similar manner to TIMP-2. J Biol Chem. 272(24): 15496-500. 1997 Bigg, H. F., Morrison, C. J., Butler, G. S., Bogoyevitch, M. A., Wang, Z., Soloway, P. D. and Overall, C. M. Tissue inhibitor of metalloproteinases-4 inhibits but does not support the activation of gelatinase A via efficient inhibition of membrane type 1-matrix metalloproteinase. Cancer Res. 61(9): 3610-8. 2001 Birkedal-Hansen, H. Role of matrix metalloproteinases in human periodontal diseases. J Periodontol. 64(5 Suppl): 474-84. 1993 Birkedal-Hansen, H., Moore, W. G., Bodden, M. K., Windsor, L. J., Birkedal-Hansen, B., DeCarlo, A. and Engler, J. A. Matrix metalloproteinases: a review. Crit Rev Oral Biol Med. 4(2): 197-250 1993 Blenis, J. and Hawkes, S. P. Characterization of a transformation-sensitive protein in the extracellular matrix of chicken embryo fibroblasts. J Biol Chem. 259(18): 11563-70 1984 Bode, W., Fernandez-Catalan, C., Tschesche, H., Grams, F., Nagase, H. and Maskos, K. Structural properties of matrix metalloproteinases. Cell Mol Life Sci. 55(4): 639-52. 1999 Bond, M., Murphy, G., Bennett, M. R., Amour, A., Knauper, V., Newby, A. C. and Baker, A. H. Localization of the death domain of tissue inhibitor of metalloproteinase-3 to the N terminus. Metalloproteinase inhibition is associated with proapoptotic activity. J Biol Chem. 275(52): 41358-63 2000 99 Boudreau, N. J. and Jones, P. L. Extracellular matrix and integrin signalling: the shape of things to come. Biochem J. 339(Pt 3): 481-8. 1999 Brew, K., Dinakarpandian, D. and Nagase, H. Tissue inhibitors of metalloproteinases: evolution, structure and function. Biochim Biophys Acta. 1477(1-2): 267-83 2000 Brooks, P. C , Silletti, S., von Schalscha, T. L, Friedlander, M. and Cheresh, D. A. Disruption of angiogenesis by PEX, a noncatalytic metalloproteinase fragment with integrin binding activity. Cell. 92(3): 391-400. 1998 Butler, G. S., Will, H., Atkinson, S. J. and Murphy, G. Membrane-type-2 matrix metalloproteinase can initiate the processing of progelatinase A and is regulated by the tissue inhibitors of metalloproteinases. Eur J Biochem. 244(2): 653-7. 1997 Butler, G. S., Butler, M. J., Atkinson, S. J., Will, H., Tamura, T., van Westrum, S. S., Crabbe, T., Clements, J., d'Ortho, M. P. and Murphy, G. The TIMP2 membrane type 1 metalloproteinase "receptor" regulates the concentration and efficient activation of progelatinase A. A kinetic study. J Biol Chem. 273(2): 871-80. 1998 Butler, G. S., Apte, S. S., Willenbrock, F. and Murphy, G. Human tissue inhibitor of metalloproteinases 3 interacts with both the N- and C-terminal domains of gelatinases A and B. Regulation by polyanions. J Biol Chem. 274(16): 10846-51. 1999a Butler, G. S., Hutton, M., Wattam, B. A., Williamson, R. A., Knauper, V., Willenbrock, F. and Murphy, G. The specificity of TIMP-2 for matrix metalloproteinases can be modified by single amino acid mutations. J Biol Chem. 274(29): 20391-6. 1999b Caterina, N. C., Windsor, L. J., Bodden, M. K., Yermovsky, A. E., Taylor, K. B., Birkedal-Hansen, H. and Engler, J. A. Glycosylation and NH2-terminal domain mutants of the tissue inhibitor of metalloproteinases-1 (TIMP-1). Biochim Biophys Acta. 1388(1): 21-34 1998 Cawston, T. E., Galloway, W. A., Mercer, E., Murphy, G. and Reynolds, J. J. Purification of rabbit bone inhibitor of collagenase. Biochem J. 195(1): 159-65 1981 Celiker, M. Y., Wang, M., Atsidaftos, E., Liu, X., Liu, Y. E., Jiang, Y., Valderrama, E., Goldberg, I. D. and Shi, Y. E. Inhibition of Wilms' tumor growth by intramuscular administration of tissue inhibitor of metalloproteinases-4 plasmid DNA. Oncogene. 20(32): 4337-43 2001 100 Chesler, L, Golde, D. W., Bersch, N. and Johnson, M. D. Metalloproteinase inhibition and erythroid potentiation are independent activities of tissue inhibitor of metalloproteinases-1. Blood. 86(12): 4506-15. 1995 Cheung, P. Y., Sawicki, G., Gross, S., Van Aerde, J. and Radomski, M. Differential expression of matrix metalloproteinases and the tissue inhibitor in human milk. Proc West Pharmacol Soc. 44(97-8): 2001 Collier, I. E., Wilhelm, S. M., Eisen, A. Z., Marmer, B. L, Grant, G. A., Seltzer, J. L, Kronberger, A., He, C. S., Bauer, E. A. and Goldberg, G. I. H-ras oncogene-transformed human bronchial epithelial cells (TBE-1) secrete a single metalloprotease capable of degrading basement membrane collagen. J Biol Chern. 263(14): 6579-87. 1988 Crabbe, T., Zucker, S., Cockett, M. I., Willenbrock, F., Tickle, S., O'Connell, J. P., Scothern, J. M., Murphy, G. and Docherty, A. J. Mutation of the active site glutamic acid of human gelatinase A: effects on latency, catalysis, and the binding of tissue inhibitor of metalloproteinases-1. Biochemistry. 33(21): 6684-90 1994 Creemers, E. E., Cleutjens, J. P., Smits, J. F. and Daemen, M. J. Matrix metalloproteinase inhibition after myocardial infarction: a new approach to prevent heart failure? Circ Res. 89(3): 201-10 2001 Curran, S. and Murray, G. I. Matrix metalloproteinases: molecular aspects of their roles in tumour invasion and metastasis. EurJ Cancer. 36(13 Spec No): 1621-30. 2000 DeClerck, Y. A., Yean, T. D., Ratzkin, B. J., Lu, H. S. and Langley, K. E. Purification and characterization of two related but distinct metalloproteinase inhibitors secreted by bovine aortic endothelial cells. J. Biol. Chern. 264(29): 17445-53 1989 DeClerck, Y. A., Yean, T. D., Lee, Y., Tomich, J. M. and Langley, K. E. Characterization of the functional domain of tissue inhibitor of metalloproteinases-2 (TIMP-2). Biochem J. 289(Pt 1): 65-9 1993 Devlieger, R., Deprest, J. A., Gratacos, E., Pijnenborg, R., Leask, R. and Riley, S. C. Matrix metalloproteinases -2 and -9 and their endogenous tissue inhibitors in fetal membrane repair following fetoscopy in a rabbit model. Moi Hum Reprod. 6(5): 479-85 2000 Docherty, A. J., Lyons, A., Smith, B. J., Wright, E. M., Stephens, P. E., Harris, T. J., Murphy, G. and Reynolds, J. J. Sequence of human tissue inhibitor of metalloproteinases and its identity to erythroid-potentiating activity. Nature. 318(6041): 66-9. 1985 101 Dollery, C. M., McEwan, J. R., Wang, M., Sang, Q. A., Liu, Y. E. and Shi, Y. E. TIMP-4 is regulated by vascular injury in rats. Circ Res. 84(5): 498-504 1999 Douglas, D. A., Shi, Y. E. and Sang, Q. A. Computational sequence analysis of the tissue inhibitor of metalloproteinase family. J Protein Chem. 16(4): 237-55 1997 Edwards, D. R., Beaudry, P. P., Laing, T. D., Kowal, V., Leco, K. J., Leco, P. A. and Lim, M. S. The roles of tissue inhibitors of metalloproteinases in tissue remodelling and cell growth. IntJObes Relat Metab Disord. 20(Suppl 3): S9-15. 1996 Egeblad, M. and Werb, Z. New functions for the matrix metalloproteinases in cancer progression. Nature Rev. Cancer. 2(3): 161-74 2002 Einspanier, R., Gabler, C, Bieser, B., Einspanier, A., Berisha, B., Kosmann, M., Wollenhaupt, K. and Schams, D. Growth factors and extracellular matrix proteins in interactions of cumulus-oocyte complex, spermatozoa and oviduct. J Reprod Fertil Suppl. 54: 359-65 1999 Elliott, S. and Cawston, T. The clinical potential of matrix metalloproteinase inhibitors in the rheumatic disorders. Drugs Aging. 18(2): 87-99 2001 Fager, N. and Jaworski, D. M. Differential spatial distribution and temporal regulation of tissue inhibitor of metalloproteinase mRNA expression during rat central nervous system development. Mech Dev. 98(1-2): 105-9 2000 Fernandez-Catalan, C., Bode, W., Huber, R., Turk, D., Calvete, J. J., Lichte, A., Tschesche, H. and Maskos, K. Crystal structure of the complex formed by the membrane type 1-matrix metalloproteinase with the tissue inhibitor of metalloproteinases-2, the soluble progelatinase A receptor. Embo J. 17(17): 5238-48 1998 Fishman, D. A., Bafetti, L. M. and Stack, M. S. Membrane-type matrix metalloproteinase expression and matrix metalloproteinase-2 activation in primary human ovarian epithelial carcinoma cells. Invasion Metastasis. 16(3): 150-9 1996 Fridman, R., Fuerst, T. R., Bird, R. E., Hoyhtya, M., Oelkuct, M., Kraus, S., Komarek, D., Liotta, L. A., Berman, M. L. and Stetler-Stevenson, W. G. Domain structure of human 72-kDa gelatinase/type IV collagenase. Characterization of proteolytic activity and identification of the tissue inhibitor of metalloproteinase-2 (TIMP-2) binding regions. J Biol Chem. 267(22): 15398-405. 1992 102 Fridman, R., Toth, M., Pena, D. and Mobashery, S. Activation of progelatinase B (MMP-9) by gelatinase A (MMP-2). Cancer Res. 55(12): 2548-55. 1995 Frolichsthal-Schoeller, P., Vescovi, A. L, Krekoski, C. A., Murphy, G., Edwards, D. R. and Forsyth, P. Expression and modulation of matrix metalloproteinase-2 and tissue inhibitors of metalloproteinases in human embryonic CNS stem cells. Neuroreport. 10(2): 345-51 1999 Gasson, J. C., Golde, D. W., Kaufman, S. E., Westbrook, C. A., Hewick, R. M., Kaufman, R. J., Wong, G. G., Temple, P. A., Leary, A. C, Brown, E. L. and et al. Molecular characterization and expression of the gene encoding human erythroid-potentiating activity. Nature. 315(6022): 768-71. 1985 Gearing, A. J., Beckett, P., Christodoulou, M., Churchill, M., Clements, J., Davidson, A. H., Drummond, A. H., Galloway, W. A., Gilbert, R., Gordon, J. L. and et al. Processing of tumour necrosis factor-alpha precursor by metalloproteinases. Nature. 370(6490): 555-7. 1994 Gervasi, D. C, Raz, A., Dehem, M., Yang, M., Kurkinen, M. and Fridman, R. Carbohydrate-mediated regulation of matrix metalloproteinase-2 activation in normal human fibroblasts and fibrosarcoma cells. Biochem Biophys Res Commun. 228(2): 530-8 1996 Gjannelli, G. and Antonaci, S. Gelatinases and their inhibitors in tumor metastasis: from biological research to medical applications. Histol Histopathol. 17(1): 339-45 2002 Gohji, K., Fujimoto, N., Fujii, A., Komiyama, T., Okawa, J. and Nakajima, M. Prognostic significance of circulating matrix metalloproteinase-2 to tissue inhibitor of metalloproteinases-2 ratio in recurrence of urothelial cancer after complete resection. Cancer Res. 56(14): 3196-8. 1996 Goldberg, G. I., Marmer, B. L, Grant, G. A., Eisen, A. Z., Wilhelm, S. and He, C. S. Human 72-kilodalton type IV collagenase forms a complex with a tissue inhibitor of metalloproteases designated TIMP-2. Proc Natl Acad Sci USA. 86(21): 8207-11. 1989 Goldberg, G. I., Strongin, A., Collier, I. E., Genrich, L. T. and Marmer, B. L. Interaction of 92-kDa type IV collagenase with the tissue inhibitor of metalloproteinases prevents dimerization, complex formation with interstitial collagenase, and activation of the proenzyme with stromelysin. J Biol Chern. 267(7): 4583-91. 1992 Gomez, D. E., Alonso, D. F., Yoshiji, H. and Thorgeirsson, U. P. Tissue inhibitors of metalloproteinases: structure, regulation and biological functions. EurJ Cell Biol. 74(2): 111-22 1997 103 Gomis-Ruth, F. X., Maskos, K., Betz, M., Bergner, A., Huber, R., Suzuki, K., Yoshida, N., Nagase, H., Brew, K., Bourenkov, G. P., Bartunik, H. and Bode, W. Mechanism of inhibition of the human matrix metalloproteinase stromelysin-1 by TIMP-1. Nature. 389(6646): 77-81 1997 Greene, J., Wang, M., Liu, Y. E., Raymond, L. A., Rosen, C. and Shi, Y. E. Molecular cloning and characterization of human tissue inhibitor of metalloproteinase 4. J Biol Chem. 271(48): 30375-80. 1996 Groft, L. L., Muzik, H., Rewcastle, N. B., Johnston, R. N., Knauper, V., Lafleur, M. A., Forsyth, P. A. and Edwards, D. R. Differential expression and localization of TIMP-1 and TIMP-4 in human gliomas. BrJ Cancer. 85(1): 55-63 2001 Guedez, L., Courtemanch, L. and Stetler-Stevenson, M. Tissue inhibitor of metalloproteinase (TIMP)-1 induces differentiation and an antiapoptotic phenotype in germinal center B cells. Blood. 92(4): 1342-9 1998 Gunja-Smith, Z., Morales, A. R., Romanelli, R. and Woessner, J. F., Jr. Remodeling of human myocardial collagen in idiopathic dilated cardiomyopathy. Role of metalloproteinases and pyridinoline cross-links. Am J Pathol. 148(5): 1639-48 1996 Hagemann, T., Gunawan, B., Schulz, M., Fuzesi, L. and Binder, C. mRNA expression of matrix metalloproteases and their inhibitors differs in subtypes of renal cell carcinomas. Eur d Cancer. 37(15): 1839-46 2001 Hashimoto, G., Aoki, T., Nakamura, H., Tanzawa, K. and Okada, Y. Inhibition of ADAMTS4 (aggrecanase-1) by tissue inhibitors of metalloproteinases (TIMP-1, 2, 3 and 4). FEBS Lett. 494(3): 192-5 2001 Hayakawa, T., Yamashita, K., Tanzawa, K., Uchijima, E. and Iwata, K. Growth-promoting activity of tissue inhibitor of metalloproteinases-1 (TIMP-1) for a wide range of cells. A possible new growth factor in serum. FEBS Lett. 298(1): 29-32. 1992 Hernandez-Barrantes, S., Shimura, Y., Soloway, P. D., Sang, Q. A. and Fridman, R. Differential roles of TIMP-4 and TIMP-2 in pro-MMP-2 activation by MT1-MMP. Biochem Biophys Res Commun. 281(1): 126-30 2001 Hoekstra, R., Eskens, F. A. and Verweij, J. Matrix metalloproteinase inhibitors: current developments and future perspectives. Oncologist. 6(5): 415-27 2001 Imai, K., Hiramatsu, A., Fukushima, D., Pierschbacher, M. D. and Okada, Y. Degradation of decorin by matrix metalloproteinases: identification of the 104 cleavage sites, kinetic analyses and transforming growth factor- betal release. Biochem J. 322(Pt 3): 809-14. 1997 Inderdeo, D. S., Edwards, D. R., Han, V. K. and Khokha, R. Temporal and spatial expression of tissue inhibitors of metalloproteinases during the natural ovulatory cycle of the mouse. Biol Reprod. 55(3): 498-508 1996 Itoh, T., Tanioka, M., Yoshida, H., Yoshioka, T., Nishimoto, H. and Itohara, S. Reduced angiogenesis and tumor progression in gelatinase A-deficient mice. Cancer Res. 58(5): 1048-51 1998 Itoh, Y., Kajita, M., Kinoh, H., Mori, H., Okada, A. and Seiki, M. Membrane type 4 matrix metalloproteinase (MT4-MMP, MMP-17) is a glycosylphosphatidylinositol-anchored proteinase. J Biol Chern. 274(48): 34260-6 1999 Jiang, Y., Wang, M., Celiker, M. Y., Liu, Y. E., Sang, Q. X., Goldberg, I. D. and Shi, Y. E. Stimulation of mammary tumorigenesis by systemic tissue inhibitor of matrix metalloproteinase 4 gene delivery. Cancer Res. 61(6): 2365-70 2001 Johnson, M. D., Kim, H. R., Chesler, L, Tsao-Wu, G., Bouck, N. and Polverini, P. J. Inhibition of angiogenesis by tissue inhibitor of metalloproteinase. J Cell Physiol. 160(1): 194-202 1994 Kashiwagi, M., Tortorella, M., Nagase, H. and Brew, K. TIMP-3 is a potent inhibitor of aggrecanase 1 (ADAM-TS4) and aggrecanase 2 (ADAM-TS5). J Biol Chern. 276(16): 12501-4. 2001 Kato, T., Kure, T., Chang, J. H., Gabison, E. E., Itoh, T., Itohara, S. and Azar, D. T. Diminished corneal angiogenesis in gelatinase A-deficient mice. FEBS Lett. 508(2): 187-90 2001 Kawamata, H., Kameyama, S., Kawai, K., Tanaka, Y., Nan, L., Barch, D. H., Stetler-Stevenson, W. G. and Oyasu, R. Marked acceleration of the metastatic phenotype of a rat bladder carcinoma cell line by the expression of human gelatinase A. Int J Cancer. 63(4): 568-75 1995 Khokha, R. Suppression of the tumorigenic and metastatic abilities of murine B16-F10 melanoma cells in vivo by the overexpression of the tissue inhibitor of the metalloproteinases-1. J Natl Cancer Inst. 86(4): 299-304 1994 Kleiner, D. E. and Stetler-Stevenson, W. G. Matrix metalloproteinases and metastasis. Cancer Chemother Pharmacol. 43(Suppl): S42-51 1999 Knauper, V., Wilhelm, S. M., Seperack, P. K., DeClerck, Y. A., Langley, K. E., Osthues, A. and Tschesche, H. Direct activation of human neutrophil procollagenase by recombinant stromelysin. Biochem J. 295(Pt 2): 581-6. 1993 105 Knauper, V., Lopez-Otin, C, Smith, B., Knight, G. and Murphy, G. Biochemical characterization of human collagenase-3. J Biol Chern. 271(3): 1544-50. 1996a Knauper, V., Will, H., Lopez-Otin, C, Smith, B., Atkinson, S. J., Stanton, H., Hembry, R. M. and Murphy, G. Cellular mechanisms for human procollagenase-3 (MMP-13) activation. Evidence that MT1-MMP (MMP-14) and gelatinase a (MMP-2) are able to generate active enzyme. J Biol Chern. 271(29): 17124-31. 1996b Knauper, V., Smith, B., Lopez-Otin, C. and Murphy, G. Activation of progelatinase B (proMMP-9) by active collagenase-3 (MMP-13). Eur J Biochem. 248(2): 369-73. 1997 Knight, C. G., Willenbrock, F. and Murphy, G. A novel coumarin-labelled peptide for sensitive continuous assays of the matrix metalloproteinases. FEBS Lett. 296(3): 263-6. 1992 Kojima, S., Itoh, Y., Matsumoto, S., Masuho, Y. and Seiki, M. Membrane-type 6 matrix metalloproteinase (MT6-MMP, MMP-25) is the second glycosyl-phosphatidyl inositol (GPI)-anchored MMP. FEBS Lett. 480(2-3): 142-6 2000 Kolkenbrock, H., Orgel, D., Hecker-Kia, A., Noack, W. and Ulbrich, N. The complex between a tissue inhibitor of metalloproteinases (TIMP-2) and 72-kDa progelatinase is a metalloproteinase inhibitor. Eur J Biochem. 198(3): 775-81. 1991 Kolkenbrock, H., Orgel, D., Hecker-Kia, A., Zimmermann, J. and Ulbrich, N. Generation and activity of the ternary gelatinase B/TIMP-1/LMW- stromelysin-1 complex. Biol Chern Hoppe Seyler. 376(8): 495-500. 1995 Laemmli, U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 227(259): 680-5 1970 Langton, K. P., Barker, M. D. and McKie, N. Localization of the functional domains of human tissue inhibitor of metalloproteinases-3 and the effects of a Sorsby's fundus dystrophy mutation. J Biol Chern. 273(27): 16778-81. 1998 Langton, K. P., McKie, N., Curtis, A., Goodship, J. A., Bond, P. M., Barker, M. D. and Clarke, M. A novel tissue inhibitor of metalloproteinases-3 mutation reveals a common molecular phenotype in Sorsby's fundus dystrophy. J Biol Chern. 275(35): 27027-31. 2000 Leco, K. J., Apte, S. S., Taniguchi, G. T., Hawkes, S. P., Khokha, R., Schultz, G. A. and Edwards, D. R. Murine tissue inhibitor of metalloproteinases-4 (Timp-4): 106 cDNA isolation and expression in adult mouse tissues. FEBS Lett. 401(2-3): 213-7 1997 Lee, E., Vaughan, D. E., Parikh, S. H., Grodzinsky, A. J., Libby, P., Lark, M. W. and Lee, R. T. Regulation of matrix metalloproteinases and plasminogen activator inhibitor-1 synthesis by plasminogen in cultured human vascular smooth muscle cells. Circ Res. 78(1): 44-9. 1996 Lepre, C. A. and Moore, J. M. Microdrop screening: a rapid method to optimize solvent conditions for NMR spectroscopy of proteins. J Biomol NMR. 12(4): 493-9. 1998 Li, H., Simon, H., Bocan, T. M. and Peterson, J. T..MMP/TIMP expression in spontaneously hypertensive heart failure rats: the effect of ACE- and MMP-inhibition. Cardiovasc Res. 46(2): 298-306 2000 Li, Y. Y., Feldman, A. M., Sun, Y. and McTiernan, C. F. Differential expression of tissue inhibitors of metalloproteinases in the failing human heart. Circulation. 98(17): 1728-34 1998 Li, Y. Y., McTiernan, C. F. and Feldman, A. M. Proinflammatory cytokines regulate tissue inhibitors of metalloproteinases and disintegrin metalloproteinase in cardiac cells. Cardiovasc Res. 42(1): 162-72 1999 Liotta, L. A. and Stetler-Stevenson, W. G. Tumor invasion and metastasis: an imbalance of positive and negative regulation. Cancer Res. 51(18 Suppl): 5054s-9s 1991 Liu, Y. E., Wang, M., Greene, J., Su, J., Ullrich, S., Li, H., Sheng, S., Alexander, P., Sang, Q. A. and Shi, Y. E. Preparation and characterization of recombinant tissue inhibitor of metalloproteinase 4 (TIMP-4). J Biol Chem. 272(33): 20479-83. 1997 Maquoi, E., Munaut, C., Colige, A., Collen, D. and Lijnen, H. R. Modulation of adipose tissue expression of murine matrix metalloproteinases and their tissue inhibitors with obesity. Diabetes. 51(4): 1093-101 2002 Marvin, K. W., Keelan, J. A., Eykholt, R. L, Sato, T. A. and Mitchell, M. D. Use of cDNA arrays to generate differential expression profiles for inflammatory genes in human gestational membranes delivered at term and preterm. Mol Hum Reprod. 8(4): 399-408 2002 Matsumoto, H., Ishibashi, Y., Ohtaki, T., Hasegawa, Y., Koyama, C. and Inoue, K. Newly established murine pituitary folliculo-stellate-like cell line (TtT/GF) secretes potent pituitary glandular cell survival factors, one of which corresponds 107 to metalloproteinase inhibitor. Biochem Biophys Res Commun. 194(2): 909-15 1993 McDonnell, S., Morgan, M. and Lynch, C. Role of matrix metalloproteinases in normal and disease processes. Biochem Soc Trans. 27(4): 734-40. 1999 McQuibban, G. A., Gong, J. H., Tarn, E. M., McCulloch, C. A., Clark-Lewis, I. and Overall, C. M. Inflammation dampened by gelatinase A cleavage of monocyte chemoattractant protein-3. Science. 289(5482): 1202-6. 2000 McQuibban, G. A., Butler, G. S., Gong, J. H., Bendall, L., Power, C , Clark-Lewis, I. and Overall, C. M. Matrix metalloproteinase activity inactivates the CXC chemokine stromal cell-derived factor-1. J Biol Chern. 276(47): 43503-8. 2001 Morgunova, E., Tuuttila, A., Bergmann, U., Isupov, M., Lindqvist, Y., Schneider, G. and Tryggvason, K. Structure of human pro-matrix metalloproteinase-2: activation mechanism revealed. Science. 284(5420): 1667-70. 1999 Morrison, C. J., Butler, G. S., Bigg, H. F., Roberts, C. R., Soloway, P. D. and Overall, C. M. Cellular activation of MMP-2 (gelatinase A) by MT2-MMP occurs via a TIMP-2-independent pathway. J Biol Chern. 276(50): 47402-10. 2001 Mott, J. D., Thomas, C. L., Rosenbach, M. T., Takahara, K., Greenspan, D. S. and Banda, M. J. Post-translational proteolytic processing of procollagen C-terminal proteinase enhancer releases a metalloproteinase inhibitor. J Biol Chern. 275(2): 1384-90. 2000 Murphy, A. N., Unsworth, E. J. and Stetler-Stevenson, W. G. Tissue inhibitor of metalloproteinases-2 inhibits bFGF-induced human microvascular endothelial cell proliferation. J Cell Physiol. 157(2): 351-8 1993 Murphy, G., Cockett, M. I., Stephens, P. E., Smith, B. J. and Docherty, A. J. Stromelysin is an activator of procollagenase. A study with natural and recombinant enzymes. Biochem J. 248(1): 265-8. 1987 Murphy, G., Atkinson, S., Ward, R., Gavrilovic, J. and Reynolds, J. J. The role of plasminogen activators in the regulation of connective tissue metalloproteinases. Ann N Y Acad Sci. 667: 1-12. 1992a Murphy, G., Willenbrock, F., Ward, R. V., Cockett, M. I., Eaton, D. and Docherty, A. J. The C-terminal domain of 72 kDa gelatinase A is not required for catalysis, but is essential for membrane activation and modulates interactions with tissue inhibitors of metalloproteinases. Biochem J. 283(Pt 3): 637-41 1992b 108 Murray, G. I., Duncan, M. E., O'Neil, P., Melvin, W. T. and Fothergill, J. E. Matrix metalloproteinase-1 is associated with poor prognosis in colorectal cancer. Nat Med. 2(4): 461-2. 1996 Nagase, H., Enghild, J. J., Suzuki, K. and Salvesen, G. Stepwise activation mechanisms of the precursor of matrix metalloproteinase 3 (stromelysin) by proteinases and (4- aminophenyl)mercuric acetate. Biochemistry. 29(24): 5783-9. 1990 Nemeth, J. A. and Goolsby, C. L. TIMP-2, a growth-stimulatory protein from SV40-transformed human fibroblasts. Exp Cell Res. 207(2): 376-82 1993 Netzer, K. O., Suzuki, K., Itoh, Y., Hudson, B. G. and Khalifah, R. G. Comparative analysis of the noncollagenous NC1 domain of type IV collagen: identification of structural features important for assembly, function, and pathogenesis. Protein Sci. 7(6): 1340-51. 1998 Newby, A. C., Southgate, K. M. and Davies, M. Extracellular matrix degrading metalloproteinases in the pathogenesis of arteriosclerosis. Basic Res Cardiol. 89(Suppl 1): 59-70. 1994 Nguyen, M., Arkell, J. and Jackson, C. J. Thrombin rapidly and efficiently activates gelatinase A in human microvascular endothelial cells via a mechanism independent of active MT1 matrix metalloproteinase. Lab Invest. 79(4): 467-75. 1999 Nguyen, M., Arkell, J. and Jackson, C. J. Human endothelial gelatinases and angiogenesis. Int J Biochem Cell Biol. 33(10): 960-70. 2001 Ogata, Y., Itoh, Y. and Nagase, H. Steps involved in activation of the pro-matrix metalloproteinase 9 (progelatinase B)-tissue inhibitor of metalloproteinases-1 complex by 4- aminophenylmercuric acetate and proteinases. J Biol Chem. 270(31): 18506-11. 1995 Okada, Y. and Nakanishi, I. Activation of matrix metalloproteinase 3 (stromelysin) and matrix metalloproteinase 2 ('gelatinase') by human neutrophil elastase and cathepsin G. FEBS Lett. 249(2): 353-6. 1989 Okada, Y., Morodomi, T., Enghild, J. J., Suzuki, K., Yasui, A., Nakanishi, I., Salvesen, G. and Nagase, H. Matrix metalloproteinase 2 from human rheumatoid synovial fibroblasts. Purification and activation of the precursor and enzymic properties. Eur J Biochem. 194(3): 721-30. 1990 Olson, M. W., Gervasi, D. C., Mobashery, S. and Fridman, R. Kinetic analysis of the binding of human matrix metalloproteinase-2 and -9 to tissue inhibitor of metalloproteinase (TIMP)-1 and TIMP-2. J Biol Chem. 272(47): 29975-83 1997 109 Olson, T. M . , Hirohata, S., Ye, J., Leco, K., Seldin, M . F. and Apte, S. S. Cloning of the human tissue inhibitor of metalloproteinase-4 gene (TIMP4) and localization of the TIMP4 and Timp4 genes to human chromosome 3p25 and mouse chromosome 6, respectively. Genomics. 51(1): 148-51. 1998 Opdenakker, G., Van den Steen, P. E. and Van Damme, J. Gelatinase B: a tuner and amplifier of immune functions. Trends Immunol. 22(10): 571-9. 2001 Overall, C. M . and Sodek, J. Concanavalin A produces a matrix-degradative phenotype in human fibroblasts. Induction and endogenous activation of collagenase, 72-kDa gelatinase, and Pump-1 is accompanied by the suppression of the tissue inhibitor of matrix metalloproteinases. J Biol Chem. 265(34): 21141-51.1990 Overall, C. M . , Wrana, J. L. and Sodek, J. Transcriptional and post-transcriptional regulation of 72-kDa gelatinase/type IV collagenase by transforming growth factor-beta 1 in human fibroblasts. Comparisons with collagenase and tissue inhibitor of matrix metalloproteinase gene expression. J Biol Chem. 266(21): 14064-71. 1991 Overall, C. M . Regulation of tissue inhibitor of matrix metalloproteinase expression. Ann N Y Acad Sci. 732(51-64 1994 Overall, C. M . , King, A. E., Sam, D. K., Ong, A. D., Lau, T. T., Wallon, U. M . , DeClerck, Y. A. and Atherstone, J. Identification of the tissue inhibitor of metalloproteinases-2 (TIMP-2) binding site on the hemopexin carboxyl domain of human gelatinase A by site-directed mutagenesis. The hierarchical role in binding TIMP-2 of the unique cationic clusters of hemopexin modules III and IV. J Biol Chem. 274(7): 4421-9. 1999 Overall, C. M., Tarn, E., McQuibban, G. A., Morrison, C., Wallon, U. M., Bigg, H. F., King, A. E. and Roberts, C. R. Domain interactions in the gelatinase A.TIMP-2.MT1-MMP activation complex. The ectodomain of the 44-kDa form of membrane type-1 matrix metalloproteinase does not modulate gelatinase A activation. J Biol Chem. 275(50): 39497-506. 2000 Overall, C. M. Matrix metalloproteinase substrate binding domains, modules and exosites. Overview and experimental strategies. Methods Mol Biol. 151: 79-120 2001 Parks, W. C. Matrix metalloproteinases in repair. Wound Repair Regen. 7(6): 423-32. 1999 110 Patterson, M. L, Atkinson, S. J., Knauper, V. and Murphy, G. Specific collagenolysis by gelatinase A, MMP-2, is determined by the hemopexin domain and not the fibronectin-like domain. FEBS Lett. 503(2-3): 158-62 2001 Paulson, D. R., Addison, A. W., Dolphin, D. and James, B. R. Preparation of ruthenium(ll) and ruthenium(lll) myoglobin and the reaction of dioxygen, and carbon monoxide, with ruthenium(ll) myoglobin. J Biol Chern. 254(15): 7002-6 1979 Pei, D. and Weiss, S. J. Furin-dependent intracellular activation of the human stromelysin-3 zymogen. Nature. 375(6528): 244-7. 1995 Pei, D. CA-MMP: a matrix metalloproteinase with a novel cysteine array, but without the classic cysteine switch. FEBS Lett. 457(2): 262-70. 1999 Pei, D., Kang, T. and Qi, H. Cysteine array matrix metalloproteinase (CA-MMP)/MMP-23 is a type II transmembrane matrix metalloproteinase regulated by a single cleavage for both secretion and activation. J Biol Chern. 275(43): 33988-97. 2000 Peterson, J. T., Li, H., Dillon, L. and Bryant, J. W. Evolution of matrix metalloprotease and tissue inhibitor expression during heart failure progression in the infarcted rat. Cardiovasc Res. 46(2): 307-15 2000 Petitclerc, E., Boutaud, A., Prestayko, A., Xu, J., Sado, Y., Ninomiya, Y., Sarras, M. P., Jr., Hudson, B. G. and Brooks, P. C. New functions for non-collagenous domains of human collagen type IV. Novel integrin ligands inhibiting angiogenesis and tumor growth in vivo. J Biol Chern. 275(11): 8051-61 2000 Reno, C, Boykiw, R., Martinez, M. L. and Hart, D. A. Temporal alterations in mRNA levels for proteinases and inhibitors and their potential regulators in the healing medial collateral ligament. Biochem Biophys Res Commun. 252(3): 757-63 1998 Riley, S. C , Leask, R., Chard, T., Wathen, N. C, Calder, A. A. and Howe, D. C. Secretion of matrix metalloproteinase-2, matrix metalloproteinase-9 and tissue inhibitor of metalloproteinases into the intrauterine compartments during early pregnancy. Moi Hum Reprod. 5(4): 376-81 1999a Riley, S. C, Leask, R., Denison, F. C, Wisely, K., Calder, A. A. and Howe, D. C. Secretion of tissue inhibitors of matrix metalloproteinases by human fetal membranes, decidua and placenta at parturition. J Endocrinol. 162(3): 351-9 1999b Riley, S. C , Gibson, A. H., Leask, R., Mauchline, D. J., Pedersen, H. G. and Watson, E. D. Secretion of matrix metalloproteinases 2 and 9 and tissue inhibitor 111 of metalloproteinases into follicular fluid during follicle development in equine ovaries. Reproduction. 121(4): 553-60 2001 Robinson, L. L, Sznajder, N. A., Riley, S. C. and Anderson, R. A. Matrix metalloproteinases and tissue inhibitors of metalloproteinases in human fetal testis and ovary. Mol Hum Reprod. 7(7): 641-8 2001 Sasaki, K., Takagi, M., Mandelin, J., Takei, I., Santavirta, S., Ida, H., Ogino, T. and Konttinen, Y. T. Quantitative analysis of mRNA expression of TIMPs in the periprosthetic interface tissue of loose hips by real-time PCR system. J Biomed Mater Res. 58(6): 605-12 2001 Sato, H., Takino, T., Okada, Y., Cao, J., Shinagawa, A., Yamamoto, E. and Seiki, M. A matrix metalloproteinase expressed on the surface of invasive tumour cells. Nature. 370(6484): 61-5. 1994 Sato, H., Kinoshita, T., Takino, T., Nakayama, K. and Seiki, M. Activation of a recombinant membrane type 1-matrix metalloproteinase (MT1-MMP) by furin and its interaction with tissue inhibitor of metalloproteinases (TIMP)-2. FEBS Lett. 393(1): 101-4. 1996 Schonbeck, U., Mach, F. and Libby, P. Generation of biologically active IL-1 beta by matrix metalloproteinases: a novel caspase-1-independent pathway of IL-1 beta processing. J Immunol. 161 (7): 3340-6. 1998 Seeland, U., Kouchi, I., Zolk, O., Itter, G., Linz, W. and Bohm, M. Effect of ramipril and furosemide treatment on interstitial remodeling in post-infarction heart failure rat hearts. J Mol Cell Cardiol. 34(2): 151-63 2002 Selman, M., Ruiz, V., Cabrera, S., Segura, L., Ramirez, R., Barrios, R. and Pardo, A. TIMP-1, -2, -3, and -4 in idiopathic pulmonary fibrosis. A prevailing nondegradative lung microenvironment? Am J Physiol Lung Cell Mol Physiol. 279(3): L562-74 2000 Smith, M. F., Mclntush, E. W., Ricke, W. A., Kojima, F. N. and Smith, G. W. Regulation of ovarian extracellular matrix remodelling by metalloproteinases and their tissue inhibitors: effects on follicular development, ovulation and luteal function. J Reprod Fertil Suppl. 54: 367-81 1999 Smith, M. R., Kung, H., Durum, S. K., Colburn, N. H. and Sun, Y. TIMP-3 induces cell death by stabilizing TNF-alpha receptors on the surface of human colon carcinoma cells. Cytokine. 9(10): 770-80 1997 Sottrup-Jensen, L. and Birkedal-Hansen, H. Human fibroblast collagenase-alpha-macroglobulin interactions. Localization of cleavage sites in the bait regions of five mammalian alpha-macroglobulins. J Biol Chem. 264(1): 393-401. 1989 112 Sternlicht, M. D. and Werb, Z. How matrix metalloproteinases regulate cell behavior. Annu Rev Cell Dev Biol. 17: 463-516 2001 Stetler-Stevenson, W. G., Krutzsch, H. C. and Liotta, L. A. Tissue inhibitor of metalloproteinase (TIMP-2). A new member of the metalloproteinase inhibitor family. J Biol Chern. 264(29): 17374-8. 1989 Stetler-Stevenson, W. G., Aznavoorian, S. and Liotta, L. A. Tumor cell interactions with the extracellular matrix during invasion and metastasis. Annu Rev Cell Biol. 9: 541 -73 1993 Stetler-Stevenson, W. G. and Yu, A. E. Proteases in invasion: matrix metalloproteinases. Semin Cancer Biol. 11(2): 143-52 2001 Stratmann, B., Farr, M. and Tschesche, H. MMP-TIMP interaction depends on residue. 2 in TIMP-4. FEBS Lett. 507(3): 285-7. 2001a Stratmann, B., Farr, M. and Tschesche, H. Characterization of C-terminally truncated human tissue inhibitor of metalloproteinases-4 expressed in Pichia pastoris. Biol Chern. 382(6): 987-91 2001b Streuli, C. Extracellular matrix remodelling and cellular differentiation. CurrOpin Cell Biol. 11(5): 634-40. 1999 Stricklin, G. P., Jeffrey, J. J., Roswit, W. T. and Eisen, A. Z. Human skin fibroblast procollagenase: mechanisms of activation by organomercurials and trypsin. Biochemistry. 22(1): 61-8. 1983 Stricklin, G. P. and Welgus, H. G. Human skin fibroblast collagenase inhibitor. Purification and biochemical characterization. J Biol Chern. 258(20): 12252-8. 1983 Strongin, A. Y., Collier, I., Bannikov, G., Marmer, B. L., Grant, G. A. and Goldberg, G. I. Mechanism of cell surface activation of 72-kDa type IV collagenase. Isolation of the activated form of the membrane metalloprotease. J Biol Chern. 270(10): 5331-8. 1995 Sun, Y., Kim, H., Parker, M., Stetler-Stevenson, W. G. and Colburn, N. H. Lack of suppression of tumor cell phenotype by overexpression of TIMP-3 in mouse JB6 tumor cells identification of a transfectant with increased tumorigenicity and invasiveness. Anticancer Res. 16(1): 1-7 1996 Takino, T., Sato, H., Shinagawa, A. and Seiki, M. Identification of the second membrane-type matrix metalloproteinase (MT- MMP-2) gene from a human 113 placenta cDNA library. MT-MMPs form a unique membrane-type subclass in the MMP family. J Biol Chem. 270(39): 23013-20. 1995 Tang, H. L, Chance, B., Mauk, A. G., Powers, L. S., Reddy, K. S. and Smith, M. Spectroscopic, electrochemical, and ligand binding properties of the horse heart metmyoglobin His64-Tyr variant. Biochim Biophys Acta. 1206(1): 90-6 1994 Thomas, P., Khokha, R., Shepherd, F. A., Feld, R. and Tsao, M. S. Differential expression of matrix metalloproteinases and their inhibitors in non-small cell lung cancer. J Pathol. 190(2): 150-6 2000 Toth, M., Bernardo, M. M., Gervasi, D. C., Soloway, P. D., Wang, Z., Bigg, H. F., Overall, C. M., DeClerck, Y. A., Tschesche, H., Cher, M. L, Brown, S., Mobashery, S. and Fridman, R. Tissue inhibitor of metalloproteinase (TIMP)-2 acts synergistically with synthetic matrix metalloproteinase (MMP) inhibitors but not with TIMP-4 to enhance the (Membrane type 1)-MMP-dependent activation of pro- MMP-2. J Biol Chem. 275(52): 41415-23. 2000 Tsafriri, A. and Reich, R. Molecular aspects of mammalian ovulation. Exp Clin Endocrinol Diabetes. 107(1): 1-11 1999 Tummalapalli, C. M., Heath, B. J. and Tyagi, S. C. Tissue inhibitor of metalloproteinase-4 instigates apoptosis in transformed cardiac fibroblasts. J Cell Biochem. 80(4): 512-21 2001 Tuuttila, A., Morgunova, E., Bergmann, U., Lindqvist, Y., Maskos, K., Fernandez-Catalan, C., Bode, W., Tryggvason, K. and Schneider, G. Three-dimensional structure of human tissue inhibitor of metalloproteinases-2 at 2.1 A resolution. J Mol Biol. 284(4): 1133-40 1998 Vaalamo, M., Leivo, T. and Saarialho-Kere, U. Differential expression of tissue inhibitors of metalloproteinases (TIMP-1, -2, -3, and -4) in normal and aberrant wound healing. Hum Pathol. 30(7): 795-802 1999 Vincenti, M. P. The matrix metalloproteinase (MMP) and tissue inhibitor of metalloproteinase (TIMP) genes. Transcriptional and posttranscriptional regulation, signal transduction and cell-type-specific expression. Methods Mol Biol. 151: 121-48 2001 Wallon, U. M. and Overall, C. M. The hemopexin-like domain (C domain) of human gelatinase A (matrix metalloproteinase-2) requires Ca2+ for fibronectin and heparin binding. Binding properties of recombinant gelatinase A C domain to extracellular matrix and basement membrane components. J Biol Chem. 272(11): 7473-81 1997 114 Wang, M., Liu, Y. E., Greene, J., Sheng, S., Fuchs, A., Rosen, E. M. and Shi, Y. E. Inhibition of tumor growth and metastasis of human breast cancer cells transfected with tissue inhibitor of metalloproteinase 4. Oncogene. 14(23): 2767-74. 1997 Ward, R. V., Atkinson, S. J., Reynolds, J. J. and Murphy, G. Cell surface-mediated activation of progelatinase A: demonstration of the involvement of the C-terminal domain of progelatinase A in cell surface binding and activation of progelatinase A by primary fibroblasts. Biochem J. 304(Pt 1): 263-9 1994 Waterhouse, P., Denhardt, D. T. and Khokha, R. Temporal expression of tissue inhibitors of metalloproteinases in mouse reproductive tissues during gestation. Moi Reprod Dev. 35(3): 219-26 1993 Weber, B. H., Vogt, G., Pruett, R. C, Stohr, H. and Felbor, U. Mutations in the tissue inhibitor of metalloproteinases-3 (TIMP3) in patients with Sorsby's fundus dystrophy. Nat Genet. 8(4): 352-6. 1994 Wilde, C. G., Hawkins, P. R., Coleman, R. T., Levine, W. B., Delegeane, A. M., Okamoto, P. M., Ito, L. Y., Scott, R. W. and Seilhamer, J. J. Cloning and characterization of human tissue inhibitor of metalloproteinases-3. DNA Cell Biol. 13(7): 711-8 1994 Wilhelm, S. M., Collier, I. E., Marmer, B. L, Eisen, A. Z., Grant, G. A. and Goldberg, G. I. SV40-transformed human lung fibroblasts secrete a 92-kDa type IV collagenase which is identical to that secreted by normal human macrophages. J Biol Chern. 264(29): 17213-21. 1989 Will, H., Atkinson, S. J., Butler, G. S., Smith, B. and Murphy, G. The soluble catalytic domain of membrane type 1 matrix metalloproteinase cleaves the propeptide of progelatinase A and initiates autoproteolytic activation. Regulation by TIMP-2 and TIMP-3. J Biol Chern. 271(29): 17119-23. 1996 Willenbrock, F. and Murphy, G. Structure-function relationships in the tissue inhibitors of metalloproteinases. Am J Respir Crit Care Med. 150(6 Pt 2): S165-70. 1994 Williamson, R. A., Marston, F. A., Angal, S., Koklitis, P., Panico, M., Morris, H. R., Carne, A. F., Smith, B. J., Harris, T. J. and Freedman, R. B. Disulphide bond assignment in human tissue inhibitor of metalloproteinases (TIMP). Biochem J. 268(2): 267-74 1990 Williamson, R. A., Martorell, G., Carr, M. D., Murphy, G., Docherty, A. J., Freedman, R. B. and Feeney, J. Solution structure of the active domain of tissue inhibitor of metalloproteinases-2. A new member of the OB fold protein family. Biochemistry. 33(39): 11745-59 1994 115 Woessner, J. F., Jr. TIMPs as multifunctional molecules. Inhibitors of Metalloproteinases: Tissue Inhibitors of Metalloproteinases in Development and Disease., Banff, Alberta, Canada, Harwood Academic Publishers 73-7 (1996). Wu, I. and Moses, M. A. Molecular cloning and expression analysis of the cDNA encoding rat tissue inhibitor of metalloproteinase-4. Matrix Biol. 16(6): 339-42 1998 Yamamoto, H., Itoh, F., Iku, S., Adachi, Y., Fukushima, H., Sasaki, S., Mukaiya, M., Hirata, K. and Imai, K. Expression of matrix metalloproteinases and tissue inhibitors of metalloproteinases in human pancreatic adenocarcinomas: clinicopathologic and prognostic significance of matrilysin expression. J Clin Oncol. 19(4): 1118-27. 2001 Yamanishi, Y. and Firestein, G. S. Pathogenesis of rheumatoid arthritis: the role of synoviocytes. Rheum Dis Clin North Am. 27(2): 355-71. 2001 Yang, T. T. and Hawkes, S. P. Role of the 21-kDa protein TIMP-3 in oncogenic transformation of cultured chicken embryo fibroblasts. Proc Natl Acad Sci USA. 89(22): 10676-80. 1992 Young, D. A., Phillips, B. W., Hogan, A., Lundy, C. J., Nuttall, R. M., Schultz, G. A., Leco, K. J., Clark, I. M. and Edwards, D. R. Identification of an Initiator-like element essential for the expression of the Tissue Inhibitor of Metalloproteinases-4 (Timp-4) gene. Biochem. J. (In Press): 2002 Yu, Q. and Stamenkovic, I. Cell surface-localized matrix metalloproteinase-9 proteolytically activates TGF-beta and promotes tumor invasion and angiogenesis. Genes Dev. 14(2): 163-76. 2000 Yu, W. H., Yu, S., Meng, Q., Brew, K. and Woessner, J. F., Jr. TIMP-3 binds to sulfated glycosaminoglycans of the extracellular matrix. J Biol Chem. 275(40): 31226-32 2000 Zhou, Z., Apte, S. S., Soininen, R., Cao, R., Baaklini, G. Y., Rauser, R. W., Wang, J., Cao, Y. and Tryggvason, K. Impaired endochondral ossification and angiogenesis in mice deficient in membrane-type matrix metalloproteinase I. Proc Natl Acad Sci USA. 97(8): 4052-7 2000 116 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090224/manifest

Comment

Related Items